Mike Chaney's Tech Corner
December 03, 2020, 11:24:32 PM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: Feb 2013: Qimage Ultimate Challenges... have fun and explore features!
 
  Home Help Search Login Register  

Professional ICC Profiling Software for Windows
Create custom ICC profiles with
Profile Prism for accurate color!
  Show Posts
Pages: 1 ... 209 210 [211] 212 213
3151  Technical Discussions / Articles / March 2007: High Def Talk Part II: Content on: May 27, 2009, 02:02:34 PM

High Def Talk Part II: Content


Background

Last month we talked about high definition TV monitors and how to get started by choosing the right HDTV equipment.  This month in part 2 of 2, let's take a quick look at high definition content to see what is available as far as getting high definition signals to your new HD home theater.

 

Through the airwaves

Perhaps the most accessible way to get HD content is to subscribe to a satellite provider that offers HD content.  Unless you live in an apartment (and sometimes even if you do), you're likely to be able to get satellite TV because satellite TV is available to almost anyone provided you are not surrounded by tall trees that block the satellite signal based on where on your house (or in your yard) you can install the receiving dish.  Since satellite TV depends on "line of sight" transmission, it doesn't depend on other services (like cables) being installed to your house.  While federal law grants you the right to install up to an 18 inch dish even if you are in a subdivision with covenants/bylaws, there may be some local restrictions, so check with the provider you are considering to see if the satellite TV service can be installed in your location.  Of course, if you already have one of these services, just call or visit their web site to inquire about getting upgraded equipment and subscription plans for HD content.  Below are the two major satellite TV providers:

  • Dish Network: Dish Network offers one of the most comprehensive high definition packages on the market as of this writing as they (relatively) recently took over VOOM satellites which are dedicated to HD content.  If you are looking for the most HD channels, Dish may be your best bet at this time regardless of what other services may be available to you.  To see what Dish Network has to offer in the way of HD content, click this link.  You can contact Dish Network from their web site to find out if you can receive Dish Network programming from your location.  A visit to your home may be required to determine if the satellite dish(es) can be located/aimed from your home/property.  You may also wish to inquire as to whether or not Dish Network carries your local TV channels in high definition and if they don't, whether it is possible to receive these channels OTA (over the air) via an antenna at your location in addition to the satellite dish(es).  While visiting the web site, you can check plans/pricing as well.

  • DirecTV: DirecTV has fewer HD channels than Dish, but they do plan to add more and even launch more satellites to carry more content at some point.  Details on timing are sketchy/speculative at best.  To see what DirecTV has to offer in the way of HD content, click this link.  You can contact DirecTV from their web site to find out if you can receive their service from your location.  A visit to your home may be required to determine if the satellite dish(es) can be located/aimed from your home/property.  You may also wish to inquire as to whether or not DirecTV carries your local TV channels in high definition and if they don't, whether it is possible to receive these channels OTA (over the air) via an antenna at your location in addition to the satellite dish(es).  While visiting the web site, you can check plans/pricing as well.

In addition to satellite TV providers, most areas that have access to broadcast TV via the old "rabbit ears" also now have digital broadcast TV.  OTA (over the air) channels will be limited to what you can receive as far as the major networks from local stations (ABC, CBS, NBC, Fox, PBS, etc.).  See last month's article for information on receiving HD over the air and to find out whether or not there are digital channels in your area.  While over the air choices/channels are limited, once you buy the HD tuner box (if your TV doesn't already have one), you can receive content for free.  Believe it or not, over the air HD signals are some of the highest quality you'll experience because they are typically less "processed" than HD content that goes up to a satellite or to a cable/fiber provider and is re-encoded for viewing.  Don't think noise and snow when you think about over the air high definition broadcasts as they are just as "digital" and often even cleaner than other providers.

 

Wired/fiber HD content

If you live in an apartment or other location where satellite dishes are not an option, the next logical option is a cable/fiber service that delivers HD content via a physical cable to your residence.  Most cable companies offer high definition packages, but be aware that many cable providers only offer a handful of channels so check their web site or call them to be sure exactly what channels are offered in HD.  Also be aware that with cable, some of your (non HD) channels may be "analog", meaning they are not digital and can be susceptible to noise/snow just like an old TV with rabbit ears.  "Digital Cable" often means that only a portion of the channels are digital and noise-free while many are still the old analog format.  If you want to know which channels are digital versus analog, just check with the cable company or their web site.

One of the most promising "cable" services is Verizon FIOS.  FIOS is Verizon's fiber optic answer to local cable providers.  They offer a very good selection of HD channels and also offer (extremely) high speed internet that beats just about anything else available for residential customers.  Unfortunately, if you are reading this article any time close to when it was written (March 2007), there is more than a 90% chance that you cannot get Verizon FIOS where you live as its availability is very sparse right now due to the necessity to install fiber optic cables to every location where it is being offered.  You can enter your phone number on the web site above to find out whether or not you can get FIOS.  Note that FIOS TV and internet are actually separate services but if you can get FIOS, it is likely you'll be able to get both: high speed internet and TV.

 

Hardware for satellite/cable HD content

These days any company who offers HD content via satellite or cable also offers equipment that allows you to receive the HD content and send it to your TV (monitor).  The most common setup is an HD DVR (digital video recorder).  For those familiar with Tivo, you'll know what these are.  They are simply a decoder box with a hard drive inside that allows you to record HD content for later playback.  Many offer nice features like passes where you can record all episodes of a certain show automatically, even if the times change, etc.  An HD DVR is the best way to enjoy broadcast HD content as you can pick the shows/movies you want and you get to pick the time you watch them.  Be aware that most services offer to rent the DVR for a small fee assessed on your monthly bill so many times you don't have to pay a lot for the equipment.

 

Other hardware: HD DVD and Blu-Ray

HD DVD and Blu-Ray are the next generation digital media format.  Both the size of a standard DVD, they hold as much as 10x or more data and therefore support HD content.  Unfortunately, the "format war" is still ongoing and there is no clear winner amongst these two competing HD disc formats.  That's part of the reason why not all newly released movies are available in HD DVD or Blu-Ray even though most of the mail rental services do allow you to rent the ones that are available at no extra charge over what you pay for renting standard DVD's.  At the moment, HD DVD players are about $500 and Blu-Ray disc players are about double at $1000.  If you buy a player that only supports one of the two formats, you'll only be able to watch movies released in that format and some movies are only released in one of the two formats.  This is especially true for HD-DVD since Sony owns Blu-Ray and will therefore not allow movies from Sony Pictures to be released on HD DVD.

There is finally one "hybrid" player at about $1100 that can play both HD-DVD and Blu-Ray discs but from what I understand, this first hybrid player is really a Blu-Ray player with HD-DVD support added almost as an afterthought.  Apparently it doesn't support the full HD-DVD spec as far as being able to use all the menus/features of HD-DVD.  Unless you have deep pockets and just want to play with a new toy, the word here is still... wait and see.  Good hybrid players with less problems and faster boot up times should be available within the next year so it still probably isn't the best time to get into HD-DVD or Blu-Ray.

 

Do all HD channels carry HD content?

Well, yes and no.  All HD channels carry true HD content from time to time, but that doesn't mean that the channel continuously broadcasts nothing but HD content!  Take the major networks for example.  Nearly all prime time (that would be 8:00pm or later here on the east coast) shows on major networks are now broadcast in high definition including dramas like CSI Miami, Jericho, etc. and even half hour sitcoms.  The exceptions are reality shows like Survivor or The Apprentice or news/documentary shows like Dateline or 20/20 where footage is shot with non HD cameras as the shooting environment is less controlled out in the field.  It is worth mentioning then, that just because you are watching an HD channel does not necessarily mean that you'll be watching HD content 24/7.  In fact, while many football games are broadcast in HD during football season, you'll occasionally find some broadcast in SD (standard definition).  Of course, major sporting events like the Super Bowl and the Daytona 500 are broadcast in HD and the best part is, if you have an OTA (over the air) tuner, you can receive these broadcasts in all their glory for free with just a table top or rooftop antenna provided you are close enough to a TV station/tower.

 

Summary

In this short article, we've taken a look at the major players involved in getting the high definition content you need delivered to your home.  A big part of enjoying your new high definition home theater is being able to get actual high definition content on your system.  More and more HD content is becoming available and it is certainly no longer true that having a high definition TV doesn't make sense because there isn't any good HD material.  The material is there for the taking and it's getting better all the time!

 

Mike Chaney

3152  Technical Discussions / Articles / February 2007: High Def Talk Part I: HD Displays on: May 27, 2009, 01:59:58 PM

High Def Talk Part I: HD Displays


Background

Let's take a break from cameras, computers, and printing for a bit and take a look at high definition television and home theater.  After all, those of us who are interested in the best photographs, camera equipment, computer equipment, and printers are often interested in getting the best picture when it comes to home entertainment as well.  I find it interesting when I meet photographers who have some of the most expensive photographic equipment, claim to be home theater or AV (audio-video) fans, yet are still watching a 54 inch big screen TV from the late 80's or early 90's.  I've even heard the phrase: "My DVD's look great.  How much better can it be?"  I'll try to answer that question in this article that is geared toward those of you who have not yet made the leap to HDTV and might be wondering if it is time.  If you are an "HD nut" who frequents avsforum, you'll likely get very little out of this article as you are already ahead of the curve.

 

SD versus HD

What is "high definition" and how does it compare to "standard definition"?  Broadcast TV, based on a display format that was conceived in the 1930's in black-and-white and later modified to carry color video in the 1950's, offers a resolution of about 330 x 245 pixels.  SD or "standard definition" is a relatively new term that refers to a digital video format of approximately 640 x 480 pixels interlaced (480i).  While this is better than the old broadcast standard (and is the reason why standard DVD's look better than TV), 640 x 480 is not nearly enough resolving power for the larger (36 inch+) sets used in home theaters.  Just walk up to anyone who is digital photography savvy and tell them that you are thinking about printing a 36 x 24 inch print from a 640 x 480 shot taken with a cell phone.  When they're finished laughing and finally get up off the floor, realize that you're doing almost the same thing by watching that 640 x 480 DVD on your old 54 inch TV!

If you live in an area where you are able to get local TV stations on a television with a "rabbit ear" antenna, chances are you already have high definition signals coming through the airwaves to your home.  You simply cannot watch them because you don't have a high definition set.  High definition comes in two basic formats: 720p and 1080i.  720p is 1280 x 720 resolution and pixels are displayed progressively so that all 920,000 pixels are displayed in each frame.  1080i is 1920 x 1080 resolution but it is interlaced, meaning that only the odd or even lines are displayed with each frame.  A newer format, 1080p, is now being supported in many displays but as of this writing, there is very little content that is actually available in the 1080p format so most of the 1080p sets just take a 1080i signal and deinterlace it.  Still, 1920 x 1080 is 2 million pixels of resolution.  Compare that to the 640 x 480 (300,000 pixels) available on a standard DVD.  The best high definition material is over 6 times the resolution of a standard DVD.

 

In the eye of the beholder

It's difficult to describe how much better HD is when compared to SD.  You simply have to see it for yourself.  Many compare HD to the feeling of looking through a window at the actual scene.  The clarity, texture, and dynamic range are amazing.  So much so that once you get used to HD, it is difficult to watch SD!  I'm so used to HD now that whenever a football game is broadcast in SD, I literally cringe.  SD looks so out of focus and so devoid of detail that it almost gives me a headache when I watch something with a lot of action like a sporting event in SD.  Of course, these don't look nearly as good as a DVD due to compression artifacts and the re-processing involved with broadcast SD signals on satellite or digital cable.  Upsampled DVD on an HD monitor can actually look nearly as good as high definition even though, technically it is not high definition: most likely 480p (640 x 480  or 720 x 480 progressive).  When asked to describe HD versus SD, you'll likely get a different answer from everyone you ask but there's nothing like the first time you see true HD material on a true HD set.  It's something you'll never forget!

 

"Fake" HD

As we know, "high definition" refers to a video format that consists of either 1280 x 720 pixels or 1920 x 1080 pixels.  If you walk into a store and they are piping video from a standard DVD player through their sets, you'll know immediately that you are not looking at anything high definition because DVD's are 480p (640 x 480 resolution) and are therefore not considered high definition.  Technically, 480p is considered EDTV (enhanced definition TV).  EDTV is actually a major advance over SDTV, but it still falls short of high definition.  Be aware of sets that are marked EDTV.  If it is marked EDTV, it is not a high definition set.  Some smaller plasma TV's are EDTV as are some sets marked "high definition compatible".  High definition compatible is often used to indicate that the set can decode an HDTV signal, not that the set itself is HD!  Suffice it to say that if you are in the market for an HDTV, make sure the store where you are evaluating the sets is piping a true HD signal to the sets and that the set itself is truly HD.

 

Display types

Different types of displays (LCD's, plasmas, DLP's) have their advantages and disadvantages.  Any true HD set you buy today will certainly look many times better than a standard TV, but what type of set is right for you?  Let's take a quick look at a few of the most popular display types and look at their advantages and disadvantages.

Plasma: Plasma televisions are actually not that dissimilar from older tube sets.  They use phosphor just like the old tube sets.  The difference is in how the phosphor is excited (lit).  In a CRT (tube television), a beam of electrons scans the phosphor, lighting "pixels" in sequence.  In a plasma TV, the phosphor is lit with individual electrodes under each pixel.  The obvious advantage is size.  Since plasma TV's don't need any projection, they are made as flat panels that are very thin and can be hung on walls or used in cramped spaces.  Plasma TV's are some of the most vibrant sets with excellent dynamic range (rich blacks and bright whites) and are currently the best technology on the market for viewing at an angle, as plasma televisions don't fade when viewed off-center.  Almost all consumer plasma sets as of this writing use the 720p HD format.  That is, they have a resolution of about 1280 x 720 pixels for plasma sets 46 inches and larger.  Due to limitations in the plasma technology, 1080p (1920 x 1080 resolution) plasma sets are just now being introduced to the marker in sizes under 60 inches, so if you are looking for a plasma set that is 60 inches or smaller, you'll be getting a 720p set unless you want to pay $10,000 or more.  Burn-in, where static images can be "burnt" into the screen permanently, is not much of an issue on the latest plasmas but you should avoid static images as much as possible during the first ~100 hours of operation.  In addition, if you plan to do a lot of video gaming, an LCD may be a better choice because static images like radars, scores, and health meters can eventually be burned in if left on a plasma screen for extended periods of time.

LCD: For years, LCD (liquid crystal display) televisions have struggled to compare to plasma TV's.  The most recent LCD models have all but caught plasma TV's on every front except off-angle viewing.  Today's LCD's offer less fade when viewing from an angle, are brighter, have better blacks (better dynamic range) and have better response times, meaning that they don't "blur" fast moving objects like older LCD sets.  In addition, it's easier to manufacture smaller LCD screens (under 60 inches) with a higher pixel count compared to plasma TV's, so there are a fair number of 1080p (1920 x 1080 resolution) LCD sets available at the 55 inch and smaller sizes, making them boast higher resolution than plasma sets.  There is still some color and contrast fade when moving to off-angle viewing, but the very latest models show a lot of promise here, with very little noticeable fade when viewing from an angle.  Things tend to change quickly as far as display technology, but as of this writing, some of the latest 1080p 50 to 55 inch LCD sets top the list as the highest quality HD sets available with the least number of drawbacks (such as uneven lighting, ghosting, "hot spots", seen with many projection type TV's)!  Like plasma displays, LCD's are flat panels that can be hung on a wall and they are often even lighter/thinner than plasmas.  Screen burn-in is not an issue with LCD displays, making them the display of choice for gamers.

DLP: DLP (digital light processor) televisions have been around for about a decade.  They are a form of rear projection television that uses tiny mirrors to throw light on the front screen.  These sets are not flat panels and are therefore a bit bulkier than plasma or LCD TV's, but they can be very cost effective.  They tend to be cheaper than plasma or LCD sets of comparable size and they do offer an excellent picture.  I'm not a big fan of projection televisions as they tend to produce a less evenly lit picture and my eyes are quite sensitive to blooming, color inconsistencies, fade, and the "rainbow effect" that you can sometimes see on DLP sets.  Any projection set will tend to be a bit less sharp than a flat panel set due to the fact that light is being "thrown" at a distance rather than being created at specific points on a static (non moving) panel.  As with LCD panels, screen burn-in is generally not an issue with DLP displays.

SXRD: SXRD, short for Silicon X-tal Reflective Display, is a Sony acronym for a technology known as LCOS (liquid crystal on silicon).  It is similar to DLP in that it uses a reflective surface but instead of using mechanical mirrors, liquid crystals are used to reflect the light from the projector onto the front screen.  Again, this is a projection TV so it is not a flat panel and will take up more space than a plasma or LCD TV.  Sony SXRD sets typically have a better picture than DLP sets and some people believe they have a "film" or "movie theater" look unrivaled by any other display type.  As I mentioned under DLP, I'm not a fan of projection TV's just because I'm sensitive to the contrast and color fade that occurs when viewing projection TV's at an angle.  I also miss the silky smooth uniformity of plasma and LCD sets when I have to move (walk) in front of a projection set as I can always detect the bright spot from the projection lamp(s) moving across the screen with me.  Others may prefer SXRD technology over plasma and LCD due to the film-like look appearing less like "pixels".  Here again, beauty is in the eye of the beholder.  Since SXRD displays are basically LCD on silicon, they generally do not suffer from burn-in.

Bottom line: The bottom line on choosing a display is to first determine whether or not the display is truly high definition.  To be high definition, the display must have 1280 x 720 pixels or more and should have an "HD" logo.  Stick to models marked HDTV and shy away from models marked EDTV.  When you have limited your search to HD displays, let your eyes be the judge.  Different people look for different things in a display.  Some pay more attention to contrast and saturation while others are more critical of resolution, pixels, and sharpness.  The real bottom line is that you should pick the set that looks best to you!  Keep in mind that different display types work better in different environments with plasma and LCD displays generally being better in rooms with bright lighting (sunlight entering a window for example).  Once you've picked your favorite set in your price range, it wouldn't hurt to leave the store and do a little research.  Try Googling the model number or even the model number and the word "problem" to see if other users are experiencing any common issues with that set.  Sometimes you'll find complaints of color blooming, ghosting, banding, or other issues and that may give you some things to double check before you buy.  Any common problems are usually described or displayed with enough detail that you'll be able to look for the problem in the set you picked to see if it is an issue for you.

 

HD Content

Before buying an HD display for your home theater, it would be wise to be aware of the HD content that is actually available to you.  Broadcast HD is accessible in most locations in the form of digital cable, satellite, or fiber services.  If you have cable TV, chances are your cable provider offers "digital cable" that includes at least a few HD channels.  Note that the fact that a channel is "digital" does not necessarily mean that it is HD (it could be SD just broadcast in digital format), so be sure to ask your cable provider how many/what channels are offered in true HD and/or check out their web site.  If you are interested in cable HD, some displays offer a cable card feature, so you might want to check compatibility of the TV with your cable service, although that is not a necessity since your cable provider can provide you with an (external) cable box. 

Verizon FIOS is a promising fiber TV and internet service that has a lot of promise, but it isn't likely to be available in your area as coverage right now is extremely limited.  To find out if FIOS is available in your area, check here.

If you live in the boonies and don't have cable or FIOS service and/or you don't like the selections offered by those services, there's always satellite TV.  Right now, Dish Network has the greatest selection and number of true HD channels available in any service offering high definition content since Dish has taken over the Voom HD satellite service.  DirecTV also offers HD channels and while they plan to offer many new HD channels in the next year, their selection of true HD channels is more limited as of this writing.  If you already have one of these services but you don't get high definition channels right now, you may need new equipment and you may be required to pay an install fee and a small monthly fee to access the HD channels.  Check the web sites of the service in question for more info.  All services that offer HD content also offer DVR's that can record HD content as well.  When choosing a satellite provider, find out whether or not they offer your local (ABC, CBS, NBC, Fox) channels in high definition via satellite.  If not, you may be required to get these OTA (over the air) with an antenna connected to the satellite receiver.

If you live within say 30 miles of a city that has local TV channels, chances are you can get some HD content for free (or at least with no monthly charges).  If your display didn't come with a built in tuner that can handle "over the air" TV broadcasts, you'll need to buy a tuner that is capable of receiving high definition broadcasts.  What you're looking for is a tuner that is labeled as an ATSC tuner.  In most cases, a good indoor antenna is sufficient to receive these channels but if you need an antenna, check to see whether you need a VHF or UHF antenna as HD can be used on both frequency ranges.  To see how far you are from various TV stations and what type of antenna you need, try visiting antennaweb.  You can simply enter your ZIP code and click "Submit" to see the TV stations near you and the antenna type/size needed.  Note that channels with a sub-channel (2.1, 11.1, etc.) are "DT" or digital television.  Those are the channels that are broadcast in digital format and are likely to offer high definition content although the DT designation does not guarantee that the channel broadcasts HD.

After broadcast HD comes HD content on other media such as HD-DVD and Blu-Ray.  Movies are starting to be released in the HD-DVD and Blu-Ray formats now and most online rental services offer the formats at no additional charge over standard DVD's, but selections are limited and the two formats are still competing with no clear winner in the format war.  In addition, current HD-DVD and Blu-Ray players can be expensive (about $500 for HD-DVD players and $1000 for Blu-Ray players) and are relatively slow to start up.  There's also talk of players in the future that may be able to play both formats, but no "hybrid" players exist as of this writing.  Simply put, it may be best to wait a year to see where these technologies are going unless you have deep pockets and just want some HD to show off because you are limited as far as HD content.  A good upsampling (standard) DVD player may be a more cost effective investment until the HD media markets settle a bit.

 

Summary

While this article focused on high definition displays to be used in home theater applications, there is certainly a lot left untouched.  If you are left wondering about which connections to use, HDMI versus component, audio options, HD-DVD versus Blu-Ray, backlighting, or other aspects of home theater, those aspects will have to be covered in a future article.  And again, this article is aimed at those of you who have been wondering if it might be time to make the leap to HD and some things to look for at the starting line.  Just be warned that if you are easily obsessed, home theater and "high definition" can be quite expensive.  While my own setup is quite modest, I've seen people spend $85,000 or more on true home theaters!  On the other end of the spectrum, it is possible to set up a good HD theater for $3,000 with $1,500 being about the bottom entry point, so there's something for everyone.  To start, try to pick a store that has a home theater section with a large selection of displays and knowledgeable/helpful staff.  Let them help you but don't let them push this week's sale on you.  Take your time, use the information from this article, and make the decision that is best for you and I believe you'll find that HD will quickly become a necessity in your home theater.  If you're reading this article soon after release, you may still have time to get that HD set for the big game!  :-)

 

Mike Chaney

3153  Technical Discussions / Articles / January 2007: Profiling a Camera with an IT8 Target on: May 27, 2009, 01:57:17 PM

Profiling a Camera with an IT8 Target


Background

I am often asked about camera profiling in one context or another, and even challenged by other professionals as to whether or not it is even possible to develop ICC profiles for digital cameras.  As I often say, the answer can be complex and may depend on many factors, but let's break it down into a few key points that are relatively easy to understand.

 

What is a camera profile?

I've done numerous articles in the past about color management and color profiling, so if you need a refresher on the subject of color management and ICC profiles, please check out my articles from August 2004 and February 2005.

First, it is important to realize that the only time a custom or "home made" camera profile is needed is when the camera or raw developing software needs a little help to get more accurate color.  Due to differences in lighting and viewing conditions, the term color "accuracy" is a subject of some debate since you are unlikely to be able to reproduce the exact lighting (color temperature) and exact colors from the original scene.  To those who will be viewing your photos on screen or in print, color "accuracy" can best be defined as the reproduction looking as much like the original scene as possible to eyes of the observer.  Sometimes cameras add a little "pop" by increasing contrast and saturation a bit, and that is normally not objectionable unless it is extreme.  What most viewers object to are noticeable hue shifts: color shifts that make a blue sweater look purple, a red flower look orange, green grass look yellow, and so forth.

When hue shifts are significant enough, viewers may remark that the subject is the "wrong color".  In my 5+ years of developing Profile Prism, software that can discover the color characteristics of almost any device to produce ICC profiles, I have developed methods that allow for accurate profiling of digital cameras, with some caveats (under some conditions).  These profiles can be used to improve color "accuracy" and reduce or eliminate complaints about color problems when your raw conversion software falls a bit short.  I mention "raw" because it is nearly impossible to create a usable profile that works under a variety of shooting conditions when shooting in JPEG/TIFF mode with your camera.  When the camera stores a JPEG/TIFF, the image has already been "profiled" (in a sense) by the camera and producing a profile that second-guesses the camera is usually of little use due to the fact that results/colors are often inconsistent when shooting in JPEG/TIFF mode under different lighting and exposures.

 

Profiling a camera: the process

In the early days of digital cameras, it was possible to produce a profile for cameras shooting in JPEG/TIFF mode mainly due to the fact that some cameras produced gross errors that could benefit from correction, even if the result wasn't completely "accurate".  Now, most cameras comply reasonably well with the sRGB color space and many more advanced cameras even offer an option of sRGB or Adobe RGB as the color space used by the camera.  When we have a relatively recent camera model and/or a color space selection, it is rarely beneficial to try to develop ICC profiles for the camera shooting in JPEG/TIFF mode because it is difficult to impossible to produce corrections that result in any consistent improvement.  If we shoot in raw mode, however, most raw conversion tools offer an option to turn off color management so that custom ICC profiles can be created/used.  With color management turned off, the raw data offers a much more consistent starting point, and profiling becomes not only possible, but often quite beneficial.

The process, at least conceptually, is very simple.  Take a shot of a color target in raw mode, develop the raw image with color management turned off in the developing software, and use a profiling tool to create a profile from the image of the target.  The profile can then be activated in the raw developing tool.  That said, the actual process itself can get a bit complex if we want to ensure a quality profile.  You need to get a good shot of the target under good lighting, and you need to use a profiling tool like Profile Prism that was designed with camera profiling in mind as camera profiling requires specialized options like the ability to normalize tone curves and let the device dictate white balance.  There are other high-end (read expensive) tools that allow you to develop camera profiles.  These tools offer specialized targets and software, but I find that with some care, it is possible to match or even exceed the performance of these "high dollar" tools with Profile Prism and a standard IT8 target!

 

The problem, the solution

Before we start with the details, it is appropriate to inject a bit of reality here.  Many raw developing tools, while they are designed to produce the best color possible, just weren't built using any real "scientific" means for color accuracy.  Some use a simple color matrix to tweak color so that it looks acceptable and many don't employ reasonable tone curves to ensure good shadow detail.  In layman's terms, this is the reason that it is often possible to develop ICC profiles for raw images that result in better color reproduction than the raw tools offer out-of-the-box.

If we can develop a profile that improves color over the "default" color reproduction of the raw developing tool, we can say we have a successful/useful profile.  Some may question whether or not it is possible to develop a single profile that works under all lighting conditions, or whether it is imperative to develop one profile for each lighting condition: sunlight, fluorescent, incandescent, mercury vapor, etc..  Again, the true scientific answer here can get complex, but I've found that when profiling the true raw data, a "generic" profile can be developed using direct sunlight.  As lighting conditions (color temperature) shift from direct sunlight to warmer lighting such as incandescent lighting, the profile will become less accurate but the shift is not normally so extreme as to cause gross errors.  This is, in part, because the color filters used on the image sensor aren't changing under different lighting.  Their overall response is the same under different lighting and color temperature only affects the proportions of red, green, and blue recorded by the sensor.  A good profiling tool can discover the overall color characteristics of the sensor which tend to be valid over a wide range of lighting conditions.  Here, the closer you can get to the actual raw data the better, because up-front color corrections only tend to multiply color shifts, so a raw tool that offers the ability to process the raw data without injecting color corrections will work best.

While some may choose to develop different profiles for different lighting, and that's certainly optimal, a generic profile for sunlight should work under a variety of conditions.  Shooting the IT8 target in direct sunlight helps to reduce any metamerism of colors on the target and ensures a good match to the data file that tells the profiling software what the color on the target should look like.  Shooting in direct sunlight also offers the ability to eliminate glare as the IT8 target is a glossy target that, when not shot under the proper conditions, can certainly produce glare which will make the profile useless.  Shooting with the light hitting the target at an angle is imperative to eliminate glare/reflections and due to the fact that our light source (the sun) is so far from the target, we don't have to worry about the light being brighter on the side of the target closest to the sun as we would with angled studio lighting!  Here's how to shoot an IT8 with no reflections or glare:

  1. Of course, a lot depends on your location and the time of year, but in general, the best time to shoot the target is either 1-2 hours before mid-day or 1-2 hours after mid-day.  Try to shoot on a day with minimal clouds so the sun isn't changing intensity/color as you shoot.

  2. It is helpful to attach your IT8 target to a piece of thick cardboard using small tacks or pins at the corners or even tape at the corners as an IT8 will tend to curl and bend when it heats up in sunlight.

  3. Try to find a room where light is entering a window/door at a sharp angle and hitting a wall adjacent to the window.  If you can open the window to reduce lighting variations caused by the glass, all the better!  Here in the northern hemisphere, a south facing window often works well in the afternoon.  If the sun doesn't hit a wall, a palette, chair, or other object may be used to place your cardboard w/IT8 in the sun.

  4. Make sure the room is as dark as possible and that the only light entering the room is coming from the window.  Also try to avoid the direct sunlight hitting bright colored (non-neutral) surfaces such as red walls, blue floor tiles, etc. as these reflections can cause color shifts on the target.

  5. Place your target in the sunlight so that the sun is hitting the target at an angle and you can sit in the shadows while taking the shot.  The following is a typical setup for shooting an IT8 target in direct sunlight.  Notice how the sun hits the target at a sharp angle so that the camera can sit in the shadows, thereby eliminating glare on the target:

  6. If your camera has a custom white balance feature, using a white/gray card or a white sheet of copy paper (don't use photo paper with brighteners), place the card at about the same location as the IT8 and make sure it is in the sunlight.  Use the custom white balance on your camera to white balance on the card.

  7. Take several shots of the target in raw mode.  Take one "normal" shot and then increase exposure incrementally, taking several more shots with brighter exposures making sure to stop just before the exposure gets "blown out" in the highlights.  Camera settings like aperture usually have little influence, but smaller apertures often produce more even lighting across the frame.  Note that camera lens and ISO speed can make a slight difference in profiling, so be sure your ISO speed is set appropriately and you are using your most-often-used lens.  If you and/or the camera are sitting in the shadows of the room, you can take the photo straight-on at the target and you should get no glare or reflections.  When taking the photos, fill only about 3/4 of the frame with the target.  Don't zoom in so far that the target covers the entire frame because light falloff from the edges of the lens can be a factor here.

Once you have the shots of the target, turn off color management in your raw developing tool and develop the photos.  Depending on the raw tool you are using, turning off color management may entail selecting a color management tab and selecting "Embed camera profile", or selecting "None" in the "color management" dropdown.  Whatever you do, the important thing to remember is that you need to be able to turn off color management to develop the profile.  Then, once you are done creating the profile, the profile can be activated in the raw tool by selecting the ICC profile that you created.  Of course, this assumes that the raw tool you are using allows selection of custom profiles.  Not all tools allow use/application of custom profiles so be sure the tool you are using has this feature.  The more popular third party tools like Bibble, Capture One, and (the now discontinued) RawShooter allow the use of custom profiles.  When developing the images, develop to TIFF (you can use 8 or 16 bit/channel TIFF format).

In Profile Prism, click "File", "Open" and open one of the developed images of the IT8 target.  Next, make the following selections on the Profile Prism main window (description and file name are just an example):

Parameter Set to
Type of device to profile Camera/scanner
Reference target Choose the file for your IT8 target
Profile description Something like "Canon 5D Generic"
Printer target N/A
File name Choose a name like canon-5d.icm
Profile for Highest Accuracy
White balance Device dictates WB
Tone reprod. curves Normalize
All other options "Normal" or zero (0)

The above parameters are appropriate for profiling a camera.  Once you have set all the parameters, mark the 4 corners of the target on the image of the IT8 target.  The step by step procedures for profiling a camera or scanner in the Profile Prism help will show you how and where to place the crop markers on the IT8 target.  Once placed, there should be a white punch-out in each of the color squares on the IT8 including the gray scale at the bottom.  If the punch-outs don't align inside each color square on the target, the corner markers have not been placed properly.  Finally, click "Create Profile" at the bottom left and Profile Prism will create your camera profile.  You can then test the profile by selecting the profile in your raw developing tool using the file name you used in the table above.  Once the new custom profile has been set, simply redevelop the photos and evaluate them for color accuracy/appearance.

Since some raw tools like Capture One and RawShooter apply some "pre-curves", it isn't possible to profile based on truly raw data.  As such, you may have to create a profile for each of the exposures (the one normal exposure and several brighter ones) and then pick the profile that has the tone curve (shadow and highlight detail) that you prefer.  Usually, the best result occurs when the curves displayed in Profile Prism (after clicking "Create Profile") end as close as possible to the upper/right corner of the graph.  If the curves end on the top edge or the right edge of the graph, you may need to try a different/better exposure.  Note that it is best to pick a different shot with a different exposure as opposed to tweaking the exposure of a single shot in the raw developing tool!  With a little practice, the above process can produce excellent profiles for any camera shooting in raw mode.  The above are the procedures we used to develop our own camera profiles for numerous raw tools.  These profiles have gotten many positive reviews and are often compared to profiles produced with much more expensive equipment/targets from other sources.

 

Summary

This article can be described as a "secrets revealed" on how to create IT8 based ICC profiles for digital cameras shooting in raw mode using my inexpensive but highly effective Profile Prism software.  With the right tools and a little experience, it is possible to develop excellent ICC profiles for digital camera raw photos using a standard IT8 target.  It is possible to rival or even beat results of software/targets costing 10 to 20 times as much as Profile Prism.  Although there are thousands of satisfied Profile Prism users out there who have created excellent scanner and printer/paper profiles, some users may have been reluctant to try Profile Prism for profiling their cameras in raw capture mode.  I hope that this article will be helpful in getting people started who wish to create custom camera profiles for raw developing tools.  While this article has been tailored to my own Profile Prism software, the techniques can certainly be used by anyone using any software capable of creating camera profiles.  Regardless of the profiling tool you use, the saying "it can't hurt to try" applies here.  Just remember that the whole point of creating a profile is to improve color in the developed images, so always evaluate your results against the "default" color produced by the raw developing tool.  You want to make sure you aren't going backwards, which is a possibility when developing camera profiles!

Since the initial release of Profile Prism in 2001, it has been shipped with two targets: a glossy IT8 and a matte target.  We will be dropping the matte target soon and will be shipping Profile Prism with only the IT8 target as we have found the IT8 to be the most accurate under all conditions and with the tips in this article, the matte target should no longer be necessary for camera profiling.  We use some of the most accurate IT8 targets in the industry and I feel that with a little care, the IT8 can and should be used to profile all devices: cameras, scanners, and printers.  With the proper setup, a matte surface is not necessary to eliminate glare.  While you may hear words of disbelief regarding the ability to profile a camera, especially using a standard IT8, just follow the steps outlined in this article and you may be surprised at the results.  Many times, camera profiling difficulties come from using a tool that doesn't offer the features (like "device dictates WB" and normalization of tone curves) needed for camera profiling.  We've proven with our own custom raw profiles that with the right tools and the right setup, camera profiling can be beneficial and cost effective.

 

Mike Chaney

3154  Technical Discussions / Articles / December 2006: Hype or Hero Take 2: 16 Bit Printers on: May 27, 2009, 01:52:34 PM

Hype or Hero Take 2: 16 Bit Printers


Background

The digital photography market seems to be heating up with products that boast "groundbreaking" features not found on previous models.  Are these groundbreaking products all they are cracked up to be, or are they just hype to get you to buy a new product?  Last month we took a look at the potential of full color capture offered by the Sigma SD14 slated for release within the month.  This month we take a look at 16 bit printing and the new printers that offer specialized 16 bit print plug-ins.  Do they really offer improved color and/or control?  What are the benefits and drawbacks?  If you wanted to print 16 bits/channel, would you know how to do it in order to get the best results?  Let's take a look.  Like most things, the answer isn't simple.  In the digital world, it can take a little hype to make a hero... or vice versa!

 

8 versus 16 bits

The most common image format is a format that offers 8 bits per channel.  For the typical RGB encoded image, that would mean 8 bits to store the brightness for red, green, and blue for each pixel in an image.  Given 8 bits per channel, you get values from 0 to 255 to record the darkest to lightest intensities.  Since values of 0 to 255 can be recorded for all three color primaries (red, green, and blue) separately, the total number of colors that can be recorded is 256 * 256 * 256 or about 16.8 million colors.  Images that contain 8 bits of information per channel are often referred to as 24 bit images.

A less popular format, but one that is gaining recognition, is the 48 bit image format.  This format is similar to the 8 bit/channel format, except 16 bits are used to divide the intensities from dark to light.  With this format, you get values from 0 to 65,535 to record the brightness of the primary red, green, and blue colors for each pixel.  That equates to 65536 * 65536 * 65536 or approximately 280 trillion colors!  All this at a cost of only doubling your storage requirements for each image.  Is it worth it?  It certainly can be, but let's dig a little deeper and try to discover whether or not printers really need this capability.

 

The benefits of 16 bits per channel

By far the greatest benefit to 16 bit/channel images and 16 bit/channel processing comes from the initial image capture, particularly when working with digital cameras that must conform to a variety of lighting conditions.  Having more "quantization" points in the capture range (65536 versus only 256) allows for finer gradations between each color and allows the photographer to adjust for capture issues like underexposure and even overexposure.  Ability to adjust for exposure and still end up with a usable image is very limited when capturing only 8 bits/channel.

Once the initial image capture is done and exposure, white balance, and other factors have been corrected, the performance gap between 8 bit and 16 bit imaging decreases dramatically.  Once any initial exposure and white balance issues have been corrected and the image has been developed (from raw format), 8 bits/channel is almost always enough to get you from the developed image to print (or to screen) with no ill effects.  I refer to raw processing here because if you record in any other format in-camera: JPEG or TIFF, you'll only be getting 8 bits/channel from the capture image to start with and at that point, the benefits of printing your 8 bit images to a 16 bit printer are almost nil.  Obviously the best advice is to shoot in raw capture mode and keep your developed images in 16 bits/channel if you intend to make the most of your 16 bit printer!  That's not to say that 16 bit printer drivers can't offer any benefits when printing 8 bit images, but the utility of the 16 bit printer is all but lost if you intend to send it 8 bit/channel images.

 

16 bits at print time

So the big question is whether or not 16 bits/channel is really needed at print time.  All drivers in all Windows operating systems are 8 bits/channel as 16 bits/channel is "foreign" to the Windows operating system.  That means that you will always require a special plugin to be able to print 16 bits/channel to your 16 bit printer and the normal "File", "Print" command that you use from your standard photo editor or printing tool will not be able to utilize the 16 bit functionality of the printer.

It is worth pointing out that your monitor will still be running at 8 bits/channel, and you've likely never had any problems with displaying images on your monitor so why worry about the printer?  The push behind the new 16 bit printers is the fact that your printer is likely capable of printing some colors outside the range of your monitor's capabilities and due to this extended color gamut, you may need more gradations (bits) to render colors without banding or color posterization.

In reality, there are colors that your monitor can reproduce that are not reproducible by your printer as well.  It is generally believed that the human eye can recognize about 10 to 11 million colors.  So shouldn't 8 bits be enough since that gives us 16.8 million colors?  Like most things, it is a lot more complicated than that, as the 16.8 million colors in 8 bit/channel images are not optimized to match the 10 to 11 million that our eyes see.

 

Comparing gamuts

"Gamut" simply refers to the range of colors that can be reproduced.  Your PC monitor loosely conforms to a color gamut called sRGB.  sRGB is a relatively small gamut and due to its size, 8 bits/channel is enough to represent all colors in the sRGB gamut without any noticeable banding between colors.  While sRGB is good enough to capture almost all colors that can be rendered by your monitor, your printer can likely reproduce colors outside the sRGB gamut: colors that we can see and the printer can reproduce, but will be "clipped" by the sRGB gamut.  If you capture your images in sRGB color space or develop your raw photos into sRGB color space, that means you won't be able to print all possible colors that your printer can reproduce.

Adobe RGB is probably the most popular gamut being used by professionals.  It is a larger gamut and can therefore capture a wider range of colors, and it is still small enough that 8 bits/channel is enough bit depth to render smooth color throughout the gamut.  Adobe RGB is easily large enough to accommodate your monitor, but your printer will still be able to reproduce some colors that are beyond even Adobe RGB.

When you go beyond Adobe RGB and start using color spaces with very large gamuts (such as ProPhoto RGB or Wide Gamut RGB), the gamut is so large that 8 bits/channel may not be enough and you may start to see banding in smooth but gradually changing colors such as a blue sky just before sunset.  Here, 16 bits can help because you have more gradations to work with.  To put it simply, spreading 16.8 million colors across a large color gamut may be spreading things too thin and you may end up with noticeable difference between "adjacent" colors and that, in a nutshell, is what causes color banding in areas that should be smooth.

How bad is the problem to start with though?  Is Adobe RGB really inadequate to reproduce your photos on your printer?  The answer to that question depends on many factors including the colors in the image being printed and the printer you are using.  Generally printers with more ink colors produce larger gamuts, so printers like the Canon i9900 and Epson R1800 have larger gamuts just because they have a wider range of ink (colors).  Let's take a look at the color gamut of the i9900 on Canon's Photo Paper Pro compared to Adobe RGB:

As you can see, there are many colors in Adobe RGB (represented by the wire frame above) that the i9900 printer (represented by the solid shape) cannot reproduce, but there are some "slivers" of color that the printer can reproduce that Adobe RGB will clip.  These problem areas where the color space isn't large enough to hold the color reproducible by the printer are represented by the small sections of solid surface that poke through the wire frame above.  The biggest problem area is the swatch of mid-brightness cyan/green on the bottom left above.  As you can see by the area of cyan/green that pokes through the Adobe RGB wire frame, there are some cyans and greens that cannot be printed using Adobe RGB.  Whether or not this is a problem in your photographs depends on how many photographs you print that happen to have that shade of super-saturated mid-brightness cyan and/or green.  But wait.  It gets even more complicated.  Can your camera even record that information to begin with?  We'll get to that in a minute.

 

Matching gamuts

The biggest selling point for 16 bit printers/drivers is that you need more bits to support the larger gamut of the printers.  Given that the color gamut of the new 16 bit printers isn't really any larger than current 8 bit 8+ ink printers, it doesn't follow that 16 bits would be required to support the full gamut of the printer.  In the end, it comes down to selecting a color space that has a gamut big enough to support all printed colors but not so large that it requires 16 bits/channel to cover the gamut "smoothly".  Yes, if you shoot in raw mode, convert to the super-large ProPhoto RGB, and keep all your developed images in 16 bits/channel all the way to print, the 16 bit printer/driver may help.  Part of the reason it helps, however, is that there's a lot of overkill in that workflow.

At first it might appear that you are losing a good chunk of highly saturated cyan/green colors if you decide to use Adobe RGB for the color space of your developed images, along with a small sliver of magenta and yellow.  When we dig a little deeper, however, we find that the color gamut of the camera's image sensor is even more limiting than Adobe RGB on the cyan/green edge.  This is the color gamut of a Canon 5D Professional dSLR camera.  The color gamut that the camera is capable of recording is the wire frame and the color gamut of the i9900 printer is the solid shape.

Due to the way CMOS and CCD sensors are constructed and the light filters that they use, other cameras like Nikon Professional dSLR cameras have the same limitation on the cyan/blue edge of the gamut, meaning that you gain almost nothing from developing your photos into a super large color space like ProPhoto RGB because your camera cannot capture much more data than Adobe RGB anyway, at least where it is needed!

What all this boils down to is the fact that you need to compare the color capabilities of the camera combined with the reproduction capabilities of the printer itself and when you do that, Adobe RGB is an excellent match and using anything larger is really just overkill.  For the purist who is worried about losing a tiny sliver of highly saturated yellow or magenta that will likely go unnoticed in the few photos that actually contain those colors, I have developed a color space slightly larger than Adobe RGB that is designed to cover the entire gamut of today's printers without being excessively large and requiring the jump to 16 bits/channel.

This printer-optimized color space, called pRGB (for "Printer RGB") automatically installs in the Qimage program folder (usually \program files\qimage) when you install Qimage, so give the Qimage demo a try as it may help you with color managed printing anyway, and as a benefit, you automatically get the printer-optimized color space that works well with any 8 bit printer.  If you want to use it for your other work (like using it in your raw conversion tool or your photo editor), simply right click on the pRGB.icm file in your Qimage install folder and select "Install".  At that point, you can use that color space in any Windows application and you can use it the same way you would any other color space like Adobe RGB, ProPhoto RGB, etc..

Given the fact that 8 bits/channel is enough for finished/developed Adobe RGB images and enough to reproduce almost the entire color range that can be captured by your camera and later reproduced by the printer, I'm going to have to call 16 bit printers/drivers mostly hype at this time, at least given the current state of printing and display technology.  Shooting in raw capture mode, correcting exposure/color issues there, developing into 8 bit Adobe RGB or pRGB images, and printing to an 8 bit driver is all anyone, even the most critical professional should need.

 

What about the reviews of 16 bit printers?

I've seen a handful of reviews on the new 16 bit capable printers and some reviewers do claim to see some differences in the 16 bit versus 8 bit output of the new printers.  I've seen some claims of "more vibrant" or "smoother" colors for example.  I'm quite skeptical at this point at the notion that these differences are really the result of 16 bit/channel capability!  I believe there are a lot of potholes in trying to review these printers.  As an example, I asked one professional photographer to send me prints from his Canon iPF5000, one done in 16 bit mode and one in 8 bit mode because he claimed he could see benefits to the 16 bit mode in several more demanding shots.  I did see that the 16 bit version looked a little smoother in a few places so I asked him how he printed the two versions.  He told me that he started from a raw image, converted to ProPhoto RGB, and then printed.  Knowing that ProPhoto RGB can show some banding for 8 bit images, I asked him to go back and convert the original raw image to Adobe RGB and reprint the 8 bit version.  The banding was gone.  This was simply a case of needing to know how to best utilize both technologies (8 bit and 16 bit) and how to make the most of the 8 bit technology.  I wonder if some reviewers may have fallen into the same pothole and come to the same (misleading) conclusion. 

I will have to say that the 8 and 16 bit versions still looked a bit "different" with respect to slight color casts and certain colors, but one really didn't look "better" than the other to me.  I attribute the minor differences in look/feel to the fact that the 8 bit and 16 bit drivers are two completely different drivers and may handle color just a bit differently.  I also have to wonder if slightly different optimizations in the 16/8 bit drivers alone lead to some reviewers giving the 16 bit specialized driver the nod over 8 bits.  As a matter of interest, the same raw file when developed into Adobe RGB in 8 bits/channel and then printed to an older Canon i9900 (which is not capable of 16 bit printing) produced a print every bit as good as the iPF5000 print in either 8 or 16 bit mode.  While these tests are hardly definitive, at this point, logic has to step in and you have to wonder how we've been using 8 bit/channel printers for decades, profiling them in raw (no color adjustment) mode, using different papers, etc. and have never had a problem.  Yes, sometimes it's hard to realize what you were missing until you see the new technology, but I'm not seeing any real benefit to 16 bit printing at the moment.  As technology on both ends (camera to printer) improves over time, I may have to revise my outlook in a future article.  :-)

I do think 16 bit printers can make workflows easier if you choose to go the overkill route all the way (raw to ProPhoto RGB color space, keeping the 16 bit/channel image format all the way) because you don't really have to worry about being careful.  That does have some appeal, but as long as you shoot in raw, do any "heavy handed" manipulation like large changes to exposure and/or white balance at the raw stage before you develop, you can still develop to Adobe RGB, print to the standard 8 bit driver, and get results that are as good as the 16 bit driver plugin on the same printer.  One final thought to keep in mind is that storing developed images at 16 bits/channel doesn't just fill up your hard drive faster.  It creates added burden at the processing stage as well by doubling the amount of memory needed to process (interpolate, sharpen, spool, etc.) and that can result in problems when doing things like printing very high resolution scans or photo montages or printing large prints.  I think that 16 bit printers are new enough that the jury is still out as far as the total benefits offered by 16 bit printing.  I would simply caution that changing your entire workflow to 16 bits at this point simply because you own a 16 bit printer may be a bit premature and may lead to unnecessary side effects.

 

Summary

Certainly, we live in a "more is better" world.  Just look at how manufacturers are still able to sell consumer level cameras with more pixels and pixel counts continue to increase every year despite increased image noise and a general decline in overall image quality.  Right now, with the current state of technology considering cameras, monitors, and printers, I really don't see any real benefit to 16 bit printing over 8 bit printing when 8 bit printing is done properly.  That said, it can be easier to foul up 8 bit printing and end up with artifacts like banding and color posterization if heavy editing like exposure correction or white balance is done at the wrong stage or one tries to use a super large color space like ProPhoto RGB in 8 bit mode.

The bottom line is that I believe there will be little or no difference between 8 bit and 16 bit printing provided you follow an acceptable workflow for both.  If you've been thinking of shelling out a few thousand dollars on a new printer because it is touting 16 bit printing, my advice is to hold on to your money for a little while longer.  As with anything in the digital imaging industry, a general consensus will emerge in the next 6 to 12 months about how useful the 16 bit printing really is, and certainly these new printers (which I'm sure you have noticed will go unnamed in this article just to be "politically correct") have benefits above and beyond just being able to print at 16 bits/channel so as more and more people use them, the benefits and costs will become clear over time.  The handful of 16 bit capable printers offered at the time of this writing are excellent printers, just don't buy them solely for their 16 bit print capabilities.  In closing, I do believe 16 bit printing capability is a good feature and wouldn't mind seeing it on all printers, but it certainly should be low on the priority list when evaluating what you need in a photographic printer as the real world benefits are quite limited.

 

Mike Chaney

3155  Technical Discussions / Articles / November 2006: Full Color Capture: Hype or Hero? on: May 27, 2009, 01:49:33 PM

Full Color Capture: Hype or Hero?


Background

You may have heard about the upcoming Sigma SD14 that offers full color capture, but do you know what full capture is and what it can do for your photos?  Will the full color capture SD14 set a new standard for digital cameras or will it be a mere curiosity like it's older siblings the SD9 and SD10 which developed a loyal following but never quite turned the tables on sensor design as originally hoped.  As of this writing, the Sigma SD14 is not yet out, but the technology is already in place so let's take a look at the technical details of full color capture versus single color capture.

 

Single Color Capture

The vast majority of digital cameras including high end professional dSLR's use an image capture sensor that can record only one color per pixel.  Most sensors use what is often referred to as a "Bayer mosaic" pattern where the sensor only records one of the three primary colors (red, green, or blue) at each photo site (pixel).  A six megapixel dSLR, for example, may have a sensor with 3000 x 2000 resolution.  One thing that is often overlooked is the fact that each of those "pixels" on the sensor only records a single color: red, green, or blue.  To make matters even more complicated, single color capture sensors do not divide their pixels evenly, recording 1/3 red, 1/3 green, and 1/3 blue.  Instead, half of the pixels on the sensor are green while only 1/4 are red and 1/4 are blue.  More green sensors are used because having greater sensitivity/resolution in green mimics how the human eye captures color.  The RGBG layout of a standard digital camera sensor looks something like this:

Since a 3000 x 2000 (six megapixel) dSLR returns a full color image with all three colors present at each pixel, the most obvious question at this point is how we can end up with a full color image when only one color was recorded for each pixel on the sensor!  The answer lies in interpolation.  Digital cameras and raw processing software use sophisticated algorithms to predict the missing two colors at each photo site (pixel).  As an example, take a look at a blue photo site somewhere in the middle of the above graphic.  Notice that at every blue photo site, there are four red photo sites adjacent (diagonally) to the blue photo site.  If all four of those adjacent red photo sites have high red brightness, it can be "assumed" that the blue pixel will also have high red brightness.  This is a simple example but similar prediction-based algorithms are used at all other pixels to recover the two missing primary colors for each pixel until each pixel has all three colors (one actual, and two predicted).  Obviously the algorithms get much more complicated when surrounding photo sites are not the same brightness, but the general idea is to "guess" the missing two primary colors at any given pixel by looking at the color of surrounding pixels.  Once both of the missing primaries have been interpolated for each pixel, the final full color image has been reconstructed.

 

Problems with single color capture

The above single capture Bayer Mosaic sensor is used in nearly all digital cameras as of this writing.  If you are familiar with interpolation, you probably already know that interpolation comes with certain drawbacks.  Because a single color capture sensor only captures one of the three needed colors at each photo site, two thirds of the information in your photos is being "guessed" while only one third is "real" data!  By the numbers, you'd have to wonder how this even works at all!  The answer lies in the fact that our eyes are more sensitive to changes in detail, edges, and brightness than changes in color.  In addition, the interpolation algorithms used to reconstruct the missing colors at each pixel have become so advanced that they actually do a very good job at predicting the missing colors under most circumstances.

The real issue with single color capture sensors arises when you have subjects that have colors close to the primary red, green, and blue colors used for the photo sites on the sensor.  For areas of detail that are black/white, all photo sites on the sensor will be reacting similarly (will have similar brightness).  This makes it easier for the interpolation algorithm to reconstruct the image because each photo site will be recording near the same values.  This is why, when reviewers shoot resolution charts, the cameras return resolution numbers comparable to what you'd expect if the sensor were actually a full color capture sensor recording all three primary colors at each photo site.

When the balance of color starts to shift however, particularly toward red or blue, things start to go downhill.  When shooting a bright red flower with dark red veins that only "excites" the red photo sites on the sensor for example, you can see by the graphic above that your resolving power quickly drops to near 1/4 resolution. This is because the green and blue sensors simply offer no data (they are black) and only the red sensors contribute data.  The same would be true of a bright blue sweater or blue fabric.  While black/white subjects may be resolved at near full resolution, some red/blue subjects may fall to near 1/4 resolution and other colors like yellow, green, orange, etc. fall somewhere in between.  Of course, you don't see this difference as missing pixels: only a loss of detail/sharpness.  The result is that you end up with an inconsistency in sharpness in photos that makes some colors less sharp/detailed than other colors, and the visual result is a bit "flatter" look that some would see as less three dimensional.

The only saving grace for the single color capture sensor is the fact that it is often difficult to find a subject that has a color so closely matched to the red, green, or blue filters on the sensor that the other two primaries receive no data whatsoever.  As an example, the red photo sites on the sensor will certainly be affected more than the green and blue sites, but most shades of red will still invoke some type of response from the green and blue sensors.  It is rare to find a shade that matches so well that the sensor records no information whatsoever at the green/blue sites.  Granted, the lower the brightness recorded on the green/blue photo sites, the lower detail you'll have to work with for that red subject and (potentially) the higher the image noise levels.

For more information on "sharpness equalization" as a means for correcting loss of sharpness/detail in single capture sensors, please read my article at Digital Outback Photo or try the "sharpness equalizer" in my Qimage software.

 

Full color capture and what it can do for us

Released in 2003, the Sigma SD9 was the first camera to offer full color capture.  The sensor, manufactured by Foveon, was touted to be the next generation in digital camera sensors.  Using three sensor "layers", the SD9 (and soon-to-follow SD10) offered the ability to capture all three primary colors (red, green, and blue) at each photo site on the sensor.  Since no interpolation was necessary, the typical problem with sharpness/detail consistency across different colors was solved and to most people, the result was a more 3D feel to images.  The new technology didn't come without problems though...

The first problem faced in mass marketing this new technology was that, while the SD9 and SD10 were marketed as 10 megapixel cameras, the final images were "only" a little over 3 megapixels.  The Sigmas were competing with 6 megapixel dSLRs that, to the "unwashed" appeared to have twice the resolution even though the full color capture Sigma was actually capturing more data, and doing it in a more sensible fashion.  Because many reviewers base resolving power on test shots of a black/white resolution target, the Sigma performed poorly compared to the single color capture 6 megapixel dSLR competition because black/white detail is handled nicely on standard cameras. Had those resolution test shots been black/red or black/blue instead of black/white, it would have been a different story.

It didn't help matters that you can't stop the age-old rule of thumb that you need 300 PPI of detail to get a good print.  The die hard 300 PPI camp would argue that they could print bigger prints using a standard single color 6 megapixel dSLR because the final image was 6 megapixels compared to the 3.4 megapixels recorded by the full color capture SD9/SD10.  It also didn't help that the SD9/SD10 could only shoot in raw format and pictures had to be developed after-the-fact and that the camera body wasn't the best on the market at the time and being a Sigma body, it needed Sigma lenses which gave Nikon and Canon followers pause.

The final tether that kept full color capture from reaching escape velocity in the SD9/SD10 is the fact that it did have some problems recording consistent, noise free color.  People familiar with the camera and raw developing software could produce some gorgeous photos but it did, on average, take a little more work than standard single color capture dSLRs.  It turns out that the layers used in the Foveon full capture sensor made it more difficult to get consistent/accurate color fidelity compared to the arguably simpler design of the single color capture sensor.  The result was that the full color capture Foveon based SD9/SD10 were a little harder to keep under control with respect to color accuracy and they suffered from a bit of metamerism (colors shifting under different light sources) that was not accounted for by the hardware/software.

 

Looking for a bottom line: is full color the future?

Right now, the SD14 appears to be the new contender in the next attempt to get full color capture into the mainstream of digital photography.  The camera has not yet been released, but you can find information about it here.  At first glance, the SD14 seems to step into the ring with some of the same handicaps that held back its older siblings.  While it will be advertised as 14 megapixels because it records three colors at each photo site, it will return final (non-interpolated) images that are under 5 megapixels, less than half the final resolution being returned by the single color capture competition.

It remains to be seen if Foveon has improved color fidelity of the full color capture chip and if Sigma have made improvements to the body, but at least the SD14 is capable of returning developed (JPEG for example) photos and doesn't require raw developing tools.  While I always shoot in raw mode by choice, some jobs actually require shooting finished images for the sake of time and I'm sure the ability to shoot in a "finished form" will improve sales.  Final price still has not been set to my knowledge, so I'm sure that will be a factor as well.

Technically, the SD14 is an interesting camera and I applaud Sigma/Foveon for keeping the concept alive!  It really has potential as it does correct some image quality flaws inherent to single color capture devices.  In this respect, the SD14 is an important entry in the world of dSLR cameras!  Mathematically speaking, the SD14 will record 40% more "real" data than a 10 megapixel dSLR even though the final images will have half the pixels.  It sounds confusing at first, until you realize that the SD14 is investing the data in color capture rather than added pixels.  Whether or not the "masses" will recognize that extra data as a benefit or a detriment remains to be seen, but if it didn't happen the first time (with the SD9/SD10), I have my doubts this time around.

 

Summary: The future of full capture

Full color capture resolves a number of issues related to today's single color capture sensors.  Single color capture has been around for decades, however, and the sensors and the interpolation algorithms that make them work have been refined over time.  Many of the pitfalls of single color capture can be addressed with advanced color interpolation algorithms.  As a result, to really get noticed, I believe full color capture has to take a leap forward that would make it a clear winner in the eyes of the consumer.  In my opinion, to do that, the final image resolution needs to be comparable to today's dSLRs.  Regardless of how good you are with math, some will see the SD14 as a 4.6 megapixel camera competing in a 10+ megapixel market.  Even if you grant that the SD14 actually records 1.4 times the amount of data compared to a typical 10 megapixel dSLR, 12-14 megapixel dSLRs are on the horizon that will match the amount of data recorded by the SD14.  Anyone familiar with digital sampling and integration will realize that if you make the pixels small enough and abundant enough, it won't matter that you can't record all colors at once.  Case in point: inkjet printers, audio CD's, DVD's, etc.  At some point, when the pixels get small enough, it won't matter whether they are on top of each other or not!

Due to the consumer perception of "more pixels = better camera", it is my belief that had Sigma released an SD30 that returned 10 megapixel non-interpolated final full color images, it may have made a big dent in the digital camera market and may have turned the tide provided the technology worked as advertised.  As is, it may end up being nothing more than another curiosity.  Personally, I wish Sigma/Foveon had made a big leap forward like an SD30, but I also have to realize that true technical marvels take time and often come in small steps.  Either way, for me, the SD14 will be an interesting camera that I hope, if nothing else, will help move us forward in the arena of full color capture!

 

Mike Chaney

3156  Technical Discussions / Articles / October 2006: In-Camera Color Spaces on: May 27, 2009, 01:46:39 PM

In-Camera Color Spaces


Background

So you've been fumbling through the custom menu settings on your new dSLR or high end camera and you've found a setting called "Color Space" or something similar, with choices like "Adobe RGB" and "sRGB".  What do these settings mean, and when are they used?  Let's take a look and try to make some sense of this, because it can alter your images and frankly, can really foul things up if you don't know how to set this option and how to properly view/print the photos taken with the selected color space.

 

What is a color space

While I've covered color spaces and profiles in previous articles, a brief description of the term "color space" is probably worth summarizing here, if only briefly.  A "color space" is like a language that describes what the red, green, and blue values in your images really mean.  You might be tempted to think that knowing the RGB value of a pixel gives you it's exact composite color.  Such is not the case, however, as different shades of red, green, and blue can be used as primaries which means that a particular RGB value can indicate an entirely different color on two different devices (or images).  Tweaking the red, green, and blue primaries gives the ability to store images in a color space that better matches the device(s) that will reproduce the photos later.  A monitor, for example, is capable of reproducing a different range of colors than a printer, so using a different color space for the monitor and printer will allow both of those devices to achieve closer to their full potential. 

Visually, you can think of a color space as a container that holds all RGB values possible in an image from 0,0,0 to 255,255,255 including all combinations of vibrant, saturated colors in between.  The larger the container, the more colors that can be reproduced but the further apart the RGB values become since they are spread across the entire container.  The trick becomes trying to match the size/shape of the container that holds the image to the size/shape of the container that is used by the monitor/printer.  This matching of containers (color spaces) is what color management is all about.

 

sRGB versus Adobe RGB

High end consumer cameras and dSLR cameras usually offer two choices for color space: sRGB and Adobe RGB.  sRGB is what most PC's and monitors use and it will display reasonably well on emails and web pages without the need for any color management software (web browsers and the like do not offer color management).  While sRGB is generally well matched for your average PC monitor, the "container" is rather small with this color space: it doesn't cover some of the more vibrant and saturated shades that might possible to capture with the camera and reproduce on your printer.  That brings us to Adobe RGB.  Adobe RGB is a larger color space than sRGB, meaning that the container is large enough to hold colors that would be "clipped" in sRGB space due to those colors being too bright/saturated to be reproduced in the smaller sRGB container.  Shooting/storing images in the Adobe RGB color space will allow you to capture and therefore later reproduce vibrant, saturated colors like deep yellows, cyans, and magenta colors found in subjects like flowers, some clothing dyes, and other subjects with very deep and saturated color.

 

sRGB and Adobe RGB in practical use

By now, you're probably thinking, why even bother with sRGB if Adobe RGB can record a wider range of colors?  Good question!  The simple answer is that, unfortunately, the whole world is not yet ICC (color management) aware.  By that I mean, sRGB is a good middle ground if you are placing images in a public venue such as the web or email, not knowing whether or not the recipient can "decode" the more specialized Adobe RGB color space.  If he/she doesn't have color managed software, the Adobe RGB image will probably look washed out because it's "container" is not as well matched as sRGB to a standard monitor.  Simply put, the use of Adobe RGB color space requires specialized software to view/print the resulting images accurately.

When using fully ICC aware software such as Qimage or PhotoShop, the software will know how to take the colors from the larger container (Adobe RGB) and map them properly into the smaller containers used by your monitor/printer.  Since your monitor covers certain colors that your printer cannot print and vice versa, using a larger color space up front and then converting doesn't "penalize" either device and makes the most of your images.

If you have/use ICC aware software, there is a strong argument for using Adobe RGB in that it is a larger color space and can store a wider range of color.  You can't get back what you didn't record in the first place!  After all, if you are familiar with ICC aware software, you can easily convert from Adobe RGB to sRGB should you need to email someone some photos or upload photos to a web site, so using Adobe RGB doesn't mean that you can never use those photos for web/email display!  In addition, when printing photos, your ICC aware software will know how to translate the wider range of colors that Adobe RGB color space offers so that they can be reproduced in print (provided you have an ICC profile for your printer and the paper you are using).

 

When using Adobe RGB, be aware...

Be aware that the sRGB/Adobe RGB selection on your camera applies to in-camera JPEG/TIFF images only.  If you are shooting in raw mode, your raw images will not be altered or stored in any color space so the color space selection will not be a limiting factor: you'll choose the "converted" color space in whatever raw decoder you use to develop the raw images.  Shooting in raw format really offers the widest gamut (color coverage) because raw images record data straight from the image sensor and that data covers an even wider gamut than the larger Adobe RGB color space!  For frequent shooting of subjects with very vibrant and saturated colors, this can be important because there are likely some areas of color that your printer can reproduce that not even Adobe RGB can record.  For example, most inkjet printers can reproduce some shades of yellow and cyan that are beyond the Adobe RGB color gamut.  This is not normally an issue with "general" shooting, but can become a factor when shooting subjects that fill the frame with vibrant colors such as might be the case if you are shooting sunflowers in bright sunlight.

Also be aware that not all so-called ICC aware software can discern when your camera JPEG's have been stored with the Adobe RGB color space selection.  The above mentioned Qimage and PhotoShop can automatically decode the color space properly, but many other photo editing and printing tools will not.  The bottom line here is: make sure you check the software you are using to ensure that it is picking up the fact that your images are stored in Adobe RGB color space.  If the software you are using opens the images in sRGB color space, you'll know that the software isn't properly decoding the embedded color space tag(s) in the images.  In that case, you may need to manually assign the Adobe RGB color space to tell the software that the images are in that space.  Unfortunately, camera manufacturers still aren't embedding the actual profile in the images even though doing so would only add about 500 bytes to the size of the file.  What we are left with are a handful of programs that are smart enough to decode the proprietary embedded tags used by the manufacturers, so be careful when shooting in Adobe RGB color space that your software actually recognizes the photos as Adobe RGB photos!

 

Summary
 

The short story here is that I recommend using the Adobe RGB color space when shooting JPEG's or TIFF's with your camera if your camera offers the option AND you are familiar with color management and ICC profiles.  Because the sRGB color space is smaller and cannot record as many colors in the vibrant and saturated range, it should be used only on more limited platforms such as specialized applications that do not (or cannot) make use of color management.  For example, you may be forced to use sRGB color space if you must rely on super-fast or super-simple output that requires printing directly from the memory card using a printer that can print without a computer attached.  In this case, the printer will likely not recognize Adobe RGB photos and will probably assume the photos are in sRGB color space.  The result will be dull and inaccurate color if the standalone printer assumes sRGB color space but is "fed" Adobe RGB photos.  So if you have the time, the software, and the know-how, Adobe RGB is the way to go unless you are shooting in raw mode, which gives you even more flexibility as the color space decision can be made later, when you develop the photos.

 

Mike Chaney

3157  Technical Discussions / Articles / September 2006: Working With Aspect Ratios on: May 27, 2009, 01:44:25 PM

Working with Aspect Ratios


Background

This month we tackle the simple but often misunderstood topic of aspect ratios and how to handle cases where the aspect ratio of the image doesn't match the aspect ratio of the print.

 

Aspect ratio: the simple definition


Aspect ratio is nothing more than width divided by height.  The higher the aspect ratio, the wider the image (or screen).  For example, standard televisions have an aspect ratio of 1.33.  That is because the screen is 1.33 times as wide as it is tall.  This 1.33 aspect ratio can be written as 1.33, 1.33:1, or 4:3.  HDTV sets have an aspect ratio of 1.78, sometimes displayed as 1.78:1 or 16:9.  The higher number (1.78 versus 1.33) indicates that the HDTV set has a wider, more rectangular screen than the more "square" standard set.

Standard TV

4:3
1.33:1

HDTV

16:9
1.78:1

This same concept applies to digital photographs.  Most consumer digital cameras capture a picture that is about the same aspect ratio as a standard television: 4:3 or 1.33:1 while professional dSLR cameras often use the 3:2 or 1.5:1 standard that matches the typical 35mm negative, slide, and 4x6 photograph.  As you can see below, a dSLR produces a picture that is a little more rectangular (wider) than the more square photo from the consumer camera.

Consumer Camera

4:3
1.33:1

Pro dSLR Camera

3:2
1.5:1

 

Matching aspect ratios

Now that we know the definition of an aspect ratio, it becomes clear that we have a problem.  First consider aspect ratios that match.  For example, the 3:2 photo from a pro dSLR camera (displayed above right) can be printed at the popular 4x6 photo size because the aspect ratio of the image (3:2) matches that of the print which is also a 3:2 ratio.  That means that the entire photograph from the pro dSLR camera can be printed as a 4x6 print with no cropping and the final print will be exactly 4x6.  Here, we have no problem because we have a match between the aspect ratio of the image and the print size we have chosen.

The problem occurs when we have a mismatch.  For example, if we have a consumer camera that produces 4:3 photos, we cannot print a 4x6 photo without either distorting the image (making the subjects look wider than normal) or cropping some of the image.  Let's consider three methods for obtaining a 4x6 photograph from a consumer camera that records a 4:3 "mismatched" image.

Method 1: Fit in frame

With method 1 above, we fit the entire 4:3 photo inside a 4x6 frame.  Using this method, the actual photograph is 4 inches tall but only 5.3 inches wide.  The white bars on the left/right fill out the rest of the 4x6 photo and would show if mounted in a 4x6 photo frame.  This method is often not desirable when placing photos in a frame because the white bars show inside the 4x6 frame.  The advantage to using this method is the fact that the entire photo can be printed with no cropping.

Method 2: Crop to Size

With method 2, we crop out a portion of the center of the photo using a 3:2 crop.  Using this method, we lose a little off the top and bottom (notice the flags are missing on the bottom) but we lose nothing on the left/right.  This method is often the preferred method since the photograph will be exactly 4x6 inches and will fit in a 4x6 frame with no borders.  The compromise, of course, is that we must lose a bit of the image on the top and/or bottom.

Method 3: Distort (stretch)

The third and least preferred method is to "stretch" the image from left to right so that the entire image fits in the 4x6 photo.  Since this method distorts the image, it should not be used with photographs.  The distortion is not as obvious in the above photo as it would be with people as subjects.  We can see that the tall/skinny building near the right/center of the photo looks "fatter" in the distorted image.  This effect is more noticeable with people/faces than with buildings for which we have no internal reference in our mind.

 

Dealing with the differences
 

While any photo editor will allow you to achieve any of the above aspect-ratio-matching methods, the best way to deal with this issue is to use software specifically designed for photo printing.  Most photo printing software will allow you to easily switch between methods 1 and 2.  Method 3 is not offered in most photo printing applications as it is considered an "error" since it distorts the photo.

As an example, in my own Qimage photo printing software, you can easily switch back and forth between "fit in frame" and "crop to fit" by simply selecting photos on the page and clicking the crop button (scissors icon on the main window).  With the button in the up position, photographs will print via method 1 above.  With the button down, method 2 is used.  When using method 2, the default cropped area is the exact center of the photo (equal portions of the top/bottom are cropped in the above example) but the area that gets cropped can easily be changed.

When using method 2, it is desirable to have a quick and easy way of adjusting the part of the photo that is cropped.  For example, if the flags are important, you may want to drag the crop down a bit so that the flags are included in the photo, losing a little more of the tops of the buildings.  If the flags are not important or are considered a distraction, you need to be able to drag the cropped area up so that the flags disappear and you get more of the tops of the buildings.  In Qimage, this task can be performed simply by clicking the "Full page editor" button under the preview page on the main window and then dragging the small image on the "Cropping" tab on the right side of the page editor window.

Other photo printing software may offer similar methods of fitting/cropping and adjusting but most multiple-photo-printing programs do offer the option at some point in the user interface.

 

Summary
 

While this entire topic may be trivial to the advanced amateur or pro, I'm still surprised by how many inquiries I get on a daily basis regarding how to effectively deal with this issue.  I often get the same question, for example, asking how to print a 4:3 photo at exactly 4x6 inches without cropping.  After reading this article, hopefully the answer is clear: the only way to do this is by distorting the image.  Other than distorting the image, your only other options are to adjust the size (to 4 x 5.33) or crop some of the image (on the top and/or bottom).  Obviously, this article focused on one example but similar situations exist when printing other sizes.  For example, we have the same problem when trying to print a 3:2 photo from a dSLR at a size of 8x10 or 5x7.  Also note that depending on the orientation of the image (portrait/landscape) and the image-versus-print aspect ratios, sometimes the cropping method will require cropping from the top/bottom rather than the left/right.  I hope this article will help in the basic understanding of aspect ratios and the handling of "mismatched" aspect ratios.

 

Mike Chaney

3158  Technical Discussions / Articles / August 2006: Enable Advanced Printing Features on: May 27, 2009, 01:41:09 PM

Enable Advanced Printing Features


Background

If you right click on your favorite printer in Windows "Printers and Faxes", you will find a little check box labeled "Enable Advanced Printing Features" on the "Advanced" tab.  Of all the printing features found in your print driver and printer properties, this is perhaps the most mysterious.  Having a check in that box when printing photographs (particularly large prints) can cause a multitude of problems from error messages to missing pieces of photos or blank pages.  Remove the check and you may start to experience other issues such as longer print processing times or failure of the print driver to "release" the printing application in a timely fashion after it is finished processing.  In this article, we'll take a quick look at this mysterious printing feature, try to give it some meaning, and we'll look at how my recently released Qimage 2007 photo printing software can make working with this feature a bit easier.

 

Two printing modes: raw and EMF


When working on the "Advanced" tab of your printer properties in the Windows "Printers and Faxes" dialog, unless you check "Print directly to printer" (which is normally not recommended), Windows will spool data to your printer.  Since most printers accept data much slower than the printing application can process it, "spooling" can make life easier by capturing the data going to your print driver, putting it in a holding area (temp files on your hard drive), and then spooling it in the background later, at a transfer rate that the printer can handle.  In a sense, the spooler is the middleman between your printing application and the printer and it sits in the background "feeding" the printer as fast as it can take the data.

Windows employs two methods of feeding the printer via the print spooler: raw and EMF (enhanced meta-file).  Let's take a look at both spooling methods.

 

EMF: "Enable Advanced Printing Features" ON
 

If there is a check in "Enable Advanced Printing Features", you have turned EMF printing on and have told Windows that it can defer some of the print processing until later.  Data is saved and the spooler later feeds each page to the print driver for further processing by the driver before it is finally sent to the printer.  With "Enable Advanced Printing Features" checked, your printing application will likely finish it's processing job faster and control will be returned to the application faster.  This is because the data being sent to the spooler is simply "stored" as a meta-file that is not fully processed (actually sent to the driver) until later, when the spooler begins sending data to the printer in the background.  Sounds like a win-win, right?  Well, almost.

One major drawback to the EMF printing mode is that, while the printing application will be able to finish processing data faster, a (sometimes much) larger spool file will be created because there is simply more overhead in the EMF spool file format in most cases.  These larger spool files can cause problems if you are running low on hard drive space or you are printing to a network printer.

In addition, since EMF printing involves the spooler "talking to" the print driver at a later time to finalize data, a lot depends on the print driver being used as to how much additional space will be required for the EMF format, or even whether the EMF format will work with the printer.  While most printers can handle EMF printing, some more specialized printers may not come with standard Windows drivers and if they don't, chances are they will not work in EMF mode because, well, there is nothing for the spooler to "talk to" later.  In such cases, "Enable Advanced Printing Features" must remain unchecked.

 

EMF: "Enable Advanced Printing Features" OFF
 

If "Enable Advanced Printing Features" is turned off (unchecked), Windows will create a spool file in the raw format.  That is, the driver is invoked up front (as your printing application is processing the data/pages) and the raw data that is ready for the printer to receive is spooled into file(s) on the hard drive.  Due mostly to halftoning and the fact that most inkjet printers don't offer continuous color for each printed "dot", these raw files are usually smaller and therefore create smaller spool files on the hard drive.  This is often helpful when printing to network printers or when running low on drive space.  When printing in the raw mode with "Enable Advanced Printing Features" turned off, your printing application will likely pause at the end of every printed page while the print driver is invoked to decode the raw data that needs to go to the printer.  These pauses can sometimes be lengthy (up to 30 seconds or more on larger pages) and can really add to the amount of processing time needed by the application you are using to print.  Sound like a bad idea to print in this mode?  Well, not really.

Simply put, raw printing with "Enable Advanced Printing Features" turned off is more reliable.  While the initial processing may be slower, normally less disk space will be required and that can result in more reliable printing on drives that are low on disk space.  In addition, some older operating systems and/or older print drivers may have a limit on the amount of data that can be read by the spooler in EMF mode, meaning that printing in raw mode may allow you to print more data or larger prints than the EMF mode.  Since EMF printed data is only partially processed, large EMF print jobs sometimes fail due to the inability of the spooler/driver to finish processing data when dealing with large jobs.  Raw printing, on the other hand, can be more reliable simply due to the fact that the spooler doesn't have to continue to communicate with the print driver to finish processing the data: the raw data is already ready for output.

 

What's best in practice?
 

I've printed 44 x 96 inch prints and larger at 720 PPI without incident with "Enable Advanced Printing Features" turned on.  Because having this option checked can make life easier by allowing your printing software to finish processing faster, I'd recommend leaving "Enable Advanced Printing Features" checked unless you have problems.  If you uncheck it, you will start to notice things like pauses after each printed page and a (potentially substantial) delay between when your printing software finishes printing and when Windows returns control to that application.  In addition, turning off (unchecking) "Enable Advanced Printing Features" will disable the print preview function on Canon printers, so if you are wondering why "Preview" is grayed out in your Canon print driver, it might simply be because you don't have "Enable Advanced Printing Features" checked.

By far, the most common symptom of problems related to checking the "Enable Advanced Printing Features" option is missing print data.  If this option is checked and you start to get prints that are only partially printed, pages that are missing, hard drive space errors, or other issues that can't be tracked down to other areas, you may wish to uncheck "Enable Advanced Printing Features".  If the problem disappears, you'll know to leave that box unchecked in your printer properties.

Again, on most systems, checking "Enable Advanced Printing Features" will result in faster processing.  While that won't speed up your printer, it will definitely result in your printing software being able to process the job faster and that means returning control to you faster so that you can do more work while the printer is printing.  If you don't want to get into the details of changing these settings in Windows or you are having trouble remembering which option has which benefits, I've designed my recently released Qimage 2007 photo printing software to be able to print either way.  Simply use "Edit", "Preferences", "Printing Options" and you can set the spool type to either the default "EMF - Faster printing" or "Raw - Large prints".  Qimage will make sure that other corresponding options such as the spool data type are set optimally and that "Enable Advanced Printing Features" is checked/unchecked in your printer's properties based on your selection.

 

Mike Chaney

3159  Technical Discussions / Articles / July 2006: A Raw Lifestyle on: May 27, 2009, 01:38:48 PM

A Raw Lifestyle


Background

In my April 2005 article, I discussed the ups and downs of working with 48 bit (16 bits/channel) images.  In this month's article, we take a bit of a vacation from the technical to talk about workflows and lifestyles related to shooting in raw capture mode.  Even if you have a digital camera and happily shoot JPEG's all day long, this article may be worth a read because some day you may decide to make the jump from "cooked" to "raw".  This article, of course, assumes you have a camera that allows you the choice of shooting either JPEG or raw format images.

 

Raw mentality, raw lifestyle


In a sense, shooting raw images can be described as a lifestyle change as it affects nearly every aspect of how you capture your life and the lives of others through your photography.  At the heart of the matter is the fact that capturing raw images means that when you are finished shooting, you'll end up with a flash card containing digital "negatives" that must be developed before they can be viewed or printed.  In contrast, when you shoot in JPEG capture mode, the camera applies processing before the developed image is saved on the flash card.  Capturing raw images offers a number of benefits but at the same time imposes a bit of a lifestyle change in that an extra step is introduced into your workflow: raw image development.  Let's take a little closer look at the process of raw shooting and development.

 

Raw benefits
 

Perhaps the most obvious benefit to shooting in raw capture mode is the fact that you are truly storing a digital "original" just as the scene was captured by the camera.  In comparison to JPEG shooting where the data is massaged and manipulated prior to saving, raw capture mode stores the data as it was digitized straight off the image sensor.  This allows higher bit depth, greater dynamic range, and much greater ability to correct issues such as underexposure or even overexposure.  The only thing better than capturing the data directly off the sensor would be actually going back to the scene and taking the shot again.  To put things into perspective, in JPEG capture mode, your camera is able to capture 256 gradations at each pixel site on the sensor.  In a well exposed shot that doesn't need white balance corrections or other tweaks, 256 gradations for each color is enough.  It can begin to fall short, however, when an image is underexposed, overexposed, or shot under the wrong white balance.  Raw images have the ability to store 4096 gradations of color at each pixel site (12 bits/channel) or even higher on some cameras.  This extra depth allows for greater accuracy and reduces banding/posterization when making color or exposure corrections.

 

Raw workflow
 

Currently, the biggest problem with shooting raw images is the fact that each manufacturer has its own raw file format and that format can (and usually does) differ even between different camera models from the same manufacturer.  This keeps third party software developers scrambling to keep up with the latest undocumented incarnation of NEF, CRW, RAF, and so on, and is the reason that I discontinued development for new raw formats in my own Qimage photo printing software years ago.  The fact that most manufacturers do not document raw image formats so that they can be decoded by third party applications has prompted many software developers to stop supporting raw formats or only provide "skeleton" support for the formats, leaving the quality developing stage to the dedicated raw developing tools.  To the photographer, this means that you can't simply open the image, print the image, or send the image to someone else without first developing the raw image.  To me, this is where the lifestyle change takes place.  If you shoot raw images, you need to be comfortable with the fact that your images must be developed using a professional raw developing tool.  In the same way it isn't sensible to pull undeveloped film out of a roll fresh from the camera and expect to view it without developing it, it isn't sensible to expect to pull raw images from a flash card, pop them onto your desktop, and be able to get good quality views/prints from those raw files.

The fact that manufacturers all seem content blazing their own trails with their proprietary undocumented formats has given rise to the Open Raw concept.  The Open Raw website is dedicated to the concept that manufacturers should document their raw formats in order to make them, well, less "raw".  It offers a platform to third party software developers like myself to lobby manufacturers to stop going in different directions and coming up with new undocumented raw formats for each new model camera.  I'd actually like to see this concept taken a step further by lobbying the manufacturers to get together and come up with one internationally accepted raw file format to be used in all future cameras: a sort of raw TIFF format.  While Adobe likes to boast their own DNG format for this purpose, it really cannot work until the cameras themselves start storing data in this format on the actual flash card.  Until then, it's just another file storage format that you have to deal with and one where you'll still need an initial developing cycle to get the data from proprietary to this other "standard".

 

Raw tools
 

Where does this leave us?  Basically it leaves us with files on our camera's flash card that we hope to develop to make photos, and that prompts us to start looking for raw developing tools.  While many utility type programs like thumbnailing or image management programs can "read" raw files, these types of multi-purpose programs generally produce poor quality developed photos.  Most of them are not color managed, produce inaccurate color, and just don't produce very "clean" results as they are prone to artifacts like zipper edges, moire, aliasing, and poor resolving power.  If you use a general utility type tool to develop or print your raw files, you'd probably get better results in most cases just shooting in JPEG capture mode!  Developing raw photos is a tough job that in my opinion should be left to dedicated raw development software.

Most cameras come with raw software that can do a good job developing raw images but manufacturer software can have limited functionality and while it does come from the manufacturer, it still rarely offers the highest possible quality.  In this day of corporate buyouts (I won't mention any names), it can be hard to tell which raw tools will be around for the long haul and which ones might give you a rather short ride for your money.  One of my long lived favorites is Bibble, an advanced, hyper-featured but still easy to use raw tool that has been around since the first consumer level camera started supporting raw captures (the Nikon D1).  Where the generic image utility programs struggle to just let you "see" what is in your raw files, Bibble actually has the horsepower to process them to actually bring out the benefits of the raw format.  So if you find yourself wondering why you have to work so hard to get your raw images to look as good as the JPEG's from the camera, find yourself always having to correct color problems, or just find yourself standing on the street corner with your existing raw tool riding off on another bus all by itself, it may be time to give Bibble a try.

I'm a firm believer that the use of specialized raw developing software is an absolute necessity when developing raw images.  You really need to shoot those raw images, process them in a professional raw developing tool, and then use the processed results in your favorite photo editor and photo printing program if you want to reap the benefits and really see what raw can do for you with respect to quality.  If you shoot in raw mode and then just take whatever your thumbnailing, printing, or image management software gives you, you still benefit from having a copy of your digital negatives but in many cases you probably won't get any better quality than you would just shooting JPEG's.  In fact, you're liable to end up with something that looks worse than a camera JPEG because most generic utility type programs know nothing about your model camera and can do little more than give you a "half baked" rendition of the raw image.  Bottom line: use a quality, dedicated, professional raw development tool to process your raw images and you'll enjoy all the benefits that raw has to offer.  A good rule of thumb is: if the tool you are using to process, view, or print your raw images is designed to do more than just develop the raw images, it probably isn't going to give you stellar results.

 

Summary
 

Hopefully this article has helped those who are thinking about trying out raw capture mode on their camera.  In the "old days" of film, most people wouldn't throw away their negatives once the 4x6 photos were processed.  Similarly, there are advantages to shooting raw and keeping your digital negatives.  Keep in mind that for many casual shooters, JPEG is just fine.  If you are good with the camera and can get consistent and accurate white balance and exposure, the quality benefits of raw shooting can be marginal.  When the one good shot of the bride and groom cutting the cake turns out underexposed though or the white cake is blown out with no detail, raw can be the difference between the recycle bin and a beautiful framed 13x20!  If you do decide to give raw a try, stick with the professional standalone raw developing tools that are specifically designed and dedicated to developing raw photos.  They do the best job by far and generally offer the only way to capitalize on all the benefits of shooting raw.

 

Mike Chaney

3160  Technical Discussions / Articles / June 2006: My Camera, My Color Space on: May 27, 2009, 01:36:15 PM

My Camera, My Color Space


Background

We've covered a lot of ground in previous articles with respect to color management, profiles, and color spaces, but one area that continues to confuse many people is the origin of color management: the color space used by your digital camera.  If you have any interest in color management, that is preserving accurate color from the capture device to the monitor and printer, you probably understand that you need a printer profile to describe how to reproduce color with your printer, paper, and ink and a monitor profile so that you can see accurate color on your monitor.  Too often, however, we forget that the origin of color is just as important as the destination!  If you don't know what color space your camera is using and your camera isn't embedding a color space profile, you can end up with color problems on screen and in print.  Let's take a look at how your digital camera records color and try to get some answers.

 

The origin of color


Cameras make taking photos so easy that few people realize how complex the image capture process really is.  Unlike scanners, which have their own consistent light source, a camera must be able to record the scene under a variety of lighting conditions.  Since the lighting conditions that existed when the photo was taken are likely quite different than the lighting conditions where you'll be viewing the reproduced photo, white point adaption must be performed (white balance) so that our eyes perceive the scene as it was when photographed even though the lighting is now different.

Suffice it to say the camera is doing some number crunching before saving the finished JPEG on the flash card.  For this article, we'll limit our discussion to the JPEG/TIFF shooting mode and won't go into raw processing since most raw processing software both handles the color conversion and stores the proper color space as part of the processed image, thus eliminating the uncertainty of which color space to use for the processed images.  With cameras shooting in JPEG or TIFF capture mode, however, it can be difficult to tell exactly what the camera is using as the color space for the photos.  The camera has "done its thing" and processed the photo, but do you know whether the saved JPEG/TIFF images are in sRGB color space, Adobe RGB color space, or some other color space, and do you know whether or not the data is really accurate for that color space?  If you are not sure, using a super accurate monitor and printer profile won't help you because for the monitor/printer profile to work, you also have to know the color space (which can be specified as a profile as well) for the image itself!

 

What is color accuracy?
 

When I talk about "accuracy" for the purpose of this article, I'm using the term a bit loosely.  Technically, accurate color would be color that is identical to the original scene including the light source that illuminated the scene.  Unfortunately, if you reproduced the original scene with this type of colorimetric accuracy, it may look quite odd both on screen and on paper, because the light source in the room where you are viewing the photos is unlikely to be identical to that of the original scene, the white point of the paper is not likely to match the original scene, etc.  When we talk about accuracy in on-screen or printed photos, we must talk about a subjective type of accuracy in that our eyes perceive the photo to be true in color to the original scene.

If you are looking at a photo of someone you know was wearing a bright red shirt and the shirt looks orange in the photo, or you were looking at a blue sky that printed purple, you would say the color reproduction is "inaccurate".  In general, most complaints about color accuracy will come from hue shifts (colors shifted toward another/different color), saturation problems (colors being too vibrant or too dull) or luminance problems (too bright or too dark) in that order. Fortunately, color management knows how we see and can adapt to different illuminants so that photos still look accurate on our monitor and printer.  Once again though, to do this, we must have an accurate image profile (color space for the camera images), an accurate monitor profile, and an accurate profile for our printer, paper, and ink.

 

What color space does my camera use?
 

As discussed above, some color space (profile) must be assumed for the images created by your camera.  If you are using a monitor and/or printer profile, some color space is being assumed for your images (out of the camera) whether you realize it or not!  Let's make sure we know the assumptions being made by our camera and our imaging software.

Both the JPEG and TIFF image file specification include an option for embedding a profile that describes the color space for the images.  Unfortunately, and for reasons still unclear to me, I don't know of any camera manufacturer who chooses to utilize this feature, so there are almost no cameras that will specifically identify the color space of the image by embedding the profile for that color space even though it would require only about 500 bytes in the file header to do so.  Instead, most manufacturers include the EXIF "color space" tag in the file header, meaning that the color space is "tagged" but not "embedded".

The EXIF data in your photos includes information such as the shutter speed, aperture, flash status, and other shooting parameters, so it is logical to identify the color space via the EXIF information.  Sadly, the EXIF color space tag can only identify the color space if the color space being used is sRGB.  There are only two valid settings for the EXIF color space tag and those are sRGB (a standard color space for PC's and the web) and "uncalibrated".  Basically this means that if your camera is storing images in the sRGB color space, you should be fine since the EXIF color space tag will specify sRGB and your photo software should be able to identify sRGB as the color space of your images.  If your camera is not storing images in the sRGB color space, you really cannot tell what color space is being used by looking at the EXIF information since "uncalibrated" is all the information that will be provided.

 

Where does this leave us?
 

Fortunately, if you use a consumer camera that doesn't give you any menu option to change the color space, it is very likely using sRGB as the color space and this will be recorded in the EXIF header of the image.  These images will open in most photo editors and other imaging software with the proper sRGB color space recognized automatically.

Things can get a bit more complicated, however, if you are using a dSLR or other camera that allows you to switch your color space from sRGB to Adobe RGB.  Shooting in Adobe RGB mode allows you to capture a wider range of colors so those who use high end cameras like dSLR's often change the color space so that the camera uses Adobe RGB.  Once you do this, the EXIF information is changed so that the color space is listed as "uncalibrated".  At this point, some photo editors and other photo related software may start telling you that there is no embedded profile/color space and may ask you what color space to assume when you open the image(s).

Since the vast majority of cameras only offer two options for color space, sRGB or Adobe RGB, if the software you are using tells you that there is no embedded color space and asks you what to use, chances are the answer is Adobe RGB since if the images were in sRGB color space, sRGB would have been explicitly identified in the image file.  I have programmed my own Qimage software with logic that can automatically determine the proper color space to use, but if you are using other photo related software and you are asked about the color space or profile to use when the image is opened, follow these general guidelines:

  1. If you are using an older camera that may not support the latest EXIF data and/or your camera does not offer the ability to change the color space (say from sRGB to Adobe RGB), it is safe to assume sRGB as the color space for your photos.

  2. If you are using a camera that allows you to select sRGB or Adobe RGB as the color space in a setup menu and you are using ICC aware software, you should not be asked about which color space to use if sRGB is selected in the camera and you may or may not be asked if Adobe RGB is selected as this depends on the capability of the software you are using.  If you are asked, you have probably set your camera to Adobe RGB mode, so select Adobe RGB.

  3. If you are not using fully ICC aware (color managed) software, you may never be asked about color space because the software ignores that information, or you may be asked every time if the software is unable to read the EXIF header to determine color space.  In cases like these, use your best judgment.  Again, if you haven't taken any action to specifically change your camera to Adobe RGB color space, it should be safe to assume sRGB is the proper color space to use for your images.

 

I know now what my camera says.  Is it accurate?
 

Here we open up a whole new can of worms.  If we've read the above and we know which color space our camera is using for images it stores on the flash card (most likely either sRGB or Adobe RGB), is color really going to be "accurate" if we assume that color space for our photos?  This is a much more difficult question to answer and the answer depends on many factors such as lighting, white balance accuracy, exposure, and even the lens being used if the camera has interchangeable lenses.  Next comes the fact that the most accurate photos may not be the most pleasing photos to many people.

In reality, many consumer grade cameras offer a simple color shaping matrix that is designed to return pleasing color that results in few complaints from consumers.  Most consumers, for example, prefer a little extra sharpness and pop (contrast) in photos.  They also like green grass to look really green even when in reality it might be a little yellow/brown.  As you begin to move up to high end or dSLR cameras, we see more of a shift toward color accuracy and less of that extra "pop", but there is often still a balance between accuracy and that "wow" factor of a photo that really leaps off the paper.

In the end, if you know what color space was intended for the images, the resulting photos should look good when that color space is used.  Some people often notice small errors like some detail being lost in shadows as darkening shadows is a common technique used to increase contrast and hide image noise/grain.  People often ask whether they should try to create a custom ICC profile for their camera using their favorite profiling package.  Most often, the answer is no.  While some older cameras in the 1-3 megapixel era could benefit from custom profiles just because manufacturers weren't as good at color in those days, you'll most likely only make things worse trying to create a profile for a modern camera shooting in JPEG/TIFF mode.  Custom profiles can be a big help for raw shooting, however, since the profile can be applied in the raw software at a more stable point in the conversion process.  The trouble with trying to create an ICC profile for your camera in JPEG/TIFF shooting mode is that there are far too many adjustments taking place before the profile is applied and you end up shooting at a moving target.

 

Summary
 

The bottom line when dealing with photos from your digital camera is that you must be aware of the color space used for those images since this is the first step in color management and your monitor/printer profiles will not be accurate unless you are assuming the correct input (image) color space.  Unless you have specifically changed a setup menu to specify a color space like Adobe RGB in your camera's options, the camera is most likely storing photos in sRGB color space.  If your camera has a color space option and allows you to change the color space from sRGB to Adobe RGB and you are unsure about the mode used for some of your shots, try using Qimage to determine the color space of your images.  It has built in logic that can determine the proper color space for your camera's photos.  Simply hold your mouse pointer over the thumbnail for an image in question, and Qimage will display the color space assignment on the status bar at the bottom of the main window.  Ensuring the proper color space for your images will enable accurate color rendition from your monitor and printer by virtue of the fact that we have the right starting point.

 

Mike Chaney

3161  Technical Discussions / Articles / May 2006: Test Prints: Getting the "A" Grade on: May 27, 2009, 01:33:52 PM

Test Prints: Getting the "A" Grade


Background

Whether you are an amateur, professional, use color management, or couldn't care less about color management, at some point you may end up printing some test prints in order to evaluate color on a new printer or new type of paper.  There are some good test images floating around on the web that you can use to make test prints on your printer.  What are the pros and cons of each of these test images and what should you be looking for when you evaluate test prints?

 

Testing your printer


Before we look at individual test images, let's first discuss their purpose.  While there are a few test images that allow you to test your printer's resolution or the amount of fine detail visible in prints, nearly all printer test images (sometimes referred to as "targets") are designed to help you evaluate color and not resolution.  The reason for this is pretty simple and stems from the fact that your printer has a well defined set of algorithms that determine the resolution.  Color on the other hand, can be more difficult to dial in, especially if you are using third party paper.

Part of the problem with color matching is the fact that the image you are printing can come from a variety of equipment that uses different methods for encoding color.  Before printing any test images on your printer, first be sure you are using software that is color managed.  Most of the latest photo editing packages are color management aware.  In addition, some high quality photographic printing software packages offer color management as well.  The latest version of my own Qimage printing software, for example, reduces the potential for user errors related to color management mismatches by offering full color management support including methods that allow the software and the printer to communicate with each other to determine how best to handle color even when color profiles are not being used.  Before printing any test images, be sure you are aware of the capabilities and limitations of the software you are using to print and be sure you have that software set up properly.  You can refer to other articles I have written for this purpose.

 

What to look for


It is important to understand that printers have their limitations and that some images are designed to test those limitations.  As such, you may notice problem areas in test images that you will never encounter in "real" photographs.  For example, many test images have wide, sweeping color gradients where much of the colors in the gradient are out of range for the printer.  This forces the printer (and software) to make compromises that can show up in the form of posterization or "blockiness" of color where the test image looks smooth on screen but a bit chunky or warped in print.  One of the most important things to realize when printing test images is to recognize the fact that not all problems seen in printed tests will appear in real photographs.  How many times will you see a full rainbow of colors that covers the entire visible spectrum at full saturation?  In a real photo, probably never.  While these mathematical gradients aren't a realistic test for photographs, they can show strengths and weaknesses in color profiles and they can be a good indicator for the possibility of problems should any of your photos enter the color range represented in the trouble area of the test image.  Such can be the case in circumstances such as when printing sunsets or certain skies that have broad areas of slowly changing color.

Many people make the mistake of discarding a setup that is really quite good because they notice banding in one of the mathematically derived gradients on a test image.  Rather than looking for the extremes in the test image, you should concentrate on overall color rendition, accuracy, and then the gradients in that order.  When judging color, it is difficult to judge skin tones because more than likely you do not know the person in the test photo, their actual skin color, what time of year it is (how good their tan is), etc.  The best you can do for skin tones is to say that they look "good".  Unfortunately "good" is in the eye of the beholder and can vary widely from viewer to viewer.

Since much of the test image may be unknown and therefore hard to judge, it is always good to have a good start: an accurate monitor profile.  One of the most important steps in judging good prints is to have an accurate monitor since that is likely what you will end up using to judge your prints.  Fortunately, monitors often have less problems with color than printers due to their more "linear" nature and monitor profiling tools that include a colorimeter that attaches to the screen are relatively inexpensive and do a nice job.

Here are some things to look for in test prints:

  1. Gray gradients: Look at the areas of the test print that are supposed to be gray (neutral) to ensure that they have no color cast.  This can be difficult to judge due to lighting and the fact that our eyes often adjust to the colors around the gray area, but here we are looking for obvious color casts.  Do the gray areas look gray or to they look like they have a tint of green, magenta, or some other color?

  2. Skin tones: The next step after evaluating (and possibly correcting) neutral tones in the print is to judge skin tones.  Skin tones are usually very lightly saturated so they are the next batch of colors to evaluate after neutral tones.  Rather than judging skin tones against people you know or are familiar with, just make sure the skin tones look natural and that they look reasonable in that you would expect similar tones for a person with the complexion type shown in the photo.

  3. Known objects: Next up on the list are the more saturated colors for objects that are clearly recognizable.  Blue sky, for example, can be a good test.  Does the sky look blue, or does it shift to purple (a common problem with many printers)?  Does a red rose look red or is it shifted toward magenta?  Does grass look green or too yellow?  Objects such as these are usually recognizable enough to determine if your printer is having significant problems in those areas of color.

  4. The extremes: Last, we look at extremes such as black, white, and saturated colors.  Is white really white or can you see little dots in areas that should be pure white?  Is black truly black or does it look too dull, too green, too red, etc.  Can you see areas of shadow (dark colors) on screen that are completely blocked with no detail in the print because they printed too dark?  Are color extremes like bright red, green, blue, yellow, magenta, or cyan blocked or "blown out" in the print?  For example, does a bright red spool of thread show detail on screen but look like one solid block of red in the print?  These types of problems are some things to look for at the extremes.  Again, be aware that test images are often designed to show the biggest problems possible here, so things like highly saturated color gradients are often the "acid test" for printers.  Take them with a grain of salt unless you notice real problems in the photographic parts of the test image (as opposed to the mathematically derived color gradients or rainbow swatches).

 

Some test images to try

1: PhotoDisc Target


Links to above:

Page referencing the PhotoDisc Target
Direct link to PhotoDisc Target

This test image shows a variety of skin tones and colors and is often a good test of printer color accuracy.  This target is not heavy on mathematical gradients or highly saturated colors but it does show some recognizable objects in a well lit scene.  This image uses the Adobe RGB color space so be sure to use color managed software to print this test image.  Use the "ICM" option in your printer driver and set your printing software to allow the printer/driver to manage color if you do not have paper specific profiles that you are using.  Also note that since this test target has many small, detailed objects, it is best to print this target about 10 inches tall if possible.

 

2: Printer Test File


 

Links to above:

Page referencing the Printer Test File
Direct link to Printer Test File

Andrew Rodney's (Digital Dog) test image is another popular printer test image on the web.  It has good gradients for evaluating smoothness of color and a good B/W photo and gray gradients for evaluating gray or neutral colors.  Unfortunately, unless you happen to have a GretagMacbeth ColorChecker chart, this test image isn't exactly chock full of recognizable photographic material.  Still, it is one of the better test images on the web as it doesn't tend to confuse the viewer with slightly off-tone colors or overdone (read impossible to render on the printer) gradients.  This image uses the ColorMatch RGB color space so be sure to use color managed software to print this test image.  Use the "ICM" option in your printer driver and set your printing software to allow the printer/driver to manage color if you do not have paper specific profiles that you are using.

 

3: Fuji Calibration Image


This is an older test image that has made its way around the web in one form or another in the past.  This test image (above) was originally designed as a calibration image to help calibrate color on a Fuji Frontier printer.  While there are some "corrected" versions and other incarnations of this image available on the web, I would not recommend using this test image should you run across it in your search for printer test images.  This image, while it does have some useful gray gradients, can be misleading in numerous ways.  The tablecloth behind the plate, for example, is really a purplish blue that is likely out of gamut on your printer.  Some people "want" to see the tablecloth as blue while it really is supposed to be a shade toward purple.  Some of the color patches on the ColorChecker can also be a bit erroneous in some versions of this test image as well.  The six colors displayed at the lower left of the test image also look like they should be primary colors while they are not, again throwing off the perception of what people expect versus what the image actually shows.  Last but certainly not least, the skin tones in this test image are a bit washed out and not representative of your "average" skin tones.  I won't post a link to this test image because I don't recommend using it and there are so many variations of it, it is hard to tell exactly where it originated.  I post this example just in case you run across it in your travels on the web.

 

4: Granger Rainbow


 

Links to above:

Page referencing the Granger Rainbow
Direct link to Granger Rainbow

The Granger Rainbow is sometimes used by those who need to fine tune color profiles.  It works well for those who are trying to smooth out colors at the extremes in a custom printer profile, but is of very little use to the average user.  Many of the colors in the above rainbow are out of your printer's color range so compromises will have to be made in the print.  These compromises usually amount to reduced overall saturation or color clipping which results in banding.  Even the best printer ICC profiles will have problems with this image and it will almost never print as smooth as it displays on screen.  A print of the above image will always result in either desaturated colors or banding/warping of the color spectrum.  Again, this image can be useful for fine tuning profiles using profile generation tools but it is quite limited for general use as it focuses on problem areas rather than actual photos.

 

Summary

While the above may give you some ideas on generic test images to use for evaluating your printer, paper, or settings, be aware that you are often the best judge of your own work.  Don't hesitate to print some of your own photos showing subjects you are familiar with!  You may have to print more than one photo to be able to evaluate skin tones, bright colors of flowers, greenery, and other objects, but you probably have enough of your own material that spending an hour locating a few good examples of your own work can be helpful after you've dialed in color using a generic test such as one of the test images in this article.  Keep in mind that nearly every test print, especially those with mathematically derived color gradients, will show some of the tradeoffs that are inevitable with photographic printing.  Since your printer may not be able to reproduce all of the highly saturated colors in many color gradients, don't get "stuck" trying to correct banding or other problems if such problems only occur in the non-photographic areas of your test prints.  Always judge the big picture and how well your settings, profiles, and other procedures work on the overall print rather than focusing only on the areas that have trouble.

 

Mike Chaney

3162  Technical Discussions / Articles / April 2006: Houston We Have a [Printing] Problem on: May 27, 2009, 01:29:23 PM

Houston, We Have a [Printing] Problem!


Background

In developing and supporting my photo printing software, Qimage, I often end up helping a lot of people with printing questions and problems that aren't related to any particular software but correspond to broader printing concepts.  In this article, I'll try to cover the most common things that can go wrong with photo printing and will suggest solutions to those problems.  Due to the wide variety of printers, papers, and inks available, it is more common to get "surprises" when printing photos than when displaying them on screen, sending them via e-mail, or performing other operations.  Printing software like Qimage can make the job of printing photos much easier, but some of the variables that affect the appearance of your prints between the software and the paper can be confusing and difficult to identify since they have nothing to do with the software that printed the photos.  Let's take a look at some common stumbling blocks and see if we can take some of the mystery out of them.

 

Wrong colors


First and foremost, make sure your printer is really printing the wrong colors!  If your print driver has a "print preview" option, never trust the colors you see in the preview!  You must print a real print and evaluate it since the colors you see in the print preview (in the print driver) are raw colors intended for the printer and you are viewing these raw colors on screen.  Assuming you have really printed the prints and you have problems with the physical prints on paper, read on...

Probably the most common issue with photo printing is sending your photo(s) to your printer only to discover the colors look wrong, or at least "different" than the way they appear on screen.  To be able to fix such problems requires that you know what the colors should look like!  Color problems can be obvious, like when a blue sweater looks purple, or a bit more elusive such as when a flower looks a little redder in print than the pinkish hue on screen.  When evaluating subjects like flowers, it can be difficult to judge which version is more "accurate" especially when the subject is no longer available.  This is why it is important to make sure your on-screen image is accurate before deciding that the problem is with the printed version.  In some cases, the print may be more accurate than the monitor if you haven't profiled or at least calibrated your monitor.  Monitors are usually easier to calibrate/profile because you don't have to deal with things like different types of paper and lighting.  Making sure that your monitor is displaying the right colors is therefore the first stepping stone to accurate color.  For more info on how to calibrate or profile your monitor, see this article which has some references to monitor profiling software.

Once you have your monitor displaying the proper colors and you know that you have a real problem with printed color, you have some options for addressing printed color(s).  First you must make sure you have selected the proper paper type.  If you are using paper from the manufacturer of the printer, this is usually a simple task of dropping down the "paper" or "media" selection and selecting the paper you are using.  If you are using third party paper, however, you may have to refer to the documentation that came with the paper as far as the proper settings to use for that paper.  If no documentation for your printer model was included in the package of paper, you may have to try several selections for paper type in the driver to find the one that works best!  Normally you would start with the paper that most resembles the paper you are using.  Obviously if you have a glossy paper, limit yourself to the glossy paper selections and if you are using a matte paper, limit yourself to the matte paper selections and so on.  The best way to determine whether or not you've selected the right paper type is to print a colorful test print using different paper type selections (and with print quality set to the highest level in the driver).  If you see unexpected color shifts, graininess, or ink clumping, blotching, or pooling, keep trying until you've found a paper type that minimizes these effects.  If the effects cannot be eliminated, you may have to consider the possibility that the paper you chose is just not compatible with your printer (this is uncommon but not unheard of).

Once you've selected the best paper type for your paper, some color issues may remain.  Unfortunately without downloading or creating an ICC profile specific to your printer, paper, and ink, your only option may be to try some trial-and-error changes in the print driver.  Selecting "File", "Printer Setup" in most applications will take you to your print driver where you'll have access to things like color controls, color spaces, print quality, and other options.  You can try changing color-related options in the driver such as changing from "Color Control" to "sRGB" print mode or vice versa on Epson printers, using different "Color Adjustment" modes on Canon printers, or fiddling with different selections on the "Color" tab on your HP printer.  Each driver puts a different spin on how you control color, so the process for adjusting color is really hit-or-miss using this method.  The only way to ensure accurate color is to have a custom profile made for you or use a profiling tool such as my own Profile Prism software to create a color profile yourself.  Some printers come with ICC profiles for different papers but they are often not specific enough to render truly accurate color on your printer and are often supplied with little or no documentation on how to use them.  Since color management and application of profiles can be a complex issue, here is another article that covers the problems you may have when using color management (printer profiles) if you decide to go that route to get color dialed in.

 

Output is grainy


Another common complaint about printed photographs is that the photos look grainy.  If you are using an inkjet printer, your printer produces photos by spraying a very fine pattern of small dots on the page.  With the latest inkjet printers, these dots are normally so small and so closely spaced that you cannot detect them with the unaided eye.  Unless there is a problem that caused noise or grain in the original photo (such as using a high ISO speed), visible grain in prints can normally be traced to either a problem with the media or improper driver settings.  First be sure you are using the appropriate paper type selection for the paper you are using (see "Wrong colors" section above).  Some papers do not work well with all printers and many times when photos look too grainy, you may have simply gone too far when looking for "cheap" paper or you may be using a paper that simply doesn't work well with your printer.  Don't take "compatible with all inkjet printers" too seriously when reading the paper package as being "compatible" doesn't necessarily mean it will work well with your printer!  When all other variables like paper type and quality settings have been ruled out and you still have trouble, consider trying another paper like one of the papers made by the same manufacturer as your printer.  Often times just switching paper will cure the grainy print problem.

Next check your print driver settings using "File", "Printer Setup" and make sure you have the highest quality selected.  This may entail selecting "Best Photo" quality, selecting the highest resolution number, and/or sliding a quality slider to the highest quality.  You want to stay away from options like "Draft", "Text", "Standard" or "Normal" as those options usually indicate that you have not selected the best print quality.  Sometimes just going through the quality options in your print driver and manually selecting the highest quality options will reduce graininess in printed photos if your print driver tends to default to a more "mediocre" quality.

Also be aware that some people are more sensitive to grainy prints than others.  I've seen some prints that wouldn't even pass for photos in my opinion (I call them "printouts" rather than "prints"), and it always amazes me that some people can't see the grain in the same print!  A lot depends on eyesight (near vision), lighting (grain is more noticeable in very bright light) and other factors.  Finally, keep in mind that some portable (mostly 4x6) printers and printers that can print without a computer may not use the highest print quality available if you just insert the memory card from your camera and print directly.  I've found some of these "standalone" printers to be grainy when printing directly from the card but substantially better if connected to a computer where you can override the defaults in the print driver and select higher quality output settings.  Few if any of these standalone printers are set up to output at the highest quality when printing directly from memory cards.

 

Banding


Banding is a general term for any type of aberrant straight lines that appear on your photos whether they are vertical or horizontal.  There are actually many causes for banding but the most common are clogged nozzles and poor print head alignment.  The print head on an inkjet printer contains dozens, hundreds, sometimes thousands of tiny holes where the ink is sprayed onto the page.  If even one of these little nozzles becomes clogged, banding can occur.  Most print drivers offer a "nozzle check" option where you can see if any of your nozzles are clogged and a "clean" option where the nozzles are cleaned with a high volume ink spraying cycle.  Another cause for banding can be print head alignment.  If you've done a nozzle check and you have determined that no nozzles are clogged, you might want to check your print driver to see if it has an option for "print head alignment".  If it does, you may have to go through a process for this manually or your printer may have an automatic option for head alignment.  If the nozzle check looked good and the print head alignment didn't fix the problem, I'd recommend going through at least one cleaning cycle anyway.  Believe it or not, clogged nozzles sometimes don't show on a nozzle check and a single cleaning cycle may correct the problem.

Yet another cause for banding is the use of third party inks that have slightly out-of-spec flow rates or viscosity.  I've seen more than a few cases where banding was caused by problems with third party inks.  Sometimes the ink will flow without a problem for a while and then banding will occur at the bottom of the page, right side of the page, or even only in areas of thick, saturated color.  In cases like these, the ink may be having trouble flowing at a rate high enough to keep up with the demand at all times.  Ink flow problems and physical problems with the print head can be very difficult to diagnose.  They can occur at strange times, in strange areas of the print, and the banding can even be worse when using software that optimizes print quality because lower quality printing software may hide the problem by not taxing the printer to its limits.  In any case, when banding occurs, look for a physical problem with the printer or ink itself even if the problem seems to only appear in strange places and/or only appears when using certain software or driver settings. The printing software is almost never the cause of banding as banding is usually a sign of some type of hardware failure.

 

Partial prints, all-back, or no prints


Another common printing problem is finding out that not all of your prints printed, pieces of the prints are missing, or some/all of your prints printed as solid black rectangles.  All of these problems are caused by your print driver not being able to handle the amount of data being sent to the driver.  Some print jobs, particularly those that have been optimized for quality or those that contain large prints or many smaller prints, put a heavier burden on the print driver.  For this reason, you must ensure that your print driver is set up to handle larger jobs.  Two very common solutions are:

  1. Go to control panel, printers and faxes, and right click on your printer and select "Properties".  On the "Advanced" tab, make sure "Enable Advanced Printing Features" is not checked!  This option, simply put, was never meant for photographs and does not work properly when printing large photos or printing many photos to large pages.  Second, press the "Print Processor" button and make sure the spool type is set to "RAW" with no other options.  This tip alone accounts for over 90% of problems with missing prints or pieces of prints.

  2. If you still have problems and/or you have problems only when printing larger prints, check the "effects" type options in your print driver.  Some drivers have trouble with effects when dealing with large prints because they simply cannot handle applying those effects to larger prints.  HP printers for some reason are more prone to this problem than others.  Check the "HP Digital Photography" area and make sure that none of the effects are checked, particularly any that have the word "auto" in them.  These options often get "confused" when doing large or optimized printing.

For additional tips when printing large photos or large jobs, see this article.

 

Prints are the wrong size


If you ever print a specific size such as 4x6 and you don't get what you expect, first check the paper itself.  Did you print borderless on 4x6 paper and the print was either cropped more than expected or isn't big enough for the paper?  First check the size of the paper with a ruler.  Sometimes the paper you bought isn't exactly 4x6 (or 5x7, 8x10, etc.)!  It seems silly, but I've seen this happen more than a few times.  If you print a specific size and only one side is the correct size, you may need to crop some of the image to get the size you want.  Some software will do this for you with the click of a button while others make you do this cropping step manually.  If you print a 10 x 8 for example and your print ends up 10 x 6.67 where only one side is the wrong size, you may just need to crop because the aspect ratio of the image doesn't match the print.

If you sent one size to the printer and it ended up printing proportionally larger or smaller (in both dimensions), it may be due to an option in the print driver.  Be aware that borderless printing often results in size expansion so if you are printing in borderless mode, your prints will actually be expanded so that they print slightly larger than the size specified.  With borderless printing turned on in the driver for example, it isn't uncommon to send a 4x6 print to the driver only to get a print that is 4.2 x 6.2 inches.  This can sometimes be controlled with the "amount of extension" option in the driver but cannot always be turned off completely unless your printing software (such as Qimage) offers an override.  Options like "fit to page", "enlarge" or "reduce" also will modify the print size without you being aware of the change.  Here is an article that deals specifically with potential print sizing problems.

 

Prints fade quickly


Longevity/permanence of prints is a widely studied, complex, and often misunderstood topic.  There are sites on the web where you can find longevity data and they might have even tested your particular model printer and the paper and ink you use.  The problem with most of these tests is that there are a lot of factors involved and few of these sites tell you about what affects longevity of prints.  These longevity testing sites can and do give you information about which papers are better for ink permanence but they often don't test many combinations.  Generally, if you have a "normal" inkjet printer that isn't labeled "lightfast", "archival", or "pigment", you probably have a printer that is using dye based inks.  Dye based inks tend to fade rather quickly if the print is exposed to (in direct contact with) circulating room air and/or is in a room with bright lighting such as direct sunlight.  Your average inkjet printer just wasn't made for archival purposes so if you put some prints on your refrigerator and the sun swipes over them every morning, don't expect them to last long!  It's as simple as that.  Outgassing and exposure to light can cause the fading so if you mount your photos behind glass (like in a photo frame) and not in direct sunlight, they will last years/decades.  Take them out and expose them to the environment and you are asking for trouble.

Pigment printers are normally called archival or lightfast and use pigment inks as opposed to the faster fading dye inks.  Prints produced by pigment/archival printers will last much longer than prints from a dye based printer when exposed to the environment but even these archival prints will do better behind glass.  As a general rule, if you want your prints to last, put them behind glass or in a protective sleeve like the ones found in most quality scrapbooks.  And if you are really into scrapbooking and you want to know if your prints from your inkjet printer are acid/lignen free, the answer is no!  Even if you buy acid free paper, the ink itself will not be acid free.  In addition, normal photographic prints that you get from a photo developing outfit are far from acid free and they aren't known to cause major problems with scrapbooking, so there's really no need to overdo it trying to fool with acid free inkjet papers.  Of course, if you are a purist, I'm sure you could find acid/lignen free paper for about three times the cost of your regular inkjet photo paper but it probably won't look as good and will it really make a noticeable difference in the long run?  I'd say no, but that's just my view.

 

Summary

Hopefully this article will be a good cheat sheet for those having trouble printing their digital photos.  My goal was to try to help show the way for people who get unwanted surprises when they try to print their photos.  The photo-to-printer path is actually quite complex and there are a lot of potholes in the road to getting good digital photos.  You may get lucky and your prints may be acceptable the first time you print.  Certainly that was the intent of the manufacturers: for you to plug the printer in and print with no issues.  With the variety of printers, drivers, paper, and inks on the market, however, it is more likely that at some point you'll run into at least one of the problems listed on this page.  As you move forward and get more involved with your photography over time, you may find that what was acceptable for your first prints isn't that good any more.  You'll tend to be more discriminating the more you print.  Such is life in any hobby or profession.  I hope this page can be of some help no matter where you are on that road.

 

Mike Chaney

3163  Technical Discussions / Articles / March 2006: My Printer Profiles Aren't Working on: May 27, 2009, 01:25:07 PM

My Printer Profiles Aren't Working!


Background

I've noticed a significant increase in complaints recently about printer profiles not working properly.  Most inquiries relate to people trying to use printer ICC profiles and then complaining of green/magenta color casts, prints that are too dark, or other color issues when using printer profiles supplied with the latest printers.  Those familiar with the concept of color management and ICC profiles might even ask "Am I double profiling"?  In most cases, people end up turning off color management and just using the automatic mode in the print driver, claiming that they get much better pictures that way.  This leads many people to simply give up on color management, convinced that it must be something they don't need.  I've done a number of articles on color management but I thought this topic needed revisiting as it seems to be causing a lot of confusion recently.

 

Am I double profiling?

As a general rule of thumb, if you know enough to ask whether or not you are double profiling, you probably aren't!  There are two places that ICC profiles can be applied when printing: inside your print driver and within the software you are using to print.  Double profiling occurs when you activate the "ICM" or "ICC" option in the print driver and you also specify a printer ICC profile in the software you are using to print your photos.  The bottom line is that if you use a printer ICC profile, you need to select color management in one place only: the print driver or the printing software, not both.

The most common symptom of double profiling is overblown or garish colors with a color cast (usually magenta or green) in gray areas.  Since most profiles lighten up prints a bit, double profiling also causes most prints to appear too bright, so if your complaint is that your prints are too dark, you most likely are not double profiling but you may be using the wrong profile or the right profile in the wrong place (see below).

Many people see printer profile assignments under the properties for their printer in control panel and assume that if a profile is assigned there and they also specify color management in the driver or in the printing software, that they will be double profiling.  Such is not the case, however, because the profiles you see associated with your printer under the printer properties are simply there as placeholders so that your driver knows which profiles to use when you check the ICC/ICM option in the driver.  Do not remove these associations under "Printers and Faxes" and "Properties" for your printer.  Any profiles listed there should remain there and the presence of those profiles will not lead to double profiling.

The two methods of applying a printer profile:

  1. The first method (and the recommended method) of applying a printer profile is to identify the ICC profile to be used within the software you use to print.  Of course, for you to be able to use this method, the software you are using must be "ICC aware" in that it knows how to apply profiles.  Software such as PhotoShop, Paint Shop Pro, Qimage, and other popular editing/printing software will allow you to do this.  When using this method, you must make sure that the ICC/ICM option is not being used in the print driver.  Usually, this means selecting the "no color adjustment" or "none" print option in the driver in the color management section.  By letting your printing software handle the color management and turning it off in the driver, the printing software can read the color space of the image and convert to the printer ICC profile that you specified in your printing software for true start-to-finish color management.

  2. The second (and usually less desirable method) is to specify the "ICC/ICM" mode in the print driver and not do any color management in your photo editor or printing software.  With color management enabled via the print driver, you must not enable it in your printing software.  In PhotoShop, you would select "printer color management" and in Qimage you would select "sRGB" as your printer profile.  This will cause your printing software to send the image to the driver in a "standard" color space, allowing the print driver to do all the work with respect to profiling.  When using this method, it is often not necessary to specify which printer profile to use because the driver will pick the right profile based on the paper type selected and other factors.

 

Which method should I use?
 

The printer profile(s) supplied with your printer (or downloaded from the manufacturer's site) can be designed for use either via your printing software or inside the driver.  Some profiles simply were not designed to be used outside the driver and will only work properly when used in the driver via checking "ICC/ICM" in the driver.  Unfortunately very little (if any) documentation is supplied with these generic profiles so it can be very difficult to tell exactly how the manufacturer intended for them to be used.  In general, unless documentation is supplied with the profile telling you exactly what settings to use in the driver, it is probably only meant to be used inside the driver and you may get unexpected results when you turn off color management in the driver and try to use these profiles in your printing software.  Taking a generic printer profile that comes with your printer and trying to use it outside the driver accounts for the reason for unexpected results in many cases.

When you have someone develop a custom profile for you or you use profiling software to create your own profile, you'll want to disable color management in the driver and let your printing software handle the profiling.  This is by far the best way to get accurate profiles since you are in control of the driver settings when the profile is created.  When you use a pre-made profile like those supplied with many printers, you often have no idea what settings such as print quality, halftoning options, etc. were used when the profile was made, making those profiles next to useless when trying to use them outside the driver.  Although some of the latest Epson profiles like the ones that show file names starting with "SP" and some of the Atkinson profiles are designed to be used in printing software with color management turned off in the driver, they often don't specify settings well enough to know exactly how the profile was developed.  In cases like these, it is often best to try the profile in your printing software but if it doesn't seem to work well, just try activating color management in the driver and deactivating it in your printing software.

If this trial-and-error method seems a bit imprecise, it is.  Not knowing how to properly utilize supplied printer profiles due to lack of documentation is the number one cause of complaints about printer profiles!  Supplying printer ICC profiles (.icm or .icc files) with a printer installation with no included documentation on how to use them is like supplying an owner's manual for a car with a couple of extra screws and no documentation about them!  Since lack of documentation automatically means that you probably don't have enough information to determine how to try using the profile(s), my best advice would be to try disabling color management in the driver and use the profile in your printing software first.  To do this, you must know at least how to set the paper type and print quality in the driver.  If you know that much, try setting the paper type and print quality in the driver, turning color management off in the driver, and selecting the profile in your printing software.  If that doesn't seem to work or leads to unexpected color or tonality, try the opposite: turn off color management in your printing software and just check the "ICC/ICM" option in the driver.  Without any documentation specifying how a profile should be used, trying both methods is your only option.

 

Getting it right from the get-go


By now it should be obvious that in many cases, the cause of problems related to use of printer profiles is that printer profiles are often supplied with no documentation as to how/when they should be used.  Some manufacturers even supply generic "matrix shaper" profiles that are not even real printer profiles.  Older Canon "CNBJPRN..." ICC profiles and Epson "EE_..." ICC profiles are examples of profiles that should be avoided because they are simply too generic to be accurate and were not designed to be used outside the print driver.  One sure way around these problems, and one way to be sure you get profiles that will work for you is to have custom profiles made for your printer, paper, and ink using a service or profiling software that you use yourself to create those profiles.  This is really the way color management and ICC profiles work best.

There are many options available and these can be found by using Google to search for "printer profiling" or "custom profiling" but I'll reference two of my favorites.  If you want to print a test target, send it in, and have a custom profile generated for you, an outfit like Cathy's Profiles is a good place to try.  If you'd like to try your hand at creating your own profiles, a tool like my own Profile Prism is a low cost solution that consistently gets good reviews as well.  Whatever method you choose, creating profiles that are designed for use on your own equipment and that you know how to properly utilize (because you specify the settings to use) has many benefits.  The immediate benefit, of course, is that you get profile(s) that you know how to use properly and that you know are designed to be used with the equipment and media you are using.

You may have questions about some of the references in this article as far as the usage of print drivers and printing software.  For more information on how to apply profiles or how color management works in general, be sure to check some of my previous articles on the subject of color management:

August 2004 - Over the Gamut and Through the Woods
February 2005 - Color Management in a Nutshell
May 2005 - Using ICC Profiles with Epson Printers
June 2005 - Using ICC Profiles with Canon Printers
July 2005 - Understanding Rendering Intents
September 2005 - Soft Proofing Basics
December 2005 - Lighting, Viewing, and Metamerism

 

Summary

There are many causes for complaints about printer profiles not working properly.  Many result from trying to use "generic" printer profiles that are supplied with a new printer or are downloaded from the manufacturer's web site.  In general, if you don't have enough information/documentation to tell you exactly how to set up the print driver in order to use these profiles in your printing software, it may be best to let the driver handle color management, at least until you are able to create some custom profiles of your own that you know will work properly for your specific equipment and media.  Double profiling should not be a problem unless you activate the color management (ICC/ICM) mode in the print driver and you also specify an ICC profile in your printing software.  If you avoid that situation and you still have trouble with an ICC profile, chances are the profile is simply not meant to be used in the way you are using it and you may not have enough documentation to correct the problem.  The solution is to create your own custom profiles.  That way you can be sure what they are for and how they are to be utilized.

 

Mike Chaney

3164  Technical Discussions / Articles / February 2006: Wide Load: Tips for Printing Large on: May 27, 2009, 01:19:45 PM

Wide Load: Tips for Printing LARGE


Background

Wide format printers are getting cheaper and more economical, making large prints like 13x20, 20x30, and larger possible even for the advanced amateur.  Logic dictates that a wide format inkjet printer is nothing more than a big printer and that printing to a wide format printer should be no different than printing to your typical desktop inkjet except that it allows you to work with larger page sizes.  Unfortunately there are a lot of snags that you may encounter when printing (very) large prints.  Most involve printer/driver setup and are easily corrected while others may require a rethink on how you have your equipment connected, what type of equipment you are using to print, whether or not your current hard drive has the capacity for wide format printing, whether or not you need more RAM, etc.  Let's take a look at printing large and we'll try to cover all the common missteps in the minefield.  We'll keep both Qimage and PhotoShop in mind for this article as those are popular PC/Windows printing applications and are the applications that I deal with most in wide format printing.

 

Page Size


Wide format printers often handle paper/page size differently than your average inkjet, particularly with respect to borderless or "no margins" printing.  First, it is important to realize that there are two methods that drivers use to perform borderless printing: expand page and expand prints. In the expand page mode, the driver simply increases the size of the page so that it is larger than the physical paper size. In this borderless printing mode, the driver will actually show a printable area larger than the physical paper size.  For example, a 16 inch wide roll may show as 16.23 inches across in your printing software. What is actually happening is that the driver is printing approximately .12 inches off the left and right edges of the roll. With this method, it is important to print your large prints in the center of the page.  For example,  use "centered" or "optimal/spaced" in Qimage.  This will ensure that the "overspray" that extends off the edges of the paper is minimized and you won't lose the edges of the print because they are printing up against one edge or the other.  If you use something like "compact" or "optimal", Qimage will place the print at the left edge of the paper and .12 inches of the print will be off the page, cropping the print slightly. Epson calls this expand page mode "Retain Size" in their latest wide format drivers. In the older (7600/9600) drivers, this was the only option available so no options were visible.  What you need to keep in mind with the expand page mode is that the driver expands the page size so that it extends slightly beyond the edges of the paper.  In doing so, it is possible to print off the edge of the paper and lose some of your prints.  Try to avoid this "clipping" by not printing anything all the way against the left/right edges of the paper when aligning your prints on the page preview on screen.

The other method of borderless printing (one that is used on most standard inkjet printers) is the expand prints mode. In this mode, the printable area remains the size of the paper (16.0 inches across for example), but prints are expanded in size. With this mode, you can specify 16 inches as the width and the driver will "artificially" expand the print to 16.23 inches so that the print is large enough for some overspray to the left/right: the overspray eliminates tiny slivers of unprinted paper at the edges due to slight misalignment of the paper. This mode is more common but often more confusing because every print you send will be slightly larger than what you specified. Even if you print 4x6 prints on a 16x20 page, the 4x6 prints will be just slightly larger than 4x6 and this may confuse people or prompt them to blame the printing software for the size problem when in fact it is the driver that took the 4x6 print and expanded it after the fact. If the expand prints mode (called "Auto Expand" by Epson in their latest wide format drivers) is being used, Qimage does have an option that can defeat the size expansion so that you can obtain prints of the specified size without the driver expanding them. See "Page Formatting", "Borderless Overspray/Expansion" in Qimage.  Some drivers even allow you to turn this expansion off (or at least reduce it) by dragging the "amount of extension" slider all the way to the left in the driver.  Keep in mind that doing this produces less overspray so any "sloppiness" in the paper loading mechanism may show up as unprinted slivers on the edges of the print.

 

Understanding software differences WRT sizing

All software including PhotoShop and Qimage must work within the
limitations of the printer which are defined by the driver. If you specify an impossible task, like printing a 16x20 print on 16x20 paper without using borderless printing, different software may handle the request differently. For example, if you don't specify borderless printing, the maximum size print allowable on 16x20 paper using the Epson 4800 is 15.766 x 19.333 inches. If you try to print a 16x20 using PhotoShop, you will be told that the print size is larger than allowed but you will be given a "Proceed" option. If you proceed, PhotoShop will print at 16x20 but will clip the edges of the print and you'll end up with a 15.766 x 19.333 inch print that has the edges missing. In Qimage, you will be told that the print size is larger than one page and will be asked if you want a poster. If you say no, you'll end up with a 15.766 x 19.333 inch print (same size as PhotoShop) but without the edges cropped off. These are just two different ways of handling the same problem and in both cases you end up with (no more than) a 15.766 x 19.333 inch print: a printer/driver limitation.  It is important to recognize how different programs handle sizing tasks and in particular, what happens when you try to print sizes that do not fit on the paper.  Whatever printing software you use, be familiar with how it handles sizing discrepancies.

 

Spooling options

Qimage will almost always send more (potentially much more) data to the driver than PhotoShop or other printing programs due to Qimage's interpolation process. As such, you must make sure that the printer is set up properly for large format printing. Not having the printer/spooler set up properly may result in partial prints, no print at all, or crashes due to the system not being able to handle the [large] amount of data being handed to the driver.  First and foremost, go to control panel, select "printers and faxes", and right click on your printer. Select "Properties" from the right click menu and then click the "Advanced" tab. If "Enable Advanced Printing Features" is checked at the bottom, UNcheck this option. This is the cause for 98% of printing troubles when printing large prints as this feature can only handle a small amount of data and isn't meant for photographic printing so the option should remain UNchecked.  The other options on that tab usually make little difference but I recommend checking "Spool print documents so program finishes printing faster" and also "Start printing immediately". Those options will ensure the best use of resources on the machine. Finally, click the "Print Processor" button and make sure that the right side is set to "RAW". If any other data type is selected, it is likely your photographic printing will not work properly. Click "OK" to save the changes.

 

Maximum print sizes

Some online sources report that the maximum print length in PhotoShop CS2 is about 90 inches. I have not confirmed this, but I can tell you that when set up properly, Qimage has no length limit. PhotoShop and most other applications that print photos try to send the image all-at-once to the driver: they basically hand off the entire original image at once and simply specify a print size for that image. Depending on the initial image size and specified print size, this all-at-once printing method can overwhelm the driver/spooler and lockups/crashes may ensue. Qimage has a "smart" handoff to the driver that passes the data in smaller chunks that don't overwhelm the system and can allow for much larger prints. With Qimage, you are only bound by the amount of RAM, virtual memory, and hard drive space available and by how well your print driver handles the printing task when dealing with the large amount of data normally used for big prints. Most of the big print failures that I've seen fall into three categories:

(1) Printer is connected through a network. I have not yet seen a reliable setup when printing through a network and working with large prints. My advice here is just don't do it! In fact, if your wide format printer is connected via an Ethernet cable, switch to USB and print directly to the printer via a local machine connected directly to the printer. There are dozens of complications that arise when trying to send large amounts of print data across a network so if at all possible, print from a machine that is directly connected to the wide format printer as a local printer. You will avoid a lot of hassles this way!  It may be possible, depending on many factors, to have a reliable network printing setup to a wide format printer, but the complications are so diverse and varied that I don't dare get into that here.  When dealing with "wide loads", it is best to avoid network connections altogether!

(2) Don't upsample originals. I've seen people scan 8x10 photographs at 2400 PPI ending up with a 1.4 GB file thinking they want the most resolution possible for printing large. In this case, the original photo only holds maybe 300 PPI of real information so scanning the photo at 600 PPI and then letting Qimage handle the interpolation makes more sense (and often produces better results). Your system and the driver will have enough to do processing your 10 foot long print, so don't hand it a 1-2 GB file unless you truly have enough resolution in the original to support it.  If you have enough pixels in the original to support the resolution of the final image (like a montage or panorama using a dozen photos from a dSLR), your original images have reason to be big, but don't "oversample" lower resolution photos or "overscan" media at ridiculous PPI as this may do nothing but hurt you in the long run!  Taking a 30MB original, for example, and resampling it to 400MB might make sense if you plan to print from PhotoShop, but if you are printing with Qimage, do not upsample that 30MB image because Qimage will do all the upsampling at print time very efficiently and with much less resources (RAM and hard drive space) if you simply print the original 30MB image!

(3) When using a photo editor or other software to prepare a final image for print, use a less proprietary and more internationally accepted standard like the TIFF format or even the JPEG format for the final image to be printed.  Other formats such as Adobe's PSD format often have more overhead and put more stress on your system, not to mention that the public spec for such formats is often far behind what is used in the latest version of the software that creates those files.  You may often be dealing with very large images when printing large prints.  To decrease the overall resource requirements for the job and make the whole process go smoother, use a standard format like an 8 bit/channel TIFF file with no alpha channels and no layers! All print drivers are 8 bits/channel so there is rarely any need to carry 16 bit/channel through to the final print-ready image as it is just going to end up getting converted back to 8 bits/channel anyway for the print driver.  Using PSD or layered TIFF's can put more strain on memory resources and may cause longer print times or even an occasional crash as the system tries to read a 400 MB image and print the 3.7 GB of data needed for a 720 PPI 40x60 inch print.  For printing large, I recommend 2 GB of RAM with both the minimum and maximum virtual memory set to 4 GB.  This should avoid most disk swapping unless your originals are extremely large.

(4) You can never have too much free hard drive space when printing large prints!  Printing a 44 inch wide print that is several feet long can take 5 GB (yes, gigabytes) of hard drive space or more.  As a general/safe rule of thumb, try to keep 10 GB free on the drive where your print driver spools data.  If you print with Qimage, Qimage will not need much hard drive space to process the job but it is passing a lot of data to the driver and the driver will in turn cache that data to disk while it is spooling.  Due to Qimage's high quality interpolation, it will almost always send more data to the driver than your average printing program, so don't assume that you have enough drive space just because you were able to print a print through some other software.  Qimage rarely has a problem processing the job and will finish its printing task, but after the printing task is over (sometimes before), I've seen the print driver itself crash even when several gigabytes of free space remain on the drive so don't be fooled into thinking it isn't a drive space problem just because you have a few gigabytes free on the drive!

(5) I have assisted professionals with the above tips and have printed prints as large as 44 inches wide by as much as 10 feet long (44 x 120 inches) with no problem using Qimage.  When printing super large prints like this with Qimage, however, I would recommend setting Qimage to interpolate no higher than 360 PPI for the final print.  That means that if the page resolution (current driver base resolution) shown above the preview page on Qimage's main window shows 720 x 720 PPI or 600 x 600 PPI, set your interpolation levels to "High".  If the resolution shown above the preview page is 360 x 360 or 300 x 300, set your interpolation level to "Max".  When printing huge banners, you rarely need the maximum 600 x 600 or 720 x 720 offered by the driver, so setting Qimage's interpolation level to "High" instead of "Max" will cause it to interpolate to 1/2 the listed PPI.

When printing prints larger than about 20x30 inches:
Qimage's preview page shows Set interpolation levels to
MORE than 720 x 720 Med
600 x 600 up to 720 x 720 High
Below 600 x 600 Max

Use the above table as a good rule of thumb for printing prints larger than 20x30 inches to avoid system overload.  Under 20x30 inches: just keep the interpolation levels set to "Max".  While there should be no problem printing at "Max" interpolation level in Qimage well beyond this arbitrary 20x30 size on a capable machine, it will take longer to process and will increase the strain on the entire system (particularly with respect to hard drive space).

 

Summary

Hopefully the above tips will help clear up some of the confusion and questions being tossed around from people printing in wide format, particularly when using the latest Epson wide format printers which have some new options.  I print wide format myself and have helped others like local camera store employees who printed 44 inch wide prints 10+ feet long from Qimage to display on the front of their store using the above tips. Most of the issues with printing large involve just setting up the equipment, system, and driver properly. I occasionally run into something out of the ordinary, but most of the time the information on this page is all you need to resolve any wide format printing problems even when they occur in software other than my own Qimage.

 

Mike Chaney

3165  Technical Discussions / Articles / January 2006: Interpolation: Magical or Mythical on: May 27, 2009, 01:16:37 PM

Interpolation: Magical or Mythical?


Background

Years ago, when most of us were taking photos using cameras with 1-3 MP (megapixel) resolution, interpolation or "upsampling" was a hot topic.  To get decent photos at larger sizes of 8x10 and beyond, the ability to upsample photos seemed more of a necessity than an option.  Don't do it and you might end up with jagged edges.  Do it and it would smooth over the jaggies to make the photo a bit softer but without the pixelization artifacts that made the photo look more like a bad video capture than a good photo.  Fast forward to present time.  With cameras approaching and soon surpassing the 8-10 MP mark, is there really much call for interpolation?  How important is it and what does the best job?  It seems that specialized interpolation software and plugins have lost little steam and people are still spending $200 on packages that claim to do the best job adding pixels.  Do you really need these expensive solutions?  How much better do they do than your average photo editor?  Let's take a look.
 

 

The problem

Interpolation attempts to hide a problem that can be described simply as not having enough pixels for the amount of space where they are displayed.  The effect is similar to walking too close to your TV.  Get too close and you start to be able to see the individual pixels and these are distracting when you are trying to see the overall picture.  The same occurs when you take a limited number of pixels and try to "stretch them out" over a large area.

Let's look at a crop from a larger picture:

This tiny crop looks pretty good.  We can tell that it is the wheel of a car and we don't notice anything strange about it.  Take the exact same image, however, and display it larger (4x) and we get:

Now we can see that we simply don't have enough pixels for this larger size: the spokes on the wheel look more like saws than straight lines and the outline of the chrome part of the wheel looks jagged and not smooth.

 

The solution

To get rid of the visually distracting pixelization in the above larger image, we can use interpolation methods to add pixels to the image.  The pixels in the original (smaller photo) describe the data that we have to work with, so interpolation cannot add any true data to the image, but it can smooth over some of the rough edges and can add "apparent detail" by predicting what should appear between pixels in the original image.  Look at interpolation like making a prediction.  If I showed you the sequence A C E G, you could make a logical assumption and fill in the missing letters to get ABCDEFG.  Are B, D, and F really the missing letters though?  You were thinking of the alphabet, when the missing letters could have really been from a person's name: A CHENG.  This just goes to show that you can only "guess" so much information when you are missing a significant portion of that information.

What does interpolation do to the above large/pixelated image?

Without interpolation
PhotoShop "bicubic smoother" interpolation
Qimage "pyramid sharper" interpolation

The top image shows what the photo would look like at the 4x expanded size without interpolation.  By using interpolation, we are able to smooth out the distracting jagged look (center and bottom photo) and improve the overall appearance of the photo.  Note that in doing so, we've reduced or eliminated the coarse look of the image but the image now looks a bit soft (blurry).  This is a necessary tradeoff, since there is simply not enough data from the original to determine which edges should be sharp and which edges might be slightly out of focus due to depth of field, lens distortions, etc.  Older methods such as the bicubic methods used in PhotoShop tend to do a good job while more advanced methods like fractal resampling, edge directed resampling, or the pyramid resampling method available in Qimage (bottom image above) tend to do even better by further reducing jagged edges to produce an even smoother result.

 

Understanding the tradeoffs

The above is a 4x upsample which is considered fairly "radical".  The truth is that if you have a recent model digital camera, you will probably never need to resample to the degree shown above.  When you print your photos, a slight upsample or downsample may be needed, but you'll rarely ever need a drastic change in resolution to get a good print unless you do extreme crops or billboard size printing.  The most important thing is to use a good interpolation algorithm to interpolate to the PPI (pixels per inch) used by your printer, or an integer multiple thereof.  Some print drivers don't do such a great job of interpolation so if you send them an "oddball" size by just printing the original, you may end up with prints that have jagged edges.  This can be true even if you send the printer too many pixels, as some drivers don't even handle downsampling well!  One example showing the problem can be seen in Imaging Resource's review of the Olympus P400 dye sub printer.  Notice near the end of the page how the 400 PPI image looks much worse (more jagged) than the image that was downsampled to 314 PPI (the PPI of the printer) first.  This illustrates the importance of being able to resample to the PPI used by the printer prior to sending images to the print driver.

We can see some of the benefits and tradeoffs of upsampling in the above samples, but downsampling is just as important.  When we take an image consisting of concentric circles of increasing frequency and downsample that image with an appropriate amount of antialiasing, we get the following result showing a single set of rings emanating from the center:

Qimage default downsampling (includes antialiasing step)

If we take the same image and downsample using a standard downsampling routine without first performing the antialiasing step, we get:

PhotoShop bicubic sharper downsampling (does not include antialiasing)

As can be seen in this second example, lack of antialiasing has caused extra patterns to appear that were not in the original image.  While at first it may look like the second example has more "detail", in fact the extra detail is nothing more than artifacts caused by the resampling algorithm trying to consider data beyond the frequency limit.

 

A balanced approach

A well balanced interpolator will be able to downsample without aliasing artifacts while also being able to upsample without jaggies or over-softening the image.  Old tried and true methods such as bicubic or lanczos are usually good enough for most upsampling needs.  More advanced methods can increase visual quality for very large prints or special jobs, but be aware that there is only so much detail you can "add" to an image.  Some of the newer interpolation methods try to make all edges as sharp as possible and while these methods can make upsampled results appear sharper, they often tend to break the correlation between sharpness and depth of field and can make results look a bit like fingerpaintings.  Methods that produce smooth (jaggy free) images but don't try to sharpen, on the other hand, can appear a bit too blurry.  As with many things, there are tradeoffs to each method.

The key to the best results with any interpolation method is often to pick the appropriate amount of sharpening.  Interpolation methods that produce softer results can often handle much more sharpening before showing any artifacts, so a touch of extra sharpening can correct that soft look.  Similarly, a slight edge blur can remove that "painterly" feel of some of the sharper interpolation methods if needed.  Generally the more you stretch an image (the more interpolation you use), the more sharpening will be needed to compensate.  "Smart printing" tools like Qimage and some PhotoShop print sharpening plugins take all this into account and can automatically apply the proper amount of final sharpening based on the resolution of the original, the size of the final print, the resolution of your printer, and other factors to allow the print to be the most visually consistent at any size.

Probably the most important thing to realize in this entire article is that there is only so much you can do to "invent" data that is not there and when displaying or printing photos, you have to go to extremes in most cases to be able to see the difference between interpolation methods.  If you are captivated by some software or plugin that claims to do a much better job at interpolation, my suggestion would be to download a trial or do a search for reviews of the product before you buy.  I've seen some ridiculous samples posted on interpolation software websites showing a vast difference between their method and others, only to download the software and find out that it really does no better than the old bicubic method.  When you consider what the image "should" look like versus what we get out of various interpolators on the market, there really is very little difference between the better ones.  If you find yourself about to plunk down $200 for an interpolation program or plugin, you might want to think twice.  There are interpolation programs out there that offer a wide variety of methods for less than $50 that do as good or better than the high priced software.

Just for a sense of "calibration", one of the better methods available produced this result for the car wheel:

Some interpolation algorithms may render this image a little sharper, with a little more/less jagged edges, but the result will always be pretty similar to the above as far as the overall amount of detail that can be seen.  Now consider what this image would look like if we had taken it with enough resolution to begin with and didn't need to interpolate:

This final image clearly shows the limitations of interpolation.  Interpolation can reduce the appearance of artifacts like jagged edges but it simply cannot retrieve detail that is not there.  The 4x reduced image simply has only 1/16th the amount of data which means that 94% of the data in the interpolated result had to be "guessed" in the above samples.  Visually, the interpolated result is miserable in comparison to this last sample above regardless of the method used, but technically the result isn't too bad considering the fact that you started with only about 6% of the data you needed and you guessed at the other 94%!
 

 

Bottom line

The bottom line here is that interpolation can and does help improve the visual quality of images.  That said, don't expect magical results and beware of some of the mythical claims out there.  If you work for a magazine that normally starts with extreme crops and blows them up to 8x10 photos for print, you might be in the market for specialized interpolation software that allows you to pick the best method for each image/situation.  Just be aware that many of the "miraculous" results displayed on the web sites for some of these interpolation programs and plugins are quite exaggerated.  Better to try them first if they have a trial than to spend a significant amount of money and find out later that they really don't do much better than what you already have.  Finally, keep in mind that if you have a 5+ megapixel camera and you normally don't do much cropping nor printing above 8x10 size, interpolation method may never be a concern for you.  More important will be to find software that gives you the most benefit as far as the time it saves you and the quality of the final result.

Mike Chaney

Pages: 1 ... 209 210 [211] 212 213
Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!