Thanks Fred.  Seems easy enough.

I should have remembered the IN and OUT boxes are the same initially, just I am altering the curves so much lately they look odd.

Fwiw, some odd curves showing up.  PS is blocking up the blacks badly in 21 step wedges, and even the calibrators are doing it albeit somewhat better tonality than stock paper profiles.

Image is of the "Inkpress Cold Press 300" RGB curve I used to make it linear in QU, and the less red ink needed (red curve) in the shadows as well.  Still needs some mid-range tuning in the wedge for a green bit.  The paper may be a warm tone hence the red in the shadows, but it doesn't say on the box.


Mack

In Curves tab in QU, I have a Saved RGB filter curve for linearizing a paper's tones for B&W.

I also made a small change in the RED Curve with it to correct some color I saw creeping in.

I'd like to move the RED curve back to its linear-line Default setting and try something else there in the RED curve.  However, when I hit Delete it erases the entire RGB curve as well as the RED curve (  ).   I'd like to keep the RGB curve but it goes bye-bye too (which becomes a big pain to redo.).

How can I keep the RGB curve, while deleting some other curve in the Saved Curves filter?

Tia.


Mack

Additional:

Well, I succeeded in getting rid of my "burple shadows."  It can be done in Qimage by altering the Curve in the Blue channel with the line.

I had to go into Curve, and select "Blue" instead of the RGB that is also saved there for the linearization above.  You need to set a boundary to work within and alter the Blue curve else the entire line will shift without the anchor points.  Setting outer anchor points allowed me to alter the mid-tones to subtract the blue and add more yellow in the middle.  I don't mind a cool or blue dark black, but a yellow or warm highlight that becomes a blue in mid-tones is sort of annoying.

So I looked at the step wedge and called step 9 and step 18 my 'anchor points' so I could alter in the middle of it to more yellow.  I had to enter some number at those two steps so I entered only one point lower in the "Out" window in QU at each step found with the eyedropper which stored that value so the line would not move from 1 to 9 and 18-21.  Just everything between 9 and 18 would be shifted next.

I selected step 11 and step 15 to drop the blue.  So I entered about 5 points less than shown in the "Out" box on the Blue channel which added more yellow.  You can see the shift if you change it by 50 points to make sure you are going the right way.  Doing the same -5 points in Out at 11 and 15 dropped the middle of the line to add more yellow in the middle and got rid of my "burple" middle and dark tones.

You can see the shift in the upper right window of Curve in the attached screenshot.  Notice the four circles on the line black line.  Two outer are set as 'anchors' to lock the ends of the line line down, two inner circles (one in red) where I lowered the Blue channel RGB value (-5 points for me) in the "Out" window for step 11 and step 15.

I saved that as a paper correction name in Curve too.  "Epson 3880 Dye ink - Canon Glossy BW Correction".  That''s it.  I can call it up later when I do B&W prints on that paper.

Now if I want a nice "Sepia" tone, I can open the image (Preferably a gamma of 2.2 in B&W, but important to save in Adobe 1998 RGB so color can be used later.) and open the above Saved Curve Filter too.  Then I can go into "+Sel. Color" tab and change the N(eutral) setting to R=1.01 G=0.92 and Y=0.83 and print it.  All nice and linearized and odd ink colors neutralized out via my newly saved Curve filter above, and a Sepia tone added too.

Who needs QuadTone RIP for B&W when Qimage Ultimate can do it and maybe refine it even more - plus it is sharper too! 


Mack

Ernst,

I've given up on QuadTone RIP as Qimage Ultimate has a far better sharpening engine.  Try the same print from both and compare the two programs.  No sense dealing with "soft and fuzzy prints" from QuadTone RIP that cannot be adjusted.  Their forum is sort of slow too, as this one has jumped almost 150 reads since I restarted this thread in 2-3 days (Lots of lurkers?).  Got tired of the "Error I/O: 32" messages (Dating back to 2009?), and it hanging up in Task Manager too needing an "End Task" to kill multiple hangs of quadtone.exe too.

I believe I have addressed the "dMax" using Qimage and using the Epson driver below.  Needed to redo some stuff too.

So far I made a couple of changes.

1. The link for the 21 Step Lab stepwedge above is in a Gray Gamma 2.2 mode.  As such, it will not allow me to change any coloration to it in QU.  I saved it in Adobe RGB 1998 as a TIF and all is well there now.  I checked to make sure the L values remained at 5 points in each step increase/decrease and they do.  I also added the step numbers to the chart as I got tired of counting steps, and also Adobe RGB 1998 in red so I know I am using the correct one and not the linked one above that only deals with B&W and no color additions I may apply in QU.

2. For Jon Cone's Linearization spreadsheet above, I changed the 1-100 scale for the steps (Bottom, x-axis) to 1-21 to match the steps in the 21 step-wedge above.  Makes it easier to pick out which one to change the "Out" RGB value in QU Curve portion (Making it a more positive RGB number moves the black line up on the L scale, negative moves it down on the L scale.  If the black line is touching the magenta, 1-2 points might do it, if there is a lot of white space between them, maybe 4-5 points is needed.

3. My initial dMax values weren't that low (or black) on the Canon Glossy Paper which I used for this experiment.  However, within the Epson driver there is a "Color Density" adjustment and moving it up the percentage scale made the black Step 21 target definitely blacker.  I got the following L (Luminance) dMax values off the x-rite i1 head below for Step 21 by increasing percentages (Lower L# = blacker):

0%  L=12.113
10% L=9.753
20% L=8.29
30% L=7.626

I seem to recall that anything less than 1 difference is where you can settle on from some QuadTone RIP articles without flooding the paper with ink.  I settled on the 20% More Color Density value (to be set in the Epson Driver Properties > Paper Configuration Page window) as the 30% wasn't as great (1 value) as from 10% to 20%.  This is using my initial profile written for the Canon paper too so I suspect my initial setup is buggered and I need to redo the i1 Photo Pro 2 reading and alter the settings in the i1 Profiler software program to produce a blacker black (I think there is a couple of settings in the i1 Profiler to do that?).  I use 20% more Color Density now, although it might be a bit better at 25% density, but that's debatable.

Then I ran a new linearization reading for the step wedge at 20% more color as I could see the black getting bunched up on the right of the wedge and darker overall.  Linearization line moved to the left.

Tweaked it in using Curves and the Eyedropper.  Just raising and lowering the "Out" setting in the Curve for each step gave me the screenshot below.  I got more dMax over the above screenshots, and it is very linear (I can see a bit of the black line at step 16 on top and a bit under at step 19, but it is pretty darn good overall.).  Maybe even more so than QuadTone RIP produced one since I can tune each of the 21 steps in Qimage individually and not follow a "curve of ink grey colors" as in QuadTone RIP.

I have been using an expensive x-rite "i1 Photo Pro 2" to do this L*ab reading stuff, but their cheaper ColorMunki Photo shows the L*ab values in the upper right hand corner of the Color Picker program within it.  It should work too.

Later, I will attack the bluish color I see in the steps around 10 to 18 and see if I can negate that color.  I should be able to do this in Qimage too and maybe just address the color at those steps too.  Dunno.


Mack

Ernst:

Yes, the above screen snap was linearized in Qimage Ultimate alone based on the readings off the 21-step L*gamma chart (It is L*ab based in readings and very close.).  I used the TIFF step-wedge link above and was using the spreadsheet by Jon Cone (in the above link) and using his dye inks in the the 3880 (I'm using the normal color DYE inkset array, not the K7 gray inks, fwiw, as I want to adjust the coloration a bit at times as in toning.).  I just changed the +Color line a bit in Qimage to force it linear.  I also re-did Cone's linear regression spreadsheet to read 1 to 21 (from the 0-100 numbers) under "Steps" so I can go immediately to that step in QU and move the +Colors tabs "Out" numbers a bit.  I saved it as a "Curve filter" under the +Color tab to apply later when I mess with it.

For fun, I entered the step-wedge numbers off a QU generated print from the above method and fed them into the QuadTone RIP program and the numbers from its Linearization came out perfect too needing no adjustment.  So the QU way above and its image seem to be happy in QTR as well.  I managed to smooth out that screenshot's non-linear hump in the screenshot just using Qimage Ulitmate's +Curve tabs "Out" RGB numbers and Jon Cone's regression spreadsheet.

Ernst, if you have an image to print, try printing it in the QuadTone RIP and then print the same image in Qimage Ultimate.  The sharpness in QU is very apparent at my end  I leave the QU 'Sharpness' set to around 12 and I shoot with a Nikon D800E so the sharpness should be apparent out of that camera.  The QuadTone RIP looks to be really, really soft and a bit out of focus for whatever reason with the same image.  Not too happy with it side by side against Qimage which is far sharper so I'll stick with QU as Mike seems to keep it updated and current whereas QuadTone RIP seems to be iffy at times, plus it hangs in Task Manager > Details in Windows 8-64 as multiple Quadtone.exe and some other *.exe that needs to be shut down hard (End Task) for those multiple entries.  It's just not a well executing program, too old maybe, and support via their Yahoo! group is very slow to non-exisitant too.  It needs a lot of current TLC, and some of the Info/User files were written way back in April 25, 2005 for some of it and do not apply to current hardware.

I do find some errors that pop up in QuadTone RIP to that date back to 2009, i.e. the "Error I/O 32" one that still haunts the program.  The GUI for it is sort of kludgy too on seeing the size of the postage stamp-sized image on it too as the GUI isn't sizable in Windows 8-64.  I'm sure Mike could do a far better job on the programming if we can convince him on some B&W RIP program or addition to QU since he has already has a good one for Color with QU.  Someone should give him some Epson printer loaded with K7 gray inks (haha!) and a truckload of Canson paper.  Plus, we'd have Terry and Fred to pick on too when it blows chunks!   

(my $0.000002 ... and sundry subliminal hints too.)


Mack

Get these first:

21 Step Grayscale L* gamma target: http://cameratico.com/media/images/articles/17/21-step-Lstar-grayscale-target.tif

Linearization Spreadsheet: http://www.piezography.com/PiezoPress/blog/piezography-k7-inks-and-curves/checking-your-linearization/

Spreadsheet download for Excel or Open Office or LibreOffice (freeware): http://www.piezography.com/Linearization-Checker.zip


Entering my own L*ab values in ColorPort (x-rite software) with their "i1 Photo Pro 2" on Spot mode into the yellow window of the spreadsheet above, I got a Before and After of the linear line.  Took a bit of work in the +Curves tab in Qimage Ultimate to get it right, but the line is looking pretty darn good using the data from reading the 21-step target above.

I wasn't happy with the targets in QuadTone RIP as they appear to be uneven in density in each square (More like a graduated ND.).  Outdated to work with the Scan mode of the newer i1 spectrometer heads and software as well.  The above L* TIF step-wedge image is smoother and larger for targeting in Spot mode too.

If you see the Black Line to the right of the magenta one, apply a bit of negative RGB at that part of the line in the +Curve window in Image Ulitmate.  If it is to the left of the magenta line, apply a few points positive to the value for that density in the +Curve.  The Eyedropper will help you find which 21 step you need to target and change the "Out" value in a few steps (1 or 2 if close to the line, 5-6 if further away).  Print target again with the newly Saved Curve (Use some name) in QU, and enter new L* values into the spreadsheet yellow blocks again and see if it is closer once dry.

Now I'm off for the Color Change (Purple shadows) and see if I can clean that up a bit in +Curve window too?  Maybe use the Blue channel only and leave the L* value alone (Use all the RGB numbers in the +Curve editor pane for the linearization part.).


Attached:  Screenshot of Epson 3880 with the 4 colors of black/gray ink (Only using the 3 and not the Matte now).  You can see the two charts made from Open Office and the linearization of the three inks:  Without the +Curve mods, it is outside the linear line on the right and not straight.  With the tweaks made to the RGB numbers for the steps on the Curve line and saved as a Curve filter, it is almost perfect! You can see the slight dip in the straight line part of the +Curve window (Little circular dots where it has been tweaked downward.) in the upper-right Editor window where I applied negative RGB numbers in the "Out" pane which moved the black line to the left under the magenta one in the spreadsheet to make it linear.

Purpose of this is too keep the ICM paper profile that I generally use, and will also let me add coloration inks to the B&W images later on same paper as a filter (or maybe along with the Curve too?) such as Sepia, Selenium, Warm, Cool, etc.  With the way the ICM profiles are made in QT RIP, that ability is sometimes hampered as it messes with the RGB portion so you may not get the color you want, plus I get to keep the ability of the better sharpening engine in Qimage Ultimate which you lose with the QT RIP interface. 

Whew!


Mack

Oddly, I seem to be right behind Ernst in looking for B&W printing stuff and found this old unanswered thread in searching.  Seems it must be somewhat popular as it has had over 1,200 reads but no response until now.

The QTR stuff is sort of outdated for the new "i1 Pro Photo 2" I find.  It won't show 21 step in those 21 step targets in a strip mode scan, maybe 20.  Something to do with the way the targets were made up maybe.  It suggests using the old outdated x-rite ProfileMaker 5.0 software too over the new stuff.  Dunno.

Right now I've been fighting the "burple shadows" in the 21 step targets that transition from a warmer yellow highlight, but found I can tune the inks via the Curves in Editor.  The Eyedropper on a target step will show on the Curve as a "Loc" (Location) and a "In" (RGB value), the third window "Out" (in RGB) is the one I can set to add or subtract a given color at that selected eyedropper point.  I can see and set each point in a 21 step tablet there and Save it out as a Curve filter.  Example:  Change it to maybe 0.5 plus or minus in the "Out" window and you can see which way it needs to go to make that step neutral.  In my case, I have to have the Blue selected too since I am working on that curve.

I haven't gotten to the Linearization part yet as adding and subtracting a color of ink seems to influence the readings somewhat.  I don't know if you can change the brightness (Luminosity) of each of the 21 steps in the Curve tab and save the color & linearization both, but maybe you can?  Might be too tedious to do the 51 step wedges other than just seeing if they do have a gradual density change.

In the "QTR User Info" (Pages 25-26) PDF file, they show suggested values in a Linearization array window for dMax of the black from 1.600 to 2.400 in 0.050 steps.  Would be interesting if one could move the curves somehow and Save them as a Curve filter or general Filter to be applied when needed.  The Levels tab of QU doesn't provide for a Save as far as I see, so it may be need to be done in the Curves tab to Save it?

I'm not too happy with QTR RIP GUI interface (Clunky old Drag-and-Drop Windows 98 looking thing.) as the images I get out of it are soft and not as sharp as can be done with QU.  Really is a remarkable difference between the two there, and QU has a much better sharpening engine and adjustable too.  The tone density between the two is substantial too (i.e. the QTR RIP using their quid extensions and no OEM print driver vs. an ICM made out of it for QU/whatever using an OEM print driver.).

Too bad Mike hasn't added some B&W tab on QU for the Color and Linearization construction somehow for us B&W printers.  Should be far better than some RIP packages out there if it were there since he has the best sharpening I've seen using QU. (I did see the sepia filter buried in QU somewhere, but it is more like a gold "Sundowner filter" as it is way too much sepia if it is.)


Mack

Fred-

This is what I see in the Main window with the 323,415 KB TIFF loaded (Shift+Help > Analyze current settings):

Start: 2146 MB
Addl: 1416 MB
Now: 2146 MB

It loaded this time into the Editor without displaying the "Qimage.exe is not responding" message.  Didn't see the red x's page either this time.  Don't know where that page comes from when it does come up.

There are times when I see the image I select disappear once dragging to the right into the Selection window on the primary QU page and it picks up an alternate image to put there other than the one I picked initially in dragging.  Odd.

Even bought a new mouse as I thought the old one may be 'unswitching' (Not holding onto) with the left click button at times.  I dunno. 

Additional:  I just found that AVG was running this AM in the background as a weekly setting when the above occurred.  Maybe something there?  I saw it pop up a "No issues found" message sometime after I was working in QU and it wanted to shut down and restart for some reason too.  Updates from Windows or AVG maybe?

I'll watch it some more.

Thanks.


Mack

Large TIFF file 315MB.  Windows 8-64.  3.2GHz. 12GB RAM.  Latest QU version 116.

While trying to load the file into the Editor to darken it a bit, it was slow and not showing/loading the image and this message popped up: "Qimage.exe is not responding" with the option to Shutdown or Wait.

First time I shut it down, did it's file report thing in Windows and closed Qimage.

Second time, same thing with the same file and message popped up again.  I selected "Wait" and the image loaded a few seconds later.

Something to do with large TIFF files maybe? 

I do notice some red x checkerboard that fills the entire page that sometimes loads into the first window right pane instead of the image itself.  I drag it again and it appears in it normally.  Odd.


Mack

Ernest-

Thanks for the feedback.  Yes, the Cone dye's have been in the 3880 for some time now.  In fact, I recently had to refill them all so they are getting a workout.  I don't think there is any residual OEM ink in the system lines now.  Printer nozzle checks are fine as I run those weekly on all my printers (I got a few.) and to keep clogs down on the unused ones.

What I may try is feeding a sheet of paper "twice" through it with a black B&W image via QU and see if the dMax builds beyond the 1.8 that I get now to the values of the ink drop.  I don't get why the "i1 Photo Pro 2" head that I use to read the density goes so high on the drops of ink.  I did find a bug in their software (Seems I'm good at that. Ask Mike.) where their ColorPort software only reads 255 for any RGB readings too.  It is stuck there in their software, but the L*ab side works correctly.  They are working on it - I hope.

I can set a black at sRGB 0 0 0 in PS and the inks are not hitting that value of the ink drop.  That's why I posed the question about a "Black Boost" in QU if the QuadTone RIP can do it (Haven't gotten that far yet.).  I don't believe the "Exposure" QU checkbox raises the dMax or does anything to that value in QU if the values are seen/set as sRGB 0 0 0 ?  I did see the ICM output the QuadTone RIP does (Some quid (sp?) file extension and some other one and not an ICM.) and they are simple text files that must apply density to each color ink.  I could see maybe a way there to kick the black up via those numbers as they build up in the table without mashing up the black tonality range much.

I think the Canon printer drivers allow one to raise each ink value independently of QU in the driver menu setup, but don't know if one turns Color to "Off" in QU if it affects that setting to some neutral value?  Epson only shows a Color Boost and not an independent for all the colors?  Someplace the "dye" may be laying down less ink out of the printer than the pigment ink is my theory right now even though I have re-written all the ICM profiles for dye.

Odd part is why I get such a dark black dMax with the dye via an "ink drop onto the paper" yet cannot out of the printer via an ICM?

Fwiw, this reminds me of the old Ansel Adams and Minor White tests of putting paper in sunlight and developing it to determine the paper's dMax (max. printing black) in the old chemistry baths, just messing with ink now.  Probably where I got the idea too.  Printing a sRGB 128 128 128 gray is hard enough, much as when our color darkroom instructor made us print a Kodak gray card with those damn CC filter packs.  Ugh!  In fact, I have a sRGB 128 128 128 file loaded on a card stuck into another printer and the readings off a test print out of it vary by maybe 6 points day to day and any of the colors both.  You can see the color differences day to day easily off it too.  Has to be ink or head issues, but it is OEM only and it changes all by itself.  I dunno.

Fwiw, a local auto color paint shop has all sorts of wiz-bang color analyzers and they still have to do a test spray-out and tweak the paint after they let them set for a day to match.  Boss showed me a white car and you could see maybe 5 shades of white, maybe different factories for some panels/bumpers/handles.  Never noticed it until he pointed it out.  Much like the same print to 5 different labs comes back differently I guess.

Oh well.  Guess no way to boost the black ink in QU that I can find, if it cannot be done somehow in the printer's driver menu.


Mack

Ernst-

Yes.  I let the ink drops dry overnight outdoors and in the 100 degree daytime temps too.  Maybe 24 hours passed before I picked the paper up an read it in the i1.  Actually, the drops all appear to be a semi-gloss once dried but they are thicker than the inkjet would spray too.  The Matte Black may be more diffuse once sprayed.

I will note that I am using Jon Cone's "Dye" ink and not the Epson 3880 OEM "Pigment" K3 ink as I'm not fond of pigment on metallic papers and some super-gloss film that I sometimes do.  Might be his inks have a better dMax, but I dunno as I never dropped the ink out of the Epson OEM tanks to compare.  However, when I run it through the printer, the PK shows about 1.8 at dMax now.  Disappointing that is is quite a bit less than an ink drop test.  However, I did do some Matte black on a thick velour and it was black as charcoal, just I never measured it (yet).

I'm also playing around with the QuadTone RIP which is sort of bug-ridden (I/O 32 errors?) in Windows 8 with the GUI that is old.  So far using its old-tech GUI and dropping the image onto it and going straight to the printer, it produces a nice tonal range a nice black.  If I output the 21 step wedge they use to set it up with the same readings out of the i1 Photo Pro 2, and then pass it through their ICM script-maker executable, the ICM profile made is not as nice in the tonal range (blocks up the black end) and has some coloration added too.  It's odd that it bypasses the ICM profiles and Epson driver both on the "Drag-and-Drop" method, but using the ICM profiles generated with the Epson driver & QU it makes two very different images, no matter if I turn Color "On" or ABW (B&W) "On" in the Epson driver window either.  Best is the non-ICM one for me now and a more even tonal gradation too than the ICM made one.

I still have to play around with QuadTone RIP some more with B&W images.  Very different program and quirky setup.  I guess it makes sense if you work through it all.  I posted the odd curve image it made for the 3 gray inks out of the 3880 in the Technical Software part of this forum.  But they sure look nice on some of the "Northern Light B&W test images" where you can see well into the shadow details over the ones made using a ICM profile.

I probably should sign up for one of Jon Cone's ink classes where they even mix ink by hand for color control in B&W.  Then dump the 8 different shades of grey inks into some 3880 for fun and supposedly better tonal range and sharpness both.  The piezo ink printing guys doing all this black/grey ink stuff only are really excited over this stuff.


Mack

Only works with Epson printers, but I was sort of astounded in a black and white that I dragged onto its (Very crude!) Windows GUI and it fired up the printer and did its thing.  Never have gotten that sort of a B&W tonal range, even with profiling out of the x-rite i1 Photo Pro 2 making an ICM to use within the Epson driver.

Seems to work somehow without the Epson 3880 driver nor does it need a ICC/ICM profile either if you set it up right.  Might need an i1 reading device or similar densitometer to set up the 21 patch your print off it once you get the ink levels tuned.

The "WOW!" part is that I see tones in the blacks and can dial the black inks even darker than going through the profile route with the Epson printer driver.  The later seems to block up the blacks quite a bit and not as many steps can be discerned in their included test targets.  You can see the included 51-step tone B&W quite well too to verify you did it right.  It would make Ansel Adams envious.

I didn't get as good a tonal range, and some colorations added, by allowing it to make a ICM profile from the read 21-step data for some reason.  The "Drag the image onto the GUI window" (And "crude GUI window" is an understatement over QU too!) made the best B&W one and clean of colorations.  As this is shareware with sundry authors, might be something different there too as to why certain things work better.

I've attached a curve generated that shows the 3 dark gray inks in the Epson 3880, and the slight Magenta/Cyan that neutralized the warm B&W mid-tonal range form whatever ink issues was going on (I'm using Jon Cone's 'Dye' ink in it and not OEM pigment ink.).  The "Lite Black" starts the tonal process at the white end, ends with the "Photo (Dark) Black," and uses the "Lite Lite Black" as a fill between them.  Or that's how I understand it.  Paper was Inkpress Glossy as it has a high dMax, imho.

If you need a B&W with a nice tonal range, might be worth checking into.  Not as easy to figure out though, almost Windows DOS at times.  I believe it is $50 for the shareware if you want to use it and it suits you.  I still need to do some playing around with it and see if I can fix the ICM curve part somehow.


Mack

Did a little experiment.  I put a drop of ink of two ink carts out of the Epson 3880 onto some glossy paper, shook it off to the side a bit and blotted with a Q-tip the excess, let it dry for 30 minutes in the sun, and then measured the black matte and photo black of the two inks.

I got a reading of 2.254 for the Photo Black, and 2.337 for the Matte Black.  Really a nice black out of either, just the Matte was a little duller and not so glossy as the Photo Black.

However, the best black that I can get out of the printer is 1.724 with a profile made for that same paper and reading it as sRGB 0,0,0 in QU.  I read that "QuadTone RIP" has some "Black Ink Boost" but it only works for the 4900 I think and it is primarily a B&W print driver software for Epson only?  Surely there is a way to boost the black ink a bit more in Qimage without destroying or compressing the contrast for the darker dMax areas and bunching them all up?

I did see some Ink Color boost in the Epson driver (under Paper Config I think), but I don't know if that is the same thing and would need to be set each time.

Also been wondering if the Paper Thickness (or Platen/Printhead distance) would have any bearing on the black?  Head closer to the paper and maybe more ink into the pores and less spread out elsewhere?  Dunno.


Mack

I have a Canon 9000 II and print canvas all the time through the rear feed slot using Qimage Ultimate.  The 9500 II shouldn't be that different? 

I print on 17x25.5 and some odd 16.8x36 inch (cut from roll) canvas panoramas.  Biggest issue is getting the canvas to feed right on the load else I'll get the "Feed/Paper Skew" warning.  If I open the top panel while loading, I can see the canvas load about 1/4" ahead of the black guide bar.  If it is off a little bit, the "Skew" error pops up after the head passes and reads the white canvas edge for alignment.  What I do is tug on the canvas so that the canvas is parallel to the black bar before the head can travel the entire width to align it so it has about 1/4" even and parallel to the black bar across the width so the "Skew" error won't pop up.

Some have suggested taping a piece of leader paper onto the canvas to get a better grab and alignment to the scan of the leader, but I had the tape separate and the mess got wrapped up inside the printer requiring dismantling.  Not cool, so I'd avoid doing it that way.

Fwiw, I'd invest in some x-rite profile maker hardware for the canvas too.  Either the ColorMunki Photo or the i1 Photo Pro 2 and use the profiles from them.  Should be far ahead of any commercial printer doing it that way too.


Mack

Thanks Mike.

Fwiw, I also ran the i1 Photo Pro 2 against the ColorMunki Photo using their respective software.  The 3D gamuts are very similar with the i1 favoring the blues a bit better and maybe due to its UV filter for optical brightners in papers.  The one I got in Profile Prism is very "blocky" in appearance and much smaller overall.  I fear the blockiness of the Profile Prism 7 generated ICM is pushing some colors off into a different coloration in some portions of a colors 0-256 density scale (especially the green/yellows where I can see density issues in the color ribbon of the green vertical ribbon in the Northern Light test image.).  I'm suspicious of IT8 target as the culprit since the x-rite's scans use 100 colors for the ColorMunki Photo test charts (2 pages) and 2033 colors (4 pages) for the i1 Photo Pro 2 where the IT8 has less colors for its initial scan base?

As to the 128 gray, I just don't get why 128 in a border in QU is different than a bucket fill of 128?  This has been driving me insane and boxes of paper and ink (Reminds me of the college assignment in the color lab of printing a Kodak Neutral Gray card in CP5/Ektaprint chemistry which took weeks!  Teacher was a sadistic SOB for that!).  I used sRGB as the colorspace in PS when I saved the image (ProPhoto is another matter!), but I cannot get a "QU 128 Border" and a "PS 128 Bucket Fill" to match out of the printer without some fine tuning in QU and altering the border 128 values - and QU is very good for that fine-tuning part.

I can also see the visual difference on the (x-rite calibrated) monitor as well.  However, I can alter the 128 in the QU border value settings to make it 128 in the print and agree with the bucket fill, just that it's a lot of work (e.g. QU might take a 132, 124, 116 for a Border, and PS needs a 127, 126, 121 for a match out of the printer - if that makes any sense?  At which point the spectrometer will read 128 for everything on the print.    )

Attached are the i1 Photo Pro 2 and ColorMunki Photo combined, and the Profile Prism 3D gamut alone (extracted out of prior image above).

I'll add that an image made with the ICM profile from the ColorMunki Photo tends to the warmer side where the i1 Photo Pro 2 tends to the cooler side.  The i1 also provides a smoother transition from 0-255 in the grays too.  The two x-rite devices also tend to be about 5 points different in agreeing with each other too.  The HP scanner reading the IT8 who knows?


Mack  (...Talking to myself this AM as I might get back to this in the future and have an "Ah Ha!" moment - I hope.   )