Hello,

I bought the PP software about a week ago and I was trying to create some printer profiles with two different scanners. I read all the instructions but no matter what I do I still having problems with the scanned image and PP. I'm using Vuescan for the Canon MP980 and the Epson Scan software for the Epson Perfection V500. I used the recommended settings for the scanners.

When I open the image with the PP sotware a message appears asking to select yes or no. When I select yes, after cropping the printer target I always have more than 40 patches and a warning message. If I select "no" to the message I have perfect results. For what I understand from the instructions for a printer profile I need to always select "yes" to the message "Is this a 48 bit a linear (gamma 1.0) image and it is being used to profile a printer?".

I will appreciate if someone can clarify this point.

Thanks for the help.

Fred,

Thanks for the information. I scanned the targets using vuescan raw mode with both scanners and the resulted images still have the same problem with the PP software. But based in your feedback, I think that if I answer "No" to the PP sotware I can use the image created with the Epson Scan program (48bit, tif).

Thanks again for your help.

Regards,
Wilfredo

If you are profiling a printer, there really is no "correct" answer as both will work because both targets are in the same image: whatever you do to one will have the same effect on the other target.  I prefer the "Yes" answer if you are profiling a printer because it makes the targets bright enough to see.  There really is no other reason.

Also, I have yet to see a scanner where 48 bit data is of any real use.  I've tested quite a few of them and anything beyond 24 bits is "in the noise" so to speak.  Now if you start doing multiple scans, that's another story.  If you run a half dozen passes, there is actually some real (image) data in the higher bits.  Not much... but a little.  If it's any help, I never use 48 bit scans when scanning for a profile.  24 bit is quite good enough.

Mike

Mike,
I tried your suggestion of the 24 bit scan instead of 48 bits. My PP setup is an Epson V500 scanner and VueScan Pro - profiles for a Canon ip4500. I ran comparisons on two printer targets I had made a while ago - Costco Kirkland Glossy and Staples Matte. I ran the 48 bit test with answering both Yes and No to the 48 bit linear question. I ran VueScan at 4 passes for both 24 and 48 bit scans. All scans were Raw which gave me uncompressed .tif scans.
The main difference I saw between the 24 and 48 bit scans had to do with Gray Patch clipping on the Prism target. In 48 bit mode (Yes to Question) I had to play with the brightness setting on the scan to get no clipping for both the Glossy and Matte papers. No to Question gave me a Poor - 3 or more clipped. With 24 bit scans both papers showed no Gray patch clipping with brightness set to 1 (default). The other values were close - 1%, 2% or 3%, and clipped patches between 2 and 6 for the 24 bit scan and the 48 bit scan (Yes). The 48 bit scan (No) had many more clipped patches for the Printer target (15 for Matte and 40 for Glossy). The Prism and Printer target Exposures were a little higher (236, 240) for the 24 bit scan compared to 219 to 230 for the 48 bit scan.
I didn't complete the generation of the profiles for each scan type but wanted to compare the target reviews for the different settings.
Looks like 24 bit scans will be the way I go in the future. Am picking up a Canon Pro9000 MkII this weekend.

Steve W.

Peter,
I just reran all my tests using Epson Scan instead of ViewScan. Some of the results were rather interesting. I ran Epson Scan in both 24 and 48 bit mode and saved the image as .tif with no compression. Think that would be pretty similar to RAW - maybe I am wrong. The file sizes for both the 24 and 48 bit scans were almost the same size as the RAW scans saved as .tif in VueScan. The only thing I seemed to lose with Epson Scan was the multipass capability.
The target analyses for the 24 bit scan from VueScan and Epson Scan were almost identical including the histograms. For the 48 bit scan (Yes to Question) the results were actually better than the 48 bit VueScan scan, but the printer target histogram was like you described - mound shaped which Mike says isn't desirable. But the 48 bit scan with No answer was much better than the VueScan version with many fewer Xed patches and Excellent Shadow Detail.
One thing that is bothering me is that the 48 bit VueScan scan was very much darker than the 24 bit scan. The only thing I changed was setting the Input Bits Per Pixel to 48 bit RGB and the Output Raw File Type to 48 bits RGB from 24 bits I had initially set in each. The two scans (24 and 48 bit) from Epson Scan were identical in brightness to each other as best I could tell. From these tests I seem to like the Epson Scan results a very little better than VueScan. Just wish Epson Scan had a multipass feature.
I am hopeing that Mike chimes in here with his wisdom on these results - especially to see if I am doing something wrong with VueScan to get such different brightness scans.

Oh yes Peter - it is a Pro 9000 MkII I'm getting, not the Pro 9500. Got a good deal (I hope) on the local Craigslist.

Steve W.