Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: David Budd on June 04, 2018, 09:22:32 am
-
I am using a Canon Pro 2000 and unlike my HP z3100 the new printer does not have a spectrophotometer so I invested in a i1Studio.
I profiled Ilford Gold Fibre Silk, using Canon's Studio Print Pro (as suggested in Mark Segal review). Color management was turned off and both targets
were allowed to dry for 24 hours. The targets were scanned using the X-rite software to V.4 profile.
Using the Apple Color Synch Utility I compared this profile with the one I downloaded from the Ilford website. It shows a marked difference between the two, with the Ilford
showing a wider gamut.
Is this likely due to the accuracy of the i1Studio, the result of the color management pipeline not turning off or some other obscure reason?
I'm using a Mac Pro 2009, OS 10.11.6 and NEC 2690WUXi (profiled) monitor.
Thanks
-David
-
Gamut volume and accuracy are two different factors and profiles with narrower gamut are not necessarily less accurate for all colours that are within gamut. How have you assessed accuracy?
-
Mark... I haven't assessed the accuracy as of yet. I suppose making prints using the two profiles would be a start..would you agree.
-
Yes, but use a printer evaluation target with very familiar colours or "memory" colours to enable decent visual judgment. The Outback printer evaluation image would be a good one. If you can use a test target with patches whose colour values you can read accurately (the Outback image includes some), and can then read the values of the printed patches with your spectrophotometer and compare that would be even better, but I don't know exactly what the i1 Studio facilitates.
-
Thanks Mark...I have that Outback target. Not sure if the i1Studio will read those patches, that's something I will investigate.
I have done a soft proof with the two profiles in Lightroom using the Outback target and the i1Studio profile showed a overall "warmer" result.
See what the print look like.
-D.
-
Not all ICC profiles are created equally
In this 23 minute video, I'll cover:
The basic anatomy of ICC Profiles
Why there are differences in profile quality and color rendering
How to evaluate an ICC output profile
Examples of good and not so good canned profiles and custom profiles on actual printed output.
High resolution: http://digitaldog.net/files/Not_All_Profiles_are_created_equally.mp4
Low resolution (YouTube): https://www.youtube.com/watch?v=TNdR_tIFMME&feature=youtu.be
-
Thanks Mark...I have that Outback target. Not sure if the i1Studio will read those patches, that's something I will investigate.
I have done a soft proof with the two profiles in Lightroom using the Outback target and the i1Studio profile showed a overall "warmer" result.
See what the print look like.
-D.
Useful to do that.
As for reading the patches, if the Studio has a function that allows for spot reading of patches you should be able to compare printed values with image file values.
-
Thanks Mark...I have that Outback target. Not sure if the i1Studio will read those patches, that's something I will investigate.
I have done a soft proof with the two profiles in Lightroom using the Outback target and the i1Studio profile showed a overall "warmer" result.
See what the print look like.
-D.
The difference is soft proofing "warmness" is probably due to differences in the profile mode. I1Studio creates M2 profiles (uV cut) and most profiles one gets on the web are M0 which has some uV. This will make soft proofs using M2 profiles look warmer than M0 (or M1) profiles when selecting show paper white. You should see little difference with show paper white deselected.
-
The difference is soft proofing "warmness" is probably due to differences in the profile mode. I1Studio creates M2 profiles (uV cut) and most profiles one gets on the web are M0 which has some uV. This will make soft proofs using M2 profiles look warmer than M0 (or M1) profiles when selecting show paper white. You should see little difference with show paper white deselected.
I just examined that using my M0 and M2 profiles for Red River San Gabriel Baryta, which is extremely similar to IGFS. The visual and measured difference is so trivial that you have to look very hard and exercise some imagination to see it. For example, on one of the mid-scale gray patches the R value changed from 129 to 130 between the two, while the G and B channels remained in the 128/129 range. So at least for this kind of paper I don't think M0 vs M2 is an obvious explanatory variable.
-
I just examined that using my M0 and M2 profiles for Red River San Gabriel Baryta, which is extremely similar to IGFS. The visual and measured difference is so trivial that you have to look very hard and exercise some imagination to see it. For example, on one of the mid-scale gray patches the R value changed from 129 to 130 between the two, while the G and B channels remained in the 128/129 range. So at least for this kind of paper I don't think M0 vs M2 is an obvious explanatory variable.
From this I gather that there is not significant OBAs in the paper or substrate. Ilford Gold Fibre Silk s/b similar (I presume) and M0/M2 will show little difference. So what would cause a noticeable tint in soft proofing. The paper's white point may vary a small amount between instruments but not typically a noticeable amount. So the OP's observation together with the rather large gamut differences indicates something is messed up or perhaps a defective device.
-
From this I gather that there is not significant OBAs in the paper or substrate. Ilford Gold Fibre Silk s/b similar (I presume) and M0/M2 will show little difference. So what would cause a noticeable tint in soft proofing. The paper's white point may vary a small amount between instruments but not typically a noticeable amount. So the OP's observation together with the rather large gamut differences indicates something is messed up or perhaps a defective device.
Quite so; evidence from spectro readings indicates low UV reactivity i.e. much less than one finds for example in Epson Premium Luster or Exhibition Fiber. Let's see whether David can report some comparative measurements. Recall, we don't know what software and instruments Ilford was using to make their profiles, so there could be differences in those areas causing differences of outcomes, but even if we knew what profiling hardware/software they were using that alone may not be determinative. Tough one.
-
Quite so; evidence from spectro readings indicates low UV reactivity i.e. much less than one finds for example in Epson Premium Luster or Exhibition Fiber. Let's see whether David can report some comparative measurements. Recall, we don't know what software and instruments Ilford was using to make their profiles, so there could be differences in those areas causing differences of outcomes, but even if we knew what profiling hardware/software they were using that alone may not be determinative. Tough one.
It would help of the OP posted copies of the image, with the profiles attached to each image. Then they could be extracted with Argyll, PatchTool, or Matlab and the differences examined. This would also allow detailed analysis of the gamut differences.
-
Thanks to everyone for their input. I will print the image and upload it with the two profiles later today.
-D.
-
Both images were printed in Lightroom Classic 7.3, rendering intent Perpetual. Also attached are the two .icc profiles.
The warm tint of the i1Studio profile that I mentioned earlier is far less dominant in the printed image compared to what
Lightroom's soft proofing showed.
I may be wrong, but I couldn't find the option in i1Studio that would allow patches to be measured for comparision.
Thanks
-D.
-
The X-rite .icc profile didn't load on the above post.
-
David,
There's something screwball about both profiles. The gamut volumes on both are way too small for that printer/paper combination. The Ilford profile has a volume of only 651K and your i1Studio profile 375K. For that printer and paper both should be in the range of 800+. On my Epson SC-P5000 for IGFS paper my gamut volume is just over 980K. made with an i1Pro2 and i1Profiler. As well, the profile made with i1Studio is malformed in the yellow-green area.
-
The warm tint of the i1Studio profile that I mentioned earlier is far less dominant in the printed image compared to what
Lightroom's soft proofing showed.
Either the soft proof table is out of wack so to speak, or more likely your display calibration is.
-
Hi Mark
Thanks for your comments. The profile volume of the i1Studio and malformed-yellow green area would that be a indication that I have a faulty device.
You mention that your IGFS profiles are over 980K. Is that the result of the i1Profiler measuring a larger number of patches, where as the i1Studio is limited to just 100 patches.
-D.
-
Andrew...when you say the "soft proof" table is out of whack is that related to Lightroom.
I profiled my monitor last week using Spectraview 2 and i1 Display... the warm tint only appeared when I soft proofed the i1Studio profile, the Ilford one was neutral.
-D.
-
Hi Mark
Thanks for your comments. The profile volume of the i1Studio and malformed-yellow green area would that be a indication that I have a faulty device.
You mention that your IGFS profiles are over 980K. Is that the result of the i1Profiler measuring a larger number of patches, where as the i1Studio is limited to just 100 patches.
-D.
It's not necessarily a faulty device. It could be faulty settings, faulty target printing or faulty software.
Part of the reason for the huge gamut volume for the Epson SC-P5000/IGFS combination is the inkset - the orange and green inks do expand gamut volume considerably and is probably what makes most of the difference between the 800K+ range for the inksets without those colours versus the 900K+ range for the inksets with them. More concerning is the far below 800K volumes you have for both profiles with the Canon Pro-4000 and IGFS paper.
My profile for Epson SC-P5000/IGFS is made from a 2371 patch reading. I think what may matter more, however, is what the math does with whatever number of patches it crunches. I don't know the algorithms under the hood that create these profiles so I can't say for sure to what extent the patch count difference matters here.
-
I think I will do another round of tests. Instead of using Canon's Print Studio Pro to print the target I will try ACPU and Color Synch and also make v2 and 4 profiles.
At least then I can narrow down the variables causing issues such as the "malformed yellow-green area".
Thanks for your insight.
-D.
-
In Print Studio Pro, did you make sure that "No Color Correction" was turned on? If not, that could be an important cause of problems. Have a look at my Canon Pro-2000 review (much of which should apply to the Pro-4000 as well). PSP should work well according to that research. ACPU may not be all that useful, but no harm trying it. Can you print targets from i1Studio? If so, may be worth trying that.
Anyhow, you are right to start the troubleshooting at the level of printing the targets, so will be interesting to see what you come up with.
-
Yes, "No Color Correction" was turned off in PSP. I have printed some more targets using ACPU and Colorsync. Looking at those there is considerable difference in the rendition of the color patches compared to the one printed with PSP.
I will also print a set of targets from i1Studio as well. The print driver under Color Matching is greyed out, so I assume that is the case.
I will leave the targets to stabilise for 24 hours and then come back with some results.
Have read your informative review of the Pro 2000 a number of times and it make interesting reading. Its the Pro 2000 that I have. llford lumped together the 2000 & 4000 together for their GFS profile.
Cheers
-D.
-
Yes, "No Color Correction" was turned off in PSP. I have printed some more targets using ACPU and Colorsync. Looking at those there is considerable difference in the rendition of the color patches compared to the one printed with PSP.
Cheers
-D.
Oops "No Color Correction" is supposed to be turned ON for printing profiling targets.
-
I will also print a set of targets from i1Studio as well.
If you can print from the application that built and will measure the patches, do so!
-
Sorry Mark...I meant to say that "No Color Management" was turned On...too much "midnight oil" burning!
I just received this from X-rite support so will give this a shot as well.
Dear David Budd,
Many thanks for your request.
Please note that the i1Studio (as well as the last versions of CM Photo and Design 1.2.4) has all color management disabled in the print driver when printing directly out of the application. The only things that you will need to select are the paper size, paper/media type, and the print quality (should be the same as their printed images).
-D.
-
That sounds good - worth trying!
-
After receiving confirmation from Xrite that the i1Studio software disables color management I made another profile for Gold Fibre Silk using the targets printed directly from i1Studio and the Canon Pro-2000 using the same setting as before. Target prints were left to dry for 24 hours.
Compared to the targets printed with ColorSync this latest profile show a marked difference to the previous profile.
The current Outback evaluation print, compared to the previous one, show's more defined yellow-green patches, somewhat less saturation and on the far left the RGB color's are out of "whack" (my technical term).
-David
-
What Rendering Intent did you use for printing the Outback test image?
-
Perceptual...that's the only choice with the Canon driver.
-
Huh? There must be a way to print with Rel Col in the Canon driver. Is this from PSP or straight from the driver? I don't have it installed any longer so I can't check directly. Are you using an Airprint driver or the full-featured driver?
-
I printed it from LR as perceptual...its in i1Studio when the targets are printed that the Canon driver has no option the than perceptual.
Sorry for the confusion.
Should I try Relative?
-
Yes.
-
Thanks Mark...will do