[Elliot Puritz Silver Member Posts: 77]

Papers & Media

on: July 21, 2021 at 9:37 pm

As always, great work Mark.

It is interesting to note your remarks concerning the soft proofing using the OEM profile NOT matching the print.

Many of us might chose to have bespoke profiles made for one of these new papers, but custom profiles for all four of these papers would obviously be costly.

I wonder if a custom profile for one of the papers might suffice for use when using ANY of these new papers:  That is, are the measured values close enough so that one custom profile will still be “superior” to the OEM profile?

Elliot

 

 

Elliot

[Mark D Segal Silver Member Posts: 951]

Re: New Canson Infinity Arches Paper

Reply #1 on: July 22, 2021 at 8:04 am

Hi Elliot, thanks glad you enjoyed the article. The main issues for producing an acceptable match are mismatches in gamut volume and maximum Black level. The importance of these mismatches will vary from printer to printer and paper to paper. I think the cheapest and easiest thing you could do for soft-proofing would be to print a good printer evaluation target image (many are downloadable from the Internet) on whichever of those Canson papers you want to try, using Canson’s profile. Let it dry for a few hours, set-it up in your normal viewing conditions for looking at prints, and then go into your Custom Proof Set-Up menu in Photoshop, scroll to the Epson or Canon OEM profiles you have there (depending on whether you are using an Epson or Canon printer) and as you activate each proofing condition on your properly profiled and calibrated monitor, look to see which one most closely replicates the appearance on your monitor of the print you just made, and use that one for softproofing. Once you have edited your photo under those softproof conditions you can use the Canson profile for printing and it should render what you expect.

[Mark McCormick-Goodhart Participant Posts: 14]

Re: New Canson Infinity Arches Paper

Reply #2 on: July 29, 2021 at 12:22 pm

Hi Mark S.,

Well, it gets curiouser and curiouser! I had downloaded the various new media profiles from the Canson website last May for my Canon Pro-1000, not an Epson SC P5000. They softproofed as I expected, thus no indication of being built with M3 (polarized) illuminant condition, and I didn’t give it another thought. After reading your article, I checked the internal tag info, and it says they were built with M2 (uv filtered but no polarized) illuminant condition. This got me wondering about your test results, so I just downloaded the BFK Rives Pure white paper profile for an Epson SC P5000. Sure enough, just as you indicated in your article, it was built with M3 illuminant condition!

So it would seem Canson may be subcontracting some ICC profiling work to more than one vendor, and they aren’t conforming to the same specifications, or Canson may be slowly converting to M3 over time (I hope not because I think it’s a mistake).  I guess we will have to wait and see unless perhaps you have some contacts at Canson who can clear up this confusion. Either way, M3 built profiles muddy the waters for endusers. While some folks may welcome the more open shadow rendering M3 profiles can provide for matte papers, it comes at the serious expense of accurate softproofing, gamut volume calculations, and even some hue and chroma color accuracy. Thus, I personally believe M3-built profiles are one step forward, two steps backward at best.  Amateurs who typically don’t even know what soft proofing is may well be pleased with the output from these M3 derived ICC profiles. However, advanced users who routinely rely on soft proofing will need to determine what “flavor” of profile they are dealing with, or just roll their own custom profiles and ignore generic profiles built with the M3 illuminant condition.

Lastly, I noted your comments about opening shadow details in an M0/M1/M2 profile to try to match M3 output as not being as easy to do as it might seem.  A relatively straight-forward approach I recommend in this situation is to apply a PS curves adjustment layer to fine tune the shadow area of the tone curve, but be sure to set the curve behavior from “normal” mode to the “luminosity” mode setting. The curves adjustment luminosity setting in PS confines the subsequent RGB output changes such that only the lightness of the image is getting altered by the curve, not hue and chroma at the same time.  In other words, one can directly open the image shadows to achieve a comparable M3 open shadow appearance without introducing other unwanted color and saturation shifts in the image which might otherwise need further additional edits.

cheers,
Mark

http://www.aardenburg-imaging.com

[Mark D Segal Silver Member Posts: 951]

Re: New Canson Infinity Arches Paper

Reply #3 on: July 29, 2021 at 1:14 pm

Hi Mark, thanks for the substantial feedback on this work – much appreciated.

I agree with your assessment of the pros and cons of M3 profiling. It’s not an unambiguous win-win – there are the trade-offs you mention and whether they’re worthwhile is something readers will need to decide for themselves, having seen the evidence and considerations I laid out here and the results of their own experimentation. I think at least we have enough “meat on the plate” to table the right questions and we’ve now done that.

I look upon the BVDM Roman #16 target photo as the best and most scientific resource for visual and objective examination of the deep tone separation issue and base my comments on my work with that target, complemented with my own 35D neutral patch set and accompanying graphical representation.  The extent to which M3 profiles open deep-tone separation is limited and subtle. And yes, you’re right, the obvious recourse for opening deep shade detail with say an M1 profile is to use a Curves adjustment layer to steepen the curve in the relevant segment, which I experimented with not only for the current article, but years ago when the issue arose as well for another paper – a Red River product if I remember correctly. It’s doable but the operational challenge (regardless of the Blend Mode) is that the relevant area of the curve is so short that getting this adjustment just right is somewhat tricky. That’s why I commented on it being less simple than it may appear.

Of course I was not aware of the difference in OEM profiling Measurement Condition between the Canon P1000 and the Epson SC-P5000 – that is indeed puzzling and good you discovered it. I no longer have the Canon printer so I could only test with the Epson. I do have contacts that I can ping to learn more about the inconsistency of approach. I do know who made their profiles for the Epson printers, but have not discussed any other profiles with them. I shall raise it.

Best regards, Mark

[Elliot Puritz Silver Member Posts: 77]

Re: New Canson Infinity Arches Paper

Reply #4 on: July 29, 2021 at 1:35 pm

It is fortunate that we have access to comments from two of the best informed and recognized experts regarding digital printings and the use of ICC profiles!

IMO Mark is completely correct in his comments about using the appropriate blend mode

Please allow me to ask a question:  When soft proofing using ICC profiles completed using the M3 setting I notice that checking the “simulate paper” box does NOT make the print look muddy and terrible on the monitor.  Indeed, checking the box using most other manufacturer’s profiles makes it difficult to match luminance and color on the monitor to the print.

Another question:  I note that many skilled digital printers are adamant in their ability to exactly match the monitor to the print when using a D50 print booth.  I admit to not having the same skill and wonder if it is ever possible to be completely confident that concordance between the appearance of an image on the screen and on a print held in the hand is possible….

Elliot

[Mark D Segal Silver Member Posts: 951]

Re: New Canson Infinity Arches Paper

Reply #5 on: July 29, 2021 at 2:01 pm

The main thing about the Blend Mode is to prevent hue distortion when making what is intended to be only a tone adjustment, and that’s fine, but whether you use Normal or Luminosity Blend mode isn’t germane to the main purpose of steepening the curve.

On your first question, the simulation uses a table in the profile. An M3 profile in and of itself contains a wider gamut and deeper Black point (closer to a luster paper or your display) than say an M1 profile particularly with matte papers. So that’s why you see little difference whether that Simulate box is checked or unchecked. But seen without polarizing, the print will more resemble M1 condition, so that’s why softproofing a matte paper using an M3 profile is unreliable. Now, if you were to use a standard, say, M1 profile, it will give you a better prediction of what your print will look like when seen under normal viewing conditions. With that simulation on display, you can then quite reliably adjust your image to achieve the best you can get out of those printing conditions, and the print will resemble what you see on your display if your colour management set-up is quite correct.

On your second question, yes a D50 viewing booth is the best way to achieve the most apparently accurate match between the print appearance and the display appearance when your display is also calibrated for D50 and the correct brightness level. Notice I used the words “apparently” and “appearance” here, because once we start looking at prints, we are leaving the world of numbers and graphs, which computer programs calculate with merciless consistency, and we enter the world of human visual perception and its merciless inaccuracy and inconsistency. When I hear people talk about “exact” matches, my antennae immediately perk-up. There is a world of commercial photography in which accuracy is expected and achieved – for example, the reproduction of corporate logos, shades of paint, shades of textiles. Many companies insist that the colours produced under various press conditions match identically to the official images and specifications of their materials. And this is all achieved using top-of-the-line profiles and measuring instruments, not visual representation. But once we get into other kinds of photography – say portraits, landscapes and the like, you can stick-up a print in a viewing booth and have the same photo on your display, and experienced viewers with good eyesight will be able to see whether they correspond reasonably well, but to claim exactitude I think may be in most cases an over-reach.

[Author]

Posts