Unexpected result after converting image profile to softproof profile

[Mark D Segal]

I don't understand what this is meant to show, because neither sRGB nor "printer space" are output device profiles. As far as I know, these printers should need a properly created output device profile to render the colours correctly, assuming the upstream colour management is set right.

[digitaldog]

I don't understand what this is meant to show....

That makes at least two of us.

[Stephen Ray]

The Walgreen's tech can switch to "Printer Space" or "sRGB" space within the driver tabbed menu (very different driver interface from Walmart's). I did a print test printing an sRGB encoded image in both spaces (having the Walgreens manager set it while I watched her select each in the Fuji driver interface) and the level of saturation between the two prints was VERY PRONOUNCED meaning "Printer Space" has a much larger gamut than "sRGB". The European ICC printer profile whose gamut size is shown in the Colorsync Utility screengrab I had to convert the sRGB image to and print in "Printer Space" fixed the saturation issue.

Hi Tim,

The conditions you have described are very obvious to me and I hope I can explain what you have experienced.

The "Printer Space," (as you've designated it) does not have a wider gamut. What you are seeing is just balls-deep color; a print with no control other than basic exposure and who knows if it was too light or too dark or neutral. Did you include a gray scale in that particular file? If the situation would be compared to an inkjet as an analogy, the print had no ink restrictions, the most basic of calibration (exposure in this case), no ink limits and, as you already know, no ICC output profile. It was the setting that would be used in order to print color targets from which to produce an ICC profile.

The "sRGB Space" print utilized the machine's color lookup tables to make a good print. It's smart that way.

If you actually had imaged a test file with somewhat large, discrete patches of known colors, you would see both color spaces of the machine reach the same color gamut. One no more than the other. You just could not differentiate saturated colors at the time of your study because they were crowded together. A good ICC profile will separate and smooth those clustered colors and even neutralize what should be grays.

I hope this helps. It's clear to me what you have encountered.

[Tim Lookingbill]

Hi Tim,

The conditions you have described are very obvious to me and I hope I can explain what you have experienced.

The "Printer Space," (as you've designated it) does not have a wider gamut. What you are seeing is just balls-deep color;

How do you know that? Have you operated and profiled that particular model of Fuji Frontier DL printer? Why would a printer manufacturer offer a a color space selection that's useless? If it's not useless, what's its purpose?

[Tim Lookingbill]

I don't understand what this is meant to show, because neither sRGB nor "printer space" are output device profiles. As far as I know, these printers should need a properly created output device profile to render the colours correctly, assuming the upstream colour management is set right.

Where did I say "printer space" and sRGB are device profiles? I said it is a state to put the printer into to print a color target to build a profile from.

You're referring to optimized state of color rendering. Clearly "Printer Space" or "PD" setting on the Fuji Frontier is a setting that is used for something. I didn't say it was a profile.

Mark, do you have any idea what is the function of "Printer Space" or "PD" setting on a Fuji Frontier Dry Lab printer?

[Stephen Ray]

Why would a printer manufacturer offer a a color space selection that's useless? If it's not useless, what's its purpose?

It's the calibration mode, Man.

I just looked at your image attachment. I would say the machine is perfectly calibrated, unlike so many other machines. Perform the same exercise on your inkjet and then upload the attachment.

[Tim Lookingbill]

It's the calibration mode, Man.

I just looked at your image attachment. I would say the machine is perfectly calibrated, unlike so many other machines. Perform the same exercise on your inkjet and then upload the attachment.

I already knew it was the calibration mode, but obviously your previous statement wasn't referenced from actually looking at the print depicting the two states the printer can be set at.

Now you know "Printer Space" has a purpose and if you read my entire posts you would've read that the European ICC profile was built from that printer setting. But if you're not going to read what I said, you really aren't giving helpful info.

The Dry Creek Photo profiles were built from the sRGB setting. That compared to the European icc profile are quite different from each other.

 

[Stephen Ray]

Tim's image attachment from his post #79 is very good example of a machine that is actually calibrated to a known metric, a common gray scale, familiar to photographers. It may surprise many that a vast amount of image printers do not understand the necessity of the tool.

From Tim's upload, I can see the machine is balanced. The "Printer Space" (no ICC color management) gray scale is the same as the other "sRGB" space. In other words, the sRGB space does not need to correct for a color cast or density.

In so many other machines and workflows I encounter, I first print a control image without color management to see the state of calibration to known elements of the control. If all is good as far as density, neutrality, gradation and mechanics of the machine, I know the ICC profile has that much less to fix and then the profile can simply concentrate on optimization. On the other hand, I have seen many machines with gross problems of many sorts and it's amazing what an ICC profile can fix, such as heavy nozzle deflections contaminating the image but the ICC can neutralize the color.

[Stephen Ray]

But if you're not going to read what I said, you really aren't giving helpful info.

Tim, have you learned the two color spaces have the same gamut?

[digitaldog]

Tim, have you learned the two color spaces have the same gamut?

Learning isn’t a flower that grows in everyone garden.

[Tim Lookingbill]

And just to make my point more clear here's what assigning to sRGB data a FujiDL profile built from the sRGB printer setting renders color vs assigning the European ICC profile built from the "PD" or "Printer Space" setting.

[Stephen Ray]

And just to make my point more clear

Would you please put your point into a sentence, Tim?

[digitaldog]

Topic hijacked at post #35... 

[Tim Lookingbill]

Would you please put your point into a sentence, Tim?

OK...You're not helping! How's that?

[Stephen Ray]

I'm involved in a project of some extreme colors for reproduction using the machines mentioned. My searches found this thread. I made these photographs this morning after breakfast. Some of the profiles I find online concern me.

[Doug Gray]

I'm involved in a project of some extreme colors for reproduction using the machines mentioned. My searches found this thread. I made these photographs this morning after breakfast. Some of the profiles I find online concern me.

Yeah. They depend on the stability of the machine process and how much change has occurred since the profile was made. Quite apart from some profiles looking like they may not be well formed.

I just refined my profile target for the 9800. I make a profile then cross check it with independent colors printed in 16 bit tiffs using ProPhoto. The approach most use is to just run Abs. Col. on the target RGB values and compare against the scanned LAB values. That greatly overstates accuracy because it depends only on the A2B1 tables which are the most precise. It's a tough nut but I've finally come up with a target that produces excellent results in both the neutrals and the rest of the gamut. Neutrals come in at ave dE00 of .33 and colors: ave dE .27.  dE00 is more sensitive than dE76 on the neutrals but much less sensitive as the patches' chroma increases. The 9800 has a couple sharp gradient changes in L* and abrupt shifts on a* and b* as well. That extra page of near neutrals helped quite a bit.

It's been quite hard to get the neutrals below .5

I wound up with a target made out of the original, default single page  of 957 iSiS patches. Then added 1914 (two more pages) made out of I1Profiler's optimization process. Then I added another 957 patch page of neutrals and near neutrals with 3x redundancy on the near neutrals. The near neutrals had RGB intervals that matched the 37 point I1P grid which I speculate should provide somewhat better accuracy in the A2B1 tables which have evenly spaced RGB points.

This provides a good, 4 page, target set for my 9800. At least on glossy type media.

The general rule of thumb is an average dE76 of 2.5 in a complex image is roughly at the "perceptible" point. It's a lot lower for certain images though and I'm mostly curious just to find out what is possible.

Oh, and the other thing I do is scan the pages both forward and backwards. I make targets with reference bars on the bottom as well as the top to enable this with the iSiS. Averaging these provides significant improvements over just re-reading them in the same direction. Quite a bit more than I anticipated. I have some theories as to why based on some other tests to determine the effective scan area within a patch. That's another topic though.

[Tim Lookingbill]

I'm involved in a project of some extreme colors for reproduction using the machines mentioned. My searches found this thread. I made these photographs this morning after breakfast. Some of the profiles I find online concern me.

Check out the profiles on this Fuji Color Management page: https://www.fujifilm.eu/eu/support/photofinishing/color-management

The PD/No Convert is the one shown above but I don't know why they would say not to convert because the file will go very saturated if you send it in sRGB. Maybe convert the file to AdobeRGB and test.

Any Pantone color such as the safety yellow you posted that is that high in luminance is going to go dark no matter what custom profile you use but at least a custom one will retain the greenish hue or allow an edit during soft proof in a larger gamut working space to override the reddish shift.

You'll need to find a Frontier that has the PD setting and convert to that printer profile or maybe convert to AdobeRGB and cross your fingers.

[digitaldog]

OK...You're not helping! How's that?

Pot calling the kettle black!

[Mark D Segal]

Guys, this conversation is disintegrating. Why not check back to the beginning of the thread and see whether the O/P's initial problem has been addressed and if so, move on; if not, revert to a technical probe of his particular issue.

[Tim Lookingbill]

Guys, this conversation is disintegrating. Why not check back to the beginning of the thread and see whether the O/P's initial problem has been addressed and if so, move on; if not, revert to a technical probe of his particular issue.

The OP doesn't seem to be engaged in his own thread or even interested in offering specifics and answering questions on where he sees "horrible results" in the print or soft proof preview.

What I've posted is more relevant to the OP's issue concerning Fuji Frontier Dry Lab printers, but the OP doesn't even acknowledge what printer Costco is using.

Hey Walter! How about some feedback here for all those that tried to help you?