Color Management Issues

….

It seems like most people I know are getting similar results in print as what they see, which is why I thought I was supposed to buy a calibration system for my monitor, but as things are right now I get much closer results using sRGB as my default monitor profile instead of the one my Huey makes for me. I would assume based on that that it’s the Huey’s fault, except wouldn’t that affect everything that is viewed on my monitor, including all the Windows app.’s?

Anyway, I’m lost and don’t know where else to turn. Hopefully someone here has enough experience and time to help me out.

BTW, I really think my problem is very similar to what this person is trying to figure out on this thread: <http://www.adobeforums.com/webx/.59b4e6df/44>

Thanks,
Travis Minnig

Several observations:
You may be changing color profiles too frequently between AdobeRGB and sRGB and expecting that what you see in 16 bit AdobeRGB will be matched in an sRGB jpeg.
I am not sure what you mean by blue banding unless it is an out of gamut color that cannot translate from 16 bits to a lower bit count image. Also, despite being in sRGB, non color managed programs can and will display images with varying degrees of fidelity. Photoshop color management only can try to get you WYSIWYG color printing from within Photoshop and nothing else.
Despite calibration if you are using an LCD panel it can be difficult to match saturation, contrast and brightness values in the print because the LCD panel is so bright and, unless you have a dedicated graphics panel, brightness and contrast of the panel are essentially fixed. There are ways to work around this that are easiest to use if you are doing your own printing. One method is to use a modified test strip technique of applying fixed gradations of brightness/contrast adjustment until you find a setting that prints reliably.
Another method is to somehow find a CRT monitor, calibrate and use it until it dies . . .

Travis
Try reading this thread on color management. It is long but has some very good information and presented in a clear concise manor.

It may help to clear some things up for you.

< http://forums.dpreview.com/forums/read.asp?forum=1039&me ssage=27177698>

Ed

A couple things to consider:

Does your lab have a custom profile for their Lightjet? The Lightjet is NOT sRGB or any other standard working space. It needs a custom profile. Has the lab recalibrated their printer lately? That can cause color and tonal shifts. How stable is their RA-4 line? Another source for shifts.

What is the luminence level of your calibrated monitor? It’s entirely possible that it’s way too bright, causing you to adjust your images darker to look "right" on your screen. Related – what is the ambient light level in your editing room and what are your print viewing conditions?

Ideally, you should be able to use a hardware calibrated screen, adjust to suit on screen, convert to custom profile and get back a print that is a very close match to what you saw, within the limits of the paper gamut and reflectivity. If your lab can’t help you sort this out, I’d say that they are probably not qualified to earn your business, but there are a few variables to go through first.

Just out of curiosity, what lab is it?

Ed, thank you for the link. I have read this information and several of the other popular links posted on this and many other forums about color management. All of them make it seem as though what I am doing is right, but obviously I’m missing something. I appreciate the link and your time and don’t mean to come across as ungrateful, but I am at a point that I really need someone to help me disect everything until I can figure out what is up.

Peter, what is the LightJet’s color space or how does it work? The lab has asked me to use sRGB embedded files so that they should match. I asked for their printer profile so I could use them as a soft proofing option, but she didn’t know what I was talking about. So, if their printer is running on its own profile, aside from a paper profile (or is that the same thing?), then how do I get hold of it for proofing my images? When talking to the lab on Friday, the lady that runs the lab went through their process a little: she downloads a paper profile from oce.com (their distributor, I think) with every new batch of paper. Then she prints a color chart of some sort that she runs through a piece of hardware and it is supposed to calibrate the printer based on the readings done from the chart in the hardware component, if that makes sense. She does this at least twice a day and on every new batch of paper. I have no idea what the RA-4 line is or how to tell if it’s stable. Can you give me more information?

How would I tell my luminance level? I did some testing at this website: < http://www.drycreekphoto.com/Learn/Calibration/monitor_black .htm>

There are a few tests here and my monitor seems to be performing at an average or acceptable level, although admittedly it’s not what the site describes as optimal.

My monitor is a LG1932 Flatron. It is a mid-level consumer flat-screen with an analog connection to an ATI Radeon 1900XT video card. If that helps. Currently, my editing room changes some with ambient light because my studio is in a temporary location awaiting a final construction project for a new location. Anyway, the light changes throughout the day because the cheap blinds and large window. I typically only edit at night or late evening under the light from a desk lamp (tungsten). I realize this can cause some problems, but the issues I see are very dramatic, not just minor. It wouldn’t matter what environment I view my prints in either at the moment, because again they are so dramatically different that there is going to be a major discrepancy regardless. I will address this however, my viewing conditions are typically under daylight in a room with a large west-facing window.

I am using a lab called Color Pro Lab. They are local and have spent lots of hours trying to help me figure this out, so I’m not ready to bail on them just yet. The other two local professional labs were reluctant to treat me like a professional and thought they should just do their ‘it’s good enough’ printing for me; so I will not be using them any more; although they print for some of the biggest names in the portrait photography world. My current lab (Color Pro Lab) is a smaller operation, but they also generated 80% of the prints that won awards in our local PPA affiliates print competition last year (about 60% of which then merited and/or went on loan at the regional level). There are a lot of people using them and having great luck. If I pay them the extra money to finalize my images, they will turn out fine. I have been doing this. However, I want ultimate control (because I’m a control freak) and I want to print cheaper (because I’m not making much money at this as it is), so I really want to figure this whole color management thing out.

Also, if it were my monitor’s luminance or brightness or gamma, wouldn’t you think it should still look the same between different software applications? I see a really close relationship between my prints and how the images look in everything except Photoshop. Except minor problems that I could accept as being viewing conditions and the light being used while editing.

Did I answer your questions?

Thanks for your time,
Travis

I see a really close relationship between my prints and how the images

look in everything except Photoshop

In this case, I’d think about applying correct monitor profile.

Are you trying to be rude? What would you deem as the correct monitor profile? Maybe I am just confused on this point as well. I currently have my custom profile made by my calibration system applied as the default for the system. If I apply the sRGB (.icm file) as my default, then yes everything looks seems to match relatively close. But doesn’t that just throw away the purpose of the calibration system?

Thanks,
Travis

Is it possible that Windows could cause conflicts with profiles?

Also, is it possible that the ATI software that is automatically installed causes issues with the profiles?

"Peter, what is the LightJet’s color space or how does it work? The lab has asked me to use sRGB embedded files so that they should match. I asked for their printer profile so I could use them as a soft proofing option, but she didn’t know what I was talking about"

Now that’s reassuring. Would she BE the one who would know at that lab? If so, then you’re pretty much out of luck. The thing is, is that there’s not a single Lightjet printer that has a gamut the same as sRGB. Your lab may be asking for all images in sRGB because the Lightjet software is doing an on the fly conversion internally to what we’ll call Lightjet RGB. As I recall, the software has options for auto image correction or just passing the data through as it comes as well as for sharpening or not. Ideally you would want to have a file pre-converted to a custom profile for that machine and paper and have them just send that unaltered data to the printer.

"So, if their printer is running on its own profile, aside from a paper profile (or is that the same thing?), then how do I get hold of it for proofing my images? When talking to the lab on Friday, the lady that runs the lab went through their process a little: she downloads a paper profile from oce.com (their distributor, I think) with every new batch of paper. Then she prints a color chart of some sort that she runs through a piece of hardware and it is supposed to calibrate the printer based on the readings done from the chart in the hardware component, if that makes sense."

That should be the linearization that you have to do at least daily if not several times a day to compensate for machine drift. After you do the linearization, you really need to make a custom profile of the paper/printer/RA-4 processing line combination, by printing out a profiling target and reading that on a spectrophotometer and then making an ICC profile which can be used in Ps both for softproofing and for converting to final output color.

"She does this at least twice a day and on every new batch of paper. I have no idea what the RA-4 line is or how to tell if it’s stable. Can you give me more information?"

RA-4 is the type of chemical process used to develop the paper that the Lightjet prints on. It’s the same process as prints made from a color neg. The stability of that processing line is as important as anything else in the Lightjet printing process and any variation in that will show in your prints.

"How would I tell my luminance level?"

Usually, in your calibration software, you can set a candelas per meter squared (cd/m2) overall brightness level. CRT screens are usually optimal at around 90 cd/m2 and LCD screen usually in the 100-130 cd/m2 range. The reason I asked is because if you perhaps had your monitor cranked as high as it would go, and were running 200 cd/m2 or so, that might cause a problem. Also, what gamma are you using?

Can you post a screen shot of your Ps color settings? Let’s make sure there’s not something awry there.

I looked for a website for that lab, and if it’s the one in Kaysville, Ut. then they don’t appear to have one. I’m curious what they’re workflow is to get an exact screen match and what auto setting they may have set in the LJ driver.

Peter,

Thanks again for spending so much of your time to help me with this.

www.colorprolab.com is their website. Yes, it’s in Kaysville, UT. It is an extremely simple site with not much on it, but it’s there.

I will post some screen shots later this evening. Also, I re-calibrated my monitor with a black coat over my monitor and my head to try and cut out any bias from the ambient lighting. It did make some difference, but not much. Mostly, it made my images look more reddish. My prints were slightly more reddish in print than my monitor before doing this, but now the monitor is more reddish than the prints; so I will continue to work with this until I can get a more stable editing environment (in June). My monitor calibration software does not allow me to set my cd/m2. I am assuming I would need to pay a lot more money for a system that would do that?

The gamma is tough to tell as well. My monitor only has a +50/0/-50 options, and the calibration software has options for 1.8, 2.2, and 2.4. I had been using 2.2, but swiched it to 2.4 last night as it seemed to match my prints better.

I talked to my lab’s tech and she began to explain to me about a software that makes a new image file from my original in order to print from (if I understood her correctly), she loads my image file into a program from Onyx and makes a "Print Image" file. She is thinking that this may be where the custom printer profile is applied to my file. Is this true? Do you know what is going on in this process? I may have totally misunderstood, so maybe you can clear that up for me as well. Anyway, can you tell me where her custom profile would typically be saved so I can direct her to the file? That way she can email it to me and I can use it for soft-proofing.

Thanks again,
Travis

Travis:

Since you don’t have a profile for the LightJet you can’t soft-proof on your monitor. This means – for all practical purposes – you are color-correcting in the dark. If your lab keeps their process in-control by running the linearizations twice a day, you should be able to build your own profile for their machine. It might not be perfect, but it should be lots better than what you have now.

You would create a test-chart file using the image from whatever color-profile service you choose (I use Chromix). Have your lab print it via their standard process. Send the print to the profile company and they’ll send you the custom profile for that particualr printer, paper and chemistry setup. Use that as your working color-space and for soft-proofing.

This should get you WYSIWYG. Right now, you’re seeing sRGB and getting who-knows-what. You have to eliminate the unknown variable, which is the LightJet. A custom profile should do this for you.

I spoke with both Janine at the lab and Travis yesterday in a effort to understand what’s happening. Janine was very helpful, and she is the only one who touches either the Lightjet or the RA-4 chemistry. They use an Onyx RIP but she isn’t sure if there are any automatically applied profiles, although it doesn’t appear they have a spectro. The one thing she did say is that Travis’ files look dark on her calibrated CRT and they print that way too. Travis’ monitor calibration device does not let him set actual luminence levels and seeing that it is an LCD, there’s a pretty good chance that it’s too bright. I’m sending him this morning a printout of the PhotoDisc test image with added gray steps and a few of my own known images along with the file for him to compare, printed on a well calibrated and profile 9800. This should help determine if it’s the monitor that is causing his problems. Once that is settled, it’s a pretty easy move to deal with the lab and make a custom profile for their printer, but you also have to make sure that they are printing the target with no intervention. There are a lot of options when printing to a Lightjet.