I am sometimes, by web color management. I just posted an image on the Landscape Gallery (http://www.naturephotographers.net/imagecritique/ic.cgi?a=vp&pr=181545&CGISESSID=7493c8cd55e0840f26108d82ce959c60&u=31464) and the way people are seeing it is all over the map (whats new). On my monitor it is muted and color correct. On others it is obvious it is not. I am using a new wider gamut Dell ultra-sharp (I have found to be a pretty darn good monitor) calibrated in its "Standard" gamut mode to 130 candelas m2 which should be ideal for desktop LCD (120 is recommended by Spyder but most monitors out there are too bright by default so 130 to even 140 is more appropriate for web according to some gurus) and viewing the image through FIREFOX and the issues don't show themselves in PS or Firefox on MY system. On my system in IE it looks like cra*. I have tried processing for web in the monitors SRGB mode and its Adobe RGB mode (calibrated of course) still with mixed results. I'm no color management dummy, but am just trying to figure out how to hit that "middle of a very wide barn door" (as Bruce Fraser used to say).
Suggestions from color management savvy folks is always helpful, and thank you very much for taking time to improve my life and lower my blood pressure!
Posted by Ron Bielefeld on October 4, 2011 at 3:32 PM
Mark, I feel your pain. The biggest problem IMO is that a high percentage of people don't calibrate their monitors. They are critiquing images on uncalibrated monitors that are almost always too bright and with a less than stellar color gamut width. I would love to tell you how you can process your images so that the majority of people viewing them will see them close to the way you see them, but I don't think that is possible. There are just too many variables, of which you mention several. I think all you can do is what you are doing. Process your images on a good, calibrated monitor to your liking. I use an NEC MultiSync 2490WUXi that is calibrated with a Spyder 3 Elite. When people comment on my images about the whites being to bright or some other similar issue I usually try and find out if they are viewing on a good calibrated monitor. Most of the time they are not. Sorry I could not help you, but I hope it makes you feel a bit better knowing another person feels your pain and is dealing with similar issues.
Posted by David Bostedo on October 4, 2011 at 6:10 PM
Two questions :
What color space did you actually post the image in?
And I believe 120cd is still on the high end for monitors. Some things I've seen recommend going lower than that - like 100. You're right that most monitors are too bright, but did you mean that 140 might be better than 120?
Ron thank you for taking time. I am crazy enough to calibrate daily!
What do you think about the monitor MODES. This one has Adobe, Srgb, Standard (the default) Game, Warm and Cool. I would never use game, warm or cool, but I have tested the Adobe and it is really close to full Adobe RGB. That is what I use when I do the prints (custom profiled and soft-proofed in PS and work on them for FujiFlex Lambda, Lightjet or Chromira). It kind of makes sense to ideally (if you can) do your web stuff separately for web and in SRGB all the way through, but I am not sure what gamut mode the monitor should best be in? I tried Srgb but more recently went back to "Standard" since most will be in that anyways (but myself calibrated).
David I ALWAYS convert to SRGB and embed. And the gurus I respect say 140 for web and the Spyder equipment says 120 so I'll error off spyder ever so slightly brighter since monitors trend that way... I think 130 would be better than 120...
Thanks for your input!
Posted by David Bostedo on October 4, 2011 at 11:33 PM
The reason I asked about screen brightness is because the number one complaint with prints is that they're too dark. This is often because the screen is too bright. Calibration software used to recommend closer to 100 before LCDs got so popular. Most LCDs won't operate well at that low a brightness level, hence the 120 target.
However, that's the recommendation for screen-to-print matching. I'm not sure if that's what you're talking about here. If you're trying to make an image that looks good to as many people on the net as possible, then perhaps brighter is better.
Anyway, given the number of folks that have commented on odd colors, I'm not sure what the issue is. Given the mottled nature of what people are seeing, it may be that the image actually has mottled colors like that. Maybe it's reality that's broken.
Seriously, it could be that there are some unusual colors in the capture and color management is simply exacerbating it. If you're OK with it, I'd be happy to take a look at your RAW file, and give you something from that for comparison. It might help identify if things are funny, or if the file is just unusual.
Posted by Kent Staubus on October 5, 2011 at 8:46 AM
I'm using a LaCie 324 graphics monitor and calibrate montly with a Spyder Pro. To me, the shot looks maybe a tad dark. The colors look just a tad over saturated. I don't let this kind of thing bug me too much. I know there is a lot of variation from monitor to monitor, and when images start getting compressed and color spaces changed on the internet all kinds of yucky things can happen. I've just resigned myself that nothing is perfect at this point in the internet and let it go at that.
Kent in SD
Posted by Doug Koepsel on October 5, 2011 at 2:24 PM
Mark,
Since I commented in your ESS/Landscape post, let me elaborate.
From my images taken there with John M, on the right side and foreground there is the change from the orange-tan bluffs to a beige colored plateau above the river channel. I think that the mottled cloud shadows that fall on the border between these two colors really allow the difference to be noticed. Those colors look fine to me.
What concerned me and looked abnormal was that the left side ridges and faces (that should also have the orange/tan) have a distinct purple tint. This purple even extends to the limestone of the Kaibab plateau of the far horizon. In viewing John's Arcs image and mine, I can detect a very very slight amount of puple in the back wall before the bend starts. I have not seen this extreme purple tinting that i see in your TP image on any other images posted on NPN.
I am using 64-bit IE. In comparing my edited images to those I posted on NPN, I see that IE in general gives just a little more contrast and saturation. I do not notice any color shifts. After your previous discussion of IE and FF, I downloaded Firefox to compare to IE. To my eyes on my computer, the difference was neglible.
Doug
Posted by David Bostedo on October 5, 2011 at 2:33 PM
To my eyes on my computer, the difference was neglible.
The difference between IE and Firefox (when set to properly color manage) is dependent on the system you are working with. If calibration doesn't change the defaults much, and you're viewing sRGB images on a normal gamut monitor, things will look pretty darn close.
That's in no way always the case though - wide-gamut monitors, larger changes during calibration, and images posted in non-sRGB color spaces can all make non-color managed images (IE) look very different than properly color managed images (Firefox).
So someone on a different system could potentially see a very large difference between Firefox and IE, and some people might see no difference. That's one of the reasons it would be good to have everyone running calibration/profiling and properly color managed apps. It would put everyone on a common footing.
Discussion
9 repliesSuggestions from color management savvy folks is always helpful, and thank you very much for taking time to improve my life and lower my blood pressure!
What color space did you actually post the image in?
And I believe 120cd is still on the high end for monitors. Some things I've seen recommend going lower than that - like 100. You're right that most monitors are too bright, but did you mean that 140 might be better than 120?
What do you think about the monitor MODES. This one has Adobe, Srgb, Standard (the default) Game, Warm and Cool. I would never use game, warm or cool, but I have tested the Adobe and it is really close to full Adobe RGB. That is what I use when I do the prints (custom profiled and soft-proofed in PS and work on them for FujiFlex Lambda, Lightjet or Chromira). It kind of makes sense to ideally (if you can) do your web stuff separately for web and in SRGB all the way through, but I am not sure what gamut mode the monitor should best be in? I tried Srgb but more recently went back to "Standard" since most will be in that anyways (but myself calibrated).
David I ALWAYS convert to SRGB and embed. And the gurus I respect say 140 for web and the Spyder equipment says 120 so I'll error off spyder ever so slightly brighter since monitors trend that way... I think 130 would be better than 120...
Thanks for your input!
However, that's the recommendation for screen-to-print matching. I'm not sure if that's what you're talking about here. If you're trying to make an image that looks good to as many people on the net as possible, then perhaps brighter is better.
Anyway, given the number of folks that have commented on odd colors, I'm not sure what the issue is. Given the mottled nature of what people are seeing, it may be that the image actually has mottled colors like that. Maybe it's reality that's broken.
Seriously, it could be that there are some unusual colors in the capture and color management is simply exacerbating it. If you're OK with it, I'd be happy to take a look at your RAW file, and give you something from that for comparison. It might help identify if things are funny, or if the file is just unusual.
Kent in SD
Since I commented in your ESS/Landscape post, let me elaborate.
From my images taken there with John M, on the right side and foreground there is the change from the orange-tan bluffs to a beige colored plateau above the river channel. I think that the mottled cloud shadows that fall on the border between these two colors really allow the difference to be noticed. Those colors look fine to me.
What concerned me and looked abnormal was that the left side ridges and faces (that should also have the orange/tan) have a distinct purple tint. This purple even extends to the limestone of the Kaibab plateau of the far horizon. In viewing John's Arcs image and mine, I can detect a very very slight amount of puple in the back wall before the bend starts. I have not seen this extreme purple tinting that i see in your TP image on any other images posted on NPN.
I am using 64-bit IE. In comparing my edited images to those I posted on NPN, I see that IE in general gives just a little more contrast and saturation. I do not notice any color shifts. After your previous discussion of IE and FF, I downloaded Firefox to compare to IE. To my eyes on my computer, the difference was neglible.
Doug
The difference between IE and Firefox (when set to properly color manage) is dependent on the system you are working with. If calibration doesn't change the defaults much, and you're viewing sRGB images on a normal gamut monitor, things will look pretty darn close.
That's in no way always the case though - wide-gamut monitors, larger changes during calibration, and images posted in non-sRGB color spaces can all make non-color managed images (IE) look very different than properly color managed images (Firefox).
So someone on a different system could potentially see a very large difference between Firefox and IE, and some people might see no difference. That's one of the reasons it would be good to have everyone running calibration/profiling and properly color managed apps. It would put everyone on a common footing.
Now seriously, please do yourself a favor by using Firefox.