Photoshop CS3 color management "Save for Web" problem

>if you have not used good hardware calibrators, and I"m talking Monaco Optix Pro, Gretag EyeOne 2, Sony Artisan, Barco Calibrator

To Peter's list I would add the BlueEye if you still have a LaCie CRT.

"I will admit that I have not gone to the extreme lengths you've described."

One of the biggest problems with all this is that there are so many possible physical variables, not to mention how we all perceive color a little differently. I share the lengths I go to because it's taken about eleven years of gradual evolution to come to where I'm at now. I first started hardware calibrating monitors back in late '95 or early '96 - way before we had more than just two color spaces to worry about - RGB and CMYK and one flavor of each. As I continually strove for better screen to print matches, I learned that all of these factors had an influence on how I worked. By controlling as many of these variables as possible, and gradually understanding the digital color workflow - mostly through long mountain bike rides and back-country skiing, the puzzle finally fell into place. Not unlike learning to play a musical instrument. All of a sudden you get it.

This overall subject appears to be the one that most photographers, designers and Photoshop users seem to have the hardest time coming to grips with. It's apparent by the sheer numbers of similar threads populating this forum. On the one hand, it seems that the more you get it the easier it becomes, but it's the initial "getting it" that is so difficult. You gotta climb up the hill before you can coast down the other side. It's not helped by people contributing advice that is either confusing or just plain wrong. How is one expected to sort all of this out? Unfortunately too, the answers being sought here are not, to my knowledge really addressed directly in books like Real World Color Management or even in the Ps help menus. You sort of have to come to your own realizations after absorbing information, but have patience when doing so. It can take a while to click, but it does, you'll feel extremely empowered.

"I will say however, that the calibration image I am using appears to me to be an industry standard image (it has 4 photos: a woman with flowers and fruit, a bunch of coloured soaps, a boombox and cd, and a clock with more flowers and fruit; also has a bunch or colour bars), and it came from our printers, along with a high quality, colour accurate printed proof. I don't know what technology was used to output the proof, but it was supplied for calibration purposes."

Without knowing everything about the image and the output, there is no way to make a meaningful comment or analysis, only to say that there are, once again, many variables to consider. If you want to send me the file and the print, I'll be happy to tell you what I think.

"I personally am not seeking more than a reasonably accurate calibration...the final product is going to be viewed in daylight, tungsten, florescent...it will look different depending."

Fair enough. In that situation, I try to come to a compromise that appears as good as possible under all those situations.

must be a hard ass...

;o)

>This is why I think you're an arrogant ass.

I not sure what I can add to what yu had explained to you so clearly other than the fact you have a bad monitor profile a crappy calibrator and a cheap monitor. You don't want to hear that so I'm an ass. I give up. Good luck.

Buko: "But you have a problem and we don't."

This is why I think you're an arrogant ass. You don't have a problem, so you dismiss my problem as a creation of my own imagination or something. It's great that you don't have this problem, but I really think you should be posting useful comments rather than insults.

Buko: "thepoint you've so sadly missed..."

This must be the fourth time I'm saying this, but here goes again. I UNDERSTAND THAT EVERY ONE'S MONITOR OUT THERE WILL NOT LOOK THE SAME AS MINE. That has nothing to do with my quest to understand exactly how SFW works, or to gain a deeper understanding of colour management.

Peter Figen: I truly appreciate your effort to be detailed, much more useful I think. I will admit that I have not gone to the extreme lengths you've described. I will say however, that the calibration image I am using appears to me to be an industry standard image (it has 4 photos: a woman with flowers and fruit, a bunch of coloured soaps, a boombox and cd, and a clock with more flowers and fruit; also has a bunch or colour bars), and it came from our printers, along with a high quality, colour accurate printed proof. I don't know what technology was used to output the proof, but it was supplied for calibration purposes. I personally am not seeking more than a reasonably accurate calibration...the final product is going to be viewed in daylight, tungsten, florescent...it will look different depending.

If my monitor profile is off, I don't think it's off by very much, certainly not as drastic as the shift I'm seeing in SFW. My belief is that SFW is showing me what an sRGB image will look like when displayed untagged on my monitor. My calibrated monitor profile must be quite different than sRGB -- it's closer to Adobe RGB. But I think my monitor profile differing from sRGB does not mean it's "bad". (I know Mr. Ballard, noone said that.) I believe that the people who are not experiencing this "problem" have monitors that display accurate colours using a monitor profile that is much closer to sRGB. Perhaps the Dell monitors much be shifted much further from sRGB to display accurately.

Since I am satisfied with my monitor calibration for print purposes, and I know how to use SFW in a way that produces images as I expect them to appear on my monitor, I am basically done here. I still would like to fill in some details for my own personal enrichment, but who knows what progress I'll make there.

"Completely useless comment. What is the point of a forum, if all you're going to say is "you don't understand, and don't bother trying"? Why not clue me into why you think I can be mislead by the calibration test image? "

Here's how. You can have a monitor that is out of calibration and adjust all your images to look great on that monitor, but because you're starting with a false premise - a miscalibrated monitor, what you see on screen is not accurate and giving you a false impression of what you think you image looks like. You can then take that same image, and unknowingly, use another false output profile and actually get results that you think look great. The problem is that all you've really done is create a closed loop calibration that is only good on that monitor and that particular output device.

I don't have the time to go back and read all these nearly 300 posts to see exactly how you've calibrated your displays, but if you have not used good hardware calibrators, and I"m talking Monaco Optix Pro, Gretag EyeOne 2, Sony Artisan, Barco Calibrator - not the Huey or Colorvision Spyders, which too often give you crappy calibrations, then you are just going to be chasing your collective tails.

You have to start out with calibrating your display to a known standard, and that includes the ambient light characteristics of your editing room. That also includes not having brightly colored or even slightly colored paint on the walls. That includes blocking the light coming in from any skylight, window or doorway leading to a brightly lit area. All of these extraneous light sources and colors and affect the way you perceive images on screen and cause you to make compensatory adjustments that throw your images off. You should carefully evaluate your display calibration using industry standard test imagess and looking at how evenly neutral gradations appear and whether or not they stay neutral or shift color from light to dark. You should check to make sure that you can differentiate between the subtlest of tones - can you see all the steps between levels 250 and 255 and between 0 and 10. If you can't, your monitor is clipping highlights and shadows and not showing you what might actually be in your file.

So, how exactly is your screen calibrated? What is the ambient light level? How did you convert to your output profile and where did that come from? How are you viewing your subsequent printed output, and are you looking at inkjet or offset output? Do you have at the very least a portable Solux light to judge your output? Finally, have your taken an image that is in sRGB and used the Assign Profile command to assign Monitor RGB and toggled the preview button on and off to immediately see how different your monitor is from sRGB? Remember that sRGB is a standard based on the average of inexpensive 13 inch CRT displays, and you all are now using "modern" LCD units. They are not going to be the same as the sRGB "standard".

>Buko, you are an arrogant ass.

But you have a problem and we don't.

Raven thepoint you've so sadly missed is the one about the majority of users that will view your images have gear that is so far out of whack that it will bare no resemblance to what you have on your computer so the minute differences that I get using SFW are inconsequential in the big picture.

now for you and Peter M I have know idea what the problem is most of the people that stop by here get it on the first explanation. now we are pushing 300 posts.

Peter Figen: "...but without really understanding why it might just be coincidental..."

Completely useless comment. What is the point of a forum, if all you're going to say is "you don't understand, and don't bother trying"? Why not clue me into why you think I can be mislead by the calibration test image?

Buko, you are an arrogant ass. Another useless comment.

Yes yes I know that I can't control what images look like on other people's monitors. I also now know how to output an image in SFW that produces a predictable image on MY monitor. (Convert to Monitor RGB before saving for web, leave "convert to sRGB" unchecked.) Therefore my practical problem has been solved. However, I am still interested in gaining a better understanding of colour management and how different parts of PS handle it. You are saying something that sounds a lot like "just forget it, you don't have the capacity to understand it." What are you doing here if that's all you have to say?

Peter that's what I said in Post #21. They didn't listen then what makes you think they will listen now you've posted it in #289? B)

People are problem pushers. Off the desk and out of mind.