converting from ProPhoto RGB to Adobe RGB

David wrote:

I use photoshop at home with ProPhoto RGB and only print at my art school; they work only with Adobe RGB for color management. The results are often too dark and have too much contrast.

Question 1: is these darkness and contrast because of the different color spaces?

When I edit/convert-profile from ProPhoto RGB to Adobe RGB – the photo is becoming brighter, the histogram is clearly shifted to the right. I tried all the intents.

Question 2: why does it happen?

Thanks for any help.

The color spaces is not the problem. Most likely its caused by your monitor not being calibrated and/or too bright. From the factor most monitors are set too bright. This makes the non photographer happy as they get bright shiny colors. But to a photographer it means that what they show on the screen can never be matched on paper. Turn the brightness down on your monitor until it matches the prints. Then adjust the brightness of the photograph to were you like it with your software. When you print it again the photo should be closer to what you see on the screen.

This is the most common reason for too dark prints.

Hope it helps

John Passaneau

My monitor is calibrated but indeed many people complaint that it (Dell 2209wa) is too bright.

You answered the first question. I’ll be happy to get an answer to my second problem.

On Tue, 26 Jan 2010 01:30:02 -0800 (PST), David wrote:

[first question already answered]

When I edit/convert-profile from ProPhoto RGB to Adobe RGB – the photo is becoming brighter, the histogram is clearly shifted to the right. I tried all the intents.

Question 2: why does it happen?

Well this turns out to be a very interesting answer. (Whether it’s correct in your actual case or not is another story 🙂

Re the histogram shifting:
ProPhoto RGB has a gamma of 1.8, and Adobe RGB is 2.2. The difference in histograms is as expected. A larger gamma value requires larger channel values to achieve the same visual brightness, so the histogram would move to the right. But this does not explain the increase in apparent brightness that you are seeing. Normally that would stay the same in a profile conversion.

Here’s a possibility re the increase in visual brightness: The following explanation may apply if you are dealing with very saturated colors. On most monitors, in ProPhoto, Photoshop must "clip" saturated colors near the primaries, setting them to the purest red, green, or blue that your monitor can display. Converting to a smaller color space results in a smooth scaling of RGB values, in effect adding white to some of the pixels, and the result is a brighter image.

If you are as curious as I was, or more so, try the following experiment:

1) Create an empty ProPhoto RGB image about 256×256 pixels in size.
2) Use the gradient tool to draw a "rainbow" gradient.
3) Dupe the image, and name it "Adobe RGB"
4) Go to Edit>Convert to Profile, and select Adobe RGB. (sRGB will show the effect more clearly)
5) Toggle the Preview button to see the lightening of colors, particularly in deep blues.
6) Press Print Screen, or use your favorite method of screen capture
7) Paste the screen image into a new image, zoom to 100%, and use the info
palette to compare the screen RGB values from both images numerically

Exercise for the reader: try the same experiment converting to lab mode, and notice the very dark band in the middle of the blue stripe. Cool, huh? Thought Lab was lossles, din’cha?, LOL.

Now back to our regular program. Let us know what the answer actually turns out to be.
—
Mike Russell – http://www.curvemeister.com

[first question already answered]

Re, my first question and John’s answer, I understand why the print is darker but sometimes it also has more contrast: the dark areas are darker and the bright areas are brighter.

When I edit/convert-profile from  ProPhoto RGB to Adobe RGB – the photo is becoming brighter, the histogram is clearly shifted to the right. I tried all the intents.

Question 2: why does it happen?

Well this turns out to be a very interesting answer.  (Whether it’s correct in your actual case or not is another story 🙂

Re the histogram shifting:
ProPhoto RGB has a gamma of 1.8, and Adobe RGB is 2.2.  The difference in histograms is as expected.  A larger gamma value requires larger channel values to achieve the same visual brightness, so the histogram would move to the right.

I didn’t know it although I read a lot about color management. Where in the web can I read about this stuff? (of course I believe you).

 But this does not explain the increase in apparent
brightness that you are seeing.  Normally that would stay the same in a profile conversion.

The brightness is increased but very slightly – much less that what I expected from the histogram shift.

Here’s a possibility re the increase in visual brightness: The following explanation may apply if you are dealing with very saturated colors.  On most monitors, in ProPhoto, Photoshop must "clip" saturated colors near the primaries, setting them to the purest red, green, or blue that your monitor can display.  Converting to a smaller color space results in a smooth scaling of RGB values, in effect adding white to some of the pixels, and the result is a brighter image.

If you are as curious as I was, or more so, try the following experiment:
1) Create an empty ProPhoto RGB image about 256×256 pixels in size.
2) Use the gradient tool to draw a "rainbow" gradient.
3) Dupe the image, and name it "Adobe RGB"
4) Go to Edit>Convert to Profile, and select Adobe RGB. (sRGB will show the effect more clearly)
5) Toggle the Preview button to see the lightening of colors, particularly in deep blues.
6) Press Print Screen, or use your favorite method of screen capture
7) Paste the screen image into a new image, zoom to 100%, and use the info
palette to compare the screen RGB values from both images numerically

The images are indeed different! I feel that I can’t trust the ground on which I stand.

BTW, I firstly noticed the problem with an RGB photo with a B&W adjustment layer, which remained in RGB color model.
Must I convert it into gray color model for printing in order to get better result?

Exercise for the reader: try the same experiment converting to lab mode, and notice the very dark band in the middle of the blue stripe.  Cool, huh? Thought Lab was lossles, din’cha?, LOL.

Now back to our regular program.  Let us know what the answer actually turns out to be.
—
Mike Russell -http://www.curvemeister.com

On Tue, 26 Jan 2010 11:16:04 -0800 (PST), David wrote:

[re brightness changes on converting from ProPhoto RGB to Adobe RGB]

The images are indeed different! I feel that I can’t trust the ground on which I stand.

Luckily, this is usually such a small change that it is negligible.

BTW, I firstly noticed the problem with an RGB photo with a B&W adjustment layer, which remained in RGB color model.
Must I convert it into gray color model for printing in order to get better result?

No need for that, however if you flatten the adjustment layer before converting from ProPhoto RGB to Adobe, you will probably see little or no change in brightness. This is caused by the shift in RGB channel data, relative to the adjustment layer settings, that happens when you change gamma.
—
Mike Russell – http://www.curvemeister.com

Well, I found today that the print workflow is based on sRGB and not AdobeRGB (or ProfphotoRGB). Is this the way a good printing process should be?

On Thu, 28 Jan 2010 11:54:14 -0800 (PST), David wrote:

Well, I found today that the print workflow is based on sRGB and not AdobeRGB (or ProfphotoRGB). Is this the way a good printing process should be?

Absolutely. SRGB and Adobe RGB are both suitable for high quality work. ProPhoto RGB can cause problems, as we’ve seen in this thread. —
Mike Russell – http://www.curvemeister.com

I am confused. Do you say that the practice is not to use Prophoto RGB at all even though it is a bigger color space, or to use it only until we send it to print.

On Fri, 29 Jan 2010 23:39:49 -0800 (PST), David wrote:

I am confused. Do you say that the practice is not to use Prophoto RGB at all even though it is a bigger color space, or to use it only until we send it to print.

Unfortunately, at this state of the game, there’s no single best practice to point to in this regard. It all depends on your preferences, what the people you look up to recommend, and – eventually, after you trust your own judgement – what you’ve found out works best for your situation.

Broadly speaking, the majority of professional color practitioners (as Dan Margulis calls them) use either Adobe RGB or sRGB as their primary color space. A certain number of excellent photographers use ProPhoto RGB as their main working space, and feel that they get the best results that way.

I believe the advantages of ProPhoto are overblown, but excellent results are obtainable using that color spac. There are many who believe it is not even a close call – ProPhoto is better, period. I respect their recommendation but do not share it because I have not seen specific photographs where ProPhoto provides an advantage for a specific set of color corrections. The distinction between raw and jpeg capture, OTOH, does have examples where the raw image can give superior results.

If you do decide to work in ProPhoto RGB, then keep in mind that you’ll need to be careful when converting down to the smaller color spaces. You’ll also have to work in 16 bits while in that color space.

There are also some surprising stumbling blocks, one of which is the reason this thread was started. More color gamut is not always more better, because of problems with shrinking the gamut down again for publishing on the web, or printing. For this reason, I recommend using a more standard color space, such as sRGB or Adobe RGB.
—
Mike Russell – http://www.curvemeister.com

Mike Russell wrote:

I believe the advantages of ProPhoto are overblown, but excellent results are obtainable using that color spac. There are many who believe it is not even a close call – ProPhoto is better, period. I respect their recommendation but do not share it because I have not seen specific photographs where ProPhoto provides an advantage for a specific set of color corrections.

I’m not sure there is anybody who claims that ProPhotoRGB has an advantage for color *corrections*. That is not the point. The point people make about ProPhotoRGB is that it’s the only color space that is large enough to hold *all* the colors that can be printed by modern inkjet printers. These printers can print colors that are *way* out of gamut for sRGB. Some colors are even out of gamut for AdobeRGB. If you want to print all the colors that such a printer can possibly print, you simply *have to* start with a large gamut RGB space like ProPhotoRGB.

Of course, your photo needs to have these colors in the first place, and that is where the theory may be different from the practical situation. The average landscape image for example, probably doesn’t contain any of these super saturated colors that are even out of gamut for AdobeRGB (a sunset image with lots of saturated yellows might, however). That’s why I also do not use ProPhotoRGB for most of my images, but the theory is completely sound and doesn’t have anything to do with color corrections. It applies even if you use the image straight out of your RAW converter and do not do any editting in Photoshop.

—
Johan W. Elzenga johan<<at>>johanfoto.nl Editor / Photographer http://www.johanfoto.com

On Sat, 30 Jan 2010 17:56:47 +0100, Johan W. Elzenga wrote:

Mike Russell wrote:

I believe the advantages of ProPhoto are overblown, but excellent results are obtainable using that color spac. There are many who believe it is not even a close call – ProPhoto is better, period. I respect their recommendation but do not share it because I have not seen specific photographs where ProPhoto provides an advantage for a specific set of color corrections.

I’m not sure there is anybody who claims that ProPhotoRGB has an advantage for color *corrections*. That is not the point. The point people make about ProPhotoRGB is that it’s the only color space that is large enough to hold *all* the colors that can be printed by modern inkjet printers. These printers can print colors that are *way* out of gamut for sRGB. Some colors are even out of gamut for AdobeRGB. If you want to print all the colors that such a printer can possibly print, you simply *have to* start with a large gamut RGB space like ProPhotoRGB.
Of course, your photo needs to have these colors in the first place, and that is where the theory may be different from the practical situation. The average landscape image for example, probably doesn’t contain any of these super saturated colors that are even out of gamut for AdobeRGB (a sunset image with lots of saturated yellows might, however). That’s why I also do not use ProPhotoRGB for most of my images, but the theory is completely sound and doesn’t have anything to do with color corrections. It applies even if you use the image straight out of your RAW converter and do not do any editting in Photoshop.

Good point, Johan – the theoretical advantages of a larger space are not limited to color correction alone but to original capture. Even there, though, I have not seen an example of a yellow, or other, color, captured in a camera, that disappears when you convert to Adobe RGB.

Certainly you have demonstrated great care in the recreation of original colors. I still remember how happy you were, years ago, when you discovered you could recover the color of some terra cotta roof tiles by wetting the paper after printing, but I never heard the eventual outcome – did you capture the color in print?
—
Mike Russell – http://www.curvemeister.com

Mike Russell wrote:

I’m not sure there is anybody who claims that ProPhotoRGB has an advantage for color *corrections*. That is not the point. The point people make about ProPhotoRGB is that it’s the only color space that is large enough to hold *all* the colors that can be printed by modern inkjet printers. These printers can print colors that are *way* out of gamut for sRGB. Some colors are even out of gamut for AdobeRGB. If you want to print all the colors that such a printer can possibly print, you simply *have to* start with a large gamut RGB space like ProPhotoRGB.
Of course, your photo needs to have these colors in the first place, and that is where the theory may be different from the practical situation. The average landscape image for example, probably doesn’t contain any of these super saturated colors that are even out of gamut for AdobeRGB (a sunset image with lots of saturated yellows might, however). That’s why I also do not use ProPhotoRGB for most of my images, but the theory is completely sound and doesn’t have anything to do with color corrections. It applies even if you use the image straight out of your RAW converter and do not do any editting in Photoshop.

Good point, Johan – the theoretical advantages of a larger space are not limited to color correction alone but to original capture. Even there, though, I have not seen an example of a yellow, or other, color, captured in a camera, that disappears when you convert to Adobe RGB.

Maybe you haven’t seen it, but that doesn’t mean it’s not possible. If you try to see it on a computer monitor you won’t indeed. But that is because the computer monitor itself does not have a larger color space than AdobeRGB, so you couldn’t see those colors to begin with.

It all depends on the camera and on the subject. These websites clearly show that even older cameras like the Canon EOS 20D and 1Ds Mark II did already have a significantly larger color space than AdobeRGB, so they *can* capture colors which are out of gamut for AdobeRGB: http://www.luminous-landscape.com/tutorials/prophoto-rgb.sht ml http://www.outbackphoto.com/color_management/cm_06/essay.htm l

Of course, you could say this is all theory and that even if a camera *can* capture colors that are out of gamut for AdobeRGB, it *will* not do so because those colors do not exist in normal shooting situations. That’s not true either. It does happen in real life:
http://www.naturephotographers.net/articles1203/mh1203-1.htm l

Certainly you have demonstrated great care in the recreation of original colors. I still remember how happy you were, years ago, when you discovered you could recover the color of some terra cotta roof tiles by wetting the paper after printing, but I never heard the eventual outcome – did you capture the color in print?

Huh? I think you mistake me for someone else. I have no idea what you are talking about.

—
Johan W. Elzenga johan<<at>>johanfoto.nl Editor / Photographer http://www.johanfoto.com

On Sun, 31 Jan 2010 01:30:45 +0100, Johan W. Elzenga wrote:

….

Huh? I think you mistake me for someone else. I have no idea what you are talking about.

Right – must have been another Johan. I’ll therefore withdraw my assertion, LOL.

Your other points are very interesting, and I’ll give the articles you refer to a careful read.
—
Mike Russell – http://www.curvemeister.com