Clipping levels DPP v CS4

With Raw images I have noticed in DPP (with the highlights warning turned on) that if I make sure that there is no clipping of the whites and then convert and send to CS4 the levels clipping warning will often show on the whites. Both levels are set to 255 - how come they seem to be picking up different values?

When you say "convert and send to CS4" what are you converting to?

I thought any changes made to a RAW image in DPP had no effect on the RAW image when imported into CS4. but, I dont think this is what you are doing.

Sorry, forgot to say I am converting to a 16 bit tiff.

What working color space is set in DPP Preferences. If it is WideGamutRGB and PSCS4 is converting it to sRGB, that could be the reason.
Read this: http://www.getcolorman​aged.com/color...clipw​arninglr/
It refers to Lightroom but if DPP is set to WideGamutRGB it would apply to DPP as well.

tzalman wrote in post #9156424
What working color space is set in DPP Preferences. If it is WideGamutRGB and PSCS4 is converting it to sRGB, that could be the reason.
Read this: http://www.getcolorman​aged.com/color...clipw​arninglr/
It refers to Lightroom but if DPP is set to WideGamutRGB it would apply to DPP as well.

Both are sRGB

That's only if you automatically convert to working space in PS (which I wouldn't want)

As far as I can see, the histogram (RGB tab, not Raw Tab!) in DPP does reflect the working space set:

I have been using the histogram in the RAW tab in DPP - is this not accurate?

edit: I have just checked my settings as shown above and they are exactly the same.

I have discussed this anomoly here before. At the time the conclusion was that DPP and Photoshop have slightly different interpretations of what is meant to be a standard! I have read that LightRoom has similar issues. So why is a standard colourspace not standard?

I use AdobeRGB throughout, routinely the exported 16 bit image in PS7 will be less saturated than DPP and the histogram different. It is not a big deal, as it is a simple correction to make. I trust PS7 to print what I see on the monitor and images I've uploaded both here and to my zenfolio page compare well. Of course this means that I suspect DPP is the culprit.

Have a look at this example. Although only the luminosity graph is shown from PS7, each of the RGB channels show similar slight differences to the DPP graph. This is using a processed TIFF with no changes whatsoever other than export to PS.

HOSTED PHOTO
please log in to view hosted photos in full size.

Roy C wrote in post #9156582
I have been using the histogram in the RAW tab in DPP - is this not accurate?

AFAIK, it's based on the Raw data.

Lowner wrote in post #9156617
I have discussed this anomoly here before. At the time the conclusion was that DPP and Photoshop have slightly different interpretations of what is meant to be a standard! I have read that LightRoom has similar issues. So why is a standard colourspace not standard?

The color space is, but the rendering to another color space (your monitors for instance) may be slightly different. (No more then slightly though)

Lowner wrote in post #9156617
I use AdobeRGB throughout, routinely the exported 16 bit image in PS7 will be less saturated than DPP and the histogram different. It is not a big deal, as it is a simple correction to make. I trust PS7 to print what I see on the monitor and images I've uploaded both here and to my zenfolio page compare well. Of course this means that I suspect DPP is the culprit.

DPP should not display an image very different then PS. In fact: If there is an obvious difference, something is wrong.
I use AdobeRGB as default in DPP, and never noticed a real difference.

Rene,

I know you contributed to the discussion we had when I raised the issue originally. As you can hopefully see from the screenshots the histograms are not massively different, but a difference there certainly is.

I also have read information from Adobe that stated they used different algorithms to represent colours in LR than they used in Photoshop. Hence my doubt about how "cast in tablets of stone" these colourspaces actually are.

Roy C wrote in post #9156582
I have been using the histogram in the RAW tab in DPP - is this not accurate?

edit: I have just checked my settings as shown above and they are exactly the same.

The histogram on the RAW tab is not a histogram of the output RGB image. So it definitly will not be the same as either the PS histogram or the histogram on the RGB tab which are derived from the output. OTOH, it is not exactly a histogram of the RAW data either (which might have been a more likely assumption). Move the sliders on the RAW tab and you will see that two of them affect the shape and position of the histogram, Brightness and WB. The inference is that this is image data after demoasicing, gamma correction, and WB adjustment. It also seems likely that it is a Luminance histogram calculated according to the formula 59%G/30%R/11%B.

Originally Posted by Lowner

HTTP response: NOT FOUND | MIME changed to 'text/html'

I have discussed this anomoly here before. At the time the conclusion was that DPP and Photoshop have slightly different interpretations of what is meant to be a standard! I have read that LightRoom has similar issues. So why is a standard colourspace not standard?

The color space is, but the rendering to another color space (your monitors for instance) may be slightly different. (No more then slightly though)

DPP uses a unique variety of sRGB all its own. Go into the folder C:/Programs/Canon/DPP/​icc and you will find there a creature called sRGB v1.31(Canon). Some time ago I did a comparison of this profile and the standard sRGB IEC61966-2.1 with the MS Color Panel applet and found a slight difference. IIRC one of them was a bit bigger in the blue-cyan area, but I don't remember which.