Can do, its on my iDisk. And it's pretty darn ugly!
File name is:
PMA2007_07March09-44.JPG
My public iDisk:
thedigitaldog
Name (lower case) public
Password (lower case) public
Public folder Password is "public" (note the first letter is NOT capitalized).
To go there via a web browser, use this URL:
http://idisk.mac.com/thedigitaldog-Public
For comparison, the image you two were working on (a DNG) is also in the root folder of my public iDisk. It's called PMA2007_07March09_064.dng
If you want to totally neutralize both, by all means do so. I didn't spend a lot of time on these images but I also got frustrated quickly trying to get the JPEG to look even close to the DNG.
As for camera generated versus non camera generated JPEGs, the differences are a few. For one, there's a specific compression that's not too far over the top and the data has (as yet) undergone no editing other than the Raw to JPEG in camera conversions. There's probably less noise than say a scanned image that's been manipulated and then converted to JPEG.
Bruce had this to say about JPEGs:
A JPEG has already had a third of its luminance data discarded, and
then been tone-mapped into a gamma-encoded space. Note that this
invariably involves more complex mapping than a simple gamma
conversion-in-camea JPEG generally has a pretty steep contrast curve
imposed, and liberties have usually been taken with the endpoints.
[a href=\"index.php?act=findpost&pid=119296\"][{POST_SNAPBACK}][/a]
Andrew,
Thanks for making the original JPEG available and for the quotation from Bruce. I downloaded the JPEG and found that any easy WB correction in ACR is impossible. I think the reason for this difficulty is that the blue channel is clipped in the midtone and shadow areas, and these clipped values are lost and beyond recovery.
This view shows the woodwork (presumably white) in the background. The blue channel has a good bell shaped appearance:
[attachment=2547:attachment]
Here is the histogram from the lady's face with the blue channel clipped:
[attachment=2546:attachment]
I think that Bruce's comments are right on and explain the problem. To illustrate the point, here is a shot that was taken in raw under tungsten illumination, but rendered into 8 bit aRGB with daylight WB and saved as a JPEG. This is an attempt to duplicate an in camera JPEG with daylight WB (as compared with the default TRC of my Nikon D200, the roll off in the shadows is less with ACR defaults than with camera defaults). The blue channel is clipped at the shadow end, even though the image is slightly overexposed as shown by the clipped red channel at the highlight end:
[attachment=2548:attachment]
And here is the same shot rendered into 16 bit aRGB with a linear tone curve and the black point set to zero and saved as PSD (so that it can't be opened in ACR--apparently once a TIFF is opened in ACR, it always opens in ACR even if you want to open it in Photoshop). There is some clipping in the blue channel, but much less than the JPEG. WB correction of the saved TIFF with ACR is relatively successful with ACR, whereas WB correction from the 8 bit JPEG is poor. The clipped blues are lost and this produce a yellow cast in the clipped areas.
[attachment=2549:attachment]