« New Folio! Five Days of Tulips in Rain by Robert Hecht | Main | Newsweek Magazine to End Print Edition »



Walter Hawn

Very much worthwhile. Thanks. Posted this to Google+.

Terry McDonald

Very helpful visual comparison, Brooks, however...

I take issue with jpegs or lossy-dng being a solution for photographers who wish to do some post-capture processing. Once you go "down" to 8-bit, you have lost forever the extra colour bit depth that is so helpful in preventing colour artefacts when post-capture processing.

When teaching this, I use this metaphor: using 8-bit colour depth when post-capture processing is like playing basketball in a gym where the gym wall is the boundary of the basketball court - you simply cannot play as hard if you are constantly smacking up against a cement wall. On the other hand, 12-, 14- and even 16-bit is like playing with a lot of out-of-bounds space - you can play harder without fear of crashing into the wall. While 8-bit is the boundary created by printing or web use (although even fine art printers like the Epson now print at 16-bit), 12-bits+ gives you more wiggle room to edit prior to committing to 8-bit prints or jpegs.

Furthermore, if you want to re-visit an image in the future for further processing, keeping it in 12-bits (or more) keeps that extra headroom. This is crucial - I have revisited many digital images I made 6-, 8-, even 10 years ago using my improved editing skills and the improvements in apps such as Lightroom and I am glad that I shot 12-bit raw images at the time because it's like reprinting old negs - they are much, much better than what was possible back then, partly because I still have 12-bits to play with.


A couple of comments on the JPG compression issue mentioned in the video: first, the actual usage of a DNG file that is lossily made, will not itself involve further re-compression of the picture inside it, unless a lossy method is used to save the results of the Raw conversion one has made from it - which applies equally well for a non-lossy Raw.

Second, there was a lot of emphasis put on the fact of a 10th-generation JPG save (at the same quality). It has I believe been established that (aside from any areas of the picture that are altered in between) Photoshop repeats the content of a save precisely when the same JPG quality is used, once the initial "damage" is done at the first compression. The new data has accommodated itself to the compression and whatever remains of the original data, by definition fits with this compression perfectly - has already been forced through that particular sieve - so is not going to see any further loss at the next iteration of that.

So the only issue is the single impact of the DNG conversion from original Raw (when the lossy option is selected). This - as with JPG - is compression that is deliberately designed for negligible visibility.

In the same way that a JPG looks great until you try to edit its values strongly; the difference between a lossy and a non-lossy Raw will likely come only where some very heroic conversion is envisaged. So: where the conversion needs to lift every scrap of detail from unederexposed areas, say. But if the file is generally underexposed then I have read that the reduced-bit-depth values inside the lossy Raw are shifted to make the best allowance for that - it "auto-exposes" for best representation of the data found, and its own histogram will not leave a big empty gap at the right. IOW there's a degree of adaptation image-by-image, as I understand it, for something closer to the optimal outcome each time.

If my camera offered a lossy DNG format as an alternative to JPG or normal Raw (it has a DNG option there), I'd certainly consider using that from the beginning. But having already taken normal Raw, and accepted the filesize impact of that once, I see little need to then re-save smaller afterwards.

ios application

DNG is not an industry standard (as JPEG, for example). It looks one more format for raw files aggressively supported and promoted by Adobe. I prefer to stay with the camera manufacturer format yet.

On the other hand Adobe trying to impose their "standard" offers translation from camera manufacturers custom format to theirs (but not vice-versa). One more reason to stay with the camera manufacturer original format.

Brooks Jensen

You might want to take a look at the Creative Lab we posted today over at the membership site that demonstrates that there is, indeed, compounding loss in successive JPG saves, even if they are all at the same level of compression.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)