Home Products Download Order Contacts


Subject: Re: OpenRAW article: "DNG is not the answer"!


"don't get me wrong-- standardization is a critical step and DNG is a
path to standardization. but somewhere in this discussion i've lost the
meaning of inter-vendor compatability within an open and portable architecture."

I wrote a lengthy reply to your post, but I am not going to send it. The reason is that you pose as pseudo-incompatibilities things are simply not incompatible. They are false dichotomies.

It is a bit like the lengthy responses I've given to people who assume that a common raw format stifles innovation. But there is no incompatibility between a common raw format and innovation. The onus was really on them to show that there was, not on me to show that there wasn't. They are the ones making the claims, and first they need to support their claims, which of course they can't.

Similarly with statements like these, where first you need to support your implied assertions:

"if sony embedded the algorithim to handle it's anomolous sensor configuration--inter-vendor
compatability is served."

Who says there is "the algorithm", and if there is who says Sony knows it? In fact, of course, there ISN'T "the algorithm". We already know there isn't "the algorithm" for 3-colour demosaicing, and we already know that there is no consensus that the raw converters provided by camera manufacturers are superior to others. Perhaps Sony has AN algorithm, and supplied some software to support it. But anyway ... who says 4-colour sensors are in any way anomolous?

Thomas Knoll said: "The Sony RBGE sensor would not have required an update since four color CFA sensors have been in use for a long time. In fact, most of the first generation digital cameras used four color CFA sensors, and which were already supported by Camera Raw before the Sony camera came out. The particular set of filter colors Sony used was new, but that makes absolutely no difference to DNG since it just means different numbers in the color matrix tags."

"an advanced raw converter that uses alternative algorithims for processing
Leica's no anti-alias filter might offer those in place of Leica's default

Who says that Leica has a default method for handling no AA filter? Why couldn't they simply be assuming that the industry knows those things, therefore they are OK to do what they did? (They don't ship their own software with the DMR Back). And is there any evidence that it causes problems for any raw converters anyway?

"i own the photo. therefore i prefer a portable method of owning my results,
rather than depending on any one raw processor to supply that ownership."

But it isn't a distinction between "a portable method" and "depending on any one raw processor". That is a false dichotomy. There may be 50 or more products you can use to handle your DNGs at the moment. They exist on all platforms on which C is supported. (If DNG identified the algorithms, in effect you WOULD be depending on one raw converter - as defined by those algorithms).

"we as consumers-- who like to vote with our hard earned dollars and evangelical
energies -- wish to know whether the DNG format is designed to accomodate
total self-containment and thereby inter-vendor compatability or whether
it's designed to favour processor side dependencies and a DNG arms race."

Another false dichotomy. There is no incompatibility between self-containment and processor side dependencies, (if I understand what you mean by the term). Neither is there any logic that says that processor side dependencies lead to a DNG arms race, (whatever that means).

(Once again - this argument really has nothing to do with DNG. It would apply if we standardised on NEF, or continued with multiple formats).

"is a DNG processor designed for an open architecture?"

I don't recognise the concept of a DNG processor. DNG is an interface specification.

I'll refer you back to that last statement: "DNG is an interface specification". Don't overload DNG with things outside of the scope of an interface specification. USB is an interface specification, that can be used for connect disc drives, printers, mice, and lots of other things, to computers. It doesn't provide "algorithims, JS, referrants and other forms of guidance" for implementing disc drives, printers, and mice. CompactFlash is an interface specification. It doesn't tell you how to implement a memory card. TIFF is an interface specification. It doesn't tell you how to write the TIFF-creation firmware in a camera, or the photo-editor to process it.

If USB told people how to implement disc drives, it would be a failure as an interface specification, because it would probably stifle innovation, and probably not provide inter-vendor compatability. But it is then necessary to live with a harsh fact of life - the ideal interface specification requires people on both sides of it to do REAL work in supporting the data flow across the interface! They have to earn their money.

There is nothing to prevent any camera manufacturer with a novel sensor configuration making code available for raw converters to use if they want to. Bt don't blame DNG if they choose not to.


View All Messages in adobe.digital.negative

OpenRAW article: "DNG is not the answer"! =>


Copyright 2006 WatermarkFactory.com. All Rights Reserved.