Shoudn't OpenRAW format be created by independent developers?

Hi all! I prefer to make all raw formats open, but... shouldn't independent developers create a new, open raw format? I think that OpenRAW (as organisation) should create such format and then promote it to camera manufacturers. Until manufacturers will use it, there should be converters from current raw formats to universal OpenRAW file format (without loosing any data), and plugins to all softwares like Adobe Photoshop, GIMP.
This file format should be extensible, universal to various ccd's, etc.

Otherwise, camera manufacturers will never create one, unified file format - maybe they will use DNG, which is not open.

Regards,
Tomek

thom_ek – Tue, 2006/07/11 – 6:49am

Its probably more easy to

Its probably more easy to just document an existing format than to create a new one, which needs to be adopted by the industry. If one of the market leaders (Canon or Nikon) would start doing this, they could create a defacto standard in a very short time, both of their formats obviously are rigouros tested out in the field, so there is no need to re-invent the wheel.

Just my 2 yen,

Juergen

--
http://www.openraw.org/

Juergen Specht – Tue, 2006/07/11 – 9:46am

What independent developers?

What independent developers?

So far, there have been two attempts to devise a common raw format. ISO created TIFF/EP (ISO 12234-2), using resources drawn from the major companies worldwide involved in digital imaging. And Adobe created DNG, largely based on TIFF/EP and Adobe's own TIFF, in effect bringing TIFF/EP up to date and making it fit for purpose.

It requires knowledge in breadth and depth to devise such a standard. And there is still no guarantee that any particluar company would use it. Everyone knows that Adobe would simply release a converter from that new format to DNG, and so 3rd party software products could all safely ignore it, and just handle DNG instead.

I think the only way of stopping DNG becoming the de facto standard raw file format would have been an alliance between Nikon and Microsoft, and that hasn't happened. Nikon are busy extending NEF, and building NX on it, while Microsoft are pursuing WDP (Windows Media Photo), which doesn't have raw capability (yet).

DNG is as open as TIFF, perhaps more so. (Adobe own TIFF, of course).

Barry Pearson – Wed, 2006/07/12 – 12:28am

NEF and CR2 are unsuitable!

NEF and CR2 are unsuitable!

DNG supports conversions from about 15+ camera manufacturers, and of course a few cameras that use it as their native format. It supports not only Bayer sensors, which both Nikon and Canon use (apparently exclusively), but also 4-colour sensors, Fuji offset sensors, and (using conversion to Linear DNG) Sigma/Foveon sensors.

Neither NEF nor CR2 were designed as archival formats - they both need software to have a separate camera profile, which is why their software has to be updated when they bring out new cameras. And surely, if there is one thing OpenRAW wants, it is an archival format! DNG was designed as an archival format, which is why some software products can handle DNG files converted from the raw files of cameras they don't support directly.

And what hints have there ever been that either Canon or Nikon would fully publish their formats and make SDKs freely available? Or that if one of them did so, the other would use it? All evidence is to the contrary.

The advantage of DNG to any camera manufacturer is that Adobe don't make cameras in competition; and, in fact, Adobe are demonstrably in the business of trying to make cameras' raw images look as good as possible to photographers. (Adobe is really only a threat to camera manufacturers who are trying to make a business out of selling their own software - and that may just be Nikon).

Barry Pearson – Wed, 2006/07/12 – 12:49am

I'm offering to help with

I'm offering to help with creation of a long lasting RAW standard.

It will take quite a few areas of expertise to do so.

I have some skills at making sure the standard is comprehensive. My background includes physics, engineering and software publishing.

Cheers,
-David

D-3 – Sat, 2006/07/15 – 12:13am

How to create a raw format

How to create a raw format that is long-lasting:

1. Ensure that it will not be undermined by its competition. (This includes Adobe & perhaps Microsoft).

2. Ensure that it can cater for the requirements of many (15+) camera manufacturers, and many (120+) cameras. (Demonstrate familiarity with these).

3. Ensure that it will be recognised and considered credible by all the people and organisations concerned. (3rd party software developers; US Library of Congress; etc).

4. Ensure that all necessary tools are in place, such as converters to the format, and software development kits. (Budget for 30+ man years effort over at least 3 years).

5. Ensure that some members of ISO's Technical Committee 42 ("Photography") advocate the format.

6. It helps if a pedigree can be established. For example, is it based on Adobe's TIFF, already used as a resource by ISO? Is it based on ISO's TIFF/EP? Does it exploit Adobe's XMP, already recognised by ISO?

Barry Pearson – Mon, 2006/07/17 – 2:44am

It seems to make more sense

It seems to make more sense for someone like Adobe to propose a format than for a manufacturer to do so. I have no problem with .dng, just as long as all processing software can incorporate it without having to pay Adobe a commission. Am I missing something with .dng?

Jim Landers – Sat, 2006/07/29 – 5:31am

You might want to read this

You might want to read this article:

Notes on the future of Open RAW formats, and a look at DNG (by Stuart Nixon)

--
http://www.openraw.org/

Juergen Specht – Mon, 2006/07/31 – 7:51am

Response to Stuart Nixon

Response to Stuart Nixon article - I published this commentary on Stuart Nixon's article just after it was published, then I've edited since:
http://www.barry.pearson.name/articles/dng/commentary1.htm

Stuart Nikon does not make his case. In fact, what he claims to be flaws in DNG for the purposes of image preservation are really the well-known flaws in proprietary raw file formats, such as NEF, CR2, etc.

Of course there limits to the quality of the conversion to DNG of poorly-designed poorly-documented proprietary raw files. But poorly-designed poorly-documented proprietary raw files are not themselves suitable for image preservation. Whatever is done to improve the design and/or documentation of proprietary raw files will also improve the quality of DNG conversion and the documentation of the resultant DNGs.

All efforts should be concentrated on ensuring that we have the highest quality DNG converters, and the fullest support of DNG by software developers. Then we will have the critical mass and economies of scale to ensure that, whatever other formats are supported by future products, DNG will certainly be fully supported to a high quality.

Barry Pearson – Mon, 2006/07/31 – 5:42pm