03-12-2011, 10:28 AM
(This post was last modified: 03-12-2011, 10:29 AM by mst.)
[quote name='genotypewriter' timestamp='1299897785' post='6683']
Dcraw won't evolve by itself if you stick to a single version like I said and not install the new versions.
You also will only need to upgrade the dng converter when you're upgrading the camera used for the test, which is very infrequently. Either way, the quality of the raw files will be determined by the raw converter, draw, the dng converter won't change the quality.
[/quote]
As I said, this doesn't change anything. DNG is not the simple and fixed format many believe it is. Especially when using the compatibility option for "ACR 2.4 or later", RAW files of some cameras are converted to linear DNG (even if you unchecked this option, since some sensor RAW formats simply cannot be reproduced in DNG RAW). Linear DNG means the RAW file is run through the demosaic algorithm of the DNG converter ... which uses the same routines as ACR with the same version number. So, with newer DNG converters, you're bound to use the evolved demosaic algorithms. With different quality in different versions.
The only other option would be to embed the original RAW data into the DNG, but then of course one would need a RAW converter that knows and understands this data. Which, for a rather new camera, means you'd have to use a current converter. And of course it would be way more simple to use the initial RAW file in the first place, instead of embedding it in another format.
In addition, DNG itself is a constantly evolving format.
So, problem not solved.
-- Markus
Editor
opticallimits.com
genotypewriter
Unregistered
[quote name='mst' timestamp='1299925705' post='6686']
As I said, this doesn't change anything. DNG is not the simple and fixed format many believe it is. Especially when using the compatibility option for "ACR 2.4 or later", RAW files of some cameras are converted to linear DNG (even if you unchecked this option, since some sensor RAW formats simply cannot be reproduced in DNG RAW). Linear DNG means the RAW file is run through the demosaic algorithm of the DNG converter ... which uses the same routines as ACR with the same version number. So, with newer DNG converters, you're bound to use the evolved demosaic algorithms. With different quality in different versions.
[/quote]
Sorry but I find your reasoning hilarious in this context... even if what you're saying is true, there are only a handful of sensors out there that are different from the usual Bayer RGGB CFA design. E.g. Fuji Super CCD, Foveon, Achromatic+, Kodak CMY (DCS 620/720x), CYGM sensors from the late 90s, etc. The point being, unless you're going to use these unusual or very out-dated sensors for your tests, you have little to worry about with DNG v1.1.
Also it's so easy to check whether the RAW has been demosaic'ed... just look at the file sizes! lol
GTW
Among the affected cameras are current Panasonic cameras. Hilarious, yeah ...
-- Markus
Editor
opticallimits.com
genotypewriter
Unregistered
[quote name='mst' timestamp='1300256369' post='6854']
Among the affected cameras are current Panasonic cameras. Hilarious, yeah ...
[/quote]
I see you're not seeing the point again... the suggestion to use DNG was to minimise differences... not to eliminate them. Can't see how converting things to DNG is going to account for all the differences in sensors and internal data processing.
And even if Panasonic's affected, what's the worst outcome? You'll have to do it slightly differently for FourThirds and microFourThirds only. At the moment, aren't you guys using Apple Aperture for Canon and Adobe Camera Raw for Nikon or something like that? Is there some benefit in maximising the differences that I'm not quite seeing here?
GTW
03-16-2011, 11:41 PM
(This post was last modified: 03-16-2011, 11:46 PM by mst.)
[quote name='genotypewriter' timestamp='1300312293' post='6888']
the suggestion to use DNG was to minimise differences... not to eliminate them.[/quote]
Than why bother at all? There'd still be differences (check Klaus' post #9 again), so we wouldn't gain anything.
[quote name='genotypewriter' timestamp='1300312293' post='6888']At the moment, aren't you guys using Apple Aperture for Canon and Adobe Camera Raw for Nikon or something like that?[/quote]
Well, yes ... "something like that". It's ACR for the old Nikon DX reviews, most of Klaus' reviews (AFAIK), LR for m4/3, Raw Developer for Nikon FX, the upcoming D7000 based DX and M9 reviews.
[quote name='genotypewriter' timestamp='1300312293' post='6888']
Is there some benefit in maximising the differences that I'm not quite seeing here?
[/quote]
We don't maximise the differences. We use the tools that are available and get the job done at the time we start the reviews for a new system. There may be years between those decisions. For reasons already explained we stick to a given tool once chosen.
I chose Raw Developer over Adobe because it has a very useful batch mode that makes things a lot easier during analysis (that's the benefit). I'm sure Klaus would use it, too, if it wasn't a Mac only tool.
-- Markus
Editor
opticallimits.com
genotypewriter
Unregistered
[quote name='mst' timestamp='1300318917' post='6890']
Well, yes ... "something like that". It's ACR for the old Nikon DX reviews, most of Klaus' reviews (AFAIK), LR for m4/3, Raw Developer for Nikon FX, the upcoming D7000 based DX and M9 reviews.
[/quote]
Interesting... Which RAW converter did you use for the 8MP Canon APS-C tests and which one do you use now for the newer 15MP Canon APS-C tests?
GTW
[quote name='genotypewriter' timestamp='1300324121' post='6891']
Interesting... Which RAW converter did you use for the 8MP Canon APS-C tests and which one do you use now for the newer 15MP Canon APS-C tests?
[/quote]
To my knowledge they are both ACR-based, but with different versions. However, I don't do these tests, so I don't know for sure.
-- Markus
Editor
opticallimits.com
genotypewriter
Unregistered
[quote name='mst' timestamp='1301640775' post='7277']
To my knowledge they are both ACR-based, but with different versions. However, I don't do these tests, so I don't know for sure.
-- Markus
[/quote]
Can you be a bit more specific with the finger pointing so I can sleep peacefully knowing who compared the 70-200 f4L IS's resolution on the 15MP 50D to the 8MP 350D while using different RAW converters that, as you say, are likely to produce incomparable resolution numbers?
GTW
[quote name='genotypewriter' timestamp='1302150002' post='7460']
Can you be a bit more specific with the finger pointing[/quote]
http://www.opticallimits.com/impressum
[quote name='genotypewriter' timestamp='1302150002' post='7460']
so I can sleep peacefully knowing who compared the 70-200 f4L IS's resolution on the 15MP 50D to the 8MP 350D while using different RAW converters that, as you say, are likely to produce incomparable resolution numbers?
[/quote]
As already explained, the reason for incomparable results is not only the different RAW converters used. There are other reasons, too. And the reasons to use different RAW converters have been explained, too.
-- Markus
Editor
opticallimits.com
|
