Go Back   Steve's Digicams Forums > Digicam Help > General Discussion

Reply
 
Thread Tools Search this Thread
Old Apr 23, 2005, 10:38 AM   #11
Junior Member
 
Join Date: Apr 2005
Posts: 1
Default

:idea:

It amazes me how eager people are when it comes to having an opportunity to beat up on Nikon. Is Nikon really that big of a threat to them? So much commotion without much factual information – it seems everyone is an expert (or a wan-a-be expert).

After this all flushes out, most of those individuals slamming Nikon will be eating crow. (Nope – I do not own a Nikon).

The fact is, via Nikons FREE developer software kit (DSK), the encryption issue is not an issue at all, for those software developers that choose to use it. All in all, digital is not just about the lens and CCD/CMOS sensor. It also has a lot to do with how that information is processed and interpreted. All Nikon is trying to do is preserve the integrity of their propriety NEF file system, subsequently, preserving the integrity of the photographers original image. Maybe this is about the photographer after all.
poohba is offline   Reply With Quote
Old Apr 23, 2005, 11:04 AM   #12
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

poohba wrote:
Quote:
The fact is, via Nikons FREE developer software kit (DSK), the encryption issue is not an issue at all, for those software developers that choose to use it.
Sure it's an issue....

Based on comments I've seen, if you use the Nikon SDK, you don't get to choose how the RAW file isconverted. It's more like a "black box" approach (.NEF in, TIFF or JPEG out).

Also, the SDK is for Windows. Not everyone runs Windows. How about Linux users, OS X users, etc.?

Look, I'm no expert in what it takes to process RAW data, but it's complex enough that there are a number of approaches to it.

None of them are going to be perfect (especially with Bayer sensors, since each photosite is only sensitive to one color due to the Color Filter Array). You've also got demosaic algorithms, etc., incorporated.

Here is one study of different types of color interpolation algorithms (and there are many more papers and studies on this subject):

http://www-ise.stanford.edu/~tingchen/

That's probably one reason why some RAW conversion software works better than others. Different ways of interpreting this information.

Do you really want only Nikon's way of interpreting the data to be used in software?

It would not surprise me to see some significant advances in how RAW image data is processed in the coming years.

We may see RAW conversion algorithms get very sophisicated, perhaps even to a point where different algorithms are used for different subject types and conditions based on subject recognition to choose how portions of an image are converted from RAW.

But, if third party developers are forced to use a manufacturer's way of converting from RAW, then innovation is stifled.


Quote:
All in all, digital is not just about the lens and CCD/CMOS sensor. It also has a lot to do with how that information is processed and interpreted.
That's exactly the point. IMO, users should be able to process this data in a different manner if they choose to do so.

Quote:
All Nikon is trying to do is preserve the integrity of their propriety NEF file system, subsequently, preserving the integrity of the photographers original image.

I think you mean Nikon's interpretation of the integrity of the image. ;-)

With their SDK, they're deciding how the image is processed. They're deciding if noise is smoothed out as part of the conversion process. They're deciding how to deal withaliasing, etc.

I've seen a number of complaints about Nikon's raw decoder algorithms already.

For example, some users say that they mess up the red and blue scalar values (red and blue scalars are part of the white balance information), making highlights that aren't actually blown in the raw file turn strange colors after processing by Nikon's algorithms.

Another example, Eric at Bibble Labs created a blown highlight recovery algorithm forskin designed tosuccessfully recover from overexposure. Nikon's algorithms are reportedby some users to simply clip this infromation away, leading to blotchy skin highlights when they are blown.

The even bigger issue is who owns the data. Is it Nikon's or is it the photographer's?

If you think Nikon's conversion algorithms are superior, fine, use them.

However, I think Nikon should let the market decide (versus trying to prevent third party manufacturers from legally accessing the information contained in the RAW images, unless they are willing to convert them "Nikon's way", using Nikon's SDK).



JimC is offline   Reply With Quote
Old Apr 24, 2005, 9:07 AM   #13
DBB
Senior Member
 
Join Date: Jan 2004
Posts: 2,483
Default

I have to agree with Jim C. Protecting their data is a very minor part of the process as compared to hardware, etc, etc.

While I don't want to minimise this, the fact is that you are shelling out BIG BUCK$ for a camera, you don't want to find out, AS I DID (forgive the shouting) that I wasn't getting the best possible results because making an interpreter was so difficult.

My images as SO much better with PS than with the Nikon importer that there is not even a reasonable comparison...

Meaning that thousands of my older files are not in the same league as the newer ones

Dave
DBB is offline   Reply With Quote
Old Apr 24, 2005, 10:11 AM   #14
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

Guys...

I don't even own a Nikon DSLR. I do own some Nikon cameras (35mm SLR +an old Coolpix 950 digicam thatI keep around for closeups).

But, I'm very concerned about the precedent this kind of thing could set (especially since I've been considering buying a Nikon DSLR for my wife, who still shoots with a 35mm Nikon). She prefers a Nikon SLR, so the only way I'll likely get her to switch from film is to buy a Nikon DSLR.;-)

In fact, I recently installed Linux on a laptop I bought for her (even though I'm running XP Pro). Nikon doesn't even have a Linux based solution (but others do).

Even if that were not the case (considering a Nikon DSLR for my wife), I'd still be very concerned about it (and voicing my opinion over it).

Because of the way a bayer sensor works, the colors we see are not what the sensor captured. Much of what we see in the final result is interpreted information (color interpolation). This has a lot of inherent problems.

Of course, I've seen arguments that Nikon's algorithms may be more aware of the weaker anti-aliasing filter they use, etc.

IMO, this is not the issue (who's algorithms are best). IMO, that's up to the end users to decide.

If third party developers are not allowed to legally access the information because it's encrypted (where decryption*might* be a violation of the Digital Millenium Copyright Act, which is Adobe's primary concern), then you're stifling competition and innovation.

I'm no lawyer (and I don't play one on TV). But, I don't think Nikon would win this one (any lawsuit over decrypting the information outside of their SDK).

I think the RAW images are the property of the photographer. So, IMO,by decrypting the information on behalf of the owners of the RAW image, third party developers arenot breaking the law (since the act is designed to protect copyrighted works, and prevents decrypting information in them). I just don't see the courts going along with the concept that information contained in RAW images is Nikon's. Again, that's only my opinion, and I'm not an attorney.

I can imagine that Adobe is spending a lot of time discussing this with their attorneys though.

It's not like we're asking Nikon to reveal their processing secrets (how they process and interpret the RAW data).

All RAW conversion software is not equal. For all we know, some teenage kid may come up with a totally new approach to color interpolation that increases the accuracy of the final result (if they are legally allowed to do so).

So, my message to Nikon would be to let the market decide what works best.

If I want to have my Kodak negatives processed on Fuji equipment and printed on Fujipaper, that's my choice (even if Kodak said that their processing and paper must be used for best results). If the results are worse because I'm not using a lab and process that they certify, that's my decision, not theirs.

Ditto for RAW images. That's your Digital Negative. If I want to process it differently than the manufacturer recommends, I think that should be my choice. That's the biggest issue -- who owns the RAW image. IMO, it's the photographer's, and they should be able to do what they want with it.

So, I'll continue to voice my opinion whena manufacturer takes this kind of action. If we don't, what's next? Stronger Encryption that developers may not be able to crack? Lawsuits? Third party developers avoiding this market? More restrictions on what you can do with a RAW file? Idon't like the precedent thiskind of thing could set in the industry.



JimC is offline   Reply With Quote
Old Apr 24, 2005, 5:07 PM   #15
Member
 
Join Date: Apr 2004
Posts: 76
Default

Again, this is another recent takeover from (Nikon) marketing. The first one was to introduce an extra fine JPEG compression level (http://www.stevesforums.com/forums/v...mp;forum_id=21). I own a Nikon camera and I think that the engineering section of Nikon is very good.

The question is why the encryption was introduced. The answer seems easy. To gain marketshare of the inhouse software and was just the right size for a marketing mind: "We introduce some encryption and everyone has to use our software to get the best result." Guess they were so proud of their great idea that they took everyone out for a lunch

Fact is that within a few day the encryption was cracked. Any Nikon engineer could have predict that. Sony tried it before with the same result. But its not difficult to create a good encryption.

But thats all not important. Nikon could have even used a trivial XOR encryption. Important is to make it illegal to decrypt without going through Nikon. Does it fall under DMCA ? I don't know, and that is to decide.

I think having dcraw from David Coffin is great. What else can someone use? Some SDK or libs that just work for one platform?There will always be solutions for future platforms but with all the different Raw formats out there I would not bet too high on that.

In version 1.250 ofdcraw I saw a new Makernote processing that looked like a decryption to me but at that timethe news about the Nikon flaw wasn't published.I will update the plugin forcPicture the next days.
Jurgen Eidt is offline   Reply With Quote
Old Apr 25, 2005, 8:46 AM   #16
Super Moderator
 
Hards80's Avatar
 
Join Date: Nov 2004
Location: Cleveland, OH
Posts: 9,046
Default

JimCsaid:
"Ditto for RAW images. That's your Digital Negative. If I want to process it differently than the manufacturer recommends, I think that should be my choice."

great way to put it.. i 100% agree and this pretty much sums up my opinion on the matter.. i as well do not own a nikon dslr, but am afraid of the precedence this could set in the industry.. though i hope other manufactures are watching various photography forumns to see and try to avoid the backlash that this has created..


Hards80 is offline   Reply With Quote
Old Apr 25, 2005, 2:32 PM   #17
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

It now appears that the new D50 might use this same encryption scheme... I'll post more info when I can on this issue.
JimC is offline   Reply With Quote
Old Apr 25, 2005, 10:29 PM   #18
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

An update...

It's not just Nikon (it's justthat encryption of portions of RAW files has gone largely unnoticed by most users until Adobe madeNikon's use of encryption public).

From David Coffin (the author of dcraw.c).

http://www.cybercom.net/~dcoffin/dcraw/

David Coffin wrote:

Quote:
A note about metadata encryption:

A firestorm of controversy recently erupted when Thomas Knoll of Adobe accused Nikon of encrypting the white balance data in the D2X and D2Hs cameras, thus preventing Adobe from fully supporting these cameras.

I cracked this encryption on April 15, and updated dcraw.c and parse.c on April 17. So "dcraw -w" now works correctly with all Nikon cameras.

This is not a new problem. Phase One, Sony, Foveon, and Canon all apply some form of encryption to their raw files. Dcraw decodes them all -- you can easily find decryption code by searching for the ^ operator.

Compression is not encryption. Phase One and Sony do encryption only. Kodak does compression only. Canon, Nikon, and Foveon compress the image data and encrypt some of the metadata.

JimC is offline   Reply With Quote
Old Apr 26, 2005, 12:54 AM   #19
Junior Member
 
Join Date: Apr 2005
Posts: 1
Default

I have, what I think is a good idea; let me know what you all think. Couldn't one of the super geek computer programmers write a program that would convert NEFs to DNG format. DNG is Adobes RAW format and once converted it would take the whole issue out of the hands of the camera manufactures and let us return to a sane workflow, with only a short conversion process.

George

http://www.georgedeloache.com



amanofgod is offline   Reply With Quote
Old Apr 26, 2005, 1:16 AM   #20
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

amanofgod wrote:
Quote:
I have, what I think is a good idea; let me know what you all think. Couldn't one of the super geek computer programmers write a program that would convert NEFs to DNG format. DNG is Adobes RAW format and once converted it would take the whole issue out of the hands of the camera manufactures and let us return to a sane workflow, with only a short conversion process.
That won't solve it if you can't get the file structures to use a converter. IMO, there's little benefit to converting (since it's only putting the information contained in the original file into a different format).

Adobe already offers a free converter (but it won't support the WB information in the new .nef files for the D2x, D2Hs (and it looks like that's going to be true for the new D50, also).

http://www.adobe.com/products/dng/main.html

There many flavors of .nef files (ditto for other manufacturer's RAW formats).They are moving targets. You can already convert the ones supported by Adobe if you want to go that route. The problem is when changes occur, someone has to make the same changes in the conversion programs.

If the manufacturers refuse to publish their properietaryfile structures (and even go to some lengths to make deciphering their contents difficult), then third party developers will always be playing "catch up" (or it will get to a point where developers are unwilling or unable to write the code to decipher them).

That's one of the reasons you're seeing so much about this issue now on many digital photography sites (because Nikon decided to encrypt it's White Balance Information in .nef files for some newer models, and Adobe says it can't support the White Balance decoding because Nikon is encrypting it).

That means, they can't convert them toDNGwith this information included either.

The ball is in Nikon's court. For that matter, the ball is also in other manufacturer's courts.

Countless hours are being wasted reverse engineering RAW file formats because the manufacturers aren't publishing information about them (and are going to some lengths to prevent companies like Adobe from using the information they contain).


P.S. -- here is a new site that is dedicated to this issue (open standards):

http://www.openraw.org/



JimC is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 3:26 AM.