AWARE [SYSTEMS] Imaging expertise for the Delphi developer
AWare Systems, Imaging expertise for the Delphi developer, Home TIFF and LibTiff Mailing List Archive

LibTiff Mailing List

TIFF and LibTiff Mailing List Archive
February 2007

Previous Thread
Next Thread

Previous by Thread
Next by Thread

Previous by Date
Next by Date

Contact

The TIFF Mailing List Homepage
This list is run by Frank Warmerdam
Archive maintained by AWare Systems



Valid HTML 4.01!



Thread

2007.02.05 03:28 "Request help On OJPEG Compression In Libtiff", by Steven Lim
2007.02.05 18:19 "Re: Request help On OJPEG Compression In Libtiff", by Chris Cox
2007.02.05 19:08 "Re: Request help On OJPEG Compression In Libtiff", by Joris Van Damme
2007.02.05 19:14 "Re: Request help On OJPEG Compression In Libtiff", by Chris Cox
2007.02.05 19:29 "Re: Request help On OJPEG Compression In Libtiff", by Joris Van Damme
2007.02.05 18:49 "Re: Request help On OJPEG Compression In Libtiff", by Joris Van Damme

2007.02.05 19:29 "Re: Request help On OJPEG Compression In Libtiff", by Joris Van Damme

Chris,

Chris Cox wrote:
> That's because most (if not all) of Microsoft's implementations are
> based on the same buggy library (it has other serious problems).

I've indications that this is not the case. The newer Microsoft products 
don't seem to read the very ugly OJPEG flavours that the early KODAK-WANG 
code base wrote. Furthermore, that early code base didn't read the cleanest 
and least bad flavour that my codec currently (reluctantly) writes, and the 
newer Microsoft code does read that fine.

> We have several bugs filed with Microsoft about those problems, and
> they are supposed to be fixing the code (was scheduled for Vista, but
> we haven't heard any response on the bugs and I haven't had time to
> verify in Vista).
>
> And just because Microsoft propagates a bug doesn't mean that you
> should as well.

That's what I told my main codec customer.

But he feels he has to make a living, and his customers are demanding that 
his image library can write JPEG compressed TIFFs that can be read back in 
Microsoft Document Imaging products. And I have to make a living too, and 
thus I have to serve his needs. (And I do understand his predicament, BTW.)

I did experiment to find the cleanest possible version that Microsoft 
malware can read, and I found that to be a single-strip OJPEG with one 
totally valid JPEG stream in the single strip, no additional table markers, 
and the Strip offset and bytecount reflected in the JpegInterchangFormat and 
JpegInterchangeFormatLength tags. So this is at least the least filthy 
version imaginable.

Really, Microsoft is causing a whole food chain of vendors to follow in 
their OJPEG footsteps. Nothing much I can do about that other then starve 
myself to nobody's gain. I need to be in that food chain.

I guess my point is, if you find you have some energy you can devote to 
dealing with this issue, then marking the TIFF 6.0 document OJPEG chapter as 
overriden inside that PDF on the one hand, and pushing all your contacts in 
Microsoft on the other hand, I think could be much more effective then any 
other action, as that directly deals with both causes for the persistence of 
this problem.

> Don't write compression code 6 in TIFF.
> You can read it, but you should never write it.

I would like nothing better then to follow your advice. It is my own 
original point of view. But then there's the reality out there...


Best regards,

Joris Van Damme
info@awaresystems.be
http://www.awaresystems.be/
Download your free TIFF tag viewer for windows here:
http://www.awaresystems.be/imaging/tiff/astifftagviewer.html