[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mjpeg-users
Subject:    Re: [Mjpeg-users] RE: Mjpegtools 1.6.3 on an i386 (fwd)
From:       "Steven M. Schultz" <sms () 2BSD ! COM>
Date:       2005-08-06 3:26:44
Message-ID: Pine.BSI.4.05L.10508052020430.29277-100000 () moe ! 2bsd ! com
[Download RAW message or body]


On Sat, 6 Aug 2005, Gernot Ziegler wrote:

> since C routines always are the common denominator for porting to new
> platforms, I doubt that any of the developers will abandon these fallback
> routines altogether. That leads me to the statement that any
> compilation problems with the C only code must be a bug and was not
> intended :-)

	True for mjpegtools itself.  However, one thing that is worth keeping
	in mind is that some of the dependencies (libdv, libquicktime, etc)
	may NOT be as friendly to non-MMX (i.e. ancient ;)) systems.

	If someone wants to encode video at a several-seconds/frame rate (as 
	compared to several-frames/sec) that's entirely their business 
	(patience is a virtue I'm told :-)).   My threshold of pain is around
	1.5 frames/second - below that I'll begin dropping filters from the
	pipeline to speed things up.

	No plans at all to remove the C code - it's not just a fallback but
	also serves as a "reference" to compare the output of the SIMD routines
	against.  In fact considerable work was done a while back to allow
	the easy runtime switching (via an environment variable) of the SIMD
	routines On/Off.

	Cheers,
	Steven Schultz



-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
Mjpeg-users mailing list
Mjpeg-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mjpeg-users
[prev in list] [next in list] [prev in thread] [next in thread]