Re: Possible MPEG1 encoding regression
Nathan Pilatzke <[email protected]>
| Newsgroups | gmane.comp.video.mjpeg.devel |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 11 Dec 2006, Set Norman wrote: > I am trying to encode a very simple black & white "clock" as MPEG1. > > Using mjpegtools-1.6.0-beta2.tar.gz everything works fine. However > mjpegtools-1.8.0.tar.gz and CVS creates some type of macroblock > artifacts. > cat clock.pgm*100 > movie # pseudo shell code > cat movie | pgmtoy4m -i p | 1.6/mpeg2enc -f1 -o clock_16_f1.mpeg > cat movie | pgmtoy4m -i p | 1.8/mpeg2enc -f1 -o clock_18_f1.mpeg i have seen this as well in 1.8 and in cvs, usually around very light coloured or white areas, such as with movie studio logos at the beginning of a movie. 1.6.0-beta2 worked fine for me. i am running gentoo linux with a gcc-4.1.1 compiler but i have compiled 1.8 and later with gcc-3.4.6 and the artifacts persist. i am available for any testing that i can offer but i am not subscribed to the list. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV