Wow, welcom back, our leader! >>>>> "M" == Mark Taylor <[EMAIL PROTECTED]> writes:
M> This will disable "best_huffman_divide", which I beleive is the M> main difference between short blocks in lame 3.70 and later M> versions. (lame 3.70 is reported to work okay with hardware M> decoders which have this problem) It is also a feature which is M> used (as far as I know) by only lame and some FhG encoders M> which may explain why using other encoders will not create M> problems. uum. I agree it may cause the problem. M> I notice that best_huffman_divide() is specifically disabled M> for short blocks in MPEG 2/2.5. Does any one remember why this M> was done? >From the commit log of takehiro.c in Attic, http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/lame/lame/Attic/takehiro.c?sortby=date that was your commit to fix my "broken" commit. http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/lame/lame/Attic/takehiro.c.diff?r1=1.68&r2=1.69&sortby=date This disabled best_huffman_divide for SHORT BLOCK for all MPEG version. And later, you enabled it for MPEG1. http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/lame/lame/Attic/takehiro.c.diff?r1=1.69&r2=1.70&sortby=date I searched the mail archive arroud these days, but I cannot find out the reason at the moment :( -- Takehiro TOMINAGA // may the source be with you! _______________________________________________ mp3encoder mailing list [EMAIL PROTECTED] http://minnie.tuhs.org/mailman/listinfo/mp3encoder