>>>>> "T" == Takehiro Tominaga <[EMAIL PROTECTED]> writes:
>>>>> "R" == Robert Hegemann <[EMAIL PROTECTED]> writes:
T> I make a quick hack and do it, but it is not clever and cool:
T> 1 added the routine that reorder the MDCT coefficiences of short
T> block
T> 2 rewrite calc_xmin, calc_noise, count_bits, ... for short block
T> 3 rewrite ShortHuffmancode routine to write out count1 region.
R> Is it really necessary to rewrite calc_xmin and calc_noise?
R> What's missing there?
It's because of change 1, reodering MDCT coefficiences. we should change
MDCT index for the new order.
T> The new code still has little bug, but the snapshot will be on
T> my web page(maybe tomorrow?).
That's on web :)
http://www.isoternet.org/~tominaga/lame-beta/lame-0402.tar.bz2
note that this is very the beta.
you need -Z option to enable the short block big_values/count1 compression.
T> And bacause the new short block huffman coding algorithm is
T> very efficient, I think we should change the bit allocation for
T> short block.
I removed any addtional bit for short block. It needs many tuning for it,
I think.
---
Takehiro TOMINAGA // May the source be with you.
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
