0
I Use This!
Activity Not Available

Commits : Listings

Analyzed 8 months ago. based on code collected almost 2 years ago.
Mar 05, 2022 — Mar 05, 2023
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
Improved speed with better bit-stream code, fixed a couple of minor arithmetic bigs. Reinstated the audio time-out tweak. Added automake support. More... over 24 years ago
Got rid of the (now rather anachronistic) use of temportary files for the 1st-pass data and slotted in a faster bit-stream reading library. More... over 24 years ago
Added work-around for arithetic conversion bug... More... over 24 years ago
Removing files now auto-generated More... over 24 years ago
No serious modifications .. More... over 24 years ago
Brought mpeg2enc and aenc into the autoconf/automake framework. Tidied up selection of CPU-specific optimisations. More... over 24 years ago
Added run-time determination of SIMD routines (MMX/SSE) More... over 24 years ago
*** empty log message *** More... over 24 years ago
*** empty log message *** More... over 24 years ago
quicktime now optional. More... over 24 years ago
Slowly emerging configure options More... over 24 years ago
Small updates. More... over 24 years ago
*** empty log message *** More... over 24 years ago
automake & autoconf support ... there are more possibilities still left More... over 24 years ago
Added support for multiplexing variable-bit-rate video streams. More... over 24 years ago
*** empty log message *** More... over 24 years ago
experiments More... over 24 years ago
Believe it or not, some essentially minor modifications to fix some subtle quantisation bugs that only showed up when I started creating high data-rate streams. More... over 24 years ago
Added movtar support to lavtrans (untested yet) More... over 24 years ago
Small fixes, the second .. More... over 24 years ago
Some minor, but important, fixes ... More... over 24 years ago
*** empty log message *** More... over 24 years ago
Fixed bug in SSE dist routines caused by changes in C (parameter no longer provided) and tidied stuff up a little more... More... over 24 years ago
Made fast quantisation code robust. It is now perfect for all legal values for coefficients below 63 and gets than 0.1% wrong for big coefficients. Zoom zoom! More... over 24 years ago
Slowly coming towards a release ... lots of untested code still ! More... over 24 years ago
Completed initial experiments with gradient descent. Solved problem with inaccurate fast integer quantisation calculations (for now.... maybe) and completed porting over of available MMX routines. There's really very little left in the code that can benefit from this. One or two more routines plus extending the fast MMX stuff to the other quantisation routines. More... over 24 years ago
O.k. now the bugs are out some polish. First cut a faster C routines for quantization. The job is completely done yet - need to tidy up parameter usage vs. globals and do intra as well and non_intra. However, it works and its a lot quicker (mainly due to avoiding the dreaded integer divide (positively glacial on the x86). More... over 24 years ago
Fixed some nasty and damn subtle bugs causing corruption of the fast motion compensation data. Took *ages* to pin down... More... over 24 years ago
Found that wretched bug in the the assembler that was causing weird, rare, random misbehaviour. Guess who wasn't saving a modified register. Eeeeee-orrrr.... More... over 24 years ago
Wasted effort - further tuning of motion compensation search. Instad found my own and compiler bugs and gain .... wow ... and 8% speed-up for a days work. Should have worked on SSE code for the quantisation routines. More... over 24 years ago