Glad to hear it! Besides much more drastic bitrate fluctuations, a definitive improvement in ringing is noticeable on one of my samples at (specified) 68kbps. The old version produced a file at 62kbps (with ringing on the vocals), while the new version produced a file that was over 74kbps (with ringing much reduced on the vocals). The instrumental version (w/o vocals) of the same song came out to be nearly identical in bitrate (62 old, 63 new). Interesting.
foo_abx 1.3.4 reportfoobar2000 v1.1.14a2012/11/03 22:38:23File A: Z:\media\data\musik\temp\01-The_Decline.flacFile B: Z:\home\ickefes\Documents\01-The_Decline.flac.ogg22:38:23 : Test started.22:39:27 : 00/01 100.0%22:39:28 : Trial reset.22:40:16 : 01/01 50.0%22:40:25 : 02/02 25.0%22:40:40 : 03/03 12.5%22:40:52 : 04/04 6.3%22:41:10 : 05/05 3.1%22:41:33 : 06/06 1.6%22:42:03 : 07/07 0.8%22:42:05 : Test finished. ---------- Total: 7/8 (3.5%)
foo_abx 1.3.4 reportfoobar2000 v1.1.14a2012/11/03 22:22:45File A: Z:\media\data\musik\temp\01-The_Decline.flacFile B: Z:\home\ickefes\Documents\The Decline.opus22:22:45 : Test started.22:24:26 : 00/01 100.0%22:25:32 : Trial reset.22:26:32 : 01/01 50.0%22:26:48 : 02/02 25.0%22:27:35 : 03/03 12.5%22:27:47 : 03/04 31.3%22:28:19 : 04/05 18.8%22:28:36 : 05/06 10.9%22:29:27 : 06/07 6.3%22:30:16 : 06/08 14.5%22:31:17 : 07/09 9.0%22:31:52 : 08/10 5.5%22:32:08 : 09/11 3.3%22:32:51 : 10/12 1.9%22:33:50 : 10/13 4.6%22:34:20 : 11/14 2.9%22:34:40 : 12/15 1.8%22:35:12 : 13/16 1.1%22:35:51 : 13/17 2.5%22:36:50 : 14/18 1.5%22:37:22 : 15/19 1.0%22:37:27 : Test finished. ---------- Total: 15/20 (2.1%)
but the Opus had more smearing.
New transient detection algorithmThis one is explicitly based on a simple temporal masking model
Quote from: punkrockdude on 03 November, 2012, 05:47:14 PMbut the Opus had more smearing.BTW, from the Opus git (http://git.xiph.org/?p=opus.git;a=commit;h...33c10b5cec11649), Fri, 2 Nov 2012:QuoteNew transient detection algorithmThis one is explicitly based on a simple temporal masking model
I will try that one but do you know which command line to use to compile under Linux? Regards.
% ./autogen.sh% ./configure% make% sudo make install
Now we just need something we can encode with from this.
opus:Further cleanup of the MDCT code, fixes PLC bugAvoid copying imdct outputVarious fixes to draft-terriberry-oggopus.xmlValincomb_filter() bypass for the case where the gain is... Oops, put back the "static" for transient_analysis()Fixes a fixed-point overflow in the new transient detectorFixes a fixed-point divide-by-zero issueNew transient detection algorithmopus-tools:opusrtp now compiles properly for windows (but remains useless)
That AVX build encoded a single file around 45x speed using 128kbps VBR default in Foobar2000 under Window 7 x64 on a ASUS K53SV-SX812V (2.4GHZ).
Quote from: punkrockdude on 06 November, 2012, 05:14:31 AMThat AVX build encoded a single file around 45x speed using 128kbps VBR default in Foobar2000 under Window 7 x64 on a ASUS K53SV-SX812V (2.4GHZ).Good news, I guess! That build wasn't for AVX -- it was generic w/SSE. I was contemplating producing an AVX build but then decided against it for now. Long live genericism! As for the build speed, I guess the question is how it stacks up against the "official" release on a range of processors. Thanks for testing!
I've been reading the manpages for opusenc, but don't understand how to create multichannel opusfiles.
BTW how big is speed gain for AVX build?
Intel i7-3770 / AsRock B75 Pro3 / 16Gb DDRIII-1333 / Windows 7 Home Premiumopusenc --bitrate 24 --vbr --comp 10 --downmix-mono --framesize 40 test.wav test.opus1. opus-tools-0.1.5-win32> 52.07x2. opus_tools_11_7_2012> 45.78x3. opus_tools_11_1_2012> 45.01x4. opus_20121103\Release64> 44.26x5. opus_20121103\Release32> 40.85x6. opus-tools-0.1.5-Cygwin5_sl2_121022_04-25> 35.89x
No special method is required; wav files contain sufficient information about the channels for opusenc to just go ahead and encode it like any other file. (If you're noticing the things in the manpage about multichannel, those are there so people who are encoding from raw audio can supply the channel information themselves.)
Intel i7-3770 / AsRock B75 Pro3 / 16Gb DDRIII-1333 / Windows 7 Home Premiumopusenc --bitrate 68 --vbr --comp 10 --framesize 60 test.wav test.opus1. opus-tools-0.1.5-win32> 82.99x2. opus_tools_11_7_2012> 80.47x3. opus_tools_11_1_2012> 78.1x4. opus_20121103\Release64> 66.39x5. opus_20121103\Release32> 57.73x6. opus-tools-0.1.5-Cygwin5_sl2_121022_04-25> 42.83x
Why use --vbr and --comp 10 when they are already the default?
I suspect the official release is ICC since it runs significantly slower than the builds by The Sheep for me.
Also wondering why a i7-3770 is only producing speeds ~2x faster than my Athlon x2 on The Sheeps builds, is it even multithreaded?