QUOTE(Case @ Jul 17 2003, 05:57 PM)
I made some major improvements for Gap killer yesterday, still not perfect always but it's much better than old versions. Please try latest version.
Case, (and apologies if this doesn't belong on the :B 0.7 test thread)
I'm just trying it now (on a [cough] FhG Fastenc 128 CBR mix album (Fatboy Slim)to start with).
Can I ask if it matters where it goes in the DSP order, especially if it's making assumptions about 588 samples per CD frame. Would it therefore be fooled if it appeared after Resampler and that converted to 48000 S/s?
(My Resampler at work is actually set to 44100, because the soundcard won't go any higher and sounds badly aliased at 8000 or 11025, so it's not an issue on this machine)
Also, I've had the max gap removal set to 105 ms (4 frames) since the old version (after checking the console and noticing it was removing 2000 ms from some Musepack files, which are gapless, so the gap is intentional!).
Just trying The Very Best of Curtis Mayfield (Musepack encodes) and it's removing 4630 samples (my 105 ms limit) from both the end and the beginning. It still sounds great, however, and I don't notice the difference in length of silence.
CODE
INFO (CORE) : opening file for playback : file://F:\My Music\Curtis Mayfield\The Very Best of Curtis Mayfield\03 - Curtis Mayfield - Move On Up.mpc
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
I'm not familiar with the architecture of FB2K, but is it possible for the plugin to be aware of the filetype being played and whether it's an intrinsically gapless format? I'd have thought so, if it can get the encoder delay and padding from the LAME header on --alt-preset standard files, although with some Rachmaninov in APS, it still removed 4630 samples despite getting that extra information (this album has silence between movements, but it's one of the few albums in APS I happen to have on my PC to test gapkiller on):
CODE
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\01 - Rhapsody On A Theme Of Paganini, Op.34.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 1620
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\02 - 1st movement- Moderato; Concerto For Piano And Orchestra No.2 in C minor, Op.18.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 1344
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\03 - 2nd movement- Adagio sostenuto; Concerto For Piano And Orchestra No.2 in C minor, Op.18.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 2232
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\04 - Finale- Allegro scherzando; Concerto For Piano And Orchestra No.2 in C minor, Op.18.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 2004
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 1620
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\02 - 1st movement- Moderato; Concerto For Piano And Orchestra No.2 in C minor, Op.18.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 1344
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\03 - 2nd movement- Adagio sostenuto; Concerto For Piano And Orchestra No.2 in C minor, Op.18.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 2232
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
INFO (CORE) : opening file for playback : file://F:\My Music\Rachmaninov\Rhapsody On A Theme Of Paganini, Concerto For Piano & Orchestra No.2\04 - Finale- Allegro scherzando; Concerto For Piano And Orchestra No.2 in C minor, Op.18.mp3
INFO (foo_input_std) : got encoder delay: 576
INFO (foo_input_std) : got encoder padding: 2004
INFO (foo_dsp_nogaps) : 4630 samples removed (from end)
INFO (foo_dsp_nogaps) : 4630 samples removed (from beginning)
(Incidentally, the 128kbps Fatboy Slim track changes sounded smooth - mostly after skipping to the end of one track to just hear the changeover - so I'm pretty happy to be able to turn on gap killer for such albums. It didn't seem to matter whether I skipped forward or listened to the whole track - the removed samples were the same)
Anyway, although I usually keep it turned off because I'm mostly using MPC, this is a useful tool for times I want to recover from some gappy MP3s I still have, and audibly it seems to do approximately as good a job as the much-lauded mp3splice does in Winamp 2.x.
Thanks,
DickD