For those who haven't yet seen this: EarGuy's Digital Ear (physiological model of the ear based on work by Frank Baumgarte) has just finished rating the sound quality of Vorbis beta 4 at 128 kbs using 30 random 10-second selections of music: http://pub41.ezboard.com/fr3mixfrm4.showMessage?topicID=33.topic Two samples on which the Ear says Vorbis performed uncharacteristically poorly are on my iDrive at: https://www.idrive.com/miyaguch/files/Shared/?curr-node=8821135668693434409 as s27VorbisTest.wav and s37VorbisTest.wav. After listening to them, I can probably agree with s37, which has a nasty laserbeam-like artifact starting at about 7.5 seconds, in addition to a noisy (fluttering) voice, for example at 3.5 seconds, and pre-echo in the snaps. s27, though, is somewhat puzzling to me because I think it sounds ok. Overall at 128 kbs, Vorbis definitely performed better than FhG FastEnc, and probably performed better than Lame 3.88b using --nspsytune at the same bitrates (except for the two samples above, which pulled down its rating). From my own subjective listening tests, I'd agree that Vorbis beta 4 sounds better than the best mp3 codecs at 128 kbs for most music. ff123 --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Does anybody have any such ratings as Frank B's relative to low (40Kbps and less) data rates? Bob Widergren ----- Original Message ----- From: "ff123" <miyaguch@usa.net> To: <vorbis@xiph.org> Sent: Thursday, April 05, 2001 11:38 PM Subject: [vorbis] Digital Ear evaluation of Vorbis beta 4> For those who haven't yet seen this: > > EarGuy's Digital Ear (physiological model of the ear based on work byFrank> Baumgarte) has just finished rating the sound quality of Vorbis beta 4 at > 128 kbs using 30 random 10-second selections of music: > > http://pub41.ezboard.com/fr3mixfrm4.showMessage?topicID=33.topic > > Two samples on which the Ear says Vorbis performed uncharacteristically > poorly are on my iDrive at: > >https://www.idrive.com/miyaguch/files/Shared/?curr-node=8821135668693434409> > as s27VorbisTest.wav and s37VorbisTest.wav. > > After listening to them, I can probably agree with s37, which has a nasty > laserbeam-like artifact starting at about 7.5 seconds, in addition to a > noisy (fluttering) voice, for example at 3.5 seconds, and pre-echo in the > snaps. s27, though, is somewhat puzzling to me because I think it soundsok.> > Overall at 128 kbs, Vorbis definitely performed better than FhG FastEnc, > and probably performed better than Lame 3.88b using --nspsytune at thesame> bitrates (except for the two samples above, which pulled down its rating). > > From my own subjective listening tests, I'd agree that Vorbis beta 4 > sounds better than the best mp3 codecs at 128 kbs for most music. > > ff123 > > > --- >8 ---- > List archives: http://www.xiph.org/archives/ > Ogg project homepage: http://www.xiph.org/ogg/ > To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' > containing only the word 'unsubscribe' in the body. No subject is needed. > Unsubscribe messages sent to the list will be ignored/filtered. >--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
On Thu, Apr 05, 2001 at 11:38:51PM -0700, ff123 wrote:> For those who haven't yet seen this: > > EarGuy's Digital Ear (physiological model of the ear based on work by Frank > Baumgarte) has just finished rating the sound quality of Vorbis beta 4 at > 128 kbs using 30 random 10-second selections of music: > > http://pub41.ezboard.com/fr3mixfrm4.showMessage?topicID=33.topicThanks for pointing the list to this. Aside from the graph making things look bad, it seems like Ogg actually did pretty well in this test... And it points out again what we already know; quality is very good overall and we still have some consistency issues to deal with.> Two samples on which the Ear says Vorbis performed uncharacteristically > poorly are on my iDrive at: > > https://www.idrive.com/miyaguch/files/Shared/?curr-node=8821135668693434409 > > as s27VorbisTest.wav and s37VorbisTest.wav.Thanks. I'm stuck on a modem connection until the end of this coming week so I'll not be able to grab them until the 14th or 15th. Monty --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.