Bass1000
May 18 2004, 11:22
I know, that GT3b2 is recommended encoder, but it is BETA, so is it stable ?
Why is it better than Ogg Vorbis V1.01 ?
john33
May 18 2004, 11:50
I suggest you do some browsing through the OggVorbis fora for answers. For higher bitrates, above '-q 4', GT3b2 is probably 'better' tuned than 1.0.1, but at the expense of a slightly higher bitrate. At the lower bitrates, they are the same. So, it depends what bitrate you're targetting and how 'generous' you can afford to be with disk storage.
Lord Steele
May 18 2004, 12:38
QUOTE(Bass1000 @ May 18 2004, 09:22 AM)
Why is it better than Ogg Vorbis V1.01 ?
it mainly reduces the so called pre-echo effect at q5 and above
QuantumKnot
May 18 2004, 17:49
Currently GT3b2 is the best Vorbis encoder at -q 5 and above which gives sharper and cleaner attacks than 1.0.1. However, it does suffer from some noise issues which have been addressed in other third party tunings. The Vorbis listening test showed that aoTuV is the best Vorbis encoder at -q 4 and is now being judged at q 4.35 right at this very moment in the 128 kbps multiformat listening test. As soon as the results come out, it may be the recommend encoder, at least at -q 4 and below.
Bass1000
May 19 2004, 02:24
Can you really HEAR the difference between GT3b2 and 1.0.1 ? (for example at -q7)
QuantumKnot
May 19 2004, 04:03
QUOTE(Bass1000 @ May 19 2004, 06:24 PM)
Can you really HEAR the difference between GT3b2 and 1.0.1 ? (for example at -q7)
I probably can't hear the difference with my mortal ears but I'm sure the resident audiophiles here can
There are some preecho samples (castinets, etc.) where I've ABXed xiph vorbis (it was probably 1.0 at the time) from gt3b1 at -q6. 1.0.1 and gt3b2 are not significantly different from 1.0 and gt3b1 respectively at that -q level. I'm not postive that gt3b1 removed the audibility of the preecho completely, however, I don't recall if I was able to ABX the gt3 version from the original.
Note: I believe I'm unusually sensitive to pre-echo, but other than that, I'm no golden ear.
harashin
May 20 2004, 02:15
There actually are some audible differences between 1.0.1 and GT3b2 at -q7 at least on those samples.
1.0.1CVS vs GT3b2 at -q7 Vorbis test
KA_CH_RO
Jun 7 2004, 18:30
What does all that mean?...
I've been reading this forum all day long, and I totally get the thing that for low bitrates ogg rulez,but what happens when you rise the bar above 200kbps?(especifically 256 vbr), ogg vorbis is still better than mp3?...what encoder/settings should i use?..
thnks
shadowking
Jun 7 2004, 18:41
At >200k all modern encoders sound good.
KA_CH_RO
Jun 9 2004, 21:23
c'mon, there's gotta be a difference between them.....
The difference is for you yourself to judge. What other factors do you take into account when you choose a codec? Are you judging sound quality solely, or do you also look at file size, portable use, and active development?
There is a general consensus that mpc is the 'best' (least number of problem samples) above 170 kbps. I have been out of the loop for a while, so the zealots, ahem, I mean proponents

of other audio codecs will want to add the latest news. AFAIK, aac is practically on par with mpc quality-wise, and mpc only wins in terms of decoding speed.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please
click here.