AACplus will get much better (that is what I have heard) thanks to SBR and also the MP3 based MP3pro had much improvement in low bitrates with spectral band reduction.

Couldn't be there something comparable for our beloved ogg Vorbis at least for low bitrates?

I suggest that SBR is patented, but would it be possible in general?

The techniques used in Vorbis are already quite similair to the ones in SBR, I don't think there's much to gain by using SBR in Vorbis.

Menno

thx, that's interesting to know

Do you know anyone who has checked this statement?
Do you have checked this statement, or do you repeat it, because you believe it is true?

I don't found anything in the C code which is more or less similar to SBR.

Statistic analysis of the decoded output also don't show ANY sign of SBR.

Frank,

I did not check into Vorbis too thoroughly, but I think it uses envelope encoding and noise floor encoding, which is what SBR does as well.
Please correct me if I'm wrong, this is just my understanding of Vorbis.

Menno

AFAIK Monty made the claim the Vorbis noise normalization is 'somewhat' comparable to SBR.

Well I don't check into these algorithms but as far as I know SBR uses a kind of parametric encoding, and Vorbis don't (MDCT, adaptative psychoacoustic models...). So, high frequencies encoded with SBR will never sound very near to the original, but for low bitrates it works fine, as you get good quality (doesn't sound bad, just different) with less bitrate. For the same reason, Vorbis will be better (and will not benefit) when using medium and high bitrates. Am I wrong?

i might be way off here but just my 2 cents on the matter

would it be possible to adapt the lame plusv technology to ogg at low bit rates???

i know lame plusv sounded 'decent' to me - but i don't have golden ears. maybe bronze at best. the lame plusv needed tuned. the creator mentioned this. he didn't have the 'ears' to tune it and was hoping that someone else would take the ball and fininish the tuning.

i don't know all the particulars on the plusv but the source code was at the site. for all i know it may be like mixing gas and water. they don't.

just a suggestion. if i recall correctly, the plusv was open source but had some kind of copyright with it...not sure how all that fits in either.

sorry if i was way off...delete this post if necessary.

Lame has nothing to do with PlusV, but the answer to your question is no. Vorbis will not incorporate PlusV technology. See here:
http://www.xiph.org/archives/vorbis-dev/20...00301/0114.html

sorry

just a suggestion. i 'kinda' thought the patent issue would be a problem.

thanks for the link - i had been wondering if the plusv technology would mesh with ogg. and now i know that it won't

http://www.vorbis.com/ot/20030202.html#id2726611

Then maybe he doesn't know how Spectral Band Replication works. See http://www.audiocoding.com/wiki/index.php?page=SBR

I think he does much better than you understand how noise normalization works

I couldn't find origin of what I posted, except for a post from Monty on vorbis-dev that says that '...vorbis is basically doing it's own thing which obtains similar results'.