Why does BeSweet use Ogg Vorbis Quality range 0-1?

[metrom]

Why does BeSweet use Ogg Vorbis Quality range 0-1?

This confuses alot of users and newbies.

The standard is 0-10 with decimals (.00).

This is documented several places.
I searched for a few minutes:

http://www.vorbis.com/encoding.html

"As an alternative to dealing with bitrates, Vorbis offers a streamlined "quality" setting control. The values range from zero to ten."

http://www.vorbis.com/intro.html

"Instead of having to deal with specific bitrate settings, Vorbis encoders use a quality setting from zero to ten."

metrom

[Nic]

Because the Vorbis API uses a float between 0-1. Interfaces just divide input down into a number between 0-1, besweet could do this, but I prefer that it doesnt.

-Nic

[DSPguru]

Originally posted by Nic
Because the Vorbis API uses a float between 0-1.
yea, and i find this scale very reasonable :).
Interfaces just divide input down into a number between 0-1, besweet could do this, but I prefer that it doesnt.
in fact, if you'll set -q 2, BeSweet will handle it as if it were -q 0.2 ;)

[Nic]

Ooohhh...very clever :D

-Nic

[metrom]

I didn't know the thing about Vorbis API.

But still, BeSweet is the only Ogg Vorbis encoding program that uses the 0-1 scale that I know of.

I can understand that you "followed" the API-scale, but since the standard is documented and is used by all other programs, I can't understand why you are keeping your type of scale.

I'm not trying to make a big discussion about this :) but, I'm just concerned about newbies and standards etc.

[DSPguru]

hi metrom, don't take it too hard :)
BeSweet's ogg parser is error-friendly. any number between 1 to 10 will be divided by 10.

why do i like the 0-1 scale?
i find the normalized 0-1 scale much more common in dsp applications, simply because there are lots of dsp applications based on fixed-point math (smallest number=-1, biggest number<1).

[sillKotscha]

biggest number<1

</=1 ;)

regards

[metrom]

Originally posted by DSPguru
BeSweet's ogg parser is error-friendly. any number between 1 to 10 will be divided by 10.
Ok, then it's no problem :)
Originally posted by DSPguru

i find the normalized 0-1 scale much more common in dsp applications, simply because there are lots of dsp applications based on fixed-point math (smallest number=-1, biggest number<1).

I understand you make your programs for your dsp-users, then it all seams more logical. thats nice :)

metrom

btw: http://www.vorbis.nu is a Nordic Ogg Vorbis site, but not yet released. Here you will find very user-friendly articles, faqs, news and forums.
An OggMachine article is allready under way :)

[sillKotscha]

ahh yes...

Here you will find very user-friendly articles, faqs, news and forums.

... just a little extraction

Audacity er et åpent lydredigeringssystem for Linux, Mac og Windows. Bass Boost, Wahwah og Støyfjerning er noen av funksjonene, og den støtter også VST-plugins. Det ser virkelig bra ut så langt og den fikk nylig Ogg Vorbis-støtte.

but only for scandinavian people, I guess ;)

cheers sill

[metrom]

Originally posted by sillKotscha
but only for norwegian people, I guess ;)


well, not exactly right.
Norwegian is one of the Nordic languages, so people from Denmark, Sweden and some from Finland and Iceland will also be able to read at vorbis.nu.

:)

[sillKotscha]

sorry...

edited, see above ;)

cheers

[DSPguru]

Originally posted by metrom
An OggMachine article is allready under way :) cool :cool:.

Originally posted by sillKotscha
</=1 ;)in fact, fixed-point machine cannot represet the number "1", only "-1" :).