Help - Search - Members - Calendar
Full Version: Quality of Sonic Foundry OggVorbis Beta 3
Hydrogenaudio Forums > Lossy Audio Compression > Ogg Vorbis > Ogg Vorbis - General
TomKure
Hello there!
A short time ago I got from my friend a psytrance album encoded with Sonic Foundry OggVorbis Beta 3 (its written in tag), vendor: Xiphophorus libVorbis I 20011231, nominal bitrate 256kbps. It sounds very terribly to me with a lot of high-frequency artifacts. I want to ask you smart people here, if it could be caused by the encoder (if it isnt some buggy version or something like that...). I wouldnt wonder if it was some transcode, because my hearing is very good (I think) and thats why I like OGG... I was very surprised, when I heard those songs for the first time... And another question: is there a program that would show amounts of frequencies in a sound file (graph, text...)?
Leo 69
Ogg Vorbis has never sounded in the way you've described. This album was probably transcoded from another lossy format.
Raiden
That is indeed a very old version and i have not heard any files encoded with RC3.
You could always do a ABX test yourself, though.
http://www.vorbis.com/files/rc3/windows/vo....0rc3-win32.zip (with the '-q 8' option)

To view the spectrum of a sound file i use the 'simple spectrum' visualisation in Foobar 0.8.3. I still miss this feature in newer Foobars. smile.gif
Mo0zOoH
QUOTE(Leo 69 @ Aug 1 2006, 22:30) *

Ogg Vorbis has never sounded in the way you've described. This album was probably transcoded from another lossy format.

Oh sure.
In fact, earliest versions sucked at high bitrates much more than MP3 did at that time, until GT3b2 came out. Noise boost and smearing were not uncommon even at -q8.
TomKure
I tried to encode some music with the rc3 at q8, and it sounded very good. I think that this one album has been transcoded from another lossy format... And what does the "Sonic Foundry OggVorbis Beta 3" in coment mean? Isnt that some tweaked version?
TomKure
I have just checked one of the songs with ogginfo 1.1.0 and it floods my screen with repetitious notice: "Negative granulepos on vorbis stream outside of headers. This file was created by a buggy encoder" I think that this certainly means something... smile.gif Or not?
pepoluan
QUOTE(TomKure @ Aug 2 2006, 01:06) *
A short time ago I got from my friend an psytrance album encoded with Sonic Foundry OggVorbis Beta 3 (its written in tag), vendor: Xiphophorus libVorbis I 20011231, nominal bitrate 256kbps.

I think the information basically says:

The tool is Sonic Foundry, using an OggVorbis plugin, and the version of the plugin is Beta 3.

The plugin itself is based on libVorbis I 20011231. The 8-digits there is ISO-8601 representation of year 2001, month 12, day 31, which is extremely old. That old an encoder will lack various tunings and perfection of the latest versions.

Considering the latest version is dated 2005 (i.e. Aoyumi's magnificent aoTuVb4.51), that in itself is enough to cause some artifacts.

It's even worse if it was transcoded from a lossy format.

In any case, tell your friend to change his/her tool. If your friend wants to transcode from MP3, although not recommended, use the latest WinLame or MediaCoder or foobar2000 (with the latest Lancer-optimized version of aoTuVb4.51). Jebus of this forum made another transcoder that looks promising, but still in development stage and I haven't tested it yet.

Oh, and transcode to at least -q 4.

TomKure
Yes, it is very old version, but when I encoded one song from Alice in Chains using version 1.0 rc3 at q8 (vendor 20011217), it sounded very good and I didnt hear any artifacts... I think that Alice in Chains and other such music is more inclinable to have high-freq. artifacts than most electronical dance music. Isnt it?
pepoluan
If you think 1.0 RC3 at -q8 sounds good, try encoding with aoTuVb4.51 at -q6, then down to -q4.

You'll be amazed.
Garf
One thing to take into account is that this will still use floor0 which is not so well supported on some portable players.
pepoluan
@Garf: By 'floor0' you are referring to RC3, right?
Garf
RC3 and beta3 both use it
TomKure
QUOTE(pepoluan @ Aug 2 2006, 20:29) *

If you think 1.0 RC3 at -q8 sounds good, try encoding with aoTuVb4.51 at -q6, then down to -q4.

You'll be amazed.

I tried aoTuVb4.51 at both -q4 and -q6 and I can say that -q4 would be very easy to ABX for me and I tried to ABX -q6 and I gave up at 12th try cause it was very hard, but I got 7 of 11 anyway, thats good, or not?

QUOTE(Garf @ Aug 2 2006, 20:50) *

One thing to take into account is that this will still use floor0 which is not so well supported on some portable players.

I played those rc3 files on Winamp and foobar2000 and it sounded on both the same bad.


Cant the wrong quality be caused by the buggy encoder, as the ogginfo reports?
Garf
QUOTE(TomKure @ Aug 2 2006, 22:15) *

QUOTE(Garf @ Aug 2 2006, 20:50) *

One thing to take into account is that this will still use floor0 which is not so well supported on some portable players.

I played those rc3 files on Winamp and foobar2000 and it sounded on both the same bad.


Cant the wrong quality be caused by the buggy encoder, as the ogginfo reports?


It has nothing to do with each other.
TomKure
QUOTE(Garf @ Aug 2 2006, 22:29) *

QUOTE(TomKure @ Aug 2 2006, 22:15) *

QUOTE(Garf @ Aug 2 2006, 20:50) *

One thing to take into account is that this will still use floor0 which is not so well supported on some portable players.

I played those rc3 files on Winamp and foobar2000 and it sounded on both the same bad.


Cant the wrong quality be caused by the buggy encoder, as the ogginfo reports?


It has nothing to do with each other.

So, if I understand it right, the "Negative granulepos on vorbis stream outside of headers. This file was created by a buggy encoder" message has nothing to do with the sound quality of the file. Right?
pepoluan
QUOTE(TomKure @ Aug 3 2006, 03:15) *

QUOTE(pepoluan @ Aug 2 2006, 20:29) *

If you think 1.0 RC3 at -q8 sounds good, try encoding with aoTuVb4.51 at -q6, then down to -q4.

You'll be amazed.

I tried aoTuVb4.51 at both -q4 and -q6 and I can say that -q4 would be very easy to ABX for me and I tried to ABX -q6 and I gave up at 12th try cause it was very hard, but I got 7 of 11 anyway, thats good, or not?

That means you can safely use -q6 instead of -q8, saving bits. For some tracks you can even use -q5, or -q4.

TomKure
QUOTE(pepoluan @ Aug 4 2006, 19:16) *

QUOTE(TomKure @ Aug 3 2006, 03:15) *

QUOTE(pepoluan @ Aug 2 2006, 20:29) *

If you think 1.0 RC3 at -q8 sounds good, try encoding with aoTuVb4.51 at -q6, then down to -q4.

You'll be amazed.

I tried aoTuVb4.51 at both -q4 and -q6 and I can say that -q4 would be very easy to ABX for me and I tried to ABX -q6 and I gave up at 12th try cause it was very hard, but I got 7 of 11 anyway, thats good, or not?

That means you can safely use -q6 instead of -q8, saving bits. For some tracks you can even use -q5, or -q4.


Yea, but I didnt rip this CD, sad to say... Usually, I use -q6,5 for Vorbis, thats enough, but I use Vorbis very rarely, because my little cheap Creative Zen Nano plus doesnt play .ogg files...surprisingly... smile.gif
pepoluan
Heh, sell them and buy Cowon wink.gif
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2008 Invision Power Services, Inc.