Hi all, I just came across the following page: http://r3mix.net In the "News" section there is a fairly negative critique of Vorbis; especially it is mentioned that Vorbis "has terrible masking problems". There is also a Vorbis-encoded frequency sweep which shows strong deficiencies at high frequencies, but I suppose this is due to the quite low bitrate of the distributed encoder. I would be glad if one of the Vorbis insiders could comment on that critique. Does the guy have a point or are the problems caused by the low bitrate? Happy hacking, Martin Reinecke --- >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.
This page contains details of his analysis.
"http://users.belgacom.net/gc247244/analysis.htm"
Look at the `original signal' line to explain the low frequency cut
off in the ogg analysis.
"http://users.belgacom.net/gc247244/ns-ogg.png"
Frequencies above 16 kHz and hard for most people to hear. I don't
really think that it matters for human ears. These test are the type
that anal retentive audiophile weenies like to do.
Examine his critic of the MP3 encoders. His best test is with an
actual song. Although I would think that Ogg Vorbis would do better at
pure tones, since it uses MDCT. This is somewhat distressing to me.
The last thing to note is the human ear hears in a logarithm mode.
All of his graphs are linear, which can make things look worse than
they are, 3 or 6 db is not much of a difference.
Maybe someone else can provide better details; I am certainly not
an insider. `Ich Bin Ein Auslander'
Bill
>>>>> "Martin" == Martin Reinecke
<martin@MPA-Garching.MPG.DE> writes:
Martin> Hi all, I just came across the following page:
Martin> http://r3mix.net In the "News" section there is a fairly
Martin> negative critique of Vorbis; especially it is mentioned that
Martin> Vorbis "has terrible masking problems". There is also a
Martin> Vorbis-encoded frequency sweep which shows strong
Martin> deficiencies at high frequencies, but I suppose this is due
Martin> to the quite low bitrate of the distributed encoder.
Martin> I would be glad if one of the Vorbis insiders could comment
Martin> on that critique. Does the guy have a point or are the
Martin> problems caused by the low bitrate?
--
Ask me the DIFFERENCE between PHIL SILVERS and ALEXANDER HAIG!!
Another dimension is not a sham, I watch!
--- >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.
>>>>> "Thomas" == Thomas Marshall Eubanks <tme@21rst-century.com> writes:[snip] >> The last thing to note is the human ear hears in a logarithm mode. >> All of his graphs are linear, which can make things look worse >> than they are, 3 or 6 db is not much of a difference. Thomas> I don't understand this - his plots are in deciBels (dB), Thomas> which is a log scale. opps, I am dumb! -- Why, Lassie was unborn! Ban ten-inchers! SALESMEN. ATTACKED. SINKING. U-1. --- >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.
> Hi all, > > I just came across the following page: http://r3mix.net > In the "News" section there is a fairly negative critique of Vorbis; > especially it is mentioned that Vorbis "has terrible masking > problems".The test involved is not having masking problems, it's hitting the short-block bug. We are aware of this bug (and have been since before the beta); I'll be sending out a complete description of the bug here in a few minutes. since this is getting passed around alot. It's just a bug, and not a very serious one (although it *does* have a big effect :-) The guy who runs r3mix.net says he's happy to run new tests when I have a fixed version to send him. Summary: log sweeps (and other artificial test samples with strong, isolated tones) trigger a bug in the encoder that causes it to choose almost all short blocks. Short blocks are exactly the wrong choice for strong tones (as they're designed to represent impulses). The result is terrible fidelity. For comparison purposes, listen the following log sweep sample with short blocks disabled: http://www.xiph.org/ogg/vorbis/test.ogg ...and you'll see that the sample is near perfect, 0-22050Hz, at less than 25kbps/ch.> There is also a Vorbis-encoded frequency sweep > which shows strong deficiencies at high frequencies, but > I suppose this is due to the quite low bitrate of the > distributed encoder.Actually, the beta went out with a 16kHz cutoff enabled. There will be no cutoff in the next release.> I would be glad if one of the Vorbis insiders could comment > on that critique. Does the guy have a point or are the problems > caused by the low bitrate?They are problems caused by analysis bugs, and ones that will be fixed ASAP. 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.
Monty wrote:> Actually, the beta went out with a 16kHz cutoff enabled. > There will be no cutoff in the next release.Hello Monty. I currently use LAME/MP3 for my LPFM radio station but I will likely switch to Vorbis sometime in the future. To keep equipment costs down at my station, I'm not using the usual hardware frequency limiters that most stations use. I'm currently relying on the low-pass filter option in LAME to provide a cut-off at roughly 16.5khz. Could you please add a similar option to the Vorbis wish list. Thanks, Ross Levis. Soul FM 88.5 --- >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.
Monty wrote:> Actually, I think LAME with Ogg support will still do lowpass > filtering for you. If I'm wrong, adding this is not a problem...I've asked Mark Taylor about that. I presume you will be producing your own binary without LAME at some stage. Currently the LAMEOGG binary defeats the purpose of having a legal patent free encoder! Ross. --- >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.
Jack Moffitt <jack@icecast.org> wrote:>> Currently the LAMEOGG binary defeats the purpose of having a legal patent >> free encoder! > >No it doesn't. There's a special "ONLYVORBIS" tag. There's no mp3 >routines in Ogglame besides the decoding.Oh! You are right. I had assumed MP3 was still included due to all the MP3 help options still intact. Cheers, Ross. --- >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.