Hi. I'd like to ask a little about where Ogg stands today when it comes to streaming audio. The VBR has been the main argument in the incompatibility for streaming but has this changed somehow or will it? Is there something else besides quick and large scaled bandwith usage variations and the difficulty of buffering, that could cause problems? CBR could be used but some people say that it reduces the sound quality. When doing some test listnening I couldn't tell much differences between VBR and CBR but I might as well be half-deaf.. I'd like to get the most out of Ogg potential by using VBR but if it means problems with streaming or some listeners not being able to listen at all, then I will choose CBR. -Henri <p>--- >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.
On dom, 2003-05-04 at 14:44, Henri Jylkka wrote:> [...] The VBR has been the main argument in the incompatibility for > streaming but has this changed somehow or will it? Is there something > else besides quick and large scaled bandwith usage variations and the > difficulty of buffering, that could cause problems? > > CBR could be used but some people say that it reduces the sound > quality. [...]Vorbis VBR is very variable. Since you choose and encoding quality, you can't know the bitrate. The same -q results in very different bitrates depending on the input sound. If you cannot use VBR, you can use ABR. Oggenc also has options to set min/max bitrates.> When doing some test listnening I couldn't tell much > differences between VBR and CBR but I might as well be half-deaf.. I'd > like to get the most out of Ogg potential by using VBR but if it means > problems with streaming or some listeners not being able to listen at > all, then I will choose CBR.If required timings are very strict, you may need to set a maximum bitrate. Constant bitrate makes no sense IMO. Bye. <p>--- >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.
On Monday 05 May 2003 00:44, Henri Jylkka wrote:> Hi. I'd like to ask a little about where Ogg stands today when it comes > to streaming audio. > > The VBR has been the main argument in the incompatibility for streaming > but has this changed somehow or will it? Is there something else > besides quick and large scaled bandwith usage variations and the > difficulty of buffering, that could cause problems?In practice, VBR vorbis doesn't vary so much that it causes significant problems with streaming for most people. You can use a sort of constrained VBR where you set a maximum bitrate (but not a constant bitrate), if you really need to, of course. There are _zero_ 'incompatibility' problems with VBR streaming - it's just the variation in bandwidth usage which could cause problems in some cases. Mike --- >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.
Henri Jylkka wrote:> Hi. I'd like to ask a little about where Ogg stands today when it comes > to streaming audio. > > The VBR has been the main argument in the incompatibility for streaming > but has this changed somehow or will it? Is there something else > besides quick and large scaled bandwith usage variations and the > difficulty of buffering, that could cause problems? > > CBR could be used but some people say that it reduces the sound > quality. When doing some test listnening I couldn't tell much > differences between VBR and CBR but I might as well be half-deaf.. I'd > like to get the most out of Ogg potential by using VBR but if it means > problems with streaming or some listeners not being able to listen at > all, then I will choose CBR.My personal experience shows that streaming Vorbis is entirely trouble-free if your target audience is people with 64-128 kbps downstream. For example (I have an 128kbps upstream), streaming at -q 1.5 is really nice for 128kbps users. The average of the stream lies somewhere around 88 kbps at excellent quality and plenty of overhead. VBR is clearly not an issue there. Streaming at -q -1 (or a higher q with downsampling) for 64kbps ISDN users is working very good, too. It gets a little harder for lower bitrates, as quality degration becomes more noticeable and one really wants to shove as many bits as possible through the line - however, it's still possible to get good results with ABR (oggenc parameter: -b <target_bitrate> --managed --resample <target_sample_rate> [--downmix]) and no further hard bandwidth limits. (oggenc parameter: -M <max_bitrate>) It's mostly a matter of experimentation and finding the sweet spot for your particular needs and audience. I recommend you try this by yourself by either encoding some music to Vorbis at the bitrates you need, or combine that with a real streaming experiment (which is most useful, as you can ask some friends on modems to listen to your stream and report problems) ... Icecast 2 will allow you to do just that. (www.icecast.org) <p>Moritz --- >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.