I'm just encoding some files with the latest ogg and it's too good. I tried encoding things at q1 just to see what happened and I couldn't hear the differencee between the ogg and the original files. Maybe it was the type of music I was encoding or limitations in my listening equipment or something but it seems pointless to have a scale from one to ten and have most of it unused. Seriously... this is a real problem. What if I *want* mediocre quality, eg. for downloads where I don't want people to have the full quality version of a song? Also, I think MP3 people will feel cheated if they never need use a higher quality level than two or three. So please, please, let's have worse quality. It's psychological and maybe silly, but I think q1 should sound a *little* bit bad. <p><p> -- <\___/> / O O \ \_____/ FTB. http://www.artlum.com <p><p><p><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.
Colin Dooley wrote:> > So please, please, let's have worse quality. It's psychological > and maybe silly, but I think q1 should sound a *little* bit bad. >I was just fiddleing to see how bad I can get it. "q0.1" sounds about what I think "q1" ought to sound like. "q0.01" isn't really much different from "q0.1" (the bitrate in Winamp only went down from 66 to 65). -- <\___/> / O O \ \_____/ FTB. <p><p><p><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.
q1 sounds like someone wrapped a sock around my head. To be fair, I just encoded a test track at q1, and compared it to the original WAV. In the WAV, I can easily hear hands sliding on cello strings.. In a q1 Ogg, those come across as a hissing sound. The bass frequencies sound crumbly. What's worse is that this is a relatively forgiving test track. Not very complicated. To be sure, the output is very impressive for an average bitrate of 80kbps. I'd be happy to listen to a stream that sounded this good. I usually use q7, myself. ----- Original Message ----- From: "Colin Dooley" <colin@artlum.com> To: <vorbis@xiph.org> Sent: Tuesday, March 05, 2002 10:05 PM Subject: [vorbis] Ogg too good? <p>>> I'm just encoding some files with the latest ogg and it's > too good. I tried encoding things at q1 just to see what > happened and I couldn't hear the differencee between the > ogg and the original files. > > Maybe it was the type of music I was encoding or limitations > in my listening equipment or something but it seems pointless > to have a scale from one to ten and have most of it unused. > Seriously... this is a real problem. What if I *want* mediocre > quality, eg. for downloads where I don't want people to have > the full quality version of a song? > > Also, I think MP3 people will feel cheated if they never need > use a higher quality level than two or three. > > So please, please, let's have worse quality. It's psychological > and maybe silly, but I think q1 should sound a *little* bit bad. > > > > -- > <\___/> > / O O \ > \_____/ FTB. http://www.artlum.com > > > > > > --- >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. ><p><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 Wednesday 06 March 2002 16:05, Colin Dooley wrote:> I'm just encoding some files with the latest ogg and it's > too good. I tried encoding things at q1 just to see what > happened and I couldn't hear the differencee between the > ogg and the original files.I find that I can't tell the difference past 4 a Stereolink 2000, Marantz SR4000 and either Sennheiser HD270's or B&W 603 s2 speakers. It's definitely fairly easy to tell the difference between some of the hard samples (applause.wav, et.al.) below 3, and definitely in the 0-1 range. A certain amount of that comes from training, though, just listening to samples for particular problems that they're supposed to have until I could make the distinction. You might not want to actually do that, however :-)> Maybe it was the type of music I was encoding or limitations > in my listening equipment or something but it seems pointless > to have a scale from one to ten and have most of it unused.It's a scale from 0.00 to 10.00 as it happens.> Seriously... this is a real problem. What if I *want* mediocre > quality, eg. for downloads where I don't want people to have > the full quality version of a song?Encode at something less than three. It _will_ sound lacking to people who care. Additionally, you can do plenty of quality wrecking to the tune before you feed it to oggenc, such as downsampling to 22100hz, or applying some low and high pass filters.> Also, I think MP3 people will feel cheated if they never need > use a higher quality level than two or three.Why does this matter? If they use 3 and it sounds just as good, they end up getting a file that's about 20kbps smaller and still win.> So please, please, let's have worse quality. It's psychological > and maybe silly, but I think q1 should sound a *little* bit bad.Try 0; it's the "oh hell, we'll just throw away bits" mode, and sounds somewhat worse. Also, try listening on better equipment. Personally, I use 4.99 so as to hedge my bets - it comes out to ~128kbps, but it's quite a bit higher than than the threshold of quality that I can detect a difference for. I use that just in case, 1) my equipment improves later on, and becomes more revealing and 2) to reduce the likelihood that I run into an anomalous track that doesn't sound right at what I think my current transparency threshold is. John --- >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.
> Seriously... this is a real problem. What if I *want* mediocre > quality, eg. for downloads where I don't want people to have > the full quality version of a song?Then use q0.1 as someone else noted, or resample your track to 22050khz & then encode at q1, or, if you're making 'teaser' material, just encode a 1 min excerpt. <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.