What quality levels would I use with the aoTuV Ogg encoder to produce files of equal quality compared to LAME's V and --alt presets? I did some searches and the answers seemed to vary, so I would appreciate any input to the matter before I start a mass encoding session.
Am I correct that the general opinion is LAME extreme preset = LAME V2 = aoTuV Q7?
Full Version: What are the Ogg Vorbis and LAME equivilants?
q3 apm
q4
q5 aps
q6
q7 ape
q8
q9 api ("yeah, i know it's CBR")
q10
(a very unofficial guideline...)
later
q4
q5 aps
q6
q7 ape
q8
q9 api ("yeah, i know it's CBR")
q10
(a very unofficial guideline...)
later
QUOTE(Frostmourne @ Dec 27 2005, 02:25 PM)
Am I correct that the general opinion is LAME extreme preset = LAME V2 = aoTuV Q7?
no, -V2 is --preset standard and --preset extreme is -V0
From reading around here, I also think that Lame -V2 is equivalent to around q5 or q6 with Vorbis.
I agree with the "scale".
Notice that Vorbis will produce much lower bitrates
Notice that Vorbis will produce much lower bitrates
Hard question! From my limited testing of Q6 and above i even think you donīt need q9 to beat lame api. At ~q7 with the last Aotuv builts you will have problems to pick the original or the vorbis file. There are more problems with api i strongly believe.
This is much speculation and i only conclude this from me having problems finding problems for vorbis above q6
So i wonder why others say you need ~q9 to fit api. Some input is welcome!
This is much speculation and i only conclude this from me having problems finding problems for vorbis above q6
So i wonder why others say you need ~q9 to fit api. Some input is welcome!
I performed a comparison long time ago (mid-2004) between LAME -V3/-V2 and Vorbis called "megamix" at -q5,50/5,99/6,00. I did it on classical musical samples only without any "killers" (which are often harder to handle with MP3):
http://www.hydrogenaudio.org/forums/index....showtopic=23355
The test is now outdated, but at this time, and for this kind of music (wider dynamic), Vorbis -q6 was better than LAME -V2 but the same Vorbis at -q5,99 performed worse.
Situation has now changed, because both LAME and Vorbis has improved in the meantime. The test's results and conclusions are not valid anymore.
Just keep in mind that both formats/encoders have different issues and artifacts. Comparisons between quality in general rarely makes sense IMO.
http://www.hydrogenaudio.org/forums/index....showtopic=23355
The test is now outdated, but at this time, and for this kind of music (wider dynamic), Vorbis -q6 was better than LAME -V2 but the same Vorbis at -q5,99 performed worse.
Situation has now changed, because both LAME and Vorbis has improved in the meantime. The test's results and conclusions are not valid anymore.
Just keep in mind that both formats/encoders have different issues and artifacts. Comparisons between quality in general rarely makes sense IMO.
QUOTE(guruboolez @ Dec 28 2005, 01:28 AM)
I performed a comparison long time ago (mid-2004) between LAME -V3/-V2 and Vorbis called "megamix" at -q5,50/5,99/6,00. I did it on classical musical samples only without any "killers" (which are often harder to handle with MP3):
http://www.hydrogenaudio.org/forums/index....showtopic=23355
The test is now outdated, but at this time, and for this kind of music (wider dynamic), Vorbis -q6 was better than LAME -V2 but the same Vorbis at -q5,99 performed worse.
Situation has now changed, because both LAME and Vorbis has improved in the meantime. The test's results and conclusions are not valid anymore.
Just keep in mind that both formats/encoders have different issues and artifacts. Comparisons between quality in general rarely makes sense IMO.
http://www.hydrogenaudio.org/forums/index....showtopic=23355
The test is now outdated, but at this time, and for this kind of music (wider dynamic), Vorbis -q6 was better than LAME -V2 but the same Vorbis at -q5,99 performed worse.
Situation has now changed, because both LAME and Vorbis has improved in the meantime. The test's results and conclusions are not valid anymore.
Just keep in mind that both formats/encoders have different issues and artifacts. Comparisons between quality in general rarely makes sense IMO.
Even if this makes it a bit OT. Is there a set of some samples where a vorbis newbie like me can listen to vorbis specific problems at higher q settings? This may also help our thread starter to do own conclusions.
Of course in the end it all depends on (1) your ear, (2) your sound equipment (e.g. speakers, amps, soundcards), and (3) the ambient noise.
For instance, in my case where I have (1) quite good ear, but (2) el cheapo earplugs + PDA, and (3) commuting by bus, -q2 is good enough. The bus's diesel engine conveniently drowns all artefacts
Then at my house, where my family have (1) good ears, (2) good enough speakers (Altec Lansing), but (3) noisy fans & barking dogs, -q3 is required. When your dog starts 'harmonizing' with whatever song you play the transparency given by -q4 (or better) will be irrelevant
For instance, in my case where I have (1) quite good ear, but (2) el cheapo earplugs + PDA, and (3) commuting by bus, -q2 is good enough. The bus's diesel engine conveniently drowns all artefacts
Then at my house, where my family have (1) good ears, (2) good enough speakers (Altec Lansing), but (3) noisy fans & barking dogs, -q3 is required. When your dog starts 'harmonizing' with whatever song you play the transparency given by -q4 (or better) will be irrelevant
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.