mp3 Test

orbital · July 01, 2008, 12:58:29 AM

Quote from: drogulus on June 30, 2008, 03:42:29 PM
Whaddaya mean you failed? You identified the B file as lower in quality than the others. If it was a guess it was a correct one. AAC at comparable bitrates may have a slight advantage over mp3, but at these high rates it may not matter.

No it wasn't a guess,the open cymbal in B did sound a bit dull compared to the other two. But then again, I thought A and C were similar whereas C is supposed to sound better than A, right? Or does a 320 AAC sound closer to the original than a 320 mp3?

drogulus · July 01, 2008, 12:12:19 PM

Quote from: orbital on July 01, 2008, 12:58:29 AM
No it wasn't a guess,the open cymbal in B did sound a bit dull compared to the other two. But then again, I thought A and C were similar whereas C is supposed to sound better than A, right? Or does a 320 AAC sound closer to the original than a 320 mp3?

The AAC 315 (File A) is about the same as the 320 mp3 in total bits used. It might be better at dealing with certain problems than mp3 because it's a newer standard. Some rigorous listening tests have been done that show this, though the tests were done at much lower bit rates. It depends on which encoder is used, too.

drogulus · July 11, 2008, 03:27:37 PM

I have an idea that might be fun to explore.

If someone would provide an uncompressed file of ~1 minute in length of some very demanding music, I'll dl it and encode it using LAME 3.98. It could be any kind of music so long as it seems like it might challenge an encoder. Old analog stuff could do this but it might be better to limit this to digital sources that are 16/44 and up. This is just an example of what I mean, though it could be very different music. Heavy metal and some other pop forms can trip up encoders, partly due to high levels as well as all those nasty overtones from distortion. I think we should limit it to classical for mutual comprehension and interest.