Head-Fi.org › Forums › Equipment Forums › Sound Science › Answer for n-fect re: mp3 bitrate
New Posts  All Forums:Forum Nav:

Answer for n-fect re: mp3 bitrate

post #1 of 7
7/3/11 at 11:53am
Thread Starter 
n-fect asked me in the locked thread about DACs whether I can tell the difference between 128 and 320 mp3s.

I didn't do tests with mp3s because my main source is a Mac running iTunes and I use AAC. But here is what I found out about that...

At 128, artifacting is fairly common in complex textures. By 192, it pretty much disappears. However, I found one particular record with a very complex texture of massed strings in an orchestra that still artifacted at 192. Bumping it up to 256 completely corrected it. I couldn't determine any difference between 256 AAC and Apple Lossless or AIFF. So my standard settings for my music server are 256 AAC VBR. (a little bit of overkill, just to be safe.) I have had no problem with any files encoded this way. I imagine that would translate to 320 MP3, because I find that AAC sounds the same as MP3 at one setting higher.

I conducted the tests using two iPods line level matched hooked into my amp via line out. I used Sennheiser HD590 cans. Previously, I had dine tests between my iPod and Mac playing AIFF files against several CD players and an SACD player and couldn't determine a difference between any of those as long as the audio files were the same.
Reply
post #2 of 7
7/3/11 at 11:58am

n-fect n-precut???

 

Even my not-reading-ears tell a difference..

Reply
post #3 of 7
7/3/11 at 3:11pm
Thread Starter 
Sorry, my iPad's spell checker is irritating sometimes. I'll see if I can edit it.
Reply
post #4 of 7
7/3/11 at 5:42pm

It's worth noting that sometimes encoders do fall apart even at 320CBR.  These are what Hydrogenaudio have dubbed the "killer tracks".  They're relatively few and far between, but their existence is worth noting.

Reply
post #5 of 7
7/3/11 at 6:30pm

"So my standard settings for my music server are 256 AAC VBR. (a little bit of overkill, just to be safe.) I have had no problem with any files encoded this way. I imagine that would translate to 320 MP3, because I find that AAC sounds the same as MP3 at one setting higher."

 

Essentially, 256 VBR mp3 = 320 CBR mp3 = 256 VBR AAC, you would never be able to tell the difference, but I'm nitpicking. 

 

Hopefully anyone should be able to tell the difference between 128 and 320 kbps lossy, but it seems to me many people are happy with low or medium lossy bitrates. . . personally, I'm not. 

Reply
post #6 of 7
7/3/11 at 6:36pm
Thread Starter 
Quote:
Originally Posted by Shike View Post

It's worth noting that sometimes encoders do fall apart even at 320CBR.  These are what Hydrogenaudio have dubbed the "killer tracks".  They're relatively few and far between, but their existence is worth noting.

I've ripped thousands of CDs using iTunes and I haven't found any tracks that 256 AAC couldn't handle. The one track I found that was problematic at 192 was a mono recording by Sammy Davis Jr on Decca from the early 50s. The orchestral string sound was very complex. Just goes to show that those killer tracks turn up in unexpected places.
Edited by bigshot - 7/3/11 at 6:37pm
Reply
post #7 of 7
7/3/11 at 7:04pm
Quote:
Originally Posted by bigshot View Post




I've ripped thousands of CDs using iTunes and I haven't found any tracks that 256 AAC couldn't handle. The one track I found that was problematic at 192 was a mono recording by Sammy Davis Jr on Decca from the early 50s. The orchestral string sound was very complex. Just goes to show that those killer tracks turn up in unexpected places.


Very much so, as I said few and far between ;)

 

For those interested:

 

http://www.hydrogenaudio.org/forums/index.php?showtopic=83905

 

Should be a good start to find out what killer samples/tracks actually consist of in general.

 

Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Answer for n-fect re: mp3 bitrate