Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: ABX test results AAC VBR vs. lossless? (Read 14273 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

ABX test results AAC VBR vs. lossless?

Hello, I was reading some threads about the ABX tests - quite interesting.  I was just curious as to whether there are available statistics on reported test results to date on ABX of AAC VBR vs. lossless.  E.g., what is the max score reported so far, or any available statistics.  I'm getting ready to rip a bunch of CDs, I'm leaning toward 320 kbps which I'm pretty sure (based on what I've read) is higher than I really need, but having some test result data might be helpful (e.g. if nobody has ever been able to discern anything higher than 256 kbps AAC VBR w.r.t. lossless in ABX test, why would I need to go higher than that...)  Thanks!

ABX test results AAC VBR vs. lossless?

Reply #1
Is there some reason you're choosing to rip to a lossy format as opposed to a lossless format? Is hard drive space a constraint?

ABX test results AAC VBR vs. lossless?

Reply #2
Is there some reason you're choosing to rip to a lossy format as opposed to a lossless format? Is hard drive space a constraint?


I'm mainly doing this to play my music on the USB of new car head unit (which does not support lossless playback).

ABX test results AAC VBR vs. lossless?

Reply #3
I get the feeling that you may be misunderstanding what conclusions can be reached by finding some ABX results. There have been a few samples that some members here could positively ABX at high bitrates, on certain AAC implementations. So there's a lot of conditions in that statement alone. The fact that you haven't ABX'd them yourself notwithstanding, the data doesn't mean as much to you. Passing an ABX also doesn't say anything about the severity of the artifact. It may rate a 4.9, in other words, barely noticeable after way too much training and concentration, or unrealistically high volume playback. You're ripping them for your car. Safe to say that's a noisy environment.

To answer your question, there are not many examples of people passing ABX tests of high bitrate AAC, but there are some if you search. I'm just trying to point out that using those tests as justification for 320 kbps encoding is not something most of us would do.

ABX test results AAC VBR vs. lossless?

Reply #4
I get the feeling that you may be misunderstanding what conclusions can be reached by finding some ABX results. There have been a few samples that some members here could positively ABX at high bitrates, on certain AAC implementations. So there's a lot of conditions in that statement alone. The fact that you haven't ABX'd them yourself notwithstanding, the data doesn't mean as much to you. Passing an ABX also doesn't say anything about the severity of the artifact. It may rate a 4.9, in other words, barely noticeable after way too much training and concentration, or unrealistically high volume playback. You're ripping them for your car. Safe to say that's a noisy environment.

To answer your question, there are not many examples of people passing ABX tests of high bitrate AAC, but there are some if you search. I'm just trying to point out that using those tests as justification for 320 kbps encoding is not something most of us would do.


OK, thanks.  The question was more out of curiosity - not really looking to use for justification or reaching conclusion.  BTW the highest examples I could find in a quick search on this forum are at ~0.5q.

ABX test results AAC VBR vs. lossless?

Reply #5
I remember there was a thunder clip that always made the iTunes encoder fail and I can't remember if it also affected Nero. There was also a clip from an early Herbie Hancock album, but other than those I can't recall seeing any that affected bitrates around ~250. Personally, I use around 100 kbps and have never noticed anything wrong, with in-ear headphones, closed circumaural headphones, on the computer's crappy speakers, on my stereo's good speakers, or in the car (even a fairly quiet car).

ABX test results AAC VBR vs. lossless?

Reply #6
I remember there was a thunder clip that always made the iTunes encoder fail and I can't remember if it also affected Nero. There was also a clip from an early Herbie Hancock album, but other than those I can't recall seeing any that affected bitrates around ~250. Personally, I use around 100 kbps and have never noticed anything wrong, with in-ear headphones, closed circumaural headphones, on the computer's crappy speakers, on my stereo's good speakers, or in the car (even a fairly quiet car).


Sounds good, thanks.  I had started ripping at 320 kbps but I'm thinking of dropping down if for no other reason than it's taking so friggin long - I'll be here for weeks ripping my whole CD collection.  Assuming it would take less time to rip say 128 kbps.  But I will say the couple I've ripped so far sound unbelievably good in my car (just sitting parked in the driveway of course

ABX test results AAC VBR vs. lossless?

Reply #7
You should perform your own ABX tests w/ foobar2000, then you'll know for certain what bit rate is transparent to you. No reason to encode any higher than that threshold.

AAC @ 256 kbps (ABR) is completely transparent to me.

320 kbps strikes me as excessive.

ABX test results AAC VBR vs. lossless?

Reply #8
I'm not sure about AAC, but MP3 at around 160 VBR (V4) is perfectly transparent for me in my car. I can't imagine you'd need too much more with AAC (and indeed, if you did need higher bitrates, I'd wonder if you should be using AAC at all).

ABX test results AAC VBR vs. lossless?

Reply #9
You really should think about ripping to lossless (like flac).  After that, transcode to whatever lossy format you desire. 

This will save you re-ripping everything if you decide to switch lossy codecs or bitrates.

ABX test results AAC VBR vs. lossless?

Reply #10
You really should think about ripping to lossless (like flac).  After that, transcode to whatever lossy format you desire. 

This will save you re-ripping everything if you decide to switch lossy codecs or bitrates.


And, do make sure your ID tags are correct and as you wish!
That'll be much more difficult to fix years in the future!

ABX test results AAC VBR vs. lossless?

Reply #11
You really should think about ripping to lossless (like flac).  After that, transcode to whatever lossy format you desire. 

This will save you re-ripping everything if you decide to switch lossy codecs or bitrates.


Does transcoding take less time than ripping?

I would imagine that one could use a script to transcode everything?