Help - Search - Members - Calendar
Full Version: Any plans for 160 or 192 kb tests?
Hydrogenaudio Forums > Lossy Audio Compression > AAC > AAC - General
keiths
Based on poll results, it seems that the majority of people encode their AAC files at 192. And 160kb came in as the 2nd most popular bitrate.

I understand that it is usually harder to conduct listening tests at higher bitrates because it is harder to ABX them, obviously.

But... assuming that people decided on their chosen bitrate (i.e. 160 or 192) after performing their own listening tests (as opposed to just picking these numbers out of hat to decide which they would use), it would seem that there are a good number of people who are capable of ABXing higher bitrate encodes. I know from my own tests that there are a fair number of "real world" tracks that I am personally able to ABX at these higher bitrates.

I am particularly interested in Nero vs. iTunes at each of these bitrates (as opposed to iTunes 160 vs iTunes 192) since I'm still undecided on which performs better for ME overall (i.e. for ME, it seems that Nero performs better more often than iTunes but when Nero has a problem with a specific sample, the problem is much more blatent than when iTunes has a problem with a sample). But, I also know that, for ME, I am better at ABXing some things more than others (i.e. I am better able to pick up when an encoded file sounds "flatter" than the original even when it is very slight but not too good at picking up on when an encoded file has something like "exagerated" bass unless it is ridiculusly "exagerated" or something).

PS. My above "claims" of which sounds better, which has more blatent problems when it does have a problem, etc. are actually not meant to be "claims" but just rather to serve as an example of how different people are able to pick up on different things (i.e. "flatter" vs. "exagerated" bass vs. other artifacts, etc) to make my point that I think that "official" higher bitrate ABX testing would be very beneficial and should be doable with the right set of samples and volunteers willing to concentrate at a higher level than they might normally be used to when performing listening tests at lower bitrates.

PPS. I would be willing to help out by suggesting samples for which I am able to pick up on the "flatnest" when compared to the original. And maybe others can suggest samples where they can pick up on other things like "exageratted" bass, artifacts, etc. And, of course, I'd be willing to be a volunteer to do the testing.
SirGrey
I think you 'll not find 1000 followers to perform such a test.
To be honest, from my expirience, there are rather few people, who can distinquish such a high bitrate aac from original on many tracks.
But if we select few problem samples only, test could possibly be biased toward one of the encoders.
Second problem is, that it is very hard (became tired quickly) and time consuming to test such a high bitrate encodings...
Search the forums - such a test was asked many times, but was never (it seems) performed...
BTW, you can visit audiocoding.com, there is a person named jans-jurden there.
Some years ago he has had a really "golden" ears rolleyes.gif
Ask him, may be he can give you more info about comparing Nero and iTunes on high bitrates.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2008 Invision Power Services, Inc.