Let's see how you ABX!

Cpyder

So, I've been running some ABX testing on various types of music using Foobar. I'm sort of working my way up to telling the difference between lossless and 320kbps. 128 vs. flac is a joke. 192 vs. flac gets a little trickier but there are definite differences. If I'm not in a listening mood, I usually don't do much better than guessing. When I actually sit down and spend some time, it gets easier.

I've found that at higher compressions, you can sometimes tell a difference but are unsure which recording is which. They both sound good and very listenable.

How good are all of you? I'd like to see some uploads of ABX testing results showing how high of compression vs. lossless you can determine. Let's try to keep the probability of guessing pretty low. Say 10/12?

Also, what are some of your techniques? I know for lower compressions, the higher frequencies are a dead giveaway. And on medium compression, you can sometimes tell a difference in the sound stage. But for full 320kbps AAC vs. Flac, what are some of your ways of determining which is which?

Let's hear some feedback and see some results!

I'd love to see someone ace a 320 vs. lossless ABX test. It's WAY harder than I thought. 320 really isn't THAT bad, even for serious listening.

Airplay355

Dynamic range is also a give away for me. More compression=much less of a range in a recording that I normally find dynamic. The bass also becomes sloppier at lower volumes with compression. Deep bass stays tight in an uncompressed format even when I crank it.

TNRabbit

Depends DIRECTLY on the abilities of your system/setup.

And your EARS, of course....

BlueFox

I use lossless compression, so there is no difference between compressed and uncompressed sound.

Ender

I follow Airplay's method. Also, you have to choose the right music. There are alot of songs out there that don't have wide complex frequency passages. When you try to differentiate quality of encoding on a song that only has a few guitar chords and drum bass that doesn't go extend very deep, then you're kinda screwed on trying to figure out differences between speakers, much less encoding.

heiney9

MP3's suck for any kind of true audiophile system. End of story!!!!

ESavinon

heiney9 wrote: »

MP3's suck for any kind of true audiophile system. End of story!!!!

+1.
I can hear differences between wav and lossless.

Cpyder

ESavinon wrote: »

+1.
I can hear differences between wav and lossless.

I think you may have something wrong with you if you are telling the difference between wav and lossless. LOL I think you mean lossy.

But no, I'd like to see someone pass lossless vs. 320 and upload the results. It's no laughing matter. Really. I find it damn near impossible. Maybe I'm using the wrong type of music but it's definitely not my systems fault.

Cpyder

heiney9 wrote: »

MP3's suck for any kind of true audiophile system. End of story!!!!

Upload some 320 vs. lossless results and then make the statement again. I don't disagree, and where I can, all of my music is lossless but I still have a fair amount of lossy music.

heiney9

Cpyder wrote: »

Upload some 320 vs. lossless results and then make the statement again. I don't disagree, and where I can, all of my music is lossless but I still have a fair amount of lossy music.

I have and I stand by my statement.

But, like you I do have some lossy material that I use for background music, casual listening, etc., as well as using Comcast's Music Choice channels, which are very lossy streaming files, to discover new music.

But I don;t use my main rig or do any serious listening with any type of lossy files. Hence my original statement.

unc2701

So if I ripped a wav and 320 mp3, then converted the 320mp3 to a wav and uploaded both, you'd be able to tell which was which?

tonyb

I don't do digital downloads,I prefer concerts,live music,where the drums hit you in the chest,the women are half naked,and the beer is luke warm.Try down loading that.Rock on.

BlueFox

unc2701 wrote: »

So if I ripped a wav and 320 mp3, then converted the 320mp3 to a wav and uploaded both, you'd be able to tell which was which?

Of course. Just because a lossy file is re-encoded in WAV does not mean it has any more data than it did as an MP3.

Cpyder

BlueFox wrote: »

Of course. Just because a lossy file is re-encoded in WAV does not mean it has any more data than it did as an MP3.

I think he meant - Could you recognize the data was missing?

unc2701

Cpyder wrote: »

I think he meant - Could you recognize the data was missing?

Exactly- we can make this interesting. Upload 2 wav files, one of which has had data removed due to the mp3 conversion and see how well people can tell.

heiney9

What would that prove? Anyone can use software to compare the WAV files and tell which was derived from an original lossy file. There would be no "control" in the experiment on the listeners end. And frankly no "control" on the uploaders end either.

I've detected it several times at home. I agree they can be close but the lossy always loses.

H9

BlueFox

Cpyder wrote: »

I think he meant - Could you recognize the data was missing?

I know what he meant, and that is why I answered. No matter how a file is encoded if the data is missing it will be obvious when compared to a lossless file encoded in the same manner.

unc2701

Jesus dude, unwad your panties. These things rapidly digress into a pissing contest of who's got the most golden ears and I'm just trying to think of a unbiased means by which we can determine our thresholds. It's up to you to not run software to cheat it out.

Oh, and I can hear when the guitarist's urine PH is under 5, so I don't know WTF we're even talking about compression bit rates.

heiney9

Sorry if I gave the impression my "panties are in a wad"; they're far from it. I just don't see what it would prove as a whole. Certainly I encourage people to try it at home with their music and if your happy with lossy files.............then "don't worry, be happy".

But lossy is lossy is lossy so I tend to make my point that 320kbps lossy files are and will never be the same as true WAV files. Much like a digital conversion of an analog signal will never be 100% of the original analog signal.

Happy listening

H9

unc2701

Keiko wrote: »

You wear panties to H9? I got those ones that have the days of the week embroidered into them. Helps me keep my days straight and I love the way they feel against my sack.

awwwwwwwwwww dammit. didja have to paint that picture? Where's the... shouldn't you be posting asian chicks in the aforementioned panties???

Anyhow, H9, gotcha. If I listen for it, I can hear it on just about any recording (there's some garbage that I'm pretty certain was _recorded_ lossy), but for a lot of things, I don't care. Out back, grilling. Meh. Stuck at work reading the new SOPs that are just like the old SOPs? Meh.

'course, I've got the sober turntable and the drunk turntable. I switch over to the technics 1200 when I start thinking I might break something....

mmadden28

How about some people list some tracks that they use for comparison and what to listen for and where (time) in the track. Not everybody is in tune or knows what to listen for.