[Cyrilix]

Listen for sss-sibilance.

Listen for depth of bass, subwoofer like feeling once you get really low, and its fullness/roundness (that's a term that's a bit difficult to explain, but underamplified DT880s have a bass that doesn't feel "whole").

Listen for clarity in really fast music that has to separate lots of notes in short time frames.

Listen for details (toughest thing to do, imo and things only pop up for you from time to time).

Hmm...those are the only suggestions I can give other than to listen for tonal differences (which are more of a matter of preference).

 

[Mercuttio]

You know, it totally depends on what encoded the audio in the first place. Some of the iTunes' store AAC files are actually quite nice.

If it's poorly recorded rock, or doesn't have much going on, then it's harder to tell the difference. If it's Orchestral stuff, you've got a lot of complex things going on and you need a higher bitrate or you start to lose stuff... you'll get that nasty compression artifact noise in the background.

If your system is good enough to give you the timbre of specific instruments / voices, then that's a good way of telling too. I don't like hearing the female voice (or higher tenor parts) compressed below 224 AAC, I start to lose the tactile feeling.

 

[DevilDog]

I know this is a terrible analogy, but everything just sounds "compressed." I'd say the best way is to polarize things. Try a 96kbps mp3, compare that to the original. Then go to a 128kbps and compare to the original. To me it's the cymbals just don't have the sizzle and the guitar strings don't have the full vibration. I guess it's all in the overtones. The music also loses impact and the soundstage lessens.

 

[tola555]

1. find quiet place
2. make yourself comfortable
3. close your eyes (important) or dim light
4. enjoy

 

[gigit]

1. Get your setup ready
2. Take 3-4 months off of work/school
3. Make 500 gallons of coffee
4. Listen until your ears bleed, then listen some more (no pain..... right?)

or just be aware of the bitrate you are listening to, and the rest should come naturaly

 

[Mr Iriver]

I dont think everything below 256 is crap. I find 128 decent with some music, and 192 the sweetspot. I find the difference most noticeable in the shimmering sounds, and in the drums, where at really low bitrates (below 128) they sound more like cans. But like I said I find it depends a lot on the music. I find it more obvious on some music than on others.

 

[catscratch]

Quote:

Originally Posted by PeeeMeS Practice your techniques on lying straight in someone's face "Anything below 256 Kbps is crap" is a lie... or that person is disillusioned You pick one. 1. Lie 2. Be out of touch with reality

******

It really does depend on your music. I find that music that has a lot of textured information in the treble is usually a dead giveaway of compression. Strangely enough, electronic music is a good example if it has a lot of transparent treble sounds. On lower-bitrate recordings, these sounds take on a whistly tone that's very characteristic of compression. I can identify compressed recordings pretty easily just by listening to the quality of the treble samples in something like Shpongle's "Tales of the Inexpressible." Also, you can listen to the tonal background, listen "between the notes" so to speak, and you'll hear that lossless has a lot more air and space between each instrument. Separation is better, textures are crisper and are clearer, and everything is more in focus. It's subtle, but it's there.

Here's how I hear it:

With 128k, you're losing massive amounts of soundstage. You can very easily pick out 128 from CD just by the amount of soundstage lost. Textural information is also very much lost and textures are smeared. The whistly tone indicative of compression is very much present. 160k has some soundstage, but it is still very evident. 192k is where soundstage starts being more open, but it's still not CD-like, and you're still getting a lot of compression artifacts. With 224k, you're getting close to CD quality, but with careful listening, you can still pick out the artifacts. Soundstage is pretty good though. With 320k, compression artifacts are at a minimum, though if you know what to listen for you can sometimes spot them. Here, what matters more is instrument separation and tonal background. They are definitely lacking in 320k. But, it's pretty subtle, and you do need a good system to hear it, together with music where it matters.

My .02

P.S. If you can't hear it, then don't worry about it. Alternatively, start listening to lossless for about a week, then switch back to lossy once you've trained your ears to lossless. Perhaps you'll pick it up.

 

[PeeeMeS]

Quote:

Originally Posted by catscratch ****** It really does depend on your music. I find that music that has a lot of textured information in the treble is usually a dead giveaway of compression. Strangely enough, electronic music is a good example if it has a lot of transparent treble sounds. On lower-bitrate recordings, these sounds take on a whistly tone that's very characteristic of compression. I can identify compressed recordings pretty easily just by listening to the quality of the treble samples in something like Shpongle's "Tales of the Inexpressible." Also, you can listen to the tonal background, listen "between the notes" so to speak, and you'll hear that lossless has a lot more air and space between each instrument. Separation is better, textures are crisper and are clearer, and everything is more in focus. It's subtle, but it's there. Here's how I hear it: With 128k, you're losing massive amounts of soundstage. You can very easily pick out 128 from CD just by the amount of soundstage lost. Textural information is also very much lost and textures are smeared. The whistly tone indicative of compression is very much present. 160k has some soundstage, but it is still very evident. 192k is where soundstage starts being more open, but it's still not CD-like, and you're still getting a lot of compression artifacts. With 224k, you're getting close to CD quality, but with careful listening, you can still pick out the artifacts. Soundstage is pretty good though. With 320k, compression artifacts are at a minimum, though if you know what to listen for you can sometimes spot them. Here, what matters more is instrument separation and tonal background. They are definitely lacking in 320k. But, it's pretty subtle, and you do need a good system to hear it, together with music where it matters. My .02 P.S. If you can't hear it, then don't worry about it. Alternatively, start listening to lossless for about a week, then switch back to lossy once you've trained your ears to lossless. Perhaps you'll pick it up.

If it takes weeks of training and then *maybe*, just *maybe* he'll be able to discern very little differences (and that's *only* with very high high high end gear) and it has to be a good day... How the hell can someone claim "anything below 256 Kbps is crap"

We're not talking 64 Kbps... we're talking 256 Kbps!
This isn't about wether you can hear differences or not... this is about wether "anything below 256 Kbps is crap."

Now, reread my post because I think you misread me.
My post:

"Practice your techniques on lying straight in someone's face

"Anything below 256 Kbps is crap" is a lie... or that person is disillusioned
You pick one.
1. Lie
2. Be out of touch with reality"
I was responding specifically to the people that would claim anything below 256 Kbps is crap.



if you still disagree I'll use some logic
If: <256Kpbs == crap
Then:

"P.S. If you can't hear it, then don't worry about it. Alternatively, start listening to lossless for about a week, then switch back to lossy once you've trained your ears to lossless. Perhaps you'll pick it up."
==
"P.S. If you can't hear [the differences between the original CD and crap], then don't worry about it. Alternatively, start listening to lossless for about a week, then switch back to [crap] once you've trained your ears to lossless. Perhaps you'll pick it up [the differences between the Original CD and crap]."






Oh yea... and in response to your post

I can roll my eyes more than you!

 

[WhiteShadow]

I agree with the original poster, because sometimes I can't hear the difference between 128kbps and Flac, but I can hear the difference between different headphones, and mp3 players. Maybe I am just not looking in the right places.


*Wears flame suit*

 

[nightfire]

I agree with the other posters here who pointed out it totally depends on:

1. The original recording quality
2. The encoder used (lame owns!)

But I'd also add familiarity with material. If you listen to a lossless song exclusively (one you know and love) for a few days, then listen to even a 320kbit version, with high end gear (with no weak links - including ICs, DAC, Amp, phones/speakers) the difference can be quite obvious. Not that it always is.

For example, in no way can I tell a 320kbit Fiona Apple track from original. But with more complex music spanning more of the frequency spectrum (Metallica, Taproot, some Bjork) the difference is "night and day." I don't mean it's bad - it surely isn't, but I can nail off whether it's the MP3 or the original within 10 seconds of song to 100% accuracy.

Down to 256, it gets more obvious, and at 192 and below, it's reaaaaaaly really bad. Only on my Etys. Not my Grados or my friend's DT431.

Because the differences are almost exclusively in the highs. Everything is sharper, more aggressive, slightly distorted, and eventually (with <192kbit) full of artifacts. Like other posters noted, look for sssssibilance, cymbal finishes, distorted guitar, snare attack, dynamics, soundstage, etc.

Actually a really good test track is Taproot - I Will Not Fall For You. Listen closely to the hihat in a 256kbit mp3. Then listen to the original. The hihat is delicate in the original, but somehow that is lost in the mp3. It sounds aggressive.

 

[slinger1182]

Quote:

Originally Posted by nightfire For example, in no way can I tell a 320kbit Fiona Apple track from original. But with more complex music spanning more of the frequency spectrum (Metallica, Taproot, some Bjork) the difference is "night and day." I don't mean it's bad - it surely isn't, but I can nail off whether it's the MP3 or the original within 10 seconds of song to 100% accuracy.

I on the other hand seem to pick up differences more easily on acoustic pieces than pieces with heavy instrumentation. So for example, I can more easily pick up differences on Nusrat Fateh Ali Khan than something like Metallica. To each his own I guess.

 

[nightfire]

Quote:

Originally Posted by slinger1182 I on the other hand seem to pick up differences more easily on acoustic pieces than pieces with heavy instrumentation. So for example, I can more easily pick up differences on Nusrat Fateh Ali Khan than something like Metallica. To each his own I guess.

Hmm.. well I didn't mean acoustic vs. distorted... more lots of instruments or different pieces across the band. Like it's harder to compress simultaneous drums, guitar, vocals, and bass (or a symphony) than it is a single voice. Never heard any Nusrat Fateh Ali Khan..

What kind of composition is it?

 

[Magnum22]

Quote:

Originally Posted by nightfire Only on my Etys.

interesting. i'm a relatively late-comer to the mp3, i thought i was the last guy on earth who still carried a cd wallet around with him. I would listen to my cd's on loop for weeks at a time. i pretty much had every note and nuance memorized. i usually used my aiwa pcdp and an olschool koss ktx-pro, the old black ones (i'd kill to get em back). the last few days i was wondering if i only think i'm hearing as much as i do cuz i know it's supposed to be there.

if anyone knows where i can find those old koss headphones, just name the victim.

 

[slinger1182]

Nusrat Fateh Ali Khan . Though he was mainly a qawwali singer, he also sang folk music from India/Pakistan. He also collaborated with Eddie Vedder for the soundtrack of Dead Man Walking and I think did some soundtrack work for THe Last Temptation Of Christ.

 

[Aryolkary]

I say that have the music as well encoded as possible. If the highest available is lossless, then use that. If you have some album in MP3, not matter the bitrate, and can't get that album in better quality, that's better than not having it.
I think the only question regarding what bitrate to use, is when it comes to storage for portability. If one's storing the music into the PC, lossless all the way. Some albums only take 200mb and some even less. It isn't that much difference with an album in MP3 (320kpbs would be around 100-150mb) and MP3 cannot be reversed to original WAV.

Regarding how to listen: I say the most important thing is to listen to the Music, not to the sound. Probably I think against audiophiles in that sense. What matters is the music. Suppose a great interpretation of some piece of music is recorded awful, and a lot of hiss and background noise is there. I still listen to it (not too often if it's a pain to listen) to see what that great performer offers.
The better the equipment, the better the sound, cleaner, crisper, etc. However, the most important is the music, not the sound. It would be like caring only about the colour of the oleum in some painting (in a technical way I mean). I think it's a great pleasure to appreciate the sound, but it's not the right thing (in my view). I think one has to listen to the music with great energy, not analyzing, or meassuring. Just being one with the music. Someday I'll make a thread about how audiophiles will never listen to music like a real musician. Flame suit ON