Getting a bit (no pun intended) frustrated with the all mp3s are created equal comments occasionally made on this site, so here is a poll. Some settings are only available on the LAME encoder. Poll assumes all encoders are created equal, which we know is not the case. Poll assumes all settings are created equal, which we know is not the case (is the 160 in 160 vbr the starting point, a rough middle ground, etc.). Where is transparency to your ears (and equipment) reached?

I see that nobody really wants to admit that MP3 is any good.
I have my old collection of AC/DC recorded as LAME r3mix. They sound just as good as the original on my equipment. Which may be the crux of the problem. Those with high end CD players may notice the differences more than than us with average stuff. Also, some of us older guys have some hearing loss and we may not be able to hear the subtle differences.
I did a little test once with a song recorded at 160 cbr and r3mix then decoded, and a copy of the original song uncompressed all on one disc. I played this disc on my old poor quality Sony pcdp. On that player the 160 version did sound less dynamic. The r3mix and original were extremely close. I should try this again with the better cd player I have now.

--alt-preset standard is damn near transparent to my ears. It has trouble on mid drum and bass drum transients, as well as some heavy guitar distortion (Red Hot Chili Peppers kind of distortion). I voted 320 cbr. --r3mix seems to have a few more artifacts on the above points than --aps.

My system: Winamp 2.81+MAD -> Sonica -> Sony V6

I've been using 192 VBR and have been very, happy with it on my iPod. Now with the AAC which I guess is MP4. Things are getting more interesting. I downed some songs from the new iTunes store 128 AAC. Sounds very, good. I would say about the same as what I'm use to. I would like to try some higher bitrate with this AAC.

Depends on the music and depends on how much i listen to it. All of my Cds are recorded via EAC + Lame at 192 at least. The mp3s i listen to the most i tend to recode at 320.

Easier to dig thru Mp3s then zillions of CDs..... =P



Foobar 0.6 -> TBSC -> Reciever -> HD280 etc.

Ahaha mp3 again... hopefully this format will never die from all the attention, good or bad, that it gets...

I use 256 k cbr for most of my music and occasionally go up to 320 k for some types, usually DDD-type recordings of classical stuff.

Here's an interesting thought I'd like to present though. When I try to compare even 192 k mp3s to the original, I find it very difficult to pick out any difference on my equipment. However, after listening to the 192 stream for some time, like at least 30 minutes or so, I find my ears start to get tired whereas they don't if I listen to the CD.

I read somewhere an analogy of this. Imagine you're listening to drumming played by some fantastic drummer, doesn't matter whether it's rock or classical or whatever, just drums (cos they're kinda repetitive). Now take that drumming and synthesize it. Assume the synthesizer is of such quality that it is impossible to tell the difference between the synthesized version and the human version.

If you take a single beat or a very small extract from both recordings and compare them, it should be impossible for even the most sophisticated audiophile to detect any difference.

However, if you play back both recordings for an extended period of time, your ears will tell you the difference because subtle things come out, such as the small errors and variances the human drummer will make. I.e. the synthesis version will become repetive and boring where as the human one will not.

I am not trying to form a direct comparison between sysnthesized drums and MP3 . However, it is interesting to think about how your ears work, and maybe what you hear over an extended period of time is more important than being able to pick out definite differences.

Badmonkey you may find this interesting.

Lame preset standard is what I use, but I have found a few very rare tracks that I can tell the difference between the wav and the mp3 this way, so I voted for lame preset insane, as I have never been able to tell the difference with this.

-dd3mon

What I find interesting is the possibility that high bit-rate MP3s may actually sound better than CDs on headphones. Tyll has a theory: the MP3s cut out some spatial information that doesn't sound good on headphones anyway.

I've done blind AB/AB testing between --aps and FLAC. There were a couple CD's that I could tell the difference on, but only if I was listening hard. One was symphony, and the other was alternative soft rock. Also of note, the symphony one (Freedom, by Michael W. Smith) was HDCD-encoded. As for the other one, (Stereotype Be by Kevin Max) let's just say it's one of the best mixed CD's I've ever heard. Period. However, I recently got a HDCD encoded Newsboys album (Thrive), so it would be interesting to see how HDCD rock does.

(-:Stephonovich

For the most part I can't tell a difference between CD and --alt preset extreme on my equipment. (There are a few things Mp3 has trouble with) However the conveinance of Mp3 is what I really enjoy.
M-Audio Revolution - Corda HA1 - HD600

there is no level where mp3 (or any audio compression) is transparent to cd (in my opinion). cd and digital audio files will always be separate in my mind.

perhaps if somebody can make mp3s of various quality and then decompress them back to wave and burn all waves to cd (including the original) and then listen to them on a high-end cd player, that would be interesting. anything under that kind of testing just doesn't sit well with me.

however, for my ipod i use --alt-preset standard (extreme for my favorite artists) and i am quite pleased with the quality.

Just ripped 4 tracks off Thrive with EAC (secure mode) and encoded them using:

LAME 3.90.2 --alt-preset standard -m j
LAME 3.90.2 --alt-preset standard -k -m j
OGG GT3B1 -q6
Uncompressed WAV

I discovered that disabling the lowpass filter resulted in no quality gain that I could detect. It did, however, boost the filesize by about 200KB.

The OGG files sounded as good as the MP3's, which in turn sounded as good as the WAV files. This re-enforces my belief that for 99% of Redbook recordings, properly encoded lossy is transparent.

I think I may now get a portable and use OGG files on it, and have FLAC on the computer, if only for the archive ability. I already have all my CD's archived on CD-R's...

(-:Stephonovich

MPC using MusePack 1.15a with the mppenc --braindead --ms 0 --nmt 32 --tmn 64 command line is as good as I've heard lossy compression sound.

NGF

grinch, that type of testing has been done before.

check out r3mix.net's guide to quality of mp3's

www.r3mix.net (click on "Quality" tab)

Edit: they performed tests using $10k+ equipment (Sennheiser Orpheus and a B&W Nautilus 803 with Marantz amp) and audiophiles could not tell the difference between 256kbps mp3 and CDs

edit: r3mix.net seems to be down - mirror here:

http://users.belgacom.net/gc247244/