how are mp3's these days?

[johnnyblotter]

I don't stay too up on this stuff, but is mp3 technology improving...the lossless thing or whatever...

I guess, could you really fool an engineer in a real room, or even a real music fan on a decent stereo, by playing the best sounding mp3's versus the CD?

Thanks.

[Lie Mf B]

I don't think mp3 itself is evolving, but there are new formats coming up, and bandwidth and media changes slowly all the time too. Ogg is a nice free alternative to mp3 with a better quality/size ratio, and there's also lossless formats like Flac and Shorten which are about five times the size of an mp3, but half the size of an uncompressed wave. The latter are pretty popular among people sharing whole albums on the net and could -- I suppose -- be an alternative for radio stations too. I guess we're moving both toward smaller lossless formats and higher quality lossy formats to a point where they will meet in size/quality and continue to improve from there.

[bpatural]

AAC files sound amazing compared to mp3s. I have no problem getting emailed a missing vocal or gtr for a song and throwing them in the mix.
-brian

[johnnyblotter]

Okay, that's interesting and good to know...but what about the sound difference? Very minor or completely unacceptable to someone, way, selecting music for a major motion picture?

[kevinc]

Hopefully in a couple years computers performance will be boosted to the level where you can more easily download and store a file as big as a wav.

It`d be nice to see mp3`s go extinct to make way for something CD quality. (or better !)

[whosyourdaddy00]

'standard' 128kbps mp3's sound like Crap, i can 'hear' it on any system, the cymbals always give it away....and variable bitrate mp3's IMHO sound even worse....i'd never try to sell them as a 'real recording' nor would i ever use them in a movie. NEVER

[johnnyblotter]

Thank you, that's exactly what I was looking for.

Anyone gonna deny that?

hmm.

Thanks.

Jon

[largeunit]

Originally Posted by johnnyblotter

I don't stay too up on this stuff, but is mp3 technology improving...the lossless thing or whatever...

I guess, could you really fool an engineer in a real room, or even a real music fan on a decent stereo, by playing the best sounding mp3's versus the CD?

Thanks.

Yes, you could.

[Johnny B]

I say the sooner that the MP3 format dies, the better.

MP3 sucks. CD sucks too.

In this day and age, 16-bit? Gimme a break.

The inferior sound quality of digital is largely due to these old, wimpy formats.

We need something far better, something "new" that can deliver "Killer Sound Quality" like the best that all-analogue gear can routinely deliver.

[Hibbsy]

Anyone ever tried a blind test on a good monitoring system?

Some "audiophiles" tried it and didn't find any difference at high bitrates...

Problem is, I listen to MP3s on a crap system, so I can't say anything...

[Nut]

Originally Posted by Johnny B

I say the sooner that the MP3 format dies, the better.

MP3 sucks. CD sucks too.

In this day and age, 16-bit? Gimme a break.

The inferior sound quality of digital is largely due to these old, wimpy formats.

We need something far better, something "new" that can deliver "Killer Sound Quality" like the best that all-analogue gear can routinely deliver.

And tell me, what regular joe is going to care, let alone own a system that will produce the benefits of higher quality audio formats ?

We sound types make up the vast minority. The 16-bit 44.1KHz audio is perfectly fine for average consumer consumption, heck even the 128kbit .mp3s are !! DVD-A and SA-CDs are for those of us who care.

[zabour]

MP3 is about at good as it needs to be for the hardware that is intented to play it back. Portable DMP, Cell phones, and auto sound systems and home systems...

Fau.. IIS has other lossy formats with better retain ratio than MP3(a) One that never really got noticed was called MP3(pro) and then it got lumped back into the standard MP3 codec: but I think their focus on the pro codec was smaller file size and lower bit rates for CD transparency....

The biggest problem with MP3 is not really its ability to reproduce a signal at 44.k - 16 bits... it is that MP3 keeps really bad time, and not much has been done to solve this problem because the intended audience of MP3 is not really that worried about it, and second because they do not generally own equipment that would allow them to tell the difference.

[excellrec]

Is anybody using MP3's at 320kbps? I find this to be a pretty good compromise myself.. i mean yeah... the 128's sound like crap... but why use 128?

[cebolao]

do you know that every time you go to the cinema for a movie in dolby digital you are listening to the mp3-quality sound?

the sound in dolby digital is heavily compressed, and it's not a very good compression (sonically). on dvd it's the same, but you can have higher kbps there.

[largeunit]

Originally Posted by excellrec

Is anybody using MP3's at 320kbps? I find this to be a pretty good compromise myself.. i mean yeah... the 128's sound like crap... but why use 128?

Exactly. Blaming this compression type for bad sound is like blaming car manufacturers for slow drivers. The cars go faster, the drivers are just choosing not to.

[ttauri]

Originally Posted by excellrec

Is anybody using MP3's at 320kbps? I find this to be a pretty good compromise myself.. i mean yeah... the 128's sound like crap... but why use 128?

I've been using them for laptop DJing... When I first tried them (320kbps mp3s), I did a test and couldn't tell the difference on a very good nightclub system; but as I've worked with them a bit, feels like I'm noticing differences. Have to go pack and do another blind test.

Still, not too bad for what I'm doing, and id3 tags are useful (actually the main reason I tried mp3s in the first place was the metadata).

Peece,
T. Tauri

[smopo24]

(generally) most people don't care about the sound at all, so no additional format will make a dent in my opinion unless you make a mass amount of people aware of it (mp3s being the kleenex of compressed audio formats available on the net). if you are using the best money can buy to record, someone is still going to listen to it at 128 and won't give a shit about the neve, fairchild, ssl, or whatever you used on the session. many just want "the song" and are willing to just get it (friend, internet, whatever) rather than buy it....every time i go into someone's car these days, they have a song or album "ripped" from someone elses' cd (they don't want to know about a disk to disk copy) and it always sounds like crap. they tell me that they can't hear the diffrence (sonically) and don't mind if there is one (that goes for radio too), some people just want to fit the most amount of songs in an ipod or cdr. i just wish there was a consensus about which upgrade people are going to latch onto next....we'll see.

[dasbin]

For me the best compromise of space vs. quality is to be found in a Lame-encoded mp3 (the encoder makes a HUGE difference!) at its highest quality settings in J-Stereo VBR 192min, 320max, with 18000Hz lowpass. At this level I can't ABX on any system except (occasionally) the mains in my studio.

The encoder and settings are what it's all about. J-Stereo is theoretically a slight compromise to the phase and width of the stereo image but I can't tell the difference. Bad things happen when you do a regular stereo mp3, though... the bandwidth is split in half (ie a 128kbps stereo mp3 is actually 64kbps each side... not good for sound!)

As for the people complaining about 16-bit delivery formats... well, it depends on the music genre, but I dare you to dither a modern hard-limited rock/pop song down to 8-10 bits or so. BARELY any difference in sound quality even with this massive and supposedly-unnaceptable change. You'll only really notice it if the song has particularily quiet sections. Most of the content is really in the top 4 bits. As for delivery formats beyond 16-bits... the genre darn well better make use of the extra dynamic range, and the listener better be in an absolutely SILENT environment. What's the point when 95% of music sold is listened to in cars or on mp3 players? We don't even need 16 bits for most of these cases.

Of course tracking/mixing at higher bit depths is a different story, as are more dynamic music genres (classical etc).

[dagg]

we are establishing music selling Internet site and we picked AAC+ codec for its superior quality. I heard many good opinions about this format, and our tests contifmed its excellence. AAC+ at 48kbps is of near-CD quality

Here in the EU they conducted some rigorous tests (professionals from EBU and overall established people from industry). They have clearly chosen AAC+ as a winner over all other formats, they weren't able to differentiate AAC+ at 64Kbps from CD.

It is probably codec of the future. Major mobile phone operators are implementing it into phones in order to enable listening and download near-CD quality music and radio programming from mobile phones.

We recorded some .wav 16bit 44.1 material, encoded it to 64Kbps AAC+ and sent tracks via Internet to the other party. They converted track to wav again, and results were excellent. So, from now on it is possible for anyone to collaborate professionally and efficiently over Internet.

[nukmusic]

use 192k mp3's

[six_wax]

Originally Posted by ttauri

I've been using them for laptop DJing... When I first tried them (320kbps mp3s), I did a test and couldn't tell the difference on a very good nightclub system; but as I've worked with them a bit, feels like I'm noticing differences. Have to go pack and do another blind test.

I've seen djs use burned-to-cd mp3s on bigger systems, and watched the crowd just sort of wilt... even though the songs and sound quality were very good, it just wasn't evoking the reaction...

GML talks about doing blind tests between 96k & analog over *days*... just deciding how each makes him feel...

not all of music is perceived, imo.

[themaidsroom]

spend 2 days with some vinyl
and a good cartridge
then re-assess cd and mp3
makes you think
somethings
not working in today's audio,
it sounded better 40 years ago...............



besides which
the 125 lb shoulder - mounted
vinyl-pod not only
offers a good cardiovascular
workout, but depending
on which storage cage you
choose, upwards
of 50 songs!!

i work for these people when i work, everybody
else comes next


be well

-jack

[krs]

Maybe I'm the only one who'd rather hear a beautiful dynamic mix as a hi-res mp3 rather than a .wav mix squashed to the nines.

my 2 cents. I think limiting is worse for music than mp3's.

[SparkyCanada]

I like to describe mp3's with 2 words - "dynamically challenged"

But that's just me...

Sparky

[brucegel]

Ringtones are the future my friend...so get all your creative juices flowing in that direction and you will be a millionaire soon my son. thumbsup

[ttauri]

Originally Posted by six_wax

GML talks about doing blind tests between 96k & analog over *days*...

Yeah, I think I could've done with some more "burn-in" time with my comparisons.

That said, for a large-ish collection, id3 tags are like crack to me... Love having all the title/artist/label/bpm information stored with the track.

Peece,
T. Tauri

[carlsaff]

Originally Posted by whosyourdaddy00

and variable bitrate mp3's IMHO sound even worse

I understand this is only your opinion, but there are many levels of VBR encoding, some of them capable of delivering near-CD quality. Certainly, an MP3 with encoding that varies between 96kbps and 160kbps will not sound great. However, one that varies between 224kbps and 320kbps will sound almost indistinguishable from the source (provided one is using a quality codec for the conversion).

An interesting test is to convert back to WAV from MP3 and then phase-invert the original mix and sum it with the converted audio. By doing this, you will hear all of the garbage that MP3 introduces in stark relief, and there's surprisingly little garbage in 320kbps CBR (constant bit rate) MP3s encoded using a quality codec like LAME.

For me, the LAME codec at 192kbps CBR or better is acceptable (the higher the bitrate the better), but high-quality VBR is certainly a contender when using LAME.

[DrFrankencopter]

The sound of the encoder seems to be a function of what the source material is. I've used the LAME encoder for MP3 at 320kbps for converting most of my CD collection (to load on my ipod), and found that the encoder sounded good on most of the cleaner sounding tracks, but it just sounded WRONG on anything that had high gain distorted guitars. I've since switched to the AAC format (at max bitrate) and found it's much better at coping with distortion, and just gives me more consistent results overall.

Regarding 16 bit audio...yes I feel it's adequate for a distribution format. Geez, with today's uber-limited mastering 10 bit audio would be more than enough (ever wonder how those lossless encoders can be so efficient, its mostly because the music doesn't use all the bits).

Cheers,

Kris

[hammerfall]

I've encoded using the highly refined LAME compile found here with the "--alt-prestet standard" command, and the distortion you describe isn't there. I can notice an artifact occasionally when listening on nearfields, but it's still the best mp3 encoding i've heard. The VBR is so good that the gains by increasing to full 320 bitrate are negligible.

BTW are you using iTunes? If your files are processed with that Soundcheck enabled, I would think it could introduce gain issues (I don't use iTunes).

The problem with the other formats is the universal adoption of mp3 by hardware and software makers. As desktop storage approaches the terabyte level, FLAC looks really good, though. I think apple will go it alone with AAC and their lossless codecs.

[whosyourdaddy00]

hmmmm good points there....i like the polarity flipping idea, i'm gonna hafta try that sometime...

cheers