What bitrate do you use for music?

[ZenChick]

One reason why CD's can sound "cold and dead" is because many record companies like to take advantage of a CD's higher dynamic range and compress the living hell out of the music to make it sound "louder" than their competitors (see this Wikipedia article for more details). Although the music may seem louder, the overcompression pretty much kills most of the advantages of digital audio over analog.

[CrowdAds]

[dfkt]

Originally Posted by ZenChick

One reason why CD's can sound "cold and dead" is...

The other reason (well actually almost a legend, nowadays) dates back to the 1980ies when CDs were new technology and recording engineers were still mastering CDs the same way they did it for vinyl. Since vinyl needs more treble to sound good, they used the same EQing for CDs, making them sound harsh and bright (remember Dire Straits' "Brothers in Arms", one of the first DDD recordings?). Of course this is no longer the commonly used practice, but the preconception still steems to stick for some people.

[derf]

Originally Posted by dfkt

The other reason (well actually almost a legend, nowadays) dates back to the 1980ies when CDs were new technology and recording engineers were still mastering CDs the same way they did it for vinyl. Since vinyl needs more treble to sound good, they used the same EQing for CDs, making them sound harsh and bright (remember Dire Straits' "Brothers in Arms", one of the first DDD recordings?). Of course this is no longer the commonly used practice, but the preconception still steems to stick for some people.

But don't phono preamp have an RMAA curve (specific preset EQ for those not familiar with the term) built in to compensate for the vinyls sound rendering?

[dfkt]

I meant it the other way round: old 1980ies CDs didn't sound as good as vinyl at that time, because mixing/mastering engineers had no experience with the new medium and the need for different EQing. This of course has become a lot better in the last two decades - but the old preconception still seems to stick with some vinyl zealots.

[saratoga]

Originally Posted by sassafras

I think you're completely ignoring my point about frequency responses. the major differences between a highly compressed mp3 and say like -b 320 are the results of highpass filtering

Lossy audio typically isn't highpass filtered. Theres really no point in doing so.

Perhaps you mean lowpass filtered? Honestly, of all the complaints about lossy quality, lowpass is one of the least relevant. It simply doesn't matter.

Originally Posted by sassafras

and the results of the fourier smoothing.

What do you mean by "fourier smoothing"?

Originally Posted by sassafras

Likewise, setups that can't appropriately reproduce rapid transient attacks (like rapid cymbal attacks, etc) will a listener likely not really be able to discern all of the differences between highly compressed and otherwise.

This is preecho, and as I've already said, it has nothing to do with equipment quality. There is no disortion present in typical analog equipment remotely equivilent to preecho. Its not something that happens in analog equipment due to causality.

Originally Posted by sassafras

So I want to disagree when you say that there is no loss of temporal resolution from equipment.

Well, since casuality applies to analog equipment, i'm not sure how you will get preecho at least. Unless your speakers can time travel.

Originally Posted by sassafras

Speakers or headphones with very accurate low inertia drivers aren't going to do well with certain tracks that have a very strong DFT applied, whereas I doubt you'd notice the difference on some five dollar OEM throw ins.

DFT == Discrete Cosine Transform?

I don't understand what this is supposed to mean.

Originally Posted by sassafras

I can't imagine that you're seriously saying that a cheap pair of pack-ins will still reveal all the differences between strong and light compression.

Thats exactly what I'm saying. Analog and digitial artifacts are completely different. They don't sound alike or really influence one another unless they're of extreme magnitude.

[sassafras]

Maybe then I'm simply misguided. Mostly what I know is that I'm speaking from anecdote and applying what I know to the opinion I already have. Either way, I don't like digital artifacts and there is no reason to listen to them.

[Drive-In-Freak]

Originally Posted by dfkt

The other reason (well actually almost a legend, nowadays) dates back to the 1980ies when CDs were new technology and recording engineers were still mastering CDs the same way they did it for vinyl. Since vinyl needs more treble to sound good, they used the same EQing for CDs, making them sound harsh and bright (remember Dire Straits' "Brothers in Arms", one of the first DDD recordings?). Of course this is no longer the commonly used practice, but the preconception still steems to stick for some people.

I think you just may have something there. The last CD I bought was "A Momentary Lapse of Reason" on the day it came out..ok..ok I know..go ahead with the obvious if you wannna... After getting a new decent CD deck and checking out a few I was really disappointed, so I flat out gave up and never bothered again. Hell I gave the deck away and have not had one wired into my system from that point on.

[AFSlug]

I use CDEX with lame.
I use
VBR 32-320.
Quality 0
Samplerate44100Hz
Joint Stereo.
Normalize to 98%

[sassafras]

Don't normalize, that's a bad idea. Use ReplayGain or Mp3Gain to perform lossless & reversible volume conversions. Normalization is a bad idea all around.

[The DarkSide]

Originally Posted by sassafras

Normalization is a bad idea all around.

How come? I thought it just levels the volume of the tracks, or am I mistaken?

[d_headshot]

I either use v0(downloaded files) or 320kbps. I ripped most of my cds in that format so I'm not going to waste time and change them all back to vbr.

[dfkt]

Originally Posted by AFSlug

I use CDEX with lame.
I use
VBR 32-320.
Quality 0
Samplerate44100Hz
Joint Stereo.
Normalize to 98%

I assume you mean by that you use the -v0 preset and add your own switches to it? If that's the case, it's not a good idea... the LAME presets are already optimized at source level, tweaks you can't achieve with switches.

Originally Posted by nywytboy68

How come? I thought it just levels the volume of the tracks, or am I mistaken?

It's ok with non-destructive, reversible tools like ReplayGain or MP3Gain, but physically normalizing a file decreases quality and is mostly not needed. You can't fix hot mastered albums anyways, you can only increase the volume of albums with low recording levels. And ReplayGain does a better job at that than standard normalizing.

Especially the crappy normalizers in apps like CDex and EAC do a terrible job.

Originally Posted by d_headshot

I either use v0(downloaded files) or 320kbps. I ripped most of my cds in that format so I'm not going to waste time and change them all back to vbr.

You could try WinMP3Packer, it losslessly converts CBR to VBR. It might pack some of your 320k files smaller without re-encoding.

[d_headshot]

Originally Posted by dfkt

I assume you mean by that you use the -v0 preset and add your own switches to it? If that's the case, it's not a good idea... the LAME presets are already optimized at source level, tweaks you can't achieve with switches.



It's ok with non-destructive, reversible tools like ReplayGain or MP3Gain, but physically normalizing a file decreases quality and is mostly not needed. You can't fix hot mastered albums anyways, you can only increase the volume of albums with low recording levels. And ReplayGain does a better job at that than standard normalizing.

Especially the crappy normalizers in apps like CDex and EAC do a terrible job.



You could try WinMP3Packer, it losslessly converts CBR to VBR. It might pack some of your 320k files smaller without re-encoding.

Thanks!

[Bravin Neff]

I use CBR 320.

I will probably redo my library with pure wave files and stop using compressed files.

Frankly, I'm from the school that cringes just thinking about "lossy" anything, and it is evident within a few years we'll be laughing at our obsession with "saving harddisk space" when it just gets cheaper by the day. While that day isn't today it seems to be rapidly approaching, and I'll regret having my LP's converted to MP3s.

Maybe I'll start this weekend.

[sassafras]

Don't use WAV, use FLAC. It's compressed but lossless. You take half the space of WAV for the exact same quality. There is no reason not to.

For information, look here:
http://wiki.hydrogenaudio.org/index....act_Audio_Copy

[Bravin Neff]

Originally Posted by sassafras

Don't use WAV, use FLAC. It's compressed but lossless. You take half the space of WAV for the exact same quality. There is no reason not to.

There are reasons not to. The main one is library consistency and compatibility. I am a musician and I do a lot of professional recording. Everything in that world is wave based (or SDII based, but I don't use Pro Tools).

Wave is a ubiquitous standard that is understood and accesible by everything. This is possibly not entirely true in the consumer audio hardware world (i.e., I know the Zune can't read wave files, which is ironic), but definitely in the pro music world.

I am aware of the lossless compression schemes, like FLAC, OGG, Apple's thing, my favorite (from Steinberg -- Wavelab lossless), and others.

But I'm weary of all of them. What do they gain you? They gain you disk space. What do you risk? You risk: future compatibility, you risk buying into yet another standard to deal with, you risk yet another file extension to manage. You have yet another headache from the archivist's point of view to worry about.

All for disk space, which is no longer at a premium?

No thank you.

[dfkt]

Originally Posted by Bravin Neff

You risk: future compatibility, you risk buying into yet another standard to deal with, you risk yet another file extension to manage.

You might risk some of these things when you buy into closed-source solutions like ALAC or WMA-L, but certainly not with open solutions like FLAC. With FLAC you can always convert them back to WAV, in a matter of seconds, even for large files. And the result is always bit-perfect (I've tested it). It sure looks like a future-proof concept to me. Moreso than all lossy compressed formats, even - since there's no way back to WAV without losses.

But the needs of music producers certainly are way different than the ones of music listeners on the go. Since this is a portable audio site, and space is valuable on portable players (even on HDD players when you use lossless audio), FLAC sounds like a very good idea to me for lossless audio.

[sassafras]

Originally Posted by Bravin Neff

There are reasons not to. The main one is library consistency and compatibility. I am a musician and I do a lot of professional recording. Everything in that world is wave based (or SDII based, but I don't use Pro Tools).

Wave is a ubiquitous standard that is understood and accesible by everything. This is possibly not entirely true in the consumer audio hardware world (i.e., I know the Zune can't read wave files, which is ironic), but definitely in the pro music world.

I am aware of the lossless compression schemes, like FLAC, OGG, Apple's thing, my favorite (from Steinberg -- Wavelab lossless), and others.

But I'm weary of all of them. What do they gain you? They gain you disk space. What do you risk? You risk: future compatibility, you risk buying into yet another standard to deal with, you risk yet another file extension to manage. You have yet another headache from the archivist's point of view to worry about.

All for disk space, which is no longer at a premium?

No thank you.

I fail to see how you're risking future compatibility when lossless FLAC encodes can be quickly and perfectly unpacked back to WAV. With a DirectShow filter for FLAC, any DirectShow application (including many professional sound application will seamlessly integrate FLAC into the workflow.

If you're so adamant that WAV is the only standard, then so be it, but I'm just suggesting that you save yourself the diskspace for the sake of the fact that it's no extra work to use FLAC, it's completely futureproof and while it may get cheaper for a bigger hard drive in the future, it will never be free.

[Bravin Neff]

Originally Posted by sassafras

If you're so adamant that WAV is the only standard, then so be it,

I'm not adamant about that at all. There are quite a few audio standards, and wave is but one of them.

Originally Posted by sassafras

but I'm just suggesting that you save yourself the diskspace for the sake of the fact that it's no extra work to use FLAC

I'll investigate it further, but I haven't found anything that counts as "no extra work," though I'd be interested to see how little (or how much) work it really amounts to. When it comes to files you already have, I would guess at a minimum you would have to:

1. Encode the file.
2. Make a decision about what to do with the original file. In the case of a CD, I suppose you do nothing. For non-CD files, my guess is that if diskspace is the motivation, you should either delete the original or put it onto some other medium for archival reasons.
3. Recatalog with the new files.
4. Recatalog the new archival series (assuming you didn't delete the source files in step #2).
5. Perhaps purchase new software or hardware if your current stuff isn't compatible with reading it.

Originally Posted by sassafras

it's completely futureproof

I don't know of very many things whose future obsolescence would inspire this much confidence against its possibility, but I'll take your word for it.

[sassafras]

Well, it's no less futureproof than WAV, considering that with a trifle of work, it can become WAV in the matter of a few seconds. But future proof in the sense that you will always be able to access it? Not necessarily. I can't say that in twenty years any computer system will even read WAV files. But for the sake of common sense, then yes, it is future proof.