Abletons Sound Quality

djadonis206 · Post by **djadonis206** » Wed Jan 17, 2007 7:24 pm

dj superflat · Post by **dj superflat** » Wed Jan 17, 2007 8:56 pm

if you didn't get around to reversing phase to confirm what your ears seem to indicate, there's no real proof it sounds different (particularly in any meaningful sense). that is, it's hard for others to care if you're not relying on something objective (though you should obviously trust what your ears seem to be telling you, even recognizing that the placebo effect is way strong).

knotkranky · Post by **knotkranky** » Wed Jan 17, 2007 9:13 pm

It's important to understand the nature of Live. I really didn't get it for a very long time because i'm deeply intrenched in what I require and I'm a real bitch when it comes to audio. If your a real audio pig too, understand that it's best to use live for what you specifically need it for and do not make across the board comparisons with straight forward daw's. They are very different.

If you can use resample in realtime then do that. If you've created a big song that's testing the limits of your computer then render to disk.

This is the nature of Live being grossly flexible and a few things to all people. With that in my mind, it makes sense.

Meef Chaloin · Post by **Meef Chaloin** » Wed Jan 17, 2007 10:35 pm

i wish someone sort this out once & for all. Tests please

leisuremuffin · Post by **leisuremuffin** » Wed Jan 17, 2007 10:49 pm

there's nothing to sort.

plug ins sound different, audio interfaces sound different, but digital pcm audio files are the same regardless of the app recording or playing them back.

in terms of the magical mix bus, if you mix in the box without plugins or timestretch (warping, in ableton's case), it's gonna be the same no matter what you use. i defy someone out there to show me a current daw that uses different math for summing than live.

.lm.

leisuremuffin · Post by **leisuremuffin** » Wed Jan 17, 2007 10:55 pm

ahh, and i guess i could post this from the render thread as well:

Robert Henke wrote:Render, Realtime, CPU Usage
----------------------------------------

if your computer can do 100 calculations per second and your audio processing only needs 40 calcuations per second the CPU load is 40% and you could do twice the amount of processing in realtime but not three times since this would need 120 calculations per second.

each time the sound card wants a new sample the computer does all calcuations, starting from reading the file, filtering it mixing it, what ever... and finally handles it over to the soundcard. once all is done it waits and does nothing. in this time other apps can do other things.

If you render, the computer does not need to wait for the soundcard, because there is no idea of realtime anymore. if a process takes only 10% CPU, render will be almost 10 times faster then realtime, and even a process too complex to be done in realtime can be done.
But a 45620 sample long delay line will always be 45620 samples long, regardless if this is realtime or not.

There is no such thing as interpolations.
Render sounds the same as resampling in real time. If it does not in very special cases, and we can reproduce, this is simply a bug and will be fixed.

The reason why a well known HD recording application does bouncing in realtime is not sound quality. It is the brutal fact that their architecture cannot deal with no realtime and changing it would mean a complete re-design of their HW / software

Robert

so that's what's up with the difference between recording to a new track in real time and render in live ----> none.

.lm.

Meef Chaloin · Post by **Meef Chaloin** » Wed Jan 17, 2007 11:25 pm

no im with you, ive never heard any difference, but surely someone running a few varied tests would stop all these threads once and for all, solid proof.

leisuremuffin · Post by **leisuremuffin** » Wed Jan 17, 2007 11:39 pm

nothing will stop these threads from coming up every once in a while.

a lot of people out there talk shit. Often those people command a certain amount of respect, and even though they are 100% wrong, some people will believe them.

now, i'm going to go ahead and continue to respond to these kind of posts with the truth, but at the end of the day, it doesn't matter to me if somebody else thinks live has poor sound quality. I know what my output sounds like.

.lm.

Tarekith · Post by **Tarekith** » Wed Jan 17, 2007 11:50 pm

Read some of the lnks posted here already, I've already done quite a few tests between Live, SX3, rewire modes, resampling versus render, and the different warp modes. The data is there, from about 6-7 months ago I believe. It largely ended up not cha nging anyone's mind anyway, so I'm not going to revisit this yet again.

dj superflat · Post by **dj superflat** » Wed Jan 17, 2007 11:55 pm

i think part of the reason this keeps coming up with live is that warping audio does affect the sound, so people either are talking about changing sound when tempo changes or not realizing that having certain warp types on at the original tempo will change audio. that is, live's flexibility can make it seem like audio is degraded when it's actually a user choice (to use complex warp, etc.).

i also think this is a remnant of the analog world, where different decks and consoles sound different because of circuitry, head, etc. but digital is digital.

udp · Post by **udp** » Thu Jan 18, 2007 12:03 am

The only difference we've found in the past is that of volume. We ran some extensive tests on a section of one of my pieces to see if rendering vs. resampling made any difference. The phases never exactly cancelled each other, but we easily saw the only difference was volume. There is a bit off trickery with Live's panning laws (still wrapping my head around this...) but it's -3db at center if I remember correctly, so it's a matter of adjusting for that. Sorry if others find errors here, I'm going off memory.

mission · Post by **mission** » Thu Jan 18, 2007 12:10 am

QUESTION/ANSWER?:

So I agree with those who hear the difference. I definitely can as a native Logic user. I just like how fast I can move in Ableton since a lot of shit Im doing is on an MPC now.

So if I go MPC to, say, channels 1/2 on my desk and DAW audio out through my interface (good DACs) to 3/4 on the console ... main output back into the INs on the desk and use an audio track to "monitor" and record the input channel on the breakout box then im good to go and essentially circumventing the rendering engine of Live and merely depending on the quality of my desk and sound card (DACs).

Do I have that right?

Cuz that's what im doing...

mission · Post by **mission** » Thu Jan 18, 2007 12:11 am

udp wrote:The only difference we've found in the past is that of volume. We ran some extensive tests on a section of one of my pieces to see if rendering vs. resampling made any difference. The phases never exactly cancelled each other, but we easily saw the only difference was volume. There is a bit off trickery with Live's panning laws (still wrapping my head around this...) but it's -3db at center if I remember correctly, so it's a matter of adjusting for that. Sorry if others find errors here, I'm going off memory.

Volume meaning what?

That Im gonna be louder resampling and recording the audio live?

Or... I dont know what -3db at center means, sorry... is this of the RMS value or... ? Sorry if Im an idiot with that one.

leisuremuffin · Post by **leisuremuffin** » Thu Jan 18, 2007 12:18 am

do whatever you want, mission, but all your doing is adding another pass of a/d conversion on your DAW track.

-3db in the center refers to the "equal contour" panning law. basically, it reduces the volume of a track by 3db when panned to the center so that the volume will seem the same in the mix when panned left or right.

I'm not even sure live uses this panning law, but i know the vast majority of mixing boards do.

.lm.

leisuremuffin · Post by **leisuremuffin** » Thu Jan 18, 2007 12:21 am

just checked, yep -3db in the center.

plus (from :Craig Anderton)

"Panning laws originated in the days of analog mixers. If there was a linear gain increase in one channel and a linear gain decrease in the other channel to change the stereo position, at the center position the sum of the two channels sounded louder than if the signal was panned full left or full right.

To compensate for this, it became common to use a logarithmic gain change response to drop the signal by ‑3dB RMS at the center. You could do this by using dual pots for panning with log/antilog tapers, but as those could be hard to find, you could do pretty much the same thing by adding tapering resistors to standard linear potentiometers. Thus, even though signals were being added together from the left and right channels, the apparent level was the same when centered because they had equal power."

also, i checked with a mono track, with stereo tracks it should just be balance rather than -3d at center. --->except it's not, still -3db in center for stereo track.

.lm.