DAW Audio Quality does vary... and Live 7 sneak peak!

[disturb]

I couldn't find that sneak peek either.

i thought it was pretty much 'we have improved the audio engine, but that's so subtle that you won't be able to hear it'

or "We need our annual fix of upgrade fees".

Somehow I don't think this will be enough...

Doug

f**k that, i wanna know about the new stuff with cycling 74

[jens]

The theoretic dynamic range of 24 bit Integer is 144 Db. Anything below that isn't even theoretically possible to hear. So the errors in the -180 to -144 dB range is of no consequence whatsoever. The best specs i've ever seen on a DA converter is the Lavry Gold which has a 127 Db dynamic range so even that can't recreate a -130 dB error. It's lost in the noise floor.

that aside:

the rounding-errors can lead to random differences between the resulting files if you do the same render again on the same host. So in order to do a scientificially valid test you do not only need to compare several hosts with eachother but also those hosts with themselves. And then you have to apply a stochastic formula to calculate whether the difference (if there is any) in the rounding-errors between a) host X and host Y and b) host X and host X is significant or just random.

As long as the Ableton-guys haven't done their tests this way (and I doubt they have) they aren't scientific and we can savely disregard the results, audible or not.

[nuffink]

I'd be much more impressed if they finally fixed the retarded MIDI handling.

[Reverse Engineer]

So...which host handles inaudible frequencies the best then?

My wallet is open and my head is up my ass (there's still room for a knob or two though...as ever).

[disturb]

So...which host handles inaudible frequencies the best then?

My dog seconds that inquiry.

[jupiter8]

The theoretic dynamic range of 24 bit Integer is 144 Db. Anything below that isn't even theoretically possible to hear. So the errors in the -180 to -144 dB range is of no consequence whatsoever. The best specs i've ever seen on a DA converter is the Lavry Gold which has a 127 Db dynamic range so even that can't recreate a -130 dB error. It's lost in the noise floor.

that aside:

the rounding-errors can lead to random differences between the resulting files if you do the same render again on the same host. So in order to do a scientificially valid test you do not only need to compare several hosts with eachother but also those hosts with themselves. And then you have to apply a stochastic formula to calculate whether the difference (if there is any) in the rounding-errors between a) host X and host Y and b) host X and host X is significant or just random.

As long as the Ableton-guys haven't done their tests this way (and I doubt they have) they aren't scientific and we can savely disregard the results, audible or not.

You're kidding right ? But i don't see a punchline so i'm not sure.

[jens]

You're kidding right ? But i don't see a punchline so i'm not sure.

What part of my post you have trouble to comprehend?

[jupiter8]

You're kidding right ? But i don't see a punchline so i'm not sure.

What part of my post you have trouble to comprehend?

Random differences ? I thought a computer would calculate the same no matter what time of day/phase of the moon/days left for the acopalypse it is. Even if that isn't the case how big would those rounding errors be? In the -180 dB range perhaps ?

[jens]

how big would those rounding errors be? In the -180 dB range perhaps ?

yes, just like those results the Ableton-guys found - that was my point after all...

[jupiter8]

how big would those rounding errors be? In the -180 dB range perhaps ?

yes, just like those results the Ableton-guys found - that was my point after all...

You said their test was unscientific and should be disregarded because of random rounding errors. I say random rounding errors or not does'nt matter. You still can't hear the difference.

[jens]

No-one ever said you could hear the differences - not even Robert Henke himself.

[DevonB]

How about these for new taglines for Ableton?

Talentless wanker + Ableton 7's improved audio engine still = talentless wanker at the end of the day.

Or

Ableton - We've improved what you can't hear!

Devon

[Meffy]

:bang:

No-one ever said you could hear the differences - not even Robert Henke himself.

Audio concerns that which is audible. If the differences are not audible, I have a hard time seeing how they could be construed as differences in audio quality.

[edit: This is not directed at you, jens. Just taking your statement as a jumping-off point. And for those who don't know, I'm a reasonably happy Live user.]

[Hink]

No-one ever said you could hear the differences - not even Robert Henke himself.

Audio concerns that which is audible. If the differences are not audible, I have a hard time seeing how they could be construed as differences in audio quality.

except now that it's been said everybody and their brother will think they hear it [standing on a crowded street corner staring at nothing in the sky]

[Meffy]

A psychological difference, that would be. :-} I'm not immune to the effect, just that I'm aware of it and don't believe my ears are particularly good so I'm willing to discount vague, unmeasurable impressions when shown proof they're in my imagination.