Brown henke dither sighting.. what to do?

[hoffman2k]

Hoffman2k - no, SRC problems don't show up as sped-up or slowed-down sounds. That would just be a SR error, so to speak. Sample Rate Conversion leads to digital artefacts, more or less noticeable depending on the quality of the conversion and the type of material. What 3phase is claiming is that this is happening in instances when technically there should be no need for it, leading to unnecessary degradation of audio quality.

Thanks for making that clear.
Warping can also lead to digital artifacts. How to you distinguish between SRC artifacts and Warping artifacts?

[massenmedium]

I think the answer to that in this case is that the OP is using Repitch mode so it's entirely based on SRC rather than granular type techniques.

Not sure though, had better leave it to the OP to explain.

[fx23]

from what i understood, he was using repitch mode at original and project tempo, so supposed transparent,

then a tiny tempo change and back to original enabled the unwanted SRC process.

[hoffman2k]

from what i understood, he was using repitch mode at original and project tempo, so supposed transparent,

then a tiny tempo change and back to original enabled the unwanted SRC process.

Yeah I see where I got things mixed up now. Still, there was no need for the insults.

Been trying to test this here and with the Hi-Q interpolation on, I'm unable to reproduce any artifacts while changing the tempo from and to the original. But with the Hi-Q off you can hear it very clearly as high pitched aliasing.
So I was wrong when I said the button doesn't have an effect while warping. And what the info box of that button says also isn't correct. That info probably still applies to how it was before it got changed in Live 7.
Ah well. Serves me right for trying to understand something and remembering the wrong stuff. I never dispute people including myself can be wrong, but there was no call to be rude about it.

[nylarch]

to recap:

blah blah blah

Ableton is like an MPC on steroids and if you're savvy like DJ Shadow you can make a record on it like "Entroducing" that people will listen to forever (and was made on an MPC). Or just a good solid record like Caribou's "Swim" which was done in Live 8.

You probably don't want to record the London Philharmonic with Live or edit complex multi-tracked drums. Use Logic/Cubase/Tools for that.

blah blah blah.

[leisuremuffin]

yeah, that recap is close...
but do you want the truth? no instrument, recording medium, mixing board, or piece of software has ever made anything good to listen to. talented people do that. and they do it with whatever is at hand.


.lm.

[Palmer Eldritch]

No probs here with "Brown henke dither"

cheers, palmer

[arafel]

There are way too many "if's" with the problems 3phase has, as he does not give enough specific (and coherent) information.

For example:
What is the sample rate of his samples?
Is he using samples of multiple sample rates? (will = SRC)
Is that samplerate different from the one set in the Project preferences? (will = SRC)
What is the samplerate he has set in Project preferences?
Has he changed this mid-project? (big no-no)
All of these functions are described as 'non-neutral' by Ableton (page 519+ of manual), which = sound is changed (degraded).

We need more information, and at least one sentence from him that is not a stream-of-consciousness rant.

Posting a few sound examples would be a real help too.

Edited my error in '<32 bits=SRC - Thanks Palmer

[arafel]

from what i understood, he was using repitch mode at original and project tempo, so supposed transparent,

then a tiny tempo change and back to original enabled the unwanted SRC process.

Supposed to be.. ya.
Unless his audio clip had a different sample rate than set in his preferences = SRC
Or he automated volume levels of the audio clip = non neutral sound change (page 520 of manual)
Maybe he only noticed it (BHD?) after the tempo change?

3phase did not give enough details, to replicate what exactly he was doing.
I respect his temerity, but for a man obsessed... its not that thorough.

[twisted-space]

What is the sample rate of his samples? (anything less that 32bit = SRC)

Are you sure this is what you meant to type?

[mattadms]

The Brown-Henke Dither is explained in detail in an eponymous paper published by Misters Brown and Henke in all the
important scientific journals. READ IT BEFORE CREATING ANOTHER THREAD! Honestly.

Yeah? Where exactly?

[Palmer Eldritch]

What is the sample rate of his samples? (anything less that 32bit = SRC) .. Or he automated volume levels of the audio clip = SRC

Are you sure this is what you meant to type?

It seems arafel mixed up to different things.
Bits are used to pack the dynamic into a digital system - in theory 1 bit means 6 dB dynamic / 16 bit = 16*6 = 96 dB - and has nothing to do with the sample rate (frequency, measured in Hz). There is no need for SRC (Sample Rate Converting) when changing something on the level side of a signal or when converting to a lower bit depth.
But Here the process of "dithering" can be useful to prevent degradation partly !!!

(But never heard of" brown-henke dither", maybe 3phase can explain what he means with that or, at least, can post a useful link.)


cheers, palmer

[arafel]

What is the sample rate of his samples? (anything less that 32bit = SRC) .. Or he automated volume levels of the audio clip = SRC

Are you sure this is what you meant to type?

It seems arafel mixed up to different things.
Bits are used to pack the dynamic into a digital system - in theory 1 bit means 6 dB dynamic / 16 bit = 16*6 = 96 dB - and has nothing to do with the sample rate (frequency, measured in Hz). There is no need for SRC (Sample Rate Converting) when changing something on the level side of a signal or when converting to a lower bit depth.
But Here the process of "dithering" can be useful to prevent degradation partly !!!

(But never heard of" brown-henke dither", maybe 3phase can explain what he means with that or, at least, can post a useful link.)


cheers, palmer

Thanks Twisted & Palmer --- late night again..wrong time to type. Sry for confusion in my statement .
I'm pretty much 16/44100 or 32/48000 - so when I saw 32 I thought 32/48000 - fail on my part.
I've edited my previous post to help eliminate confusion from other readers.

But the logic still stands that if OP is using multiple samples with dif samplerates (44100 amd 48000, et al) that SRC will be happening in Live.

[arafel]

What is the sample rate of his samples? (anything less that 32bit = SRC) .. Or he automated volume levels of the audio clip = SRC

Are you sure this is what you meant to type?

It seems arafel mixed up to different things.
Bits are used to pack the dynamic into a digital system - in theory 1 bit means 6 dB dynamic / 16 bit = 16*6 = 96 dB - and has nothing to do with the sample rate (frequency, measured in Hz). There is no need for SRC (Sample Rate Converting) when changing something on the level side of a signal or when converting to a lower bit depth.
But Here the process of "dithering" can be useful to prevent degradation partly !!!

(But never heard of" brown-henke dither", maybe 3phase can explain what he means with that or, at least, can post a useful link.)


cheers, palmer

Hey Palmer, ya..its not SRC (just non-neutral in Ableton terms )
Seems totally nonsensical, but page 520 of manual has this passage about volume automation = sound distortion

31.3.3 Volume Automation
Automation of volume level results in a change in gain, which is necessarily a non-neutral operation. But certain implementations of automation envelopes can result in audible artifacts, particularly if the envelopes are not calculated at a fast enough rate. Since Live 7, volume automation curves are updated for each audio sample, resulting in extremely low levels of distortion.

The volume passage is just below the SRC passage.. so merged them in my quick read.. fak up to late again.

I'm gonna take my own advice and read and type slower..

[Palmer Eldritch]

Hey Palmer, ya..its not SRC (just non-neutral in Ableton terms )
Seems totally nonsensical, but page 520 of manual has this passage about volume automation = sound distortion

Hi arafel,

No problem to make mistakes

I just wanted to separate things because 3Phase talked about a problem with "Sample Rate Converting" (which, to be honest, I never observed till now - also not after doing the steps 3phase has described in this thread).

And I also think you are right, there are not enough information of 3Phase to resolve his issue with SRC (at least for me).

greetings, palmer