Should I use 44.1kHz or 88.2kHz sampling rate?

Discuss music production with Ableton Live.
re:dream
Posts: 4598
Joined: Fri Dec 28, 2007 9:42 am
Location: Hoerikwaggo's sunset side...
Contact:

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by re:dream » Sat Mar 30, 2013 8:34 pm

Tone Deft wrote: I'm less of a producer and more of a guy that starts songs and gets bored with them.

....

for now I have the wife staring at me waiting to go to lunch. gotta go...


LOL all of the above sounds familiar 8)

jlgrimes
Posts: 1770
Joined: Mon Jan 22, 2007 1:55 am
Location: Atlanta, Ga

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by jlgrimes » Sat Mar 30, 2013 8:43 pm

Andreilg3 wrote:Hello, I have read countless articles on this in the past, such as Dan Lavry's "White Papers" and Xiph.Org's article on sampling rate, but I have yet to find an explanation that applies more to my setup.

I am an electronic musician, I do not record anything live and I am entirely inside of the box in Ableton and mainly use vst synths. I have absolutely no interest in keeping small file sizes or putting my stuff on a CD, I just want my stuff to be top quality at the cost of any "convenience". I use a decent amount of samples however, (mostly Ableton Suite pack samples) and they are usually at 16/24 44.1kHZ. Based on my reading, Dan Lavry says that 60kHz (or 88.2kHz being the next closest option) is the optimum sampling rate for audio quality, but Xiph.Org says that anything over 44.1kHz will only introduce distortion.

My question is, is there any reason I shouldn't just set and render with Ableton Live to 88.2kHz? I don't really understand how upsampling my samples will affect them or how that affects my vsts, but from Mr. Lavry's papers and many many articles I've read, 44.1kHz isn't as good in audio quality (however insignificant) as 88.2kHz is (or even 48kHz which is dvd standard).

This has been bugging me for a really long time now and the amount of confusion and arguments surrounding something so objective astounds me, so any advice or input would really help. Thank you.
Another benefit from higher sample rates is lower latency. Keep in mind though that this also mean more CPU, memory and disks pace.

I would only take this route if u have a sound card that don't have good latency.

kitekrazy
Posts: 795
Joined: Sat Jan 16, 2010 7:16 pm

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by kitekrazy » Sat Mar 30, 2013 8:56 pm

jlgrimes wrote:
Andreilg3 wrote:Hello, I have read countless articles on this in the past, such as Dan Lavry's "White Papers" and Xiph.Org's article on sampling rate, but I have yet to find an explanation that applies more to my setup.

I am an electronic musician, I do not record anything live and I am entirely inside of the box in Ableton and mainly use vst synths. I have absolutely no interest in keeping small file sizes or putting my stuff on a CD, I just want my stuff to be top quality at the cost of any "convenience". I use a decent amount of samples however, (mostly Ableton Suite pack samples) and they are usually at 16/24 44.1kHZ. Based on my reading, Dan Lavry says that 60kHz (or 88.2kHz being the next closest option) is the optimum sampling rate for audio quality, but Xiph.Org says that anything over 44.1kHz will only introduce distortion.

My question is, is there any reason I shouldn't just set and render with Ableton Live to 88.2kHz? I don't really understand how upsampling my samples will affect them or how that affects my vsts, but from Mr. Lavry's papers and many many articles I've read, 44.1kHz isn't as good in audio quality (however insignificant) as 88.2kHz is (or even 48kHz which is dvd standard).

This has been bugging me for a really long time now and the amount of confusion and arguments surrounding something so objective astounds me, so any advice or input would really help. Thank you.
Another benefit from higher sample rates is lower latency. Keep in mind though that this also mean more CPU, memory and disks pace.

I would only take this route if u have a sound card that don't have good latency.
Seems almost contradictory. At a higher rate you would eventually have to increase the latency to reduce CPU usage.


re:dream
Posts: 4598
Joined: Fri Dec 28, 2007 9:42 am
Location: Hoerikwaggo's sunset side...
Contact:

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by re:dream » Sun Mar 31, 2013 7:38 am

Awesome interview. When I grow up I am also going to be that cool.

Andreilg3
Posts: 70
Joined: Tue Oct 11, 2011 9:00 am

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by Andreilg3 » Mon Apr 08, 2013 3:54 am

The Finn wrote:
Tone Deft wrote:the flip side is believing bad advice.

.

... speaking of which, what's the *good* advice on dithering when doing a mix down?

I hear a lot about dithering being a bad idea; then again if I believe that Monty Montgomery video on the whole 192 saga, dithering is not a problem. What do you think?
Sorry for bumping after all this time but I've been re-reading all of this and came across this question.

Dither is basically a process that removes quantization error (error that occurs in any PCM digital system due tot he fact that bit depths only have a set amount of amplitudes they can choose from; 8-bit quantization = 256 possible values; 16 bit quantization = 65,536 possible values)by replacing it with white noise. This white noise sounds quieter (even though it is technically louder) than the quantization noise since the quantization noise is distortion and adds ugly harmonics. White noise is essentially tape hiss and even if you can hear it, which I doubt you can, you have to keep in mind that people 20 years ago threw tape hiss all over everything so there is nothing bad about it. Using shaped dither instead of flat dither, like a POW-R instead of triangular respectively, make the noise even less audible since it moves the density of loudness from the audible frequency ranges to inaudible frequency ranges.

http://www.youtube.com/watch?feature=pl ... RlohQw-1DY
http://productionadvice.co.uk/when-to-dither/
http://productionadvice.co.uk/when-in-doubt-dither/
http://wiki.xiph.org/Videos/Digital_Show_and_Tell

I'm currently writing a research paper about a lot of digital audio topics and from everything I've read there seems to be 0 practical reason to not use dither every time you process...or at the very least on your last one and definitely if you go down in bit depth. Essentially though, to cover your ass I'd suggest doing 24 bit all the time which will lower the noise floor significantly to the point where most people will agree that dithered or not, the noise is inaudible. I do both.

re:dream
Posts: 4598
Joined: Fri Dec 28, 2007 9:42 am
Location: Hoerikwaggo's sunset side...
Contact:

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by re:dream » Mon Apr 08, 2013 5:32 am

Thank you, that is very informative.

I wonder, though, why e.g. Tarekith says *not* to dither when you give him a track for mastering...

Tarekith
Posts: 19065
Joined: Fri Jan 07, 2005 11:46 pm
Location: Ableton Forum Administrator
Contact:

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by Tarekith » Mon Apr 08, 2013 6:12 am

Because most dither in people's daws is noise shaped, and THAT you want to avoid until the very end. Better to just keep it 24bit and then you don't need to worry about it.
Tarekith
Ableton Forum Administrator
https://tarekith.com

Andreilg3
Posts: 70
Joined: Tue Oct 11, 2011 9:00 am

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by Andreilg3 » Mon Apr 08, 2013 6:50 am

Tarekith wrote:Because most dither in people's daws is noise shaped, and THAT you want to avoid until the very end. Better to just keep it 24bit and then you don't need to worry about it.
Don't all modern DAWs have flat noise dither like triangular? And I've read other places that flat dither should be used before the final stage but why is that?


Andreilg3
Posts: 70
Joined: Tue Oct 11, 2011 9:00 am

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by Andreilg3 » Mon Apr 08, 2013 1:38 pm

wow that izotope guide is horrible, it explained sample rate in a completely misleading way. Digital is implied to be discrete when converted back but it's actually 100% continuous on the horizontal end up until the nyquist, which is represented by the sample rate. And dithering applies to more than just going down in bit depth, the quantization error is there because it is a digital format, not because of the truncation...that only makes it worse. Whoever wrote this should get fired.

http://www.princeton.edu/~achaney/tmve/ ... eorem.html
http://productionadvice.co.uk/no-stair- ... #more-6186
http://wiki.xiph.org/Videos/A_Digital_M ... _For_Geeks

3dot...
Posts: 9996
Joined: Tue Feb 20, 2007 11:10 pm

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by 3dot... » Mon Apr 08, 2013 1:47 pm

Andreilg3 wrote:
wow that izotope guide is horrible, it explained sample rate in a completely misleading way. Digital is implied to be discrete when converted back but it's actually 100% continuous on the horizontal end up until the nyquist, which is represented by the sample rate. And dithering applies to more than just going down in bit depth, the quantization error is there because it is a digital format, not because of the truncation...that only makes it worse. Whoever wrote this should get fired.

http://www.princeton.edu/~achaney/tmve/ ... eorem.html
http://productionadvice.co.uk/no-stair- ... #more-6186
http://wiki.xiph.org/Videos/A_Digital_M ... _For_Geeks
the 1st link is a nice and thorough explanation.. although .. offline audio processing occurs bit by bit afaik
with all due respect.. a lot of ideas in the 2nd link are pure opinoin

from the 3rd :
A digital signal on the other hand is discrete in both value and time. In the simplest and most common system, called Pulse Code Modulation, one of a fixed number of possible values directly represents the instantaneous signal amplitude at points in time spaced a fixed distance apart. The end result is a stream of digits.
and what do you mean by 'Horizontal' ? time ?
Image

Andreilg3
Posts: 70
Joined: Tue Oct 11, 2011 9:00 am

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by Andreilg3 » Mon Apr 08, 2013 1:59 pm

3dot... wrote:
Andreilg3 wrote:
wow that izotope guide is horrible, it explained sample rate in a completely misleading way. Digital is implied to be discrete when converted back but it's actually 100% continuous on the horizontal end up until the nyquist, which is represented by the sample rate. And dithering applies to more than just going down in bit depth, the quantization error is there because it is a digital format, not because of the truncation...that only makes it worse. Whoever wrote this should get fired.

http://www.princeton.edu/~achaney/tmve/ ... eorem.html
http://productionadvice.co.uk/no-stair- ... #more-6186
http://wiki.xiph.org/Videos/A_Digital_M ... _For_Geeks
the 1st link is a nice and thorough explanation.. although .. offline audio processing occurs bit by bit afaik
with all due respect.. a lot of ideas in the 2nd link are pure opinoin

from the 3rd :
A digital signal on the other hand is discrete in both value and time. In the simplest and most common system, called Pulse Code Modulation, one of a fixed number of possible values directly represents the instantaneous signal amplitude at points in time spaced a fixed distance apart. The end result is a stream of digits.
and what do you mean by 'Horizontal' ? time ?
There's nothing subjective in the 2nd link other than the fact that he says he can hear distortion on 24 bits.

Monty is talking about the data...http://wiki.xiph.org/Videos/Digital_Show_and_Tell look at this one, the absolute entire point of his video is that people don't understand basic audio and think that just because data is digits and blocks that the output is the same, the output of analog audio is 100% identical to the input regarding sample rate

and by horinzontal i'm using Izotope's idiotic representation of sampling rate as opposed to bit depth. Nyquist is the essence of why digital is perfect.

3dot...
Posts: 9996
Joined: Tue Feb 20, 2007 11:10 pm

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by 3dot... » Mon Apr 08, 2013 2:04 pm

the only thing we can agree is that regarding a 'musical tune'
digital standards today... are an appropriate acceptable replication for our ears..
Image

Andreilg3
Posts: 70
Joined: Tue Oct 11, 2011 9:00 am

Re: Should I use 44.1kHz or 88.2kHz sampling rate?

Post by Andreilg3 » Mon Apr 08, 2013 2:14 pm

3dot... wrote:the only thing we can agree is that regarding a 'musical tune'
digital standards today... are an appropriate acceptable replication for our ears..
that has nothing to do with what we were just talking about, and this entire thread isn't about appropriate, it's about optimal. Finn was asking about dither and when to use it.

Post Reply