How can samplers claim to produce 192kHz sound?

[edge100]

So samplers like Kontakt and DirectWave can technically output at 192kHz 32 bit, but unless the original sample (.wav or .aif) files used are also 192kHz 32 bit, then it's not truly 192kHz 32 bit, eh?

If someone is selling 192khz samples, you should ask to see the microphone they used. I'm no expert, but I don't know of any microphone with any frequency response in the 96khz range (the maximum representable frequency of a 192khz file). For comparison, a standard prosumer-grade LDC can pick up frequencies up to about 18khz, which would only require a 36khz sample rate to reproduce.

The advantage to higher sampling rates is NOT being able to pick up higher frequencies; it IS the reduction in the severity of the brickwall filter. At 44.1kHz, you CAN pick up frequencies up to 22.05kHz without aliasing, but you'd need an infinitely steep filter. As such, 44.1kHz converters have to start rolling off frequencies well below the Nyquist limit. If you increase the sampling rate, you can ease off on the filter. Of course, if you increase the sampling rate TOO high (and some would argue that even 96kHz is too high), your converters start making too many errors, since they are required to sample too quickly.

Lavry argues that ~60kHz is the optimal sampling rate. If you havent seen it yet, his White Paper on the subject is a great read. You may have to go through it a couple of times to understand the math (I had to), but it really explains the issues with ultra-high sampling rates.

http://www.lavryengineering.com/documen ... Theory.pdf

[ethios4]

Nice. Thanks for posting that.

[dn83]

Yeah, true - I didn't think of that.

But I don't know, it just seems that samplers should be using samples higher than 44.1kHz, to allow for resampling when you render your project to a CD, or as a .wav or .mp3. But again, I suppose the quality loss is trivial really.

[dn83]

Cool, thanks for the post and link, very interesting!

[dn83]

Angstrom - Yeah, that's what I'm saying, I want my initial input to be the best quality possible.

[Tone Deft]

So samplers like Kontakt and DirectWave can technically output at 192kHz 32 bit, but unless the original sample (.wav or .aif) files used are also 192kHz 32 bit, then it's not truly 192kHz 32 bit, eh?

If someone is selling 192khz samples, you should ask to see the microphone they used. I'm no expert, but I don't know of any microphone with any frequency response in the 96khz range (the maximum representable frequency of a 192khz file). For comparison, a standard prosumer-grade LDC can pick up frequencies up to about 18khz, which would only require a 36khz sample rate to reproduce.

The advantage to higher sampling rates is NOT being able to pick up higher frequencies; it IS the reduction in the severity of the brickwall filter. At 44.1kHz, you CAN pick up frequencies up to 22.05kHz without aliasing, but you'd need an infinitely steep filter. As such, 44.1kHz converters have to start rolling off frequencies well below the Nyquist limit. If you increase the sampling rate, you can ease off on the filter. Of course, if you increase the sampling rate TOO high (and some would argue that even 96kHz is too high), your converters start making too many errors, since they are required to sample too quickly.

Lavry argues that ~60kHz is the optimal sampling rate. If you havent seen it yet, his White Paper on the subject is a great read. You may have to go through it a couple of times to understand the math (I had to), but it really explains the issues with ultra-high sampling rates.

http://www.lavryengineering.com/documen ... Theory.pdf

Just the thought of 60kHz sampling rate pisses me off, there's already too many sampling rates out there. 192kHz is just ridiculous, from my experience, 48kHz is great, 96 is also workable but IMO overkill. The good thing about higher sampling rates is that there's more room for slop, so things become easier and cheaper to design.

Lavry strikes me as a guy stuck in academia with no knowlege of the real world, no practical application, just opinions. (I don't know who he is.)

I'd disagree with those who say that 96kHz is bad because the "converters start making too many errors, since they are required to sample too quickly." it's still just the kHz range, when stuff has to work in the MegaHertz range, that gets tricky. 96Khz is SLOW, maybe 10 times that speed gets hard to design, dunno, I've never designed an A/D converter IC, but I know they're basically a ladder of resistors that gets sampled, I'm sure the wikipedia has the standard explanation.


LOL at asking about the mic they used, great point!!!