Artificially bumping up a samplerate

[rounser]

Given a snare sample at 44K, if I then continually loop it on itself at random offsets using GRM Tools Freeze until all I'm left with is a tone, can the resulting tone be recorded at 96K with meaningful information for that sample rate?

In other words, would such looping "fill in the details" such that I'd end up with more information (and hence less aliasing if I pitched it) than the original sample? (I don't care about the original snare envelope by the way, the tone is fine.)

Audio voodoo, I know...I just have no idea whether this would work or not.

[RTaylor]

Some effects {the ones that totally transform a sound... not just adjust parameters} upgrade audio pretty well if you do the processing at higher rates. ...Try it.

[rounser]

Try it.

Will do. I'm not sure how to determine if I've succeeded, though...except perhaps by pitching and listening for aliasing.

Aha. An A-to-B between pitched 44K tone and "96K" tone, see which falls apart first. Cheers for making me think.

[Lagrange]

Someone here told once told me.. You can't mix whats not there... or in this case Whats not there is not there. In other words, even if you speed up the clock you will not be adding more audio to higher frequencies in the recording. The only way you can add more is to use something like ACON Digital Studio Rebirth or pitch shifting the material and then mixing it with the original. IME - If you used a freezed audio to "fill in" a thin image be prepared for a totally different sound to the original. One way around this problem is to try to convolve the sounds together instead. Your going to have to be reaaall patient cause your going to get many many varying results.

L

[rounser]

In other words, even if you speed up the clock you will not be adding more audio to higher frequencies in the recording.

Ah, of course.

The only way you can add more is to use something like ACON Digital Studio Rebirth or pitch shifting the material and then mixing it with the original. IME - If you used a freezed audio to "fill in" a thin image be prepared for a totally different sound to the original. One way around this problem is to try to convolve the sounds together instead. Your going to have to be reaaall patient cause your going to get many many varying results.

Thanks for the info. It sounds like the effort:gain ratio is pretty low, probably best off seeing if I can hunt down some 96K or 192K snare or hat samples (which seem to be rather thin on the ground). As is usually the case, there's no such thing as a free lunch.

The reason why I'm bothering is because I like pitching my percussion, and hat tones seem to alias instantly if I do that, probably because they're so high frequency. This adds interesting artifacts, but it would be nice to have the option of less aliasing...thus the need for a high sample rate source.