Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jon Hodgson]

wavelogix wrote on Wed, 16 November 2005 08:03

ok , all said and done , can somebody in here , throw some light on why some VSti have the oversampling option ? whats are its pros and cons and why the need for such an option in the first place ? thank you , chandan ...

Because in a VSTi you are trying to emulate a signal that is not band-limited, AND the process of sampling it.

So let's say you have a sawtooth oscillator (for simplicity's sake we will make it a perfect one), you might think this was easy to generate, just ramp up your samples till they get to max_level, then drop down to -max_level again.

But this won't actually generate what you would get if you properly sampled a sawtooth, you in fact generate the equivalent of a sawtooth sampled with an ADC with no anti-alias filter, i.e. non harmonic alias components all over the place.

There are a number of ways to work around this problem, and developers may use several at once. One such solution is oversampling.

Also non-linear processes (distortion) generate additional harmonics, which if they go over the nyquist frequency will alias back, oversampling is a way to reduce this problem.

[twilightsong]

wavelogix wrote on Wed, 16 November 2005 08:03

ok , all said and done , can somebody in here , throw some light on why some VSti have the oversampling option ? whats are its pros and cons and why the need for such an option in the first place ? thank you , chandan ...

Well, I think it's the same principle as convertor oversampling. If you take a digital signal and apply non-linear processing to it, it will add errors and possibly audible alias to that signal at output. The oversampling in a plug-in removes that error above the audible range

whoops, I hadn't seen that Jon had already answered