I use them to turn a short burst of sampled and looped white noise into a subtly changing random sound to keep the sample from sounding looped by assigning the pink/white noise source to the pitch of the white noise sample.
You mean so it constantly sounds like the sound is changing ?
It adds a random variation to the pitch and therefor the frequency, this stops the short looped sample of white noise (which is made up of random frequency content) from sounding like a loop.
If the loop is short and no variation occurs your brain will detect the loop within a few cycles.
So this method reduces the need for a long loop cycle on a white noise sample.
Consider the Lag processor as something to knock of sharp edges from a source
eg. it progressively curves a square wave as you increase the lag rate.
You mean it rounds off the effect of the source to make its effect (on the Desitination) more subtle ?
Not really more subtle.
Think of it as behaving as a pot/slider would alter a sound rather than the on/off switch behaviour of the unmodified input.
Consider the square LFO waveform or a clock input as the switch being modified.
It takes the stepped values of this input waveform and progressively turns it into a sine so the value sweeps from max to min instead of jumping max to min.
If the lag value is set too low/high the effect will not be correct, this value has to be set by ear in the LFO/lag page.