Hello,
When I edited same loop of wave in sample editor,
I noticed that a sound changed by the placement of the loop.
This change seems to depend on the sample data after the end point of the loop.
A high frequency accompanies the sound.
Is there anyone knowing this reason?
Is this a kind of the bugs?
I attach a file to exemplify.
There is a single sample at ~7.5 db at the end of your loop. The looped version includes the positive value but in the cut version it’s zero. In the case of a sudden voltage change (instantly going from 0 to positive or negative voltage) I suppose a digital system must choose one or the other. Seems like zero would be a better choice in the loop case, but I don’t know.
(Just an observation, I wonder if you noticed that. Like, in the case of snap to 0 crossing not letting you choose an actual zero db sample when there is a sudden jump. Besides the stray sample making it more obvious, there does seem to be something else funny though with the repitching here…)
Thank you !
I supposed that the system refers to the next sample data while reading individual sample data.
But in this case, loop start point data do not seem to be used as next data.
I feel it slightly strange …
Yes, it’s something. Even if you fix the piece of sample data at the end of the loop so that everything looks correct in the editor (one would think it would now be exactly the same as your cut sample), the sound is better but is still not right. It acts as if the loop is a very small bit off, I agree. Or could be something with how the interpolation works. But, I don’t know…
(it does make one wonder how often this thing, whatever it is, is the cause of unpredicted aliasing/unexpected repitching noise in the case of looping a bit of a synth hit and so on…)