Vibrato Weirdness

I just noticed that the vibrato (in effects column) seems to behave differently depending on which pattern I start playing my song from. I’ve also got some pitch sliding going on, which could also be affecting this.

I’ve reduced it to a minimal example module.
Try starting it from the first pattern, then pay attention to the vibrato in the second pattern.
Start the song from the second pattern and notice the difference.
Is this wrong, or is there something I’ve missed?

edit: I’m using Renoise 1.9.1 for Linux

the vibrato command uses two parameters:
the command syntax is like 0Fxy, where x is the frequency of the vibrato, and y is the depth.

in your example module, you never specified the frequency, leaving it always blank (zero); this can result to unpredictable result (i.e.: unpredictable frequenct for vibrato).

see if setting a value for vibrato frequency even only at the first vibrato hit, leaving to zero the others, solves the problem for you

Doh, yes, I seem to have forgotten that in the example… :P
(Why would the vibrato frequency depend on where I started playing the song from anyway, if it was never set anywhere?)

It still happens even if I set all the vibrato commands to frequency 1… ???

please update your example song, just to see if you have forgotten something else :)

Uhm… I must be blind, 'cause I can’t see anywhere to attach files (or change the one in the first post) now… :blink:

Anyway, I figured out what “causes” it:
If the NNA for the instrument is set to anything other than ‘Cut’, the vibrato will act this weird way.
If the NNA is set to ‘Cut’, it will sound the same no matter which pattern you start playing it from.

So why does the NNA affect the vibrato speed?
(The last thing on the first pattern is a note-off, if that matters.)

can you send the song to it-alien @ renoise.com ?

Done. :)

I have received the song. As I expected, I think the problem is that, when setting NNA to other than “Cut”, the sound of the various notes mix together so that, when vibrato comes, it looks as it behaved differently.

try for example to set NNA to “Cut” but moving the notes of the second pattern on the second column of the track, and see if you agree with me.

Ah, yes, I hadn’t thought of trying different note columns in the track…
It then stops happening, no matter what NNA is set to!
Thus, it’s note some kind of illusion, since the previous notes are still in the track, sounding the same way. :)

What is going on here? Is it supposed to work like this? Am I missing something obvious? :P

another source of illusion can be the volume envelope sustaining the sound. please try disabling the volume envelope for the instrument and see if this clarify the issue.

Disabling the volume envelope also makes the vibrato behave normal.
And as I said earlier, it stopped happening when I moved the notes in the first pattern to a different note column, still in the same channel (i.e. they sounded the same, with envelope and all). I just don’t see how it can be an illusion. :)

So, the cause has been narrowed down to multiple active volume envelopes in the same note column.
Does it make sense for this to affect the vibrato frequency?

I instead firmly think this is an illusion, especially because of the FM nature of the sample used: when more notes are concurrently playing (especially if simple waveforms like the one in the example), such kind of artifacts can be generated by interference.

I don’t think there could be a bug-relationship between multiple envelopes running and vibrato frequency, since the vibrato command is simply an oscillating modulation of the sample frequency.

my song Overlook makes uses of this illusion on purpose using glissando instead of vibrato. Look for example at Pattern 4, Track02.

Uhm… Perhaps I am not explaining myself very well.
I’ve rendered two WAVs, made accompanying screenshots of the patterns, and put them here.
Check them out, and tell me it is just an illusion. ;)
(Nice track, though!)

don’t worry, I’ve fully understood what you meant with your previous posts: the difference is clearly audible, but as you admit, volume envelopes and concurrent notes seem to be the source of the difference.

I will notify taktik about this thread since he should be able to put a final word about the issue.

Ok, cool :) I uploaded the .xrns too, if he needs something to test with.

Thats a small but effective bug in Renoise 1.9 with note glides in one column while using a NNA other than cut. This is already fixed in 2.0. So not the Vibrato is the problem but the glide.
You can use as a workaround now either multiple columns (as you already said), or set the NNA to cut.