Does anyone else have this weird volume bug?


#1

I just got the Parva yesterday, and it is at least a decent synthesizer. It doesn’t sound too impressive so far, but I probably need to program it better. I notice that the higher I press a key, the quieter the saw and triangle waves are. Only the pulse wave does not get quieter when I play a higher note. Does anyone else have this problem too? Fortunately, modulating the pitch of the oscillators stops this oscillator level bug. So if you are going to use a patch with a triangle and/or saw wave, I recommend using an LFO on global oscillator pitch with key sync, a very slow rate, and a little depth


#2

Hi!

I have the same problem, more or less. I’ve sent word to Brad Frerguson and normaly I will receive an another Parva in the comings days. Thanks, I shall try your thing when I will be at home.


#3

Most likely related to the voice inconsistancy bug discussed elsewhere on this forum. Brad added trimming of each voice under CPU control, but under certain conditions, this leads to change in timbre/volume for now SQR waveshapes.

Your workaround looks interesting. Need to check this out on my Parva.


#4

that’s a pretty nifty workaround @futuresonus should comment


#5

This is greatly reduced in voicecard firmware v64.


#6

Did you have a chance to listen to my file? If so, is what you hear in it also better in v64?


#7

I’ve listened to your file, and am working on analyzing the differences between it and the newer builds. I don’t think there are any changes in v64 that would affect this.


#8

Ok, well just let us know when the next build is posted so we can give it a try.


#9

I am seeing this behavior on a new Parva running firmware 0.70. Here are some audio clips from a test that I ran with a single oscillator on a single voice card (VCA envelope has 0 attack/release, 127 sustain. No VCF envelope amount, VCF in 4-pole lowpass turned all the way up with no resonance or key tracking):

Parva Sawtooth
Parva Triangle
Parva Pulse 50%

As you can see, the sawtooth and triangle waveforms quickly decrease in volume relative to the pulse waveform as the pitch goes up.

I tried the workaround mentioned in the original post but that did not seem to have any effect.