Getting uneven pitch in stages sample and hold

Ran into an issue that i wanted to see if anyone else could recreate or tell me whats my error is. was running a patch with marbles x2 output into a yellow stage t2 to trigger the sample and hold, potentiometer all the way to the left so no slew. it seems that my stages produces some pretty uneven pitches through the sample and hold and for some reason it seems to be worse when the S&h is triggered by one of the t outputs of marbles. kind of momentary dips of sharp or flat voltage at the start of a new hold up to a semitone off.tried it with diferent sources i.e. blinds offset as voltage source into time/level input rings and plaits as sound sources. the problem seems to not be as noticeable when triggering with PNW or a another stage as a square LFO rather than marbles? any idea what could be going on?

huh i tried to replicate this behavior using the scope in reaktor rather then the tuner in ableton and i couldn’t get it to reproduce. so some kind of user error i guess though i still can’t figure out what would cause this.

I’ve been encountering this issue as well. I’m getting nearly unusable pitch drifting with S&H. The odd thing is that it only happens occasionally. I first encountered this like 3 months ago, then plenty of expected behavior, until today. Today I troubleshooted it to the point of sending in a static DC offset and sending steady gates in to sample. Patching that to an oscillator and then viewing in a tuner the resulting pitches span nearly a major 2nd (B to C#).

Curious if you’ve had any developments / encounters with this since this post @daveb?

I have the same issue. I’m sending quantized pitch from a sequencer through its sample and hold and the sequence changes randomly. It seems to not change it if I use glide on the sequencer, but then I’m always sampling one note behind and also it is not convenient because I send the same sequence to multiple inputs.

Does the same thing happen with another S&H module? Which sequencer are you using? I’m thinking that maybe your sequencer is sending the gate/trigger before the time the note CV has settled. This can be solved by adding a delay internally to Stages’ gate signal (at the cost of adding extra latency).

Thanks I tried adding a delay but it still happens, also it’s different notes that are being added - I tried it with a two note sequence and every few cycles some other notes would appear. It all goes through a quantizer at the end. I don’t have another s&h module to try with.

I’ve noticed this as well at times. It is indeed strange that it doesn’t always happen though. When it has happened I swapped the signals into my disting s&h and all seemed fine. I was using a metropolis as the source for pitch.

I’m also using a metropolis and a Disting mk4 as quantizer. What do you mean swapping the signals into the Disting?

I was having the issue with Stages as a s&h so I tried using the disting instead with everything else the same to make sure there wasn’t some other issue I wasn’t thinking about. Disting worked fine.

Thanks for the answer

Any other ideas on this? It’s still happening for me intermittently. When it is happening, I have stopped and restarted the sequence a few times until it is resolved. So i am not sure it necessarily has to do with a delay.

The way I’m doing it now and it sounds fine though not sure the pitch is respected accurately: metropolis pitch cv to a mult with one out going straight to an oscillator and the other two go to stages s&h and from there to a Disting mk4 quantizer set to the same scale as the metropolis. It works because everything is in tune though I’m not sure the pitch is accurate in all the instances because the gates arrive at different times and I’m also transposing both instances straight in the Disting. It sounds good though :slight_smile: Hope my explanation makes sense.

You can try this update which adds a delay to Stages’ handling of the gate signal.

1 Like

this seems to have done the trick for me! thanks so much!

From my experience with CV-to-MIDI modules that’s the case for most sequencers. The timing of the gates is prioritized over the pitch value. I know the post is old, but I think it’s worth pointing out that most manufacturers consider this as standard practice when designing sequencers.

Thanks for this firmware. I just tried it but it’s still not working form me. While the regular one samples and holds the pitch just before the trigger, which makes it confusing to follow on the sequencer, this alternative one captures the one on the trigger indeed. Unfortunately it’s still not accurate and the pitch fluctuates on every step. I am passing it through a quantiser to compensate, but the bigger the distance between two notes, the random the sequence it gets. So for example on a two step sequence playing two Cs two octave apart, it goes through all the notes in the scale instead of playing the Cs. It seems to do fine while the difference is less than one octave.

Which sequencer do you use? It might probably need a longer delay!

I was using a René version 1 quantized cv and triggering it with René or Pamela’s New Workout. Then I tried the unquantized cv out on René and it works fine with this alternative firmware. Thanks again!