Request for ears

Hiss? No-hiss? When? Other differences you notice?

(something changes at 20s)

high pitched sound until its volume decreases at around 10s, cutoff goes down (it seems), noise stops at ~19.5s to return again at 34s

The high pitched noise cuts out at around 20s, but comes back in at 34s.

how hard is it to hear it between 20s and 34s? Anything else you notice?

  • on Electrodruid’s observations (with a small “pop” when the noise cuts out). The level of the high pitched sound is pretty consistent while it’s in the signal.

It doesn’t seem to be there between 20 to 34s. I also notice that it is being gated by the VCA; ie: it triggers alongside the notes.

Edit: For the record, my listening chain is RME Multiface2->Allen&Heath 8 Buss mixwizard->yamaha HS50M monitors.

Edit 2: Just popped it open in peak, and you’re probably getting a >50dB attenuation on it when you kill it off @ the 20s. It sits at around 7Khz with some harmonics around 14Khz and 18Khz. Even when you put the filter or whatever you do to kill the high pitched noise, there is still a little 10dB bump around 7.3Khz.

Thanks all.

It’s hard to look at a spike on a spectrogram and find out whether people will hear the bumps or not. The problem did not come from my filter circuit - no matter how dodgy it is - but from the control signal itself. At the moment, the fix involves a tradeoff between the whine level and another sound parameter. I’m not happy with moving the tradeoff point in either directions, and the current compromise is kind of lame.

I have an idea of circuit that could push it down by an additional factor of 18dB, which could be traded for a bit of improvement in sound. Also, I have to investigate a bit on the origin of the problem on the digital side…

I can still hear the whine between 20-34, more pronounced 30-34

So now you know my secret plan… I’m not working on synths anymore, but on clinical things like this:

How many synths out there come with a free hearing test? :smiley:

“How many synths out there come with a free hearing test?”

All of the ones with a sine wave and panning. :wink:

Are there still things that hurt your ears in this sample?

sweet streams of jesus that sounds good. No pain, no whine…all good.

everything fine! that sounds fat! this filter’s really going to be even better than the cem type!

What’s the smiley face code for tasty? : d

I have to reveal there’s one MAJOR cheat here: this is not a 24dB/octave. This is a “18dB/octave for a while then after some low multiple of the cutoff 24dB/octave”. This shouldn’t matter much, maybe this adds character after all, but trying to be low-noise and working from -5V/+5V (a constraint I set myself to keep the power supply section simple) with cheap op-amps (clips at 3.4V) really feels like pulling the blanket in two different directions.

The resonance is really well-behaved, and the auto-oscillation sine is pure, perfect for kicks!
(you can also hear at the end that it’s decent at tracking the pitch)

Wow! That sample sounds amazing! Great job troubleshooting.

Yes , this sounds awesome

Yes! agreed! Sounds great. The self-oscillating sine sound is really pure sounding too…