I’m a noob to this subject so I can ask some of obvious questions as user. Like what good is a midi clock?
I did some experiments to look at jitter at the received end, sending midi and osc signals in various ways. In a nutshell on my setup I got jitter values typically around 4ms, sometimes quite a lot more. Different routes gave different kinds of results - osc over a crossover cable was typical stable in long runs then patches of jitter. Over wifi, similar. I’d expect that. Surprisingly a usb-midi-midi-usb cable was worse with constant jitter up to 25ms. Makes me think the quaility of the cable matters.
My point is this - a 4ms jitter on a single note is not a problem. Or even 25ms once in a while. So if were using SPi to drive another synth with midi notes (I don’t want to, but imagine) that would work for me. Or triggering a secondary clock or sequence, that’s fine.
But thinking of the midi clock, that’s 24 pulses every beat i.e. about 42ms each. A 4ms jitter on that is significant. Can a software clock ever do this job? Not just thinking SPi here - it’s all the links in the chain.
But then I’m not sure what the midi clock is actually for. I’m quite interested in getting a hardware synth just to try it - those Volca’s look good - but if they rely on a midi clock for integration then it could spell trouble.
If it were me designing the hardware, I’d assume some jitter as it looks like a gravity issue, and in consuming a clock pulse do some moving window average to get to a ‘true’ value.
Can you point me in the right direction? Thanks.