DSP Robotics Support

hello team
j I try to make a new waveform. this one works so does not have agreements. agreements with the shape becomes unstable.
Please osx testing and if you can provide the solution to stabilize.
sorry for my bad English. (I'm french).
thank you for your contribution.
Regards mimine

Hi mimine, welcome to the forum.

I am not sure what is the desired output - I don't understand the mathematics of the code so well. But there is one thing I see that may be a problem.
There is a "de-zipper" in there which has an input value of '0.001' - but the de-zipper will round this down to zero, as it only allows integer input. There can be no fractional part to the value, so the smallest acceptable value is 1ms - which I think will be too big, and will limit the output frequencies badly.

hello trog
thank you for your very quick response. very present, great sir trog mr. In fact I have to wear a dezip output to try to fix the bug. No but it has no effect. try without dezip, the result is the same. I try to stabilized output frequency max -1. but I'm stuck on something that returns the unstable al listen, if you use the agreements.
my request! how limited value between 0 and -1 output.
thank you

You have to give us a working schematic, not only the module that you are working on. Because we don't know what you try to achieve, and what is actually not working. I don't even know what input types for pitch/freq are required.

hello MyCo.
thank you for the participation. indeed I should put a more complete example. voila. j have taken the original code (without the moog) to simplify. I note that to play simply on the volume reduces the instability, of course. this is not the problem! I wish to implement a waveform that shook modify the internal current over time (top, bottom) segment with very limited time.
hoping that my explanation is clearer for you.
I know my English is very bad (sorry).

mimine

I haven't seen any instability in there. The only thing that I can find is, there is no clipper on the output, so when the volume gets to high, everything sounds weird. Maybe you mean this with instability.

sorry for the very long answer, in fact forget the module output, this one has no difference with the problem. I think my bad english door confusion. trying to test the form in support of keyboard (aert) or (eryu) simultaneously. the combined product key the problem. I think the math again (top down the freq) induces limited to Avoided saturation.

mimine wrote:support of keyboard (aert) or (eryu) simultaneously. the combined product key the problem

When you are using the "PC Keyboard"->MIDI, some key combinations cannot always be played simultaneously.

This is not a bug with FlowStone - it is because of the way that PC keyboard hardware is built.
To make keyboards cheap and simple, there are often groups of keys which share a 'circuit' inside the keyboard. It is common that only two keys per 'circuit' can be detected at the same time. For typing text, this is not a problem - but it prevents the keyboard from working well as a MIDI source.

Because this happens inside the keyboard, before the data is made into PS/2 or USB for sending into the PC, there is no way to change this using FS or Windows settings.
Some expensive keyboards, made for gamers, are less limited, but they are are more expensive than a standard keyboard.

When you play the synth using a true MIDI 'instrument' this limitation will go away, because the music keyboard is designed to allow many keys pressed at the same time.

As far I remember - only 3 keys on standard keyboard can be presset at once, and the rest is ignored?