Renamed repo GitHub - jackokring/KRTPluginA: Another later version VCV rack repo. as it seems more logical, and I think kind of expected.
Keep up the good progress and when it’s ready, it is ready. Looks like a very interesting filter. Great work takes time. Thanks.
I doubt if the buffer under runs are your fault. The filter isn’t doing anything insane. Try setting you thread priority to real-time.
I can’t really test the poly (for loop unroll into the manufacture of 16 vars for filter cutoffs …), as the real time froze last time I tried. Maybe I should remove the BOINC, even though it’s an i5 AVX512 …
How does the directory structure handle Unicode μ? An edge case? 派对中国人。
I think the file open dialog code works correctly with Unicode.
I don’t believe there is a simple delay that works in samples - there are plenty that work in ms, Xatto Time which works in microseconds, and Bacon SampDelay, which does work in samples, but only goes up to 99.
Would be a handy thing to have to time-align ITB with loopbacks going through external gear, where the I/O buffer sizes are known.
How many samples in delay do you need?
This sounds like a sync to returned trigger plus some counter of triggers later … a nice idea. Not immediate on the module list I’m making but a good feature.
I’m making KRTPluginA/ModuleIdeas.md at master · jackokring/KRTPluginA · GitHub and will follow this thread to add the cream to the Focus Areas section, while deleting anything that has been confirmed as made as a module.
I think it would be fun to see a variant of Strip module with voltage-seedable RNG. I mean, such that randomization of right-hand size is determined by some “seed” voltage input. So that you can easily sequence presets by using just voltages (cherry picked after exploring randomizations), without the need of additional modules, save/load buttons, etc.
i don’t think there’s a need for this feature when stoermelder transit basically does this with very little extra work.
The difficulty I imagine is the tiny variance to chaos, and so some modulo sizing on the chaos ring. The modulo can be compacted on a prime totient ring but if the divisions are not prime, the maximal length property is not conserved.
Do you imagine something that has slightly different parameters if seed is changed just slightly? That would be very interesting, but I would be happy even with straight basic PRNG with seed parameter exposed to be controlled using voltage. The latter can even be reinterpreted as a string or just a sequence of bits of a float when used to seed the generator. So that even a tiny change in voltage results in totally different randomization. It would still be very interesting to play with because you can connect any sequencer to that input and get repeated presets when sequence repeats.
@aliasup That’s a great module, but this feature would still be very interesting for rapid explorative/generative doodling.
The module μ now also works. A bit like a bit crush to some extent on Gibbs ripples.
I’m starting a delay effect (simple) to get used to the best use of buffers), but live guitar sync was a thing today. Maybe a metroKnome into beat to start the guitar a bar before?
So Jack will be/are the modules in the library yet or just as self builds for testing? So far you seem to be making good progress.
There’s some sha for the library thing? Is this a commit sha or a binary sha? Release v1.2.4 · jackokring/KRTPluginA · GitHub seems to be the one I completed today. One of the modules is a blank, but I’m making the version number RACK.WORKING_MODULE_COUNT.FIX_INCREMENT so that the first known number of modules work while allowing some development space of the future modules. It seems the ordering of plugin.cpp decides the display order. So if I keep the working ones first and make any indevlopment not compile in a bad manor, then it allows for people to see development to some extent.
So they are self build as yet, from source code. I’ve never built a module from source code. So I may wait until they are ‘Library’ ready. Thanks for keep me in touch with your progress.