Hello,
I have a new plugin to release and I would ask this community to test it before I submit it to VCV to be included in the library.
It’s a simple idea: generate triggers from LFO intersections to generate patterns and CV that is organic to the pattern generation. I called them cross sequencers: 2 LFOs (XSeq2) and 3LFOs (XSeq3). There are also a couple of ancillary objects.
See details in the README.
I compiled it for Windows. I tried compiling it for Mac but I am getting error messages (*** multiple target patterns. Stop). If anyone has a clue about how to fix that please let me know.
Ok, this time it should definitely give you a different error. If you could copy paste it that would be great. Thanks for your patience (it’s frustrating not to be able to reproduce the bug…)
Are you building manually for each change? Why not just add the github action to rebuild on push? (there is a thread in this forum that shows how to do it)
Right now knob params are checked at every sample. I will add a change () function to make them run smoother. Other than that they should be pretty efficient but I will double check all the code. Thanks for checking! Other than CPU usage do they work for you?
I built on windows, no obvious problems. You now, if it’s an LFO you don’t really need to generate a sample at full audio sample rate. I think my LFN runs internally at 400 Hz sample rate, and it has a multi band equalizer in it.
Thanks for this suggestion! I am now down-sampling LFOs by a factor of 8 which makes the modules much more efficient. Maybe I can do even more, I’ll keep testing. Still working on the param update but it’s sensibly better now.
Thanks. I spent a couple of hours trying to run git push from the terminal but the authentication process on Windows seems very buggy. I was using GitHub desktop for my commits, but for the automated building that is not an option.
I have installed the credential-manager app but I still get this
git: 'credential-manager-core' is not a git command. See 'git --help'.
Extensive stackoverflow research has not solved the issue.