Closed tomcombriat closed 1 year ago
As for the wavefolding itself, I have no feedback. About the git-problem:
I do not understand why there are so many commits, I tried to synchronise all my branch to upstream but they still show… I'm sorry for that.
Yeah, it's just far too easy to get into a mess with git...
I believe, the problem is that the master branch of your tomcombriat/Mozzi is not in sync with sensorium/Mozzi. For the future, you should try to avoid that situation, by doing all your work in branches. For now, you are in a bit of a mess, as git thinks, your master branch has some changes that sensorium:master does not have. Actually they have been merged some time, but for git it still looks different.
Now, to get your master branch back in sync, use:
Register sensorium/Mozzi as a separate "remote", so you can pull from it: (you may have already done this, or something equivalent; remote show
to list remotes):
remote add upstream git@github.com:sensorium/Mozzi.git
# or HTTPS-access: remote add upstream https://github.com/sensorium/Mozzi.git
Sync your master to upstream master:
git checkout master
git pull upstream master
git reset --hard upstream/master # warning: deletes any local changes you may still have on master
git push origin master --force # warning: and now they are gone in tomcombriat/Mozzi:master, too
Now, after this, any new branch you create will be based on a state that git recognizes as something already in sensorium:master. However, your existing local branches are still messed up, and getting out of that may be non-trivial (git rebase -i
should work, in theory, or you can create a new "clean" work branches and git cherry-pick
only the commits that are actually meaningful from the current branch). However, it may not be super-important to fix that, either, and, we can still squash commits while merging.
Hi,
Thanks for the git tips, I got most of it but I think I did a fetch instead of pull on the upstream. Will try to do that and if it works might resubmit a cleaner PR (there is only one commit on this branch anyway).
Will add an example soon also. Best,
Closing this in favour of the cleaner one I've just open (thanks @tfry-git , worked like a charm).
Nice one, Tom.
I expect the / and % at audio rate will be trouble for slower processors! I'll try to test it soon.
Yes, that was also my concern…
For the %, an easy optimization is to check only the least significant bit of the number. This is fairly common so I assumed the compiler might do it by itself. I can try to see if that makes a difference.
The division is a bit more tricky, but one way would be to compute the invert in the setLimits
and use a multiply in next()
will try.
Note that this has been superseeded by #159
Hello,
This is a simple wavefolder, in response to #153. Looks good on the scope but not fully tested yet. At least an example is missing. Putting it as a draft for now, in case someone sees an obvious optimization!
I do not understand why there are so many commits, I tried to synchronise all my branch to upstream but they still show… I'm sorry for that.
Cheers, Tom