One Synth Challenge #132: ToneZ by t0nit0rmx (MrKarollus Wins!)

[IV!]

@IV! I've had to do this when I was using a 10 year old quad core with 4 gigs. It's a pain coz a track nearly always fails to sit in the mix on first (or many) freezes. What I did was to load up the track with all the FX after the synth I was going to use (EQ, Delay etc). See how that sounds and sits in any prelim mix. Then, I would disable the FX, freeze, and then slide the FX over to the new frozen audio track. To my mind the audio coming out of the bottom of the frozen track is just the same as the track with midi->synth->FX … this helps a little if you need to play with the FX applied to the track, not the audio generation from the synth. Hope this makes sense and is still within the "rules" …

I using a 8 year old quad core with 8 gigs, but using 4 oscillators with at least one filter and LFO is already > 50% CPU, and this is still without effects! Thank you for your advices, I didn't even think I would have to resort to this method!

Taron, I've seen that you can do magic with as few sounds as possible! Leave a little time for music, if it doesn't go against your wishes!

[Taron]

Thanks, IV! I may try...
...right now I'm kind of curious, if I can manage to make the keytracking work properly for the filters...curious!

[] Peter:H []

Is it covered by the rulez that participants change the synth and compete with their version?
Are the changes available as pull request to be reviewed by Tonito and is he doing a official release with them? Only see one by Dakkra from 6 days ago which has not been approved yet.
Just asking, because I didn't have problems. I spend time to understand the synth and find a cpu friendly setup. My track is almost finished by now.

[Taron]

Not all too surprising that this isn't covered by the rules, hahaha, but I'm almost certain it would be against it, if one would write his own version. However, we've had plenty of rounds where the synth got updated along the way.
That's, however, why I asked for some guidance earlier on "what to do with this (my version)". I've only just started looking into Cabbage and CSound, which is why I haven't yet dared to make an official pull request. But my version is most certainly friendlier regarding envelope behavior and likely even processing, because sounds get to reach 0.0 via envelopes and it will stop the processing of the sound generation and filtering. Chances are, though, that things will sound somewhat different, while the presets all still sound great (possibly greater).

[] Peter:H []

Not all too surprising that this isn't covered by the rules, hahaha, but I'm almost certain it would be against it, if one would write his own version. However, we've had plenty of rounds where the synth got updated along the way.
That's, however, why I asked for some guidance earlier on "what to do with this (my version)". I've only just started looking into Cabbage and CSound, which is why I haven't yet dared to make an official pull request. But my version is most certainly friendlier regarding envelope behavior and likely even processing, because sounds get to reach 0.0 via envelopes and it will stop the processing of the sound generation and filtering. Chances are, though, that things will sound somewhat different, while the presets all still sound great (possibly greater).

Create a branch, commit your stuff if you fell comfortable with it, create a pull request ... a pull request is not really "official". It will only become official if Tonito accepts it, merges it to master and creates an official release of it. But I think you kew that anyway.

[IV!]

Thanks, IV! I may try...
...right now I'm kind of curious, if I can manage to make the keytracking work properly for the filters...curious!

I hope you can!) Keytracking is "a little" unstable and deafen me 20 minutes ago. Indeed, I would be happy to see a re-release or alternative version of ToneZ if Mr. tOnitO agrees. But I would not give up the previous version - despite all the shortcomings (and ringing in the ears ), it is capable of amazing things that are so dear to my heart! Yes, the filter with 1000% resonance probably no one else!

] Peter:H [, I haven't learned how to work like this yet!) I doing a sound and do and do, using everything I want to add, then play with it contentedly, look at the monitor and... What ?! 60% CPU?! Um, well, it's already created...

[] Peter:H []

Thanks, IV! I may try...
...right now I'm kind of curious, if I can manage to make the keytracking work properly for the filters...curious!

] Peter:H [, I haven't learned how to work like this yet! I doing a sound and do and do, using everything I want to add, then play with it contentedly, look at the monitor and... What ?! 60% CPU?! Um, well, it's already created...

Probably because I have some experience creating sounds without stacking to many instances ... For sure I made some compromises in my track. But actually that is at the heart of OSC in my eyes ... see what a synth can do, let the synth guide you towards it's sweet spots.

I have mentioned earlier that in my case it helped a lot to change the sample rate of my project to 44.1 Khz and 24 Bits. I usually go with 96Khz. With the 96KHz setup ToneZ showed the same symptoms that you report.

[] Peter:H []

Hey Taron, here's the link to Dakkras pull request: https://github.com/T0NIT0RMX/ToneZ/pull/6/files
Just provide your changes as he did. Then it is up to Tonito to accept or reject it.

[exponent1]

So far I'm getting very moderate CPU usage with my patches and none of the problems highlighted in this thread. But then again my patches are usually very basic and I don't use most of the features of the synths. I'll report back when I reach more number of instances and the CPU usage.

[IV!]

Probably because I have some experience creating sounds without stacking to many instances ... For sure I made some compromises in my track. But actually that is at the heart of OSC in my eyes ... see what a synth can do, let the synth guide you towards it's sweet spots.

Probably because it) Probably you're right!) I'll try to make a few main sounds and spend less time on the rest, so that it doesn't work out like last month

I have mentioned earlier that in my case it helped a lot to change the sample rate of my project to 44.1 Khz and 24 Bits. I usually go with 96Khz. With the 96KHz setup ToneZ showed the same symptoms that you report.

No, I'm already using 44.1 Khz and 24 Bits. Probably just a weak computer...

[Eauson]

Well, its an interesting synth, just the continuation of notes and there doesn't seem to be BPM sync on the vst2 version but appears on vst3??
Its very nice to detune the Oscillator's and have an LFO on the filter, but the CPU usage makes my PC grind ( my 15 yr old quad core )
Have used transpire to deal with Attack problems, even though it seems fixed at 0.7?
Still cannot bounce a track without it reverting to init!! Has anyone else encountered this or just me
I'm not sure I'm up for the fight

[Taron]

Hey Taron, here's the link to Dakkras pull request: https://github.com/T0NIT0RMX/ToneZ/pull/6/files
Just provide your changes as he did. Then it is up to Tonito to accept or reject it.

Thanks, Peter, that's cool! I've done a few things to the code...hmmm...I'll have to see how I can prepare this properly. I've never done anything on Github, so that's a nice opportunity to try it!
Thanks again!

[dakkra]

I'd rather T0NIT0 pull Taron's code than mine. My changes were simply on the parameters to make them more user friendly. Taron seems to have actually improved the DSP, which is far more valuable (I'm barely starting to learn DSP). If you need help with GIT/Github Taron, let me know and I'd be happy to guide you.

[Taron]

Well, I'm not yet sure how to add a branch to ToneZ on GitHub? ...ok, I'm a bit tired right now, but I did the whole "hello world" on GitHub without any troubles, branching, pulling, all is well. But as soon as I get on ToneZ there... ...probably should rather sleep, I guess. Maybe tomorrow...

And, Dakkra, yeah, I went into both The SuperOSC and MultiFilter along with the integration later on. It's a bunch of stuff. But I ALSO did the same things (more or less) that you did, with addition to resonance going only to 2.0 as opposed to 10.0 (which really is overkill). I looked into alternative filter models and THERE ARE TONS in CSound, including a moog approach, which seems fine, but needs more careful restrictions as factory presets can freak out. Anyway, but touching those would get really different. What I've done already is really taming ToneZ and making it act a lot less rogue, but in a good way. I didn't pull its teeth, hehe...more like squishing pimpl...nah, that's a bad analogy.

So, yeah, when I figure it out on GitHub, I'll give it a try!

Oh, I also realized that the GUI parameter for volume on OSCs actually gives back 200 rather than 100, which I found weird. Thus it has to get divided by 200 rather than just 100 for it to become 1.0. How strange is that? I figured it might be a byte thing going either from -127 to 127 or 0 to 255, but no...very strange. I need to investigate more what CSound and Cabbage are like.

[dakkra]

The general process with Github is like so:

- Fork the target repository
- Clone your fork to your machine
- Add and commit changes to your local repo
- Push to your forked origin/master repo
- Make a pull request from your fork to the original project repo

https://help.github.com/en/github/getti ... ork-a-repo

GIT is strange in that everything, both server and client side, is a repository. So in this case you'll be interacting with three: T0NIT0's master, your master fork (on github), and your local clone (on your local machine). You add and commit to your local repo, push to the remote/master fork, and then make a pull request from your fork to T0NIT0.

Git also has branches, linked repos and a whole lot of other features, but the general process is there.