One Synth Challenge #192: Six Sines by BaconPaul (johnkhf Wins!)

[doctorbob]

Thanks @baconpaul - anything to improve workflow is a bonus along with additional "features" of course. Wasn't suggesting/expecting the mouse calls to be in the audio thread, but was just musing the fact that if it required additional code (which by the sound of what you are saying is minimal anyway), then extra mouse handling could take away cpu cycles from the audio thread!

Did loads of programming RT stuff with light pen, raster and flyback interrupts on DEC GT40 systems back in the early days of my computing (as a biologist and psychologist, not computer science!). Started in 1966 - I remember the discussions about what to call the new department - Department of Computing, or Computer Science, or Informatics!

dB

[Taron]

Hehehe, I totally understand your concern, too, as I come from good old C=64 times, being as stingy as possible about cycles. But nowadays it's a well different world with multithreading, plenty of parallel processes running independently, there's no such bottleneck anymore. Fun, though, to recognize that mindset again, hehe.

[baconpaul]

Hanz, let me second Taron's thanks for the thoughtful ideas. And let me answer some of your thoughts. And let me also acknowledge that what I'm about to say may not resonate with you, but is meant to engage and address your thoughts. And finally let me say that you are absolutely correct about the importance of varied noise in general purpose sound design.

You had asked

Do you want this synth to be used just by some nerdy sound scientists like we are?

and i thought about this question a lot. It's not how I would pose the question but if I was forced to answer Yes or No, I would reluctantly answer yes. But that's because it's not how I think of it.

My rough thinking is there are loads of general purpose synths. Massive is a good paid one, and Surge is a good free one. And there's good DX7s. Dexed is a good free one and OPS7 is a great paid one (and the arturia one is fun too). And theres loads of FM synths. OpSix and Sytrus get mentioned to me a lot. Other than Massive I think I have all of these installed, but other than surge and OPS7 I haven't really used them. So should I try and make six sines be any of these?

The answer I'm sort of coming to relates quite a bit to my personal creative process, which is I find I make more interesting music when in a bit of a box. That is, 'you can do anything at all' is less inspiring to me than 'you have four tracks and each can only be a single minimoog emulator' or some such. In that sense, where I am with six sines is it really is a way to limit your options and make you understand and be creative with a pallet which is smaller and distinct from many other synths.

We all know the 'first time with a synth' idea of 'ahh grab a saw, stack the unison, apply an lpf, hit it with an envelope on the cutoff, voila'. I think it's kinda cool that you just - well - can't do that with six sines.

So the question I would have posed is "Do I want six sines to be a general synth or one which explores a subset of synthesis but still allows interesting results" and for that I think the latter. This is the same reason I've been reticent to add filters.

It's not really a technical issue (although adding noise has some technical challenges, like what does it mean for PM to be inbound on noise, but you could fix that by having noise only be available on op1 and turning op1 feedback into a filter on noise mode). its more of a creative one.

So its entirely possible for you to read this and say 'well that's dumb, you are missing a real opportunity' and to that I'd say 'yeah that's true; i can't wait to put the same FM tricks I put in six sines into surge xt2 though' (which I can't!). But my head is sort of sitting on 'Six Sines is this little focused FM/PM/AM/Modulation gem' rather than 'six sines is the be all and end all of fm anything ever with a kitchen sink inside'.

I dunno. maybe that's dumb. I reserve the right to completely change my mind and apologize for this dumb post this summer. But it's kinda where my head is now. And since you've been so thoughtful I wanted to share a (hopefully) thoughtful response. And hope you enjoy six sines nonetheless!

[doctorbob]

Nicely put. This is exactly where I thought you were with the "project". It would/will never be a general purpose synth, and I respect that. It's a great little beast for exploring in depth the possibilities (and depth) of FM (and beyond). Helping the workflow has been important for us all to work with it in all its glory. With Taron's videos it's an excellent teaching/demonstration tool for working with FM and the notions and ideas will carry over to many other free and commercial FM synths.

I often wondered what would happen if you fed audio signals into the FM modulation chain? Might be an interesting "experiment" when you have a few spare moments (ho ho ho).

It has encouraged me to grab my real DX-11 again and get into programming it again via the 2 line LCD display. So old it needed the battery replacing - gazillions of screws to get at it! Still a beautiful sound with some clever features inside as well on top of the standard FM op handling.

Keep up the good job you are doing with all your projects. So good to see a real enthusiast at work.

dB

[baconpaul]

you know the one thing i was thinking bout doctor bob when i wrote that was "but shoot, I could make it so op1 and only op1 was an audio input with a side chain". that idea is definitely in my head. The sample rate mismatch is what makes it not-trivial. I need to go from host rate up to internal engine rate which requires another SRC instance and so on.

[Taron]

Considering that this would likely be merely a curiosity, I'd imagine even a noise would take precedence over that. Practically speaking.
[Shift] + [Mouse Wheel] for fine tuning steps!

[doctorbob]

++1 for the finer control via mouse. Pretty please.

I was just curious about how audio injection would actually work in this type of synth. But, maybe, because this synth is also a testing/proving ground for ideas, curiosity is not a bad thing. Yeah, I understand how all sorts of mangling of sample rates would be needed - exercises the grey cells though.

Thanks again for the tooltips, work an absolute treat.

dB

[doctorbob]

Here is my submission for this round of the challenge. Hope you like it. I had fun with this synth so thank you Paul.

DAW: Reaper

Track Details: 16 * Six Sines, 16 * ReaEQ, 1 * OTT, 1 * ReaDelay, 2 * LFX-1310 (autopan), 2 * Bass Professor, 6 * Pecheneg Tremolo, 1 * Twin Delay, 2 * Oril River, 1 * Valhalla SuperMassive
Master Track: TDR VOS SlickEQ, Voxengo MSED, Limiter No 6.
Meter: SPAN, YouLean Loudness Meter
Stats: LUFS: -15.4, DR: 11

https://soundcloud.com/doctorbob/six-sines-of-finches

[Negoba]

I'll start by saying this is the first FM synth I've played with so this may all be rubbish.

I found the layout of the operators really straightforward and so this synth was especially good for learning FM for me. Having individual envelopes on everything was a nice alternative to an elaborate mod matrix.

Maybe others have FM synths that they think are better teaching tools but from what little I've seen some of them are a friggin mess to figure out. This one was a great starting point for me. Thanks baconpaul.

[baconpaul]

Taron, doctorbob shift mouse wheel and shift drag goes to a fine mode for me here on my Mac no problem. Do you see different? If so os etc… would be nice to know. The code makes shift make the wheel 10x slower

[baconpaul]

this synth was especially good for learning FM for me.

Me too- I learned a bunch making it! Thanks for the kind words

[doctorbob]

The code makes shift make the wheel 10x slower

It's all good for me - my bad - Win10, Reaper, latest nightly. Maybe I was trying "alt" mousewheel?
Sorry, brain malfunction.
dB

[IV!]

I absolutely understand and feel your concept and attitude towards Six Sins, and it would be great if you left it as it is. On the other hand, I would like to see the wishes that were expressed here in the synthesizer. So, a kind of dilemma?

I do not know how this will sound to you, and whether you will have the time and desire to implement it, in any case, these are just my thoughts out loud.

What if you went two ways? Create Six Sines 2 (maybe v2 is not the best idea, maybe Six Sines Advanced, or something like that), where you would start with Six Sines, add noise to op1 and filters (I would be happy to see two, HP and LP at the output, it works great for final sound forming), there may be audio input and other ideas from people which you would find useful.

It would be two brothers, one a purely engineering FM synthesizer, the other a more general purpose FM synth. I don't know if you would even want to do something like this, it just seems to me a good solution to the dilemma. Anyway, it's just my thoughts out loud)

And thank you very much for this six amazing sines!

[emikaela]

outside of osc you can just layer it with a noise source and put a filter afterwards, if that's what you want.

that said, would be cool to see six sines as a new oscillator mode in surge eventually down the road...

[Taron]

Taron, doctorbob shift mouse wheel and shift drag goes to a fine mode for me here on my Mac no problem. Do you see different? If so os etc… would be nice to know. The code makes shift make the wheel 10x slower

Won't do anything here.
Shift + Drag, yes, but Shift+Wheel nothin...
[M1 Apple Silicon] Sonoma