Great. I just need to add RebelTech/Stoicheia to my library and I will give it a go. Thanks!
Edit: Done and working. Thanks again.
Yeah, as said, not much time was spent selecting the modules. I could have chose more ‘common’ ones. But, all modules used are freely available. Also, they can easily be replaced with modules with similar functionality (maybe except Sckitam Waveguide Delay).
Especially oscillators (other then filtered noise) and different filters/eqs and distortions/wave shapers will effect the sonic (im)possibilities. You could also add as second feedback loop with some differences in modulation to get some stereo dynamics. Or even construct completely different Waveguide models.
1 Like
Well, it helped me. Yesterday I was exploring Karplus strong and making slow progress.
Wow this little patch is fantastic. Very useful concepts in here. Thank you.
1 Like
Some not so scientific notes on Waveguide/Karplus Strong/feedback loops that might help (I’m not an expert on any of this).
The simplest form is just feeding a delayed copy of output back into the input. Where it should be noted that a cable in VCV Rack already introduces a 1 sample delay. The delay time determines the pitch, which is then actually the number of delayed feedbacks per second. So…to tune this to V/Oct this would need some (added/external) ‘calculations’.
The amount of feedback determines the ‘decay’ time (this will start sceaming if set too high, generally above above 1x).
So far, this can be done with any delay module that offers short enough delay times (in the milliseconds range)., e.g. the Fundamental Delay.
But, to get control over the exact amounts of samples to delay, you need more exact control. Then there is also the potential interpolation issues and the tuning. This is simplified/offered by Sckitam Waveguide Delay. But there are other ways.
Then there’s the polarity of the feedback signal. Inverted feedback will result in odd harmonics and a drop in octave.
Low pass filtering in the feedback path is used to emulate natural dampening, starting with the higher (lower amplitude) frequencies, emulating energy lost/transferred/leaking from the ‘oscillation process’ to the physical surrounding environment.
BTW, a delayed feedback already introduces comb filter effects, simply due to the shift in phase, also effecting mostly the higher frequencies, generally being more plentiful (per octave) and generally of lower amplitude. So, modulating the delay affects the spectrum.
Adding digital filters (based on delay) will also add delay, which should be compensated for, if exact tuning/pitch is important.
Drifting further off topic…
In a broader context many models can be constructed using this Waveguide technique. Models in many different of shapes, sizes and dimensions. Thus,generally, but not exclusively, computationally efficiently modelling all sorts of exciter/resonator constructs. Physical models for strings, tubes/cones, plates and membranes are common use cases.
Julius O. Smith III at ccmra offers an overload of both relatively simple/practical and deep/research info on this and many other DSP related subjects.
It’s a huge world with huge potential, but surprisingly few actual implementations in the world of synthesizers. Partly because of…yes…patents. Just like John Chowning’s FM before, the Physical Modeling/Waveguide technology was patented and then licensed to…Yamaha. Check out Yamaha VL-1. But, as with FM/PM before, many of those patents expired in recent years. So…who knows what the future may bring.
3 Likes
I’m cross-linking to my demo patch that involves tabla emulation as well as generative raga or psuedoraga.
oh woah ! really nice all this informations !! deep to search indeed, but you open a lot of gates ! thanks a lot for your patch and your explanations !
1 Like
Here is an electric table I created. It uses no samples; all sounds are generated. The low sounds used the PSI OP while the high (tabla) sounds are created by the Starling Via Meta. The attached patch file is different from the movie as I edited it to remove all references to paid modules. Let me know what you think. Enjoy
I like the Markov chain rhythms. I’m not sure if I am hearing the drum sounds as you intended them. Mostly I hear something like handclaps and a bass note.
The patchstorage video will not play as it is marked private.
Well…since we already seem to have fallen into the rabbit hole of Physical Modeling…
I mentioned the pioneering work (and patents) on Waveguide Synthesis (and more) by Julius O. Smith III at ccrma,
One of the “byproducts” is:
The Synthesis ToolKit in C++ (STK)
https://ccrma.stanford.edu/software/stk/index.html
The Synthesis ToolKit in C++ (STK) is a set of open source audio signal processing and algorithmic synthesis classes written in the C++ programming language. STK was designed to facilitate rapid development of music synthesis and audio processing software, with an emphasis on cross-platform functionality, realtime control, ease of use, and educational example code. The Synthesis ToolKit is extremely portable (it’s mostly platform-independent C and C++ code), and it’s completely user-extensible (all source included, no unusual libraries, and no hidden drivers). We like to think that this increases the chances that our programs will still work in another 5-10 years. In fact, the ToolKit has been working continuously for nearly 20 years now. STK currently runs with realtime support (audio and MIDI) on Linux, Macintosh OS X, and Windows computer platforms. Generic, non-realtime support has been tested under NeXTStep, Sun, and other platforms and should work with any standard C++ compiler.
The Synthesis ToolKit in C++ (STK) can be found on Github:
There are several projects available there. One of wihich is:
Ragamatic
Alas, for the drums and Tabla, the readme states that they are sampled…not modeled.
- Sitar and Drones are physical models.
- Vocalize drums and Tabla drums are samples.
But the Ragamatic project and code (and for sure the STK) might contain useful info nonetheless…
2 Likes
Interesting. I wouldn’t want to go down this path without understanding the patents.
Very few of these patents are still in effect. Most are not. Not sure which. Chances of unknowingly infringing them seem low. I bet the library/code/licence files will mention these legal details where applicable.
It’s comparable to the way FM (PM) was patented and licenced to Yamaha. Patents that prevented others to innovate in that domain. Patents that, many years later, expired, so others could incorporate the domain into their innovations and products.
1 Like
I suppose the question is whether these open source SDKs and APIs are available in the Rack build tool chain, or is the expectation that a plugin would need to include the library as source code, if the licensing allows that.
I really have no interest in fooling with this, but others may.
On the subject of using ‘external’ libraries. You may (or may not) know that the somewhat Tabla, but for sure indian (drone) music, related Seaside Modular Jawari plugin makes use of the free and open source DaisySP library, that also offers Physical Modelling.
DaisySP
A Powerful, Open Source DSP Library in C++
3 Likes
I did know that but had forgotten.
Thanks, I didn’t know about the private thing. If you are hearing handclaps something is not right. Play the video (I fixed it) and tell me if that’s the same.
Yeah, the patch sounds nothing like the video. Are there any samples in the patch that require a specific sample rate?
1 Like
No as I said in the writeup there are no samples involved. I downloaded the patch and surprise it works fine on my system. I have no idea what could be happening.
1 Like
Me neither.