braintrust for ideas/features for as-of-yet uncreated VSTs

[Shane Sanders]

I hope this thread will one day become a sticky because with all the brainpower here there's a good chance that your small idea could really make a difference in the quality and features of potential plug-ins. One day, perhaps, devs will stroll through here and the light bulb in their wonderful brains will light up with a new connection concerning something they had already planned. Hardly anyone suffers from communication, feedback, and constructive collaboration...

Post your idea for what might make a typical feature more "rich". For the sake of the devs who might only want to skim the material in the future, please highlight your idea 'parts' in yellow. That way the welcomed noise of interaction and camraderie can be filtered during those times when you've come here just for the ideas. Thanks! ..hehe...who knows, maybe one day you'll be able to search for just the text that lives in the yellow tags!!!

Here are a few ideas that might make some synths even more musical, etc.

1. If you include a delay section in your synth, go ahead and include an option to feed various amounts of the wet signal to another delay line in total isolation. Make that extra feed have an auto-pan that can sync to something else, such as an LFO. Provide two sliders, for low and high freqs, that allow the user to isolate a band and then apply a new delay rhythm to it--you might even want an extra resonance control on it just for fine details. It becomes unnecessarily complex to to this manually when you're trying to do it to isolated bands (and while you're trying to be a musician and write music rather than engineer). If it existed in most synths, it would allow for a very musical result with not much extra CPU footprint. It would allow for sublime polyrhythms that live in the patch rather than in the mixer...and of course, make those sliders have their own lfo. Yummy.

2. All synths that have portamento (and they all should), should also allow you to choose whether it's on all the time or just when you've overlapped notes. As a developer, think like a player: how are melodic phrases touched by artistry? How are lines interwoven and massaged into something that moves you? One word: articulation of individual notes. Give us mono mode. Give us legato. Give us a separate ADSR just for the attack transient apart from the main ADSR. Make the extra transient ADSR scale to velocity. Players want to feel their instrument react to them. Reward the players in this world.

3. Just about everything in a synth should sync to tempo! What's stopping you?

4. Be sociable! If you recognize that freeware developer X has badass filters, talk to that guy. Make friends. Give him something of yours that would improve his 'product'. After all, you're giving it away for free, so why not go all-out and make it the best free thing it can be.

[Zedah]

Hey! We thinks the same, a couple of weeks ago I started my project (blog), I will put my old and new ideas (mostly Reaktor based) there. Can I post a link to my blog? I'm just afraid to come up as spammer/flooder in this holy forum

[Shane Sanders]

Hey! We thinks the same, a couple of weeks ago I started my project (blog), I will put my old and new ideas (mostly Reaktor based) there. Can I post a link to my blog? I'm just afraid to come up as spammer/flooder in this holy forum

Sure! Post anythig man, you gotta drink up!

[Zedah]

Okay, thanks Shane

the link is http://www.synthofmine.com and the name is Synth Of Mine, it took 2 days of brainstorming and 2 ms to get this name, it just came up within 2 ms in my brain after these 2 days

The idea of the site is posting of fresh and simple ideas which produce unexpected, interesting and fat sounds, or any new approaches in sound synthesis, sequencing and sampling. I've got some experience with Reaktor since 2000 and decide to start a blog to keep my ideas out of the dust.

But I just began and there is only one article atm. I wasn't in hurry because most search engines have not started to index the site yet and there is not much of visitors. But by end of the week I will continue to post my old and worked out ideas and files.

Feel free to register or just comment. If you got something also to share I'd be glad to make the blog author as well.

[setAI]

been waiting for a sample-based drum-designer synth forever- something like the Korg S-3- the synth would have a wide selection of seperate drum head/body/release samples that can be used to create custom sample based percussion instruments-

also- ALL synths should allow the audio output of the oscillators/filters to be patched for audio-rate frequency/feedback modulation anywhere in the synth- I don't know why so many don't get it- just about any boring synth will come alive and stand-out with a little bit of subtle frequency/feedback modulation- especially VAs

[Z3R0T0N1N]

im thinking of a complex semi modular synth controlled entirely by moving nodes around on a series of (virtual) surfaces on the screen; think of the metasurface in audio mulch being used as the routing scheme for a modular synth, but a series of metasurfaces instead of just one. a right click on a node would allow one to set whether the proximity would result in a snap on/off or a fade in/out action. one surface would be purely for MIDI functions, one purely for audio, the other two would be for mixed audio / control signals, like analogue modular.

er.... ?

[Shane Sanders]

Cool idea Z3R0T0N1N. I don't have AudioMulch but I think I understand what you're saying.

Please be sure and set your idea in yellow so devs can find it later in the wash of other stuff in the thread. Thanks.

[Z3R0T0N1N]

hi shane
just got back from a couple weeks at Tofino and Victoria, but i changed my idea to yellow.

[runagate]

Shane, once again you've aided mankind.
I'll try to stay the f**k out of this for the moment as I do tend to carry on so. Maybe dubbyah's awake.

[MacZero]

1. An fully featured eq that "slides" up and down according to a midi note input (so the eq nodes follow the pitch of a monophonic synth line like a bassline.

2. A multiband compressor that does the same.

3. An FM synth where you can have sample playing operators to FM into synth operators and visa-versa (sp?)

4. A delay that can load up other vst's into the feedback loop. Would be cool for dub type evolving delays.

5. An eq that you can use to make kick drums and basslines sit together better by analysing the spectral content of the two signals and applying eq curves to remove muddiness and discordances etc.

[dubbyah]

here's something i just thought of since i've been in an audio-visual experiential integration mood lately

mmose

(massively multiplayer online sound enviroment)

using the latest 3d technology this puts players in a persistant virtual world in which you can design your own 3d instruments with extremely unique features and control schemes (three dimensional interfaces with extremely complex operations able to be programmed into it, using some sort of code, maybe a built in modular setup like max or just code, obviously this would have a learning curve) and put them other places so people can play them and interact and have realtime playing sessions... ideally each user will be able to use their own vsts as well as creations built inside of the world... using outside sound devices would be interesting as well... even a 2d view, but some sort of persistant world.. use your imagination...

obviously bandwidth would be an issue as would the quality of the sound in terms of how much yo uwould be needing to upload/download in a specific time constraint... all i know is that there is a severe lack of virtual interactive sound envrioments, and even very low latency real time audio interaction enviroments.. ninjam is the best i know of for audio real time interaction/jamming, but it's very buggy and doesnt do well with routing on a windows platform...

everyone could build and collaborate on instruments which could hvae some sort of unique interface similar to a game, some sort of alien creation... this wouldn't just be synths, it would be audio engines controllable by 3d graphical interactions that could be inputted to simultaniously... really taking it to a level there would be some sort of custom virtual reality interface to interact in this type of virtual alien sonic world

the technology that must be employed in such software, in terms of main consideration hindering implementation and creation, would largely be bandwidth constraints. transmitting that much complex data would obviously require broadband. but if you look at the current type of interactability with WoW, or Eve Online, or any of the mmorpg's, something like this really doesnt seem that far off. the graphics certainly wouldn't be a problem, i think the only problem would be the actual act of creating these unique 3d alien control schemes and innovative brand new original systems of musical creation.. obviously visual feedback would play an important role in such a process





---


in terms of vst's i'd like to see, (these could all be done in max easier most likely),i'd like some sort of algorithmic system that generates note on and note off's based on a psuedorandom algorithm or certain patterns/repetition schemes with modulations imposed at various intervals which could also be modulated, which also would send out program changes at random intervals based on various algorithms, cc's based on various algorithms, and then would enable/disable certain vsts in a signal chain, as well as scramble their order... so you would just load up like 10 vsts, turn them all off, then start this system while either manually modulating various parameters of modulation/speed/algorithms or just letting it play by itsself. even further on a tangent, would be a bit harder to code, would be to save the order of specific vst signal chain routing combinations.. so as it was going through turning on/off various vsts and combining them in novel ways, it could add them to a database... the good thing with turning on/off vsts is that you can conserve vst usage. you can have 50 vsts loaded into the order of operations in terms of code output and you won't strain your cpu a bit, so then you can instantly enable/disable/change programs/change cc preset schemes/shuffle signal chain at will out of an ENORMOUS bank of various audio units. you may have to manually assign each vst a rough cpu usage number in terms of some coding unit, so that you wouldn't max your cpu usage. you'd have to set up an algorithm to never go over x (let's say 90) cpu usage, but to create as many novel combinations of vst signal routing cc schemes pc and algorithms used to enable/disable the vsts. you can make this in max or supercollider and expand on the idea, as it is, but it would just be nice to get this in vst format, or something similar, maybe less complex. you can acheive something vaguely similar with a combination of current plugins, most likely, at least you can with pcs/ccs, i dont know about enabling/disable/scrambling the signal chain.


interactive musical systems are interesting too

the idea of improvising with a comptuer which improvises to your input and then you improvise to it's output is a sort of feedback loop which rivals playing with a real player, as long as you supercede an amount of unpredictability into the algorithms dictating such interactive behavior, so it doesn't become too formulaic...


let my expand on the original idea, virtual enviroments not done from a computer but an actual virtual enviroment similar to the mmose idea is further off than a digital computing implemented software but still fun to concieve... imagine that virtual reality technology is much cheaper and more accessable than it is, and then design some sort of system in which you are in a virtual alien world, physical instruments that you are able to manipulate with HAPTIC feedback that are playable in terms of muscle memory associations and physical corrolation of neuronal output, the same way you would play some sort of weird ethnic instrument that you've never played before that had alot of contraptions that you had to find out what they did, you could encounter tons of these objects that you would have to discover that had been fashioned by other players in a similar sort of haptic feedback/ advanced programming enviroment...


and don't get me started on the possibilities of neural interfaces

what was music like 40 years ago?

what is going to be possible with the enormous amount of computing power that will be available in less than 10 years.. i mean, 80 core cpus in less than 5 years...


abc231

[runagate]

dubbyah, I've actually worked on such a thing in the past with 2 programmers. Turned out not to be possible at the time. Damn, I want that to happen.

[O.L.T]

tag

[Shane Sanders]

This is probably my best idea so far.

In 3D animation programs, there is often a type of macro for doing swarms and flocking when animating flying animals. For instance, you create an insect and once you have all the individual things about its wings and legs worked out, you can then create many instances of it and through the use of an algorithm, make a swarm of them move in tandem but with subtle random differences. It makes the work load less and the realism much more convincing. OK so now hold that idea in your head for moment and apply it like this:

Make it so that a subset of your synth/effect controls can be flagged and thus ganged together to be affected by a joystick on an XY pad. Perhaps another controller, such as a real knob or slider could affect the Z axis in some manner. So, for instance, you'd have a bank of filter controls affecting incoming notes but as the player makes a circular motion on the joystick, the tolerances of how much a note is affected by the subset of controls changes slightly over time. Thus, the same player motion, even if perfectly repeated, would induce a different amount of each of the ganged paraments and it would do so on a curve that is reasonable rather than simply random. Yet it could be random or else a morph between predefined value tolerances. The idea is mainly to make it so that many controls can be affected by few controls within a range. There are other applications for this, too, especially with phsymod parameters for the values of a note, but I'll leave that to your imagination.

[runagate]

Sweet, PM me when someone has Shane's latest finished.
I've got at least 3 compositions from the last decade that use this concept and I'm getting impatient to let them be sound.