I am attempting a modular (soft)

:O)

only problem I can see is that delays and such require a lotta memory and I would have to restrict the delay length to some atrocious amount. Then again maybe I will call this a feature and have some sort of downsample/fz psuedotimestretch on purpose

but like I want a module cap of 1000 at least (if ur computer can handle it) and JS has ~ 8 million memory locations so 8000 per module at the most. Or maybe I will just manage buffers/waveshapes in a different way/panic if you have 100 delay modules. yeah that's still enuf for about 200 seconds of delay in mono I think we're gud. I still think I will do the resolution thingy though

and someone is working on a JS to VST thingy so it will probably be faster/more convenient in that form

one thing I will be sure to do is my modular of choice BLOK has a feature where you can move a waveshape around and it wraps around the window, I want to have an XY wrap if u modulate it. Just seemed like an obvious feature. Also I would like to do deformations on the waveshape (like PHOTOSHOP SWIRL FX on the waveshape srsly) in realtime but that's a tall order

the point being more for me to modulate shit with the waveshaped signal more than directly using it as a distortion. But maybe the distortion would be awesome too?

osciliscope that gets written to waveshape = doable

haha then also maybe set the offset with another wave/signal. chaos

evry color of noise even colors that don't exist yet

I SERIOUSLY have to look into random/noise algorithms though and like attach the random seed to a modulator and see wtf happenes

make script to connect 2 random.org my synth's noise generator is more random than urs

I do want to do EXTENSIVE stuff with noise generation though. Like maybe limit the amount of seeds in such a way that it aliases in some specific way I can only try to imagine/draw waveshape and use as a table of which seeds can be used. And of course downsample the noise. I had an idea/implemented partial downsampling at one point (like 3.7x downsampling would downsample 4 times 70% of the time and 3x 30% of the time). Like do everything imaginable considering to the benefit of it being digital/reproduceable unlike natural noise

pink noise is interestin because there is a way to do it with iterations or you can just filter white noise

http://www.taletn.com/reaper/mono_synth/

I'll probably just implement this in the synth

Ragnarock, you should just get Massive and so you don't have to worry about all this BS!

lol?

Would you like me to recommend obscure vsts from the 90s instead?

WTF, people are still using buzz?! Cool shit!

renoise is a descendant of buzz is one way or another afaik. also yeah they rediscovered the source code like 3 yrs ago and added some coolstuff

convert frequency of letters RSTLNE to FFT frequency bins

download data of WATMM message board convert frequency of 'lel's to time

might seriously do this on a more general level again like for databending reasons. Like allow to load a file in the data folder and collect statistics about it. Interpret as sounds/parameters. Use wrong decompression algorithm on compressed data/not even compressed data

I'm gonna research puredata before I get too deep into this/not sure if I want to do something redundant

  On 5/28/2015 at 11:58 PM, Ragnar said:

also yeah they rediscovered the source code like 3 yrs ago and added some coolstuff

Damn! So not every bit of the code got crashed be that evil HDD.

n1.