Lines with 4 sound buffers?

  • Hi,

    Any chance of this happening in the near/far future?

    Been using Aleph a great deal as a live granular box with the grainfreeze scene. It would seem, considering the 4 hardware outputs, that it could be capable of storing 4 separate buffers in Lines, or am I missing something? Are there limitations that do not permit this option?

    Would make it an even awesomer little improv box :)


  • This is definitely in development, but who knows when it will happen :)

  • there are 2 buffers in the lines module. more buffers means making some tradeoffs in other processing (less filters, slew, crossfade logic, etc) or rewriting the audio buffer loop to add latency, or something on that level. experiments with writing new modules are greatly encouraged. a simple 4x delay loop with no other processing could be a good place to start.

    there are other uses for multiple ins/outs, like mixing, splitting, routing dry/wet, etc..

  • i'd love to try building a 4 x delay loop, but really haven't got a clue where to start!!!
    i tried looking at where buffers are but i don't really understand the format they are written in.
    is there some kind of tutorial (even just for audio in C or on the blackfin) that anyone could point me towards? just some tips on anything around syntax/programming structure would be a great start

  • i think sooner or later someone (maybe trent) will finsish up a well documented tutorial/template on basic module implementation. but i might need to implement an additional buffering layer in the bfin library to speed things up, before we can get too crazy.

    i made some abortive attempts last month at a 4x FDN network reverb. not happy with the sound, even in simulation; more interesting stuff with x-feedback through SFV and delay crossfading in existing lines. i'll keep working on it though.

    ultimately i just want to make a lexicon vortex clone.. it'll get there...

  • Ha, the vortex, I remember those, could never afford one at the time.
    It still feels to me like interpolated buffer reading is the holy grail for the aleph, as being able to bend the pitch would open it up to a gazillion modulation/delay/chorus/morphing possibilities :)