...... in the signal~ domain.
vhm
http://www.pdpatchrepo.info/hurleur/random-position-playback.pd
Randomized sample playback....
hey I really like the sounds that can be achieved with this, but I have very little understanding as to what's going on.
partly because everything is in the signal domain, i have to guess what numbers are being passed through each stage.
would you comment on your patch, in order to teach me?
Dual 1.8 IBM G5: Mac OSX 10.4.11 -- Asus eeePC 701: Pure:Dyne / eeeXubuntu GNU/Linux -- myspace.com/thearifd
The base for these phase modulations is the [wrap~] object after the [phasor~] and before the [*~ ] nr-of-samples.
This makes it posibble to push or drag the playhead in the signal domain and still stay inside the loop.
The left sampler shows how to quantize in time. This is done by the [*~]->[wrap~]->[samphold~]. Eg. setting the right input of [*~] to 16 will poll the value from [samphold~] 16 times each period. To quantize values in the signal domain I scale the output from [noise~] to a 0 to 1 range, scale it with the number of steps, remove the fractional part by subtracting a [wrap~]ped version of the signal and finally divide by the number of steps to get back into the 0 to 1 range. The [hradio] sets the direction of the modulation: left == subtract from phase, the middle == bypass and right == add to phase.
The right sampler gets its time quantization by scaling the phasor to the number of steps in the table and the reading it with [tabread~]. Here I also quantize the output which is necessary if the table is changed manually.
mvh
Man, this is one great patch, I must say I am absolutely including this into my live performance! Absolutely fabulous patch.
Thanks for this, a fantastic patch
Thank you so much. Its ben some month or two ive been learning thing or two about pdata and this kind of stuff gives me strenght haha. Thanks once again.
Oops! Looks like something went wrong!