Interpolation

vvvv beta general
1

hi fellas,

i´m trying to patch a module that automatically normalizes some values that could possibly have some high peaks in future. now i gonna make myself up to some interpolation techniques, but i´d liked to ask this question to you guys because i think there is a chance that you needed to have automatic rezising of mappings too…
(i.e. scope spread or fft --> numerical data out of a music stream)

so, the easiest way would be the Map node @ clamp mode. But this would cut the dynamics out of the signal, if there is a extreme peak.
then i thought some complex damping/linear filter approximation would do it, but i think i patched just crap.

i´m asking for a module (that would be very best hehe), but i also very appreciate food for thoughts by you practised

greetinx, b³

2

i tried to use some own built modules with a queue and the mean node to get the results but i still have to clamp the map node, maybe due to wrong patching.
if you wanna have a look on a gdi scope that always maximizes values ( hoping so gg) get yourself a copy of the attached zip (:

scopewvaluepolation.zip (7.3 kB)

3

sorry, no time for long answers

try Bounds (Spectral) together with queue and map,
maybe Gamma (Value) will help you prevent clipping.

4

thanks, kalle. these two nodes come very handy. i´ll give it a try 2morrow.

5

RMS should be the better choice to display the music envelope … and output is between 0 and 1 …

6

but RMS is a DShow-node
and often i had to normalize its output…

7

well, actually i indeed had the problem that i want to make transformations “dance” to music signal. the rms node is an idea, but in good programming manner i want the patch working on music signals in any level.
(at my work place i can´t have the techno sounds cranking thru the whole level hehe)

i think every1 of you used 4v for vj purposes (music information to transformation information). for that i need the signal having a “tube” of min/max levels to start searching for patterns (any ideas or hints for analysing wave form structures with 4v? i set a bet on tonfilm hehe).

so i think there is a mix up of that mentioned Bounds / Gamma / Damper / Map but i was not successful by now. But i found out that the Damper node is a bit buggy for fast changing values where the damper node outputs a MaxFloat and is not resettable :-/

cheers