This site relies heavily on Javascript. You should enable it if you want the full experience. Learn more.

VVVV and Sound

question open misc

d0this 02/02/12 - 11:45

Hello vvvvolks,

in next Semester my sound engineering Professor wants to do some research about audio Analysis and stuff and he told me that he wants to use PureData.
I was really happy about that, because Graphical Programming Languages weren't a big deal in my university in the past.

Now i have the question: is it possible to do the same stuff in VVVV as in MaxMsp or PureData? Such as building Synthesizers?
Or has VVVV not the capabilities for sound as Max has?

greets

7 replies 0 new

link | Flag this reply as a solution. tekcor 03/02/2012 - 01:10

Hey,
There are some sound possibilitys, but the engine is not as functional as max/msp or pure data.
audio

link | Flag this reply as a solution. thezer0ist 03/02/2012 - 06:35

In general... no, vvvv can't really do the same kind of signal processing that can be done in Max/MSP or PureData (or Reaktor). VVVV's support for audio is limited to directshow audio graph construction (including vst plugins) and some basic analysis , the base.net plugin stuff (which can do... some stuff, i'm not really familiar with it).
The structure of the VVVV engine isn't really suited to signal-rate audio processing. VVVV evaluates everything in a sequence of frames, in which every node effectively does its stuff at the same time. Each frame takes a bit of time to process, meaning that you generally end up with a rate around 30-100 frames/second. Every node in a VVVV patch must be processed during each frame (except for some optimizations where some nodes only do work if their inputs change).

MSP does a similar kind of frame-based processing with its signal processing stuff. However, only the signal-rate nodes are processed in each of these frames. The signal stuff runs on a separate thread. Other nodes that aren't signal nodes send messages to each other and respond to messages from other nodes. So, their activity is separated from the signal stuff. The signal thread can run at >44100 samples/second.

In case you can't tell, i like Max/MSP..
For rendering though, VVVV can't be beat.

link | Flag this reply as a solution. jzzxh 05/02/2012 - 03:39

You can try " Pdvst ".

thx

link | Flag this reply as a solution. sebl 05/02/2012 - 12:12

ouuuuu - Pdvst looks really interesting… pd as a vst in vvvv

has anyone tried that yet successfully?

link | Flag this reply as a solution. d0this 05/02/2012 - 16:15

thanks thezer0ist for that detailed answer.

Now i understand the Problem.
Iam also very interessted in pdvst.
Maybe somebody has already made a vst out of pd and wants to share it?

greets

link | Flag this reply as a solution. r4dian 05/02/2012 - 20:27

pdvst is a VST version of pd.http://crca.ucsd.edu/~jsarlo/pdvst/

Its made from an out-of-date version of pd, but still useable.

If I were you I'd use the standalone version of pd to do audio and analysis, and comunicate to vvvv with osc if you want to use vvvv to visualise it.

link | Flag this reply as a solution. sebl 06/02/2012 - 01:18

sure, that is the way to go when you're in "production mode".

another thing would be pdvst as "plugin/addon" that is directly integrated in vvvv - like already existing vsts but editable in realtime. as far as i've seen, the licensing of pd would allow us to do that.

VVVV and Sound

Shoutbox