Hi, I am somewhat new to PD, and I must say, I really love the audio processing and roouting capabilities.

I am having one issue, however, or have been trying to figure something out for the past few days. I would like to transfer raw audio streaming from my ASIO "soundcard" through Pure Data to VVVV, the visual programming toolkit. Reading through forum post and online tutorials, but I haven't found exactly what I'm looking for.

I have no problem setting up a connection between the two over UDP using localhost, but using adc~ to udpsend~ or netsend~ I got the encoded messages in VVVV and have no idea how to decode them.

I have used adc~ and pack~ to a send message with 64 slots, $1 $2 all the way to $64 (if anyone knows a simpler syntax way to write that message, like say "$1-$64," I would be interested to know that as well), to a regular netsend, and I do indeed get a spread of 64 values that react with the sound, but I'm not sure exactly what the numbers coorelate to, the visual models I connect them to don't seem to behave like I think they will..the values seem to ramp up higher when it gets silent before going back to 0.

RMS, or just single values work fine and as expected using something like env~ snapshot~ or fiddle~, but I want to get something more like fft as well.

I have heard people saying use jack a lot, and I have used it before, but it just gives me less routing possibilities and higher latency. I'm using a special driver called ReaRoute that basically gives me a virtual ASIO driver/Audio cable with 16 outputs coming straight from any audio source in my DAW; jack only gives me 4. I would go straight into VVVV, but their ASIO driver is buggy, there's no way to overcome a latency that's about 1-1.5 seconds.

Any help would be greatly appreciated, thank you!

Ok, maybe this question:

Is there anyway to make this patch (attached) more efficient/accurate? And would it be better to analyze in pure data or in vvvv, if pure data, how would I transfer a good, accurate and relevant list of data over the network?

Again, I think this works for the most part, but I still don't know if there's a way to do it more efficiently or accurately, and I still don't really know how to decode udp audio values in vvvv accurately.

Even if no VVVV knowledge, help with the pure data patch would be appreciated

http://www.pdpatchrepo.info/hurleur/MySendsFFT.pd