Ableton multi track audio into vvvv via mixer with firewire interface(asio) for live visuals- trouble shooting

i am working on a ableton set with live visual through vvvv. currently using FFT(Dshow9 4 channels) for splitting bass, himid, low mid and high. I have a problem where when vvvv is running and the audioin nodes are set to the mixer channels, ableton wont play audio and gets only audio glitches, not anywhere near the sounds i produced in ableton.

for example, a kick track driven by a max for live sequencer controlled by apc40 from ableton goes to 1/2 channel, which is then brought into vvvv. snare goes to 3/4 and so on.

yet another issue is that it seems vvvv can’t take only channel 1 as a mono.

the mixer is 16 channel mixer with firewire interface. is it not possible for the mixer and vvvv to use the same asio driver at the same time? or could this simply be a cpu resource issue? or is there somekind of compatibility issue?

I am running windows 7 with mackie onyx 1640i, ableton 8.3.

audio analysis live_MODROP clean.v4p (127.4 kB)

i have never heard of an ASIO driver which can be openend by two programs… it might be possible to use ASIO in live and WDM in vvvv. but then make sure the sample rates match, this might be your problem…

hi tonfilm, think it is possible for one asio driver to have multiple clients, though i think it is impossible for one program to use multiple asio drivers. but i don’t know how stable or how low latency, how good the sound quality would be. just found what might be an answer. http://www.pgmusic.com/forums/showflat.php?Number=339969

have to try tmw :) it might be that vvvv and live can access the same asio but maybe not the same channels at the same time. keeping the fingers crossed.

thx for the advice for sample rates! but didn’t get what you mean with wdm, if you meant windows driver model…

yes, i meant windows driver model…
but the little steinberg tool for multi ASIO clients looks like a good solution.
hope you sort it out…

fft analyse inside ableton (max for live device) and transmit the data to vvvv via udp works great too!

hm, so live runs fine with asio client, all the input and output abled,but needs more time for me to decide if this is a set up stable enough for a live live gig.

as for vvvv, asio client doesnt show up as a choice as a driver, at least not from Audioin(Dshow9). i think I should see if it is possible with audioin(bass) though i haven’t used it b4. i look into some examples using bass, and try to see if it is already possible or I need to learn more about what kind of proxy server the steinberg tool uses… another option is asio4all i guess but I have read from different forums that steinberg works/sounds somehow better.

@ guest: oh, if that works already I would like to see if this could be a solution for me! thx a bunch!

@guest: i am trying to set up what you suggested. for the method you were describing, is it possible with m4l or do you need the full max/msp? I got m4l just recently and havent built a m4l device b4 but it shouldnt be a problem. are you suggesting that I should build a m4l device that does fft (fftin, fft and so) analysis and sends audio via UPD ? I think I even found a quick youtube tutorial for it (http://www.youtube.com/watch?v=NR0F3uCzTmo)

or was there an m4l device built already for this? I did look into the max for live website and didn’t find any- ended up getting a bunch of sequencers hehe.

hi youngjae,
have a look at this device pack to transmit osc
http://resolume.com/blog/8717/max-for-live-resolume-patches
and this one for FFT analyse
http://idiron.kaen.org/av

cheers

i’m neither a max nor an ableton guy, but maybe this could be of help to you…

edit: ooops, circuitb was faster))

:)… no text …

http://livegrabber.sourceforge.net/

works briliant!

weee thank you all for suggestions. So instead of shouting IT WORKS!!! / it doenst yet for me… new to anything that has to do with network server/clients in vvvv, I wanted to share some thoughts and a little bit of evaluation first.

I checked out idiron and i like everything about it except for its cpu footprint…it might be my machine really at the moment i dont have i7. but when I have the idiron fft to midi-osc on three or four tracks (audio channels), it really pushes my cpu to the point the audio from ableton breaks…anyways, just a few hours looking around for the right nodes and so on to see it work at least, thought udp server or oscencoder were the right nodes for taking in osc from ableton and tried but no go so far. Idiron seems really geared towards resolume though… -1 for grabbing params that is set to resolume only. But still an awsome plug in!!! last time i tried to midi clock sync resolume and ableton, the bpm was always off and moved around bit much.

from cpu standpoint livegrabber seems to be leaner by three times or so and i appreciate the parameter grab features (comes with “learn” fxn, and it also grabs params semi autometically when placed after an instrument or an effect.). however, since it only can send from ableton via IP, not along with localport, it doesnt seem suitable for what I am trying to do at first - multiple tracks to multiple audio ins, ideally along with some params, live.- all because my computer has one ip address. or sorry, do you just write the local port after the ip address, under the “send to”? or is there a way for vvvv to generate somekind of dummy ip addresses ?.. well that sounds naive…

i know this kind of live audio driving/effecting triggering motion pictures has been around for many years now but i want to try it via vvvv and build a patch for it instead of using flash samples or pre recorded loops. From my architecture background, I intend to present an experience of going to a new city drawn purely with sounds, and this city has monuments that are momental songs and tracks that i grew up with and loved. I am challanged though in vvvv with 3d modelling, bc of the difference btw vvvv and grasshopper ( an extensive plug in for Rhinoceros that is about parametric 3d geometry). Sending different types of geometry - points, lines, polyline, surfaces, polysurfaces also called brep, mesh- from grasshopper to vvvv requires first translating them as a string, then rebuilding them somehow in vvvv. In short, a lot of work. Even rebuilding a mesh just from pointcloud is not so easy nor precise so one needs to import a 3d model file for precision, it seems. Ideally, as i learn more about vvvv and c++, I should try to apply some lessons learned from grasshopper and try to translate or ideally contribute geometry nodes in time, esp the ones that has to do with lines, surfaces and meshes- their embedded information such as color and indices.

Back to the topic, my ableton set has a tendency to push cpu… midi me. of course for a stable live performance, as far as i know, its recommanded to have mostly audio not midis while on stage. Even if its composed in midi, most midi tracks should be bounced as it gets towards the stage. but then what would be the point of being able to send those param changes, or to change params in music in general, as music unfolds via vvvv ? hm, i guess all the audio effects are still in - eq is a big one, stereo eq, esp… argh, it depends, i guess audio effects can be quite cpu heavy as well, esp in the high frequencies or some complicated delays/reverbs which i dont really use.

the fact that livegrabber’s param grabber can set ip addresses is a great thing for an intricately or complexly modulated sounds AS WELL AS a minimal composition (that uses many layers of sound), assuming each ip address has either 1) countlessly available local ports or 2) all the paramters can go to the same IP, just have to slice them in vvvv, granted that the composer does alot of automations-modulations to get his or her sound, instead of using a lot of ready made legal samples. Really trying to narrow down who can benefit here. most of my friends whos been doing live for more than a couple years uses alot of samples and i think most of them will be satisfied with fft analysis and some visuals.

SO, it think livegrabber is the way to go, if not asio. The benefit of figuring out the asio problem is that with some mixers, one can build a digital – analog hybrid set using the sends and returns on a mixer like mine. this can benefit from massive stock of hardwares(not that i have any modules at the moment) that are built and JUST DO great sounds. But i still haven’t found a node that relies on IP address, as the live grabber sends to ip, and no localport (4 digits) to take osc or string from ableton. I will give it a shot one more day and post a patch : )

THANK YOU ALL!!

lastly, is the only way to get waveforms (not frequencies) of a synth directly into vvvv or another to build ones own synth using max/msp ?

p.s. @ tonfilm i found a forum post about wdm here…i havent fully understand whats it all about but i will take another look. https://discourse.vvvv.org/t/783

Hey everybody! I´m new in vvvv and I have a lot of questions but one is killing me, jajajaj. I´m using abletone live 8 a lot and I want send audios and midis to vvvv but I dont´know how, I don’t uderstand it yet. I read this post and undestood that i have to use a program that sends data via red (livegrabber), is this true?
And if it is true, could you show me how? Or showme some tutorial. I’ve been investigating a lot but no answers yet.

I´m interested in show some visuals with real response, with the audio that’s genereted in abletone.
Thanks a lot.

Pd: Sorry for my bad inglish and if you like you can write in español.