There's no real answer to that question, sorry. It totally depends on factors such as the task you want them to do, and on the machine you run it on. (When you say "faster" I assume you mean "most efficient" or sth like that, by the way. Latencies between audio in and out, or between MIDI keypress and out, are all basically determined by the subsystem (jack) and so there's no difference.) For example, you could easily run a benchmark to decide (on a particular machine) which of the two could play the most sinewaves at once without glitching (assuming you've set them to the same latency etc etc). Or which of the two could granulate the most audio files at once. Or which could add the most delay taps to an incoming audio signal. But the results of each of those would (most likely) be different, and they'd not tell you that much about the actual music-making situation. I don't use csound much. But knowing that supercollider was designed from the ground up for real-time, while csound wasn't, I'd be surprised if it was generally "faster". But the more important question is which one suits your needs better. Dan 2009/8/27 cunnilinux himself <cunnilinux(a)gmail.com>om>: -- http://www.mcld.co.uk

cunnilinux himself wrote: Like Dan said, it depends on what you're running, in hardware and software. I've used both systems in realtime, each performs very well in Linux with an rt-enabled kernel. Which one you prefer really depends on other factors, e.g. language type, available opcodes, GUI, etc. I've been using Csound since the late 1980s, I'm a little biased towards it. :) Best, dp