
"Every month or so a debate erupts on the analog heaven (and perhaps elsewhere) regarding personal computers and MIDI timing accuracy. The cycle is always the same: someone experiences some real or imagined timing inaccuracy using DAW software and concludes that all personal computers do not have acceptable MIDI timing.
Instead of propagating memes which may or may not be true, I suggest people test the timing accuracy of their system themselves. Like anything else with audio, it is all about signal flow."
BTW, while you are there, be sure to check out his Dissociative Fugue video. Really great stuff.
breaking news: guy who works for MOTU (?) says DAW midi timing is good!
ReplyDeleteActually, anonymous, I provide instructions on how to test MIDI timing accuracy for yourself. MOTU is never mentioned.
ReplyDeleteyeah that's exactly what i had the problem with!
ReplyDeleteyawn, nothing new here.
ReplyDeletebasically posits the solution:
don't use midi
or invest 3,000 in g5+logic+emagic midi convertor.
then suggests clumsy workarounds.
the fact is that the industry has too much inertia to change, and no inclination to revise the cv/gate format even though clearly that is the way to go.
CV and gates, though dear to my heart, don't deal with polyphony all that well. While it is a solution for some, I wouldn't characterize it as 'clearly' the way to go.
ReplyDeleteI'm a big fan of the OpenSound Control protocol. I'd love to see more devices embrace it.
i just wish the workaround, you know, worked. i mean, you can use DP to quantize a bunch of arbitrarily-detected beat markers sample-accurately, but there's no way the beat markers were placed in the contiguous file sample-accurately in the first place. i guess if you're lucky some of the slop cancels out, but it's not exactly a solution.
ReplyDeleteto say nothing of the headache you then have on your hands from that point on managing fades in DP or PT.
Haven't tried it myself, but if you have midi timing problems you may start trying miditest
ReplyDeletehttp://earthvegaconnection.com/evc/products/miditest/index.html
Their approach seems much more straightforward to me, just connect a standard midi cable from in to out of your midi interface and run the program.
The test procedure described in Stretta's article, i.e. soldering a 1/4" plug to a midi cable and connecting it to an audio input, makes me a little bit uncomfortable.
I really like the Dissociative Fugue video (nice music too) and would love to know what software / tools / programming languages were used to create it.
ReplyDelete