OK so I've built a simple master beat clock with [phasor~], and I've implemented a swing/shuffle feature using [pow~] so that the groove can be continuously varied (it'll make sense when you open the patch)!
I put it together on my Powerbook G4 and it works like a dream.
I just tried it on my desktop PC (AMD Athlon 64 X2, Win XP) and the timing is completely buggered up. I soon found out that [pow~] was causing the trouble.
I've built in a bypass switch to isolate the problem. Could any kind souls running an AMD machine see if the swing works on their machine? It should be painfully obvious if it's broken on your machine, cos this really ought to be some very tight MSP timing and it sounds completely broken at my end. I'd also be curious to find out whether this affects Pentium-based PCs, or even Core-based MacIntels...
Thanks for looking, yours hopefully,