Hardware Wars Redux

[billroper]

So I continue to fight with the M-Audio ProjectMix I/O that gave me such entertainment at Capricon, trying to figure out how to get it to behave. I've spent a lot of time on the M-Audio forums (not a lot of help, but some) and on Google (ditto). The combination has given me the following things that I've tried:

Turn off autosave in Cubase. This prevents the massive mess that I had at Capricon when the autosave hit and the recording stopped.

Give priority to background processes. Didn't seem to help, but left it that way.

Turn off most of the fancy display doodads. Still didn't seem to help, but left it that way.

Go into the BIOS and turn off the feature that allows the CPU to run at different speeds. I'd gone into the BIOS looking for something else altogether -- which I didn't find. But when I turned this off, I stopped getting random static on the channel with the mic plugged in.

Unfortunately, this still left the occasional random pop that would appear on random channels at random intervals. Ack!

It turns out that many audio devices only work correctly with a Texas Instruments Firewire chipset. So now I've gone out and ordered a TI-based Firewire ExpressCard to plug into the expansion slot.

With any luck, this will fix the problem.

I hope.

Comments

[filkertom.livejournal.com]

You might also ask for help at KVR (http://www.kvraudio.com)'s forums. Lotsa techies there.

[billroper]

I may check it out next week if the current trial solution doesn't work. Thanks!

[kevinnickerson.livejournal.com]

It is apparent that this technology is not yet ready for prime time. g'luk.

[billroper]

Well, the only part that I can critically blame on sloppy technology is the incompatible Firewire. The rest of it probably shouldn't be a great surprise. I mean, there's nothing like varying the clock speed of your processor on the fly to mess up the clocking of your incoming data as you're trying to do an A to D conversion. :)

[kevinnickerson.livejournal.com]

Nothing except the fact that the timing should have all been done in the external device and only final data shipped back. Processor speeds, well effective processor speeds, vary on non-dedicated hardware. That's just one of the design constraints to be dealt with.