I think the answer is obvious here but after some trial and error I found that the BEST writing speed I could get was with the following setup:
HDD on Primary Master
CDR on Secondary Master
2410B on Secondary Slave
I could burn 610MB in just under 5min. The problem I found is that once the burner ramps up to 24x, the data flow can’t keep up and the smartburn kicks in all the time as the buffer drains and fills every few seconds. At the beginning two stages of the burn (16x and 20x) the buffer doesn’t drain. All I can think of at this point is that UDMA33 can’t keep up with a 24x burn. I also noticed using NeroSpeed that the best read performance I could get was just under 24x. All things point to an IDE bottleneck, right?
Am I right here or am I overlooking something? AI is off, DMA is on, I’m on a celery 400 w/128MB and a slow 3.1GB HD (yes, I’m still in the stone age).