blaine@worsel.UUCP (Blaine Gardner) (12/25/89)
Just a followup to my massive postings on drive speed testing. With all the
recent noise of CPU drive controllers claiming to be faster than DMA
controllers, it would be nice if a disk drive benchmark had both "CPU hog"
and "Chip RAM bus hog" modes.
Sure, I can run a CPU intensive program, or I can display a video bandwidth
hungry picture to test these modes, but who actually bothers to do it?
Also, if it were part of the benchmark, then it would be consistant between
all users, a very big plus.
I think real world numbers would be a big improvement over pie-in-the-sky
optimized numbers that drive and controller vendors tend to provide. Sure a
certain controller gets great numbers on the benchmarks, but what happens
when the system is actually being used?
Michael, what would you think about adding something like this to
DiskSpeed?
--
Blaine Gardner @ worsel UUCP: uunet!iconsys!caeco!i-core!worsel!blaine
utah-cs!caeco!i-core!worsel!blaine
UUCP at work: utah-cs!esunix!blgardne