In article <1507.2a474167 at mbcl.rutgers.edu> goldman at mbcl.rutgers.edu writes:
>>>I basically concur, except that I don't really think the bottleneck on PCs
>and MACs is the CPU. High-end machines PCs and MACS have fairly respectable
>compute power. What they don't have is I/O bandwidth. So they can't
>really drive their disks that fast, and they certainly can't drive (for
>instance) a stripe set at its rated speed. So that's why I think minis
>will outperform micros in data-intensive searches.
>> Adrian
Amen! Pull out that old MFM controller board and stick in a SCSI and
you're at WARP1. Add it a smart I/O cache board with 10 megs of memory and
your at WARP8. Want to get to WARP10? Ask Michael Abrash to come in
and convert some of your critical code to "his best" assembly language!
Don't know how to do that stuff? Buy a Sparc2!
For $4000 I can dedicate a soupped-up PC
full time to sequence crunching for two years and "afford" to
throw it away when the technology is made "obsolete" by natural
marketplace evolution. Can you afford to chuck your Sparc Station after
2 years? :^)
---
+------------------------------------------------------------------+
| In person: Steve Modena AB4EL |
| On phone: (919) 515-5328 |
| At e-mail: nmodena at unity.ncsu.edu |
|samodena at csemail.cropsci.ncsu.edu |
| [ either email address is read each day ] |
| By snail: Crop Sci Dept, Box 7620, NCSU, Raleigh, NC 27695 |
+------------------------------------------------------------------+
Lighten UP! It's just a computer doing that to you.
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
