herbie@watdcsu.UUCP (Herb Chong [DCS]) (05/30/85)
i haven't been following this too closely, but what is really being
compared, hardware or software? or both at once? i think that all
three are being argued and there so far seems to be no overlap.
what is the determining factor in these benchmarks? is the speed of
the processor, the speed of the floating point processor, the speed of
the mathematical library, the efficiency of the basic interpreter, the
complexity of the interpreter, the speed of the screen I/O routines?
these are all variables in the equation. also, a single program, or
even 4 programs is hardly an adequate benchmark suite. this is much
harder to do than it sounds. benchmarks on mainframes running the
identical software and hardware but with different CPU models is hard
enough. comparing two totally different systems with a small sample
leads to highly unreliable results.
Herb Chong...
I'm user-friendly -- I don't byte, I nybble....
UUCP: {decvax|utzoo|ihnp4|allegra|clyde}!watmath!water!watdcsu!herbie
CSNET: herbie%watdcsu@waterloo.csnet
ARPA: herbie%watdcsu%waterloo.csnet@csnet-relay.arpa
NETNORTH, BITNET, EARN: herbie@watdcs, herbie@watdcsu