sagen@nucthy.physics.orst.edu (Milt Sagen) (02/02/89)
A direct quote from the Byte Magazine article describing the new SE/30
...Clearly, this machine is in a whole different league
than the standard Mac SE.
In particular, note that processor-intensive tests, such
as the Sieve, String Move, and floating-point tests, show
the SE/30 sometimes outperforming the standard SE by almost
and order of magnitude. To be fair, I should note that the
standard SE does not have an FPU.
ALMOST AN ORDER OF MAGNITUDE FASTER, ALMOST...a 16 MHz 68030 with a
16 MHz 68882 is NOT EVEN AN ORDER OF MAGNITUDE FASTER than a <12 MHz 68000
WITHOUT a floating point coprocessor? Is this for real?
The floating point benchmarks for the SE/30 and SE where
SE/30 (sec) SE (sec)
math(?) 147.6 891.3
trapezoidal rule: sin(x) 73.3 598.0
trapezoidal rule: exp(x) 97.9 720.7
Milt Sagen Internet: sagen@nucthy.physics.orst.edu
Department of Physics
Oregon State University
Corvallis, OR 97331 Tele: (503) 754-4631jnh@ece-csc.UUCP (Joseph Nathan Hall) (02/02/89)
In article <8628@orstcs.CS.ORST.EDU> sagen@nucthy.physics.orst.edu (Milt Sagen) writes: >ALMOST AN ORDER OF MAGNITUDE FASTER, ALMOST...a 16 MHz 68030 with a >16 MHz 68882 is NOT EVEN AN ORDER OF MAGNITUDE FASTER than a <12 MHz 68000 >WITHOUT a floating point coprocessor? Is this for real? etc., etc... Pipe down. This is the 68030 doing (in)SANE float calculations. Someone ought to post an SE-SANE vs. SE/30-SANE vs. SE/30-68882 benchmark. If it runs as usual the 68882 will run wide open about 100 times faster (doing the FP directly) than through the marginally SANE libraries ... -- v v sssss|| joseph hall || 201-1D Hampton Lee Court v v s s || jnh@ece-csc.ncsu.edu (Internet) || Cary, NC 27511 v sss || the opinions expressed herein are not necessarily those of my -----------|| employer, north carolina state university . . . . . . . . . . .