🚀 go-pugleaf

Ted Oliver <aqtjo@imap2.asu.edu> wrote:
>
>To be fair, the poster to whom you were responding may have exaggerated
>the performance of PPC in comparison to Pentium. My sense of the consensus
>view was that 601 PPC's and Pentiums were more or less equivalent in speed
>at the same Mhz, possibly with a slight advantage to PPC. As someone noted
>in an earlier post in this thread, at Motorola's web site, you see as much
>difference on the same chip using different compilers as you see between
>processor families.

Well, these are the numbers the Byte benchmarks returned :

Note that FP EMULATION is raw FP, while NEURAL NET and FOURIER use
the transcendental functions heavily.

Byte benchmarks are CPU intensive.  The PowerMac systems do not exhibit
the full strength of the PPC because of non-native OS, etc. -- factors
which enter into the Byte benchmarks only in the math library, ( and 
perhaps in cache management ? )

-arun gupta


Machine : PPC 7100/66, 512K L2, 40 MB memory
System  : System 7.5.3
Compiler: CodeWarrior 7
Compiler Settings :  Optimize for 601, speed.
Source code : available from www.byte.com

Quote :

BYTEmark (tm) Native Mode Benchmark ver. 2 (3/95)
========== ALL STATISTICS ==========
**Thu Mar 21 22:25:03 1996
**Macintosh 7.5
**CodeWarrior
**ver 7
**Sizeof: int:4 short:2 long:4
====================================
NUMERIC SORT:  Iterations/sec.: 26.733226  Index: 0.690870
  Standard Deviation: 0.001896
  Number of runs: 5
  Number of arrays: 1
  Array size: 8111
STRING SORT:  Iterations/sec.: 2.198690  Index: 0.966457
  Standard Deviation: 0.000072
  Number of runs: 5
  Number of arrays: 1
  Array size: 8111
BITFIELD:  Iterations/sec.: 5248998.621381  Index: 0.900369
  Standard Deviation: 4564.523250
  Number of runs: 5
  Operations array size: 30
  Bitfield array size: 32768
FP EMULATION:  Iterations/sec.: 2.539755  Index: 1.221036
  Standard Deviation: 0.000507
  Number of runs: 5
  Number of loops: 1
  Array size: 3000
FOURIER:  Iterations/sec.: 438.982000  Index: 0.497058
  Standard Deviation: 0.011889
  Number of runs: 5
  Number of coefficients: 100
ASSIGNMENT:  Iterations/sec.: 0.291624  Index: 1.111074
  Standard Deviation: 0.000055
  Number of runs: 5
  Number of arrays: 1
IDEA:  Iterations/sec.: 62.549850  Index: 0.957005
  Standard Deviation: 0.001269
  Number of runs: 5
  Array size: 4000
 Number of loops: 100
HUFFMAN:  Iterations/sec.: 28.937733  Index: 0.804161
  Standard Deviation: 0.032782
  Number of runs: 5
  Array size: 5000
  Number of loops: 100
NEURAL NET:  Iterations/sec.: 0.442084  Index: 0.748027
  Standard Deviation: 0.000013
  Number of runs: 5
  Number of loops: 1
LU DECOMPOSITION:  Iterations/sec.: 17.487519  Index: 1.032443
  Standard Deviation: 0.001197
  Number of runs: 5
  Number of arrays: 1
...done...
===========OVERALL============
INTEGER INDEX: 0.935630
FLOATING-POINT INDEX: 0.726772
 (90 MHz Dell Pentium = 1.00)
==============================