Re: Why VAX Was the Ultimate CISC and Not RISC

Lynn Wheeler <lynn@garlic.com> schrieb:

long ago and far away ... comparing pascal to pascal front-end with
pl.8 back-end (3033 is 370 about 4.5MIPS)
>
Date: 8 August 1981, 16:47:28 EDT
To: wheeler
>
the 801 group here has run a program under several different PASCAL
"systems".  The program was about 350 statements and basically
"solved" SOMA (block puzzle..).  Although this is only one test, and
all of the usual caveats apply, I thought the numbers were
interesting...  The numbers given in each case are EXECUTION TIME ONLY
(Virtual on 3033).
>
6m 30 secs               PERQ (with PERQ's Pascal compiler, of course)
4m 55 secs               68000 with PASCAL/PL.8 compiler at OPT 2
0m 21.5 secs             3033 PASCAL/VS with Optimization
0m 10.5 secs             3033 with PASCAL/PL.8 at OPT 0
0m 5.9 secs              3033 with PASCAL/PL.8 at OPT 3

Interesting figures. there is a factor of 50 of 68000 vs the 3033,
which was in the ~1 MIPS range, with the same compiler technology.