How fast is that machine???

["David J. Littleboy" <Littleboy@YALE.ARPA>]

I  have  run a few timing tests on the various machines around here, including
the 3670 we have on loan from Symbolics.  As expected, the conclusion is  that
lisp  machines  rate an almost unqualified WONDERFUL.  The major qualification
is that closures, which T handles exquisitely, lose big  in  Zetalisp:   While
most  tests  have  the  3670  being  2  to  3  times a TERN, closure intensive
computations are equally fast on TERN and 3670.

THESE NUMBERS ARE NOT THE LAST WORD:  both  T  and  ZETALISP  are  subject  to
change,  with  T more likely to improve than Zetalisp.  Tests one and two will
improve when a disk appears on the TERN.  However, they ARE indicitive.  Also,
theser  tests do not include tests of garbage collection or compilation, which
in my opinion are significantly superior on the 3670.

Tests: 0)  Fibonaci of 20
       0a) Fib with the default value as a GLOBAL
       0b) Fib with the default value as a PARAMETER
       1)  Subset (compute all subsets of a set of 12 elements
       2)  Reverse: normal recursive
       2a) Reverse: perverse  using closures
       3)  ASSOCing ten times down a 10000 element list
       4)  TAK coded straightforward recursive

RATIOS: How much SLOWER the former is than the latter...
   or:  How much FASTER the latter is than the former...

2) COMPILED T (DN400)     vs.     COMPILED T (TERN)
test:   0       0a      0b      1       2       2a      3       4
        3.16    2.92    3.23    0.78    1.4     1.15    3.08    3.27

3) COMPILED T (DN400)     vs.     COMPILED ZETALISP
test:   0       0a      0b      1       2       2a      3       4
        9.21    8.9     8.6     5.375   7.0     1.15    10.9    9.8

4) COMPILED T (TERN)      vs.     COMPILED ZETALISP
test:   0       0a      0b      1       2       2a      3       4
        2.92    2.9     2.5     6.88    5.0     1.0     3.55    3.0


                                                            Love, DJL
-------