coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: Marc Bezem <Marc.Bezem AT phil.ruu.nl>
- To: coq AT pauillac.inria.fr
- Cc: Dimitri.Hendriks AT phil.ruu.nl (Dimitri Hendriks)
- Subject: reductions
- Date: Wed, 8 Apr 1998 15:41:18 +0200 (MET DST)
Hallo,
How come that sometimes Simpl takes over a minute, whereas Compute
is almost instantaneous?
For example:
1 subgoal
em : EM
ac : AC
d : D
============================
(E'' (CLAUS (f_not f_PP)))
PP < Simpl.
1 subgoal
em : EM
ac : AC
d : D
============================
(f:SKF)
((x:D)(A (pub x)))
->((x:D)(A (drunk x)))
->((x:D)(A (pub (f O (ml D x)))))
->((x:D)~(A (drunk (f O (ml D x)))))
->((x:D)(A (pub (f (S O) (ml D x)))))->False
takes almost 100 sec.
PP < Compute (E'' (CLAUS (f_not f_PP))).
= (f:nat->(list D)->D)
((x:D)(A (pub x)))
->((x:D)(A (drunk x)))
->((x:D)(A (pub (f O (cons D x (nil D))))))
->((x:D)(A (drunk (f O (cons D x (nil D)))))->False)
->((x:D)(A (pub (f (S O) (cons D x (nil D))))))
->False
: Prop
takes 10 sec. AND reduces more (e.g. SKF = nat->(list D)->D)).
By the way, we like result of Simple better, but Simple can be expected
to be too slow on larger examples.
Many thanks, Marc Bezem.
- reductions, Marc Bezem
- Re: reductions,
Patrick Loiseleur
- Re: reductions, Bruno Barras
- Re: reductions,
Patrick Loiseleur
Archive powered by MhonArc 2.6.16.