Marco Antoniotti <[EMAIL PROTECTED]> writes:

 > This may be stupid and lead nowhere, but have you tried
 > 
 >      (declaim (ftype (function () double-float) one-normal-rand))
 > 
 >      (defun one-normal-rand () ...)

Isn't (declare (values (double-float))) inside the function supposed 
to have the same effect?

In any case, it doesn't seem to affect run time.  With the fastest
code, and the ftype declaims for one-normal-rand and callpay, run time
for:

   (time (bsmc 100d0 .3d0 1d0 100d0 .1d0 100 100000))

is about 4.8 seconds.  It might be a little faster than without the
ftype, but it's close enough that I'd have to check the compiled
code.  On the other hand, inlining gets me about 3.5 seconds, whereas
maybe-inlining gets me around 3.1 seconds.

-- 
Harvey Stein
Bloomberg LP
[EMAIL PROTECTED]


Reply via email to