Hi Tomas, it may be a bit late by now, but I'd like to write some comments about your posting last year, concerning bignum performance.
On Thu, Oct 16, 2008 at 11:57:07AM +0100, Tomas Hlavaty wrote: > I was curious to try picolisp bignums and must say that for somebody > doing anything serious, it is probably rather inefficient. > ... > (<= N 100000) > > $ time ~/picolisp/p gmp-test2.l -bye > gmp-test2.log > > real 0m10.190s > user 0m10.157s > sys 0m0.008s Now I tried it here: $ time bin/picolisp gmp-test2.l >/dev/null real 0m11.938s user 0m11.780s sys 0m0.160s As you see, with roughly the same result. However, I am aware of the fact that in picoLisp the printing of numeric results is often much more expensive than the actual calculations, due to the conversions necessary for the decimal base. If I take out the final (prinl Y) from gmp-test2.l, I get: $ time bin/picolisp gmp-test2.l real 0m3.326s user 0m3.200s sys 0m0.130s It makes a considerable difference. > (<= N 1000000) > > $ time ~/picolisp/p gmp-test2.l -bye > gmp-test2.log > C-c C-cKilled > > real 17m58.856s > user 17m51.687s > sys 0m5.572s > > (killed after 18 mins!) Now if I also use 1000000 iterations (but without the printing to save time), I get: $ time bin/picolisp gmp-test2.l real 5m4.904s user 5m4.580s sys 0m0.320s This is still six times as long as the 50 seconds of your C program, but much better than the (unfinished) 18 minutes above. So I believe that for practical uses, where processing does not consist solely of arithmetics, the overhead will be negligible, and not justify extra efforts. If we'd want to optimize it, we should improve the decimal output conversion. Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
