Post Reply 
WP-34S N-Queens Benchmark Based on Patrice Torchet's BITMAP Solution
10-22-2014, 10:36 PM
Post: #19
RE: WP-34S N-Queens Benchmark Based on Patrice Torchet's BITMAP Solution
(10-22-2014 09:15 PM)patrice Wrote:  All RPN models are using the same kind of technology and the impact will be essentially the same. What are you comparing the C/Regvars with ?
With the "structured" C version that's on the benchmark, with no compiler optimizations.

(10-22-2014 09:15 PM)patrice Wrote:  What can be the meaning of a comparison when one of the compared thing can be 40% faster? I think the only thing it show is that both can solve the problem.

I think it still does offer a decent comparison, but as I said previously, since your version uses the same algorithm it should be on the list. I wouldn't put the bitmap solution on the list, but your optimizations seem reasonable and are applicable to other problems.

(10-22-2014 09:15 PM)patrice Wrote:  I think that it is by showing people efficient programing technics that you will teach to use them
If you have the knowledge that lists are slow on RPL, you are free to find another way to solve the problem.
Difference is that I don't think it's a competition, so RPL with lists is perfectly fine with me, and I think it represents how people use RPL. If somebody else wants to optimize it to use the stack, that's fine too.
My observation was not criticism, it went to show that RPN machines are not being unfairly treated, as other models are also in the same position. I'm sure a careful review will reveal that many other platforms/languages could also get a fair amount of optimization, but that doesn't necessarily invalidate the benchmark.

And please notice I have nothing to do with Xerxes, or this benchmark. I never submitted an entry (although somebody else submitted results using my code for the C/Regvars and ARM assembly for the 50g, so you could say I'm the author of those two). I just think its fine the way it is. Perfect? No, but still very useful and informative.

PS: On second thought, if I accept that you are right and this is a competition, would you say I "won"? Should I get at least a T shirt or something for authoring the fastest times on the list, undefeated since 2008? Just joking, of course... (L size, please)
Find all posts by this user
Quote this message in a reply
Post Reply 

Messages In This Thread
IMPROVED 30% over previous post - iceman - 10-15-2014, 06:39 PM
RE: WP-34S N-Queens Benchmark Based on Patrice Torchet's BITMAP Solution - Claudio L. - 10-22-2014 10:36 PM

User(s) browsing this thread: 1 Guest(s)