Search found 16719 matches

by bob
Wed Jul 18, 2018 8:10 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI Hash Usage Rules
Replies: 58
Views: 1644

Re: UCI Hash Usage Rules

VM usage has never been real accurate. It is quite difficult to determine what is duplicated and what is not. And that answer changes as the program executes and causes copy-on-write duplications. And then there is the shared memory stuff if you use processes. Looks like each process is huge when in...
by bob
Tue Jul 17, 2018 12:57 am
Forum: Computer Chess Club: General Topics
Topic: how will Leela fare at the WCCC?
Replies: 152
Views: 6377

Re: how will Leela fare at the WCCC?

Couple of points... There has been a solution for this. If program A offers a draw ITSELF, that is supposed to be passed on to program B, IF it can process the request. Then no human intervention is possible. The idea is that a program should be able to play a completely legal game of chess all by i...
by bob
Tue Jul 17, 2018 12:40 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI Hash Usage Rules
Replies: 58
Views: 1644

Re: UCI Hash Usage Rules

I don't quite follow what that "threads are more resource heavy under linux." In fact, it is the exact opposite. Threads intentionally share everything, where processes (via fork()) share very little (at least they share executable code and anything that is not modified - an artifact of the "copy o...
by bob
Mon Jul 16, 2018 3:41 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI Hash Usage Rules
Replies: 58
Views: 1644

Re: UCI Hash Usage Rules

I am quite sure that the memory usage per thread is real, but probably it came with the OS, threads are more resource heavy under Linux. This might be one of those cases that sometimes even the author may not know the engine is behaving absurd under certain circumstances. If a user does care much a...
by bob
Sun Jul 15, 2018 3:44 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI Hash Usage Rules
Replies: 58
Views: 1644

Re: UCI Hash Usage Rules

Interesting issue. Since I doubt I will ever do a UCI version of Crafty, I really don't have a dog in this hunt. But I do tend to agree that there are three parts to a chess program that are of interest. (1) basic program + basic blocks of memory required to make the program functional. IE bit board...
by bob
Mon Jul 02, 2018 7:40 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: What am I missing with respect to MTDf
Replies: 6
Views: 777

Re: What am I missing with respect to MTDf

I think you probably have a couple of issues. One is an evaluation that has a lot of terms and a wide scoring range, which makes it very difficult for MTD(f) to stabilize within a couple of tries, so that it becomes more expensive. The second is that the way you implement the convergence is also pre...
by bob
Mon Jul 02, 2018 7:25 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: sscanf question
Replies: 4
Views: 436

Re: sscanf question

I also prefer strtok() as it makes it simple if you want to re-parse something with different delimiters. IE if you notice a "/" you might want to re-parse with that to deal with FEN, where in other cases you might not want "/" to be a delimiter (I use "/" for time control settings for example... An...
by bob
Fri Mar 23, 2018 4:17 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: I'm not very happy with the do {} while() statement in C
Replies: 111
Views: 20603

Re: I'm not very happy with the do {} while() statement in C

Little endian is a retarded design that came due to backward compatibility with the original 8 bit architecture. Many machines have done it right (big endian). Little endian has one important property, fetching smaller integer from the same address. This is the reason why BE sucks and why LE is act...
by bob
Thu Mar 22, 2018 12:21 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: trick question
Replies: 25
Views: 4288

Re: trick question

Hmm, i think they are equally likely :) My line of thought is : a) Each number from 160 - 199 has a probability of 1/40 of hitting exactly 200 on the next polling test. At this point i thought, hey this (1/40) is more likely than 1/200 but nooo... b) We are in this range, i.e.160-199 only 40/200 of...
by bob
Thu Mar 15, 2018 3:22 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Reliable speed comparison: some math required
Replies: 33
Views: 7089

Re: Reliable speed comparison: some math required

You can also use intel's vtune tool and get even more information. IE total clock cycles, total instructions executed, total cache hits/misses, etc... On Linux, perf also gives that information. Another interesting application of perf is identifying (false) sharing, which easily kills performance w...