Search found 692 matches

by rbarreira
Tue Jan 07, 2014 7:12 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: How Do You Automatically Tune Your Evaluation Tables
Replies: 116
Views: 48517

Re: How Do You Automatically Tune Your Evaluation Tables

One idea I've tried, with mixed success, is to CLOP only selected squares and then perform a linear or quadratic interpolation of the squares between the selected squares. Essentially, establishing a gradient field. This has kind of worked, but I'm sure there must be a better way. I'm curious, what...
by rbarreira
Mon Jan 06, 2014 5:57 pm
Forum: Computer Chess Club: General Topics
Topic: Quantum Computing In Five to Ten Years
Replies: 4
Views: 1115

Re: Quantum Computing In Five to Ten Years

I only know a little bit about quantum computing, so I cannot make definitive statements, but I think Grover's algorithm might be useful for chess: http://en.wikipedia.org/wiki/Grover%27s_algorithm My understanding of Quantum Computing as it pertains to computer chess is that instead of seeing one m...
by rbarreira
Fri Jan 03, 2014 4:41 pm
Forum: Computer Chess Club: General Topics
Topic: Fern & Wael vs giant Smartphones from Outer-Space
Replies: 16
Views: 7242

Re: Don Dailey, 1956 -2013

fern wrote:Just to keep the thread at least over the stupid cellphone benchmark
Don does not need that to be remembered.
by rbarreira
Tue Dec 17, 2013 10:59 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Testing on time control versus nodes | ply
Replies: 28
Views: 7490

Re: Testing on time control versus nodes | ply

I completely disagree with your premise that "repeatable testing is good testing". You want as much randomness as possible in any AB test. Then you select a number of games (in advance), run the test and take any conclusions when all the games have run. Anything else is lying to yourself with stati...
by rbarreira
Tue Dec 17, 2013 10:54 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Testing on time control versus nodes | ply
Replies: 28
Views: 7490

Re: Testing on time control versus nodes | ply

I completely disagree with your premise that "repeatable testing is good testing". You want as much randomness as possible in any AB test. Then you select a number of games (in advance), run the test and take any conclusions when all the games have run. Anything else is lying to yourself with stati...
by rbarreira
Tue Dec 17, 2013 2:45 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Testing on time control versus nodes | ply
Replies: 28
Views: 7490

Re: Testing on time control versus nodes | ply

I completely disagree with your premise that "repeatable testing is good testing". You want as much randomness as possible in any AB test. Then you select a number of games (in advance), run the test and take any conclusions when all the games have run. Anything else is lying to yourself with statis...
by rbarreira
Fri Dec 13, 2013 11:03 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: How could a compiler break the lockless hashing method?
Replies: 312
Views: 33757

Re: How could a compiler break the lockless hashing method?

I suppose that next you are going to tell me you want some of the "mistake-prone" things removed from assembly language so that you can't write code there with overflow, etc? No, I am quite consistent in that aspect - I want both C and x86 Assembly to remain backwards compatible. What made you thin...
by rbarreira
Fri Dec 13, 2013 5:24 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: About pure programming threads
Replies: 20
Views: 3858

Re: About pure programming threads

There is another kind of fun idea I saw elsewhere. They have a special place (section) on the forum called "The Arena". When person A and B have an ongoing dispute with no end in sight person A may challenge person B to the Arena and vice versa. In the Arena (setup and controlled by a mod) only A a...
by rbarreira
Fri Dec 13, 2013 7:39 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: How could a compiler break the lockless hashing method?
Replies: 312
Views: 33757

Re: How could a compiler break the lockless hashing method?

So it looks like MOST of the long-term C programmers have quite a different view from those of you that are apparently a product of the Java abstract-machine crowd. What kind of language would target assembly language programming? One that lets you get to the hardware to do the things the hardware ...
by rbarreira
Thu Dec 12, 2013 7:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: About pure programming threads
Replies: 20
Views: 3858

Re: About pure programming threads

I think the opponents of the strcpy debate don't mind the topic itself but the length. A suggestion to the mod-team might be that the moment they feel such a thread is running in circles they invite each of the key debaters a one more time all-inclusive post that explains their position in detail, ...