Search found 1052 matches

by Ras
Tue Dec 27, 2016 12:15 am
Forum: Computer Chess Club: General Topics
Topic: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Core i7
Replies: 30
Views: 6552

Re: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Cor

The question is, what is the net (!) speedup from 12 to 24 with SF? Is that enough to over-compensate the loss of performance per process? The measure isn't NPS because things are calculated redundantly.

Second, comparing a two-CPU machine against a single CPU machine is misleading at best.
by Ras
Sat Dec 24, 2016 10:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On time management
Replies: 5
Views: 1766

Re: On time management

It depends a lot on TC mode. With classical (moves-per-session) TC there is little to gain. That is the tournament mode, correct. But even there, investing more time early on is what grandmasters are doing. Of course, this is most rewarding in time-per-game mode. Then divide the remaining time by t...
by Ras
Sat Dec 24, 2016 10:40 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On time management
Replies: 5
Views: 1766

Re: On time management

Michael Sherwin wrote:One thing you can try is limiting search depth to no more than one or two ply deeper than the previous depth.
Which seems to pre-pone the problem of how much time to allocate?
by Ras
Sat Dec 24, 2016 10:35 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New idea for "easy move detection"
Replies: 3
Views: 1870

Re: New idea for "easy move detection"

Michael Sherwin wrote:Does not seem (to me) that it would be sound.
It's just how I would play myself, and test games show it works fine. For the actual ID search, there is already pruning in place - however, doing that in the PV usually is too risky without special care.
by Ras
Sat Dec 24, 2016 2:07 am
Forum: Computer Chess Club: General Topics
Topic: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Core i7
Replies: 30
Views: 6552

Re: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Cor

MikeB wrote:or a 7 year old Mac Pro , running Windows 10 under Parallels
That's at 12 threads, and hyperthreading doesn't make sense for chess. How much do you get with hyperthreading switched off at six threads?
by Ras
Sat Dec 24, 2016 2:05 am
Forum: Computer Chess Club: General Topics
Topic: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Core i7
Replies: 30
Views: 6552

Re: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Cor

Milos wrote:Your system seems to be a bit slower, you should see around 11500.
Probably, the ECC RAM has some additional delay - it's more like a workstation that I got here.
by Ras
Sat Dec 24, 2016 12:56 am
Forum: Computer Chess Club: General Topics
Topic: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Core i7
Replies: 30
Views: 6552

Re: AMD's Next-Gen Ryzen CPU Benchmarked Against Intel's Cor

Milos wrote:Considering that now 7 years old 6 core AMD Phenom II X6 1100 @3.8GHz scores 14000nps at Fritz Chess benchmark
I got exactly that CPU in my system, albeit at regular 3.3 GHz where it reports 10800. Even assuming linear frequency/nps scaling, that still would only yield 12400 nps.
by Ras
Sat Dec 24, 2016 12:18 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On time management
Replies: 5
Views: 1766

On time management

One of the most rewarding issues for serious games is time management. At the 2016 dedicated chess computer tournament in Germany, the CT800 suffered from poor time management - for the game-in mode, it only used 1/3 of its total time at move 40. So I had to improve that, which will be part of the u...
by Ras
Fri Dec 23, 2016 11:32 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: move ordering especially ordering&searching of root move
Replies: 18
Views: 4147

Re: move ordering especially ordering&searching of root

1-what is the best ordering method for root moves ? Dunno whether it's the "best", but it works fine: a) order the root move list by MVV-LVA for captures/promotions, some special value for castling and some kind of centralisation/attack logic for the rest. Same as in the ID search. b) if the oppone...
by Ras
Wed Dec 21, 2016 7:35 pm
Forum: Chess Players Forum
Topic: NAKAMURA: Fischer would almost certainly lose to us
Replies: 10
Views: 4492

Re: NAKAMURA: Fischer would almost certainly lose to us

If Fischer were now as old as Nakamura is, then Fischer would have had access to computers as well, so the point is pretty moot. Given that Fischer overplayed the whole Soviet empire all alone, nothing suggests that the situation would have been different if he (and all others) would have had also c...