Search found 189 matches

by dragontamer5788
Fri Jan 10, 2020 6:45 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

The above Dr. Dobbs journal article references this paper: https://www.researchgate.net/publication/221583842_Superlinear_Speedup_in_Parallel_State-Space_Search When N processors perform depth-first search on disjoint parts of a state space tree to find a solution, the speedup can be superlinear (i...
by dragontamer5788
Thu Jan 09, 2020 9:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

The above Dr. Dobbs journal article references this paper: https://www.researchgate.net/publication/221583842_Superlinear_Speedup_in_Parallel_State-Space_Search When N processors perform depth-first search on disjoint parts of a state space tree to find a solution, the speedup can be superlinear (i....
by dragontamer5788
Thu Jan 09, 2020 8:53 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

Doesn't this indicate that the reductions are flawed? Otherwise, where does the Elo increase come from? If a fatter tree is better then maybe the reductions are incorrect or too extreme. On the other hand, it seems possible that the small extra randomness introduced by the thread local storage migh...
by dragontamer5788
Mon Jan 06, 2020 11:05 pm
Forum: Computer Chess Club: General Topics
Topic: CES 2020: New chips for the new year
Replies: 0
Views: 1185

CES 2020: New chips for the new year

AMD's CES presentation ended about an hour ago, confirming the 64-core Threadripper 3990x is confirmed at $3990 (lol), apparently available in February (!!!!) There are also some laptops and a new NAVI: Radeon 5600 announced. https://www.cnet.com/news/amd-unveils-ryzen-7-4000-series-processors-at-ce...
by dragontamer5788
Mon Jan 06, 2020 10:58 pm
Forum: Computer Chess Club: General Topics
Topic: Nvidia’s next-gen Ampere GPUs
Replies: 7
Views: 1312

Re: Nvidia’s next-gen Ampere GPUs

The news piece comes from Tom's Hardware, which was a well established and respected outfit. Was a respected outfit. But they've posted some questionable articles recently. They're probably still better than a generic website, but I wouldn't trust them to get all the details correct anymore. Thomas...
by dragontamer5788
Mon Jan 06, 2020 4:31 pm
Forum: Computer Chess Club: General Topics
Topic: Nvidia’s next-gen Ampere GPUs
Replies: 7
Views: 1312

Re: Nvidia’s next-gen Ampere GPUs

The news piece comes from Tom's Hardware, which was a well established and respected outfit. Was a respected outfit. But they've posted some questionable articles recently. They're probably still better than a generic website, but I wouldn't trust them to get all the details correct anymore. Thomas...
by dragontamer5788
Sat Jan 04, 2020 7:46 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

Has anyone ever tested SHT-2 threads vs SHT-1 thread at fixed depth like Ronald mentioned? Maybe this is a very easy way to prove/disprove if the tree is "richer" or "wider". We can do the same for YBW as a baseline ... I mean, if you just want to measure "tree richness", you only need to do two th...
by dragontamer5788
Sat Jan 04, 2020 7:12 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

For that reason, you don't see an iterative Crafty since it was MUCH messier than a purely recursive search. I dunno if its "much" messier. Here's the general prototype for an AB Iterative search. struct SavedABState{ Position position; int16_t alpha; int16_t beta; vector<uint16_t> moves; int8_t de...
by dragontamer5788
Fri Jan 03, 2020 9:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

Perhaps now is a good time for me to plug my test methodology? I have invented a "game" for my personal study. It is the simplest possible game I could imagine, so that I can focus on the pure task of finding a better search. 1. A move is represented by a 5-bit number, between 0 and 31. 2. Players ...
by dragontamer5788
Fri Jan 03, 2020 8:57 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: RMO - Randomized Move Order - yet another Lazy SMP derivate
Replies: 114
Views: 22339

Re: RMO - Randomized Move Order - yet another Lazy SMP derivate

Something is missing from the minimization of these: 1. Number of CPU instructions executed 2. Number of memory load/stores executed (in L1, L2, L3 cache, and DDR4 RAM) 3. The number of MESI messages passed between CPU cores. Consider a program that does eval() on the root node, and then waits unti...