Changes

=DetailsFeatures= Zeta is written in [[C]] (host) and [[OpenCL]] (device). The host handles the IO, but all chess related computations are done in via OpenCL (run-time-compilation) on the device.

Zeta v099 is a port of [[Zeta Dva]] to OpenCL, with classic parallel [[Alpha-Beta]] search, therefore 64 gpu -threads are coupled to one worker, used to work on the same node in parallel during move generation, move picking selection and evaluation in parallel.

Zeta v097 and v098 were designed to run thousands of independent [[Thread|threads]] on a [[GPU]], therefore they store the expanded search game tree via an randomized parallel [[Best-FirstMinimax Search]]-MiniMax search in [[Memory|memory]] . =NPS and Playing Strength= * Zeta 099k, AMD Radeon R9 Fury X, 256 workers, ~8 Mnps* Zeta 099k, Nvidia GeForce GTX 750, 16 workers, ~800 Knps, CCRL Elo est. 1932* Zeta 099k, Nvidia GeForce GTX 750, 1 worker, ~45 Knps, CCRL Elo est. 1738* Zeta 098c, Nvidia GeForce GTX 480, 4096 threads, ~5 Mnps, CCRL Elo est. 2000* Zeta 097x, Nvidia GeForce GTX 480, 4096 threads, ~5 Mnps, CCRL Elo est. 1800

* [http://www.talkchess.com/forum/viewtopic.php?t=62288 ZetaOpenCL 098ew64 running on Windows OS40493 LIFO stack based parallel processing?] by [[Srdja Matovic]], [[CCC]], September 22, 2011* [http://www.talkchess.com/forum/viewtopic.php?topic_view=threads&p=442052&t=41853 Possible Search Algorithms for GPUs?] by [[Srdja Matovic]], [[CCC]], January 07, 2012 <ref>[[Yaron Shoham]], [[Sivan Toledo]] ('''2002'''). ''[https://www.sciencedirect.com/science/article/pii/S0004370202001959 Parallel Randomized Best-First Minimax Search]''. [https://en.wikipedia.org/wiki/Artificial_Intelligence_(journal) Artificial Intelligence], Vol. 137, Nos. 1-2</ref> <ref>[[Alberto Maria Segre]], [[Sean Forman]], [[Giovanni Resta]], [[Andrew Wildenberg]] ('''2002'''). ''[https://www.sciencedirect.com/science/article/pii/S000437020200228X Nagging: A Scalable Fault-Tolerant Paradigm for Distributed Search]''. [https://en.wikipedia.org/wiki/Artificial_Intelligence_%28journal%29 Artificial Intelligence], Vol. 140, Nos. 1-2</ref>* [http://www.talkchess.com/forum/viewtopic.php?t=44165 Help with Best-First Select-Formula] by [[Norbert Raimund LeisnerSrdja Matovic]], [[CCC]], November 27June 23, 20162012* [http://www.talkchess.com/forum/viewtopic.php?t=63719 Zeta v099a46974 Kogge Stone, Vector Based] by [[Srdja Matovic]], [[CCC]], January 22, 2013 » [[Kogge-Stone Algorithm]] <ref>[https://en.wikipedia.org/wiki/Parallel_Thread_Execution Parallel Thread Execution from Wikipedia]</ref> <ref>NVIDIA Compute PTX: Parallel Thread Execution, ISA Version 1.4, March 31, 2009, gpu [https://www.nvidia.com/content/CUDA-ptx_isa_1.4.pdf pdf]</ref>* [http://www.talkchess.com/forum/viewtopic.php?t=60386 GPU chess update, local memory...] by [[Srdja Matovic]], [[CCC]], April 11June 06, 20172016* [http://www.talkchess.com/forum3/viewtopic.php?f=27&t=72684 RMO - Randomized Move Order - yet another Lazy SMP derivate] by [[Srdja Matovic]], [[CCC]], December 30, 2019 » [[Lazy SMP]] ==2020 ...==* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=71084 Looking for someone to test 76986 Zeta v099l with NNUE on RTX 2080 TI, or similar, gpuGPU?] by [[Srdja Matovic]], [[CCC]], June 24March 31, 2021 » [[NNUE]], 2019[[GPU]]

* [https://githubgitlab.com/smatovic/Zeta GitHub GitLab - smatovic/Zeta: Experimental chess engine written in OpenCL]

[[Category:GPLMIT]]