How does deep blue chess computer work
HOW DOES DEEP BLUE CHESS COMPUTER WORK SOFTWARE
After generation an appropriate number of childs and grandchilds etc., the software search per node utilize the 16 chess chips to search the final four plies plus quiescence search in hardware. The search occurs in parallel on two levels, one distributed over the IBM RS/6000 SP switching network and the other over the Micro Channel bus inside a workstation node.Ī master workstation node first starts the software search exclusively, to distribute work to all 30 workstation nodes only at a certain depth (i.e. The 1997 Deep Blue system was based on an IBM RS/6000 SP supercomputer, using 30 workstation nodes of PowerPC processors controlling 16 chess chips each, distributed over two Micro Channel boards.Ī chess chip features of a full-fledged chess machine on its own, along with move generator, a smart move stack, hardware evaluation function, andĪn alpha-beta hardware search controller. The deepest nodes in the search tree are handled by the slave search engines which usually do 4-ply alpha-beta searches. When conducting a search, the search tree near the root position is processed on the host workstation, and includes selective search extension algorithms such as singular extensions. (This is about 1/10th of the projected speed of the Deep Blue single-processor currently in fabrication.) The 14-processor Deep Thought 2 typically searches between 3 and 5 million positions per second. Each Deep Thought 2 processor searches about 500,000 positions per second standalone, or about 400,000 positions per second as a slave processor. Each processor contains a VLSI chip for move generation, as well as additional hardware for search and evaluation. Description given in 1995 from the ICGA site :ĭeep Blue Prototype consists of an IBM RS/6000 workstation with 14 chess search engines as slave processors.