Author: Dann Corbit
Date: 17:18:17 01/21/99
Go up one level in this thread
On January 21, 1999 at 19:16:05, Albert Silver wrote: >On January 21, 1999 at 18:39:00, Dann Corbit wrote: > >>The OrAnG UtAn project has been completed. About 130,000 EPD rows were >>analyzed at 12 minutes of PII 300Mhz CPU equivalent time two or more times. >>This basically includes every 1. b4 game played in a public tournament and >>many others besides. In three months, the data will become public domain. > >Interesting, though I wonder if this will actually rehabilitate this opening. >What are the Apocalypse and Heartwood project? Just as likely, it will put it to death, as yet, I have no idea. It does have a loyal following: http://www.algonet.se/~marek/ There is a group that plays nothing but 1. b4, I guess. "Apocalypse" means 'a revealing' and we are making an examination of 7700 positions from standard EPD tests like ECM, MES, etc. Each position is analyzed at 12 Minutes of PII 300 Mhz equivalent CPU time. Then we do it again, on a different machine to ensure accuracy. If something looks interesting, we might pound it at 8 hours. It might sound tough, but if we devoted the full energy of all members towards that project it would be completed in one day. For project heartwood, we take all the positions from our game database (commercial + ameteur collection) and count how often the positions occur. For the positions that occur the most frequently, we analyze them. We are starting with positions that have occurred at least 200 times in actual games. We have about 67,000 distinct positions that fulfill this requirement. Then positions that have occurred at least 100 times. We have about 100,000 of them. Then 50 times, etc. We keep going until we get sick of it, or that phase of the project ceases to become interesting. When you think about it, a very frequently occurring position is likely to be played against you. By having those frequently occurring analyzed in great detail, we are prepared for the most probable assaults and defenses. The C.A.P. FAQ explains all of this. Why not give it a read? [snip] >>Read the C.A.P. FAQ: >>ftp://38.168.214.175/pub/Chess%20Analysis%20Project%20FAQ.htm >>-- >>C.A.P. Newsgroup http://www.dejanews.com/~c_a_p >>Chess Data: ftp://38.168.214.175/pub/
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.