Author: Robert Hyatt
Date: 08:03:44 08/10/00
Go up one level in this thread
On August 10, 2000 at 02:02:27, Dave Gomboc wrote: >On August 10, 2000 at 02:00:41, Dave Gomboc wrote: > >>On August 09, 2000 at 16:10:50, Christophe Theron wrote: >> >>>On August 09, 2000 at 05:31:52, Andrew Williams wrote: >>> >>>>On August 08, 2000 at 16:00:32, Christophe Theron wrote: >>>> >>>>>On August 07, 2000 at 05:58:44, Andrew Williams wrote: >>>>> >>>>>>On August 06, 2000 at 20:10:49, Vincent Diepeveen wrote: >>>>>> >>>>>>>On August 06, 2000 at 19:17:18, Tom Kerrigan wrote: >>>>>>> >>>>>>>>On August 06, 2000 at 16:37:24, Vincent Diepeveen wrote: >>>>>>>> >>>>>>>>>On August 06, 2000 at 12:45:11, Dan Andersson wrote: >>>>>>>>> >>>>>>>>>>Vincent has had this idea of MTD and never managed/bothered to defend it. I >>>>>>>>>>believe it to be an unsupported opinion. >>>>>>>>> >>>>>>>>>No commercial program uses MTD. End of proof man. >>>>>>>> >>>>>>>>I thought the MP version of Fritz does. >>>>>>>> >>>>>>>>-Tom >>>>>>> >>>>>>>I never saw any MP version of Fritz in the shops so far, >>>>>>>perhaps someone is gonna state soon that DB used MTD too. >>>>>> >>>>>>Oddly enough, this seems to be what Hsu says in his IEEE Micro article. >>>>>>Unfortunately, he doesn't say quite enough to be clear: >>>>>> >>>>>> "The search control does not really implement the regular >>>>>> alpha-beta search algorithm [Ref: Knuth & Moore 1975]. Rather, >>>>>> it implements a minimum-window alpha-beta search algorithm >>>>>> [Ref: Pearl 1984]" >>>>> >>>>> >>>>> >>>>>The reference "Pearl 1984" clearly indicates that they are using PVS/Negascout. >>>>> >>>>> Christophe >>>>> >>>> >>>>I'm not certain that this is true. This is a reference to: "Heuristics : >>>>intelligent search strategies for computer problem solving". Addison-Wesley, >>>>1984. I'm pretty sure that this book is where he first talked about the use >>>>of MT (Memory Test) as a way investigating the performance characteristics of >>>>alpha-beta searches. >>>> >>>>Andrew >>> >>> >>> >>>I have this book right in front of me (actually the french translation of it). >>> >>>In this book, Pearl talks about SCOUT (which is also called NegaScout or PVS), >>>and about SSS*. >>> >>>SSS* is not a practical choice, as you need to store the entire search tree in >>>memory. >>> >>>MTD(f) did not exist at the time the book has been written. >>> >>>So the only choice left is SCOUT (NegaScout/PVS). >>> >>>Maybe you are mixing SCOUT and "Memory Test" because the heart of the SCOUT >>>algorithm is a procedure called "TEST". But TEST is actually just a null-window >>>search in this case. >>> >>> >>> >>> Christophe >> >>But that's what Memory-based Test Driver is... a bunch of null-window searches! >>It is directly based on Pearl's "TEST". >> >>What is important is whether mtd() existed at the time of the IEEE Micro article >>or not. If it did not, then it's clear that SCOUT is referred to. However, I >>think it did, so I am not convinced one way or the other! >> >>Dave > >Now having read Bob's post, I am convinced that it indeed refers to SCOUT in one >form or another. :-) > >Dave I'm sure mtd existed prior to the IEEE article. Don Daily and the guys at MIT were doing this at least 3-4 years ago. I did some testing over 2 years ago and gave my opinion on how it worked for me. So it was around. But with DB's hardware, I don't see how it would work very efficiently. And with their eval which produces some big swings from iteration to iteration, it would seem to fail miserably there. And finally, mtd + singular extensions seems to be a very problematic mix of algorithms...
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.