Author: Robert Hyatt
Date: 11:42:10 11/09/98
Go up one level in this thread
On November 09, 1998 at 14:19:45, Albert Silver wrote: >On November 09, 1998 at 09:11:02, Roberto Waldteufel wrote: > >> >>On November 08, 1998 at 15:51:31, John Wentworth wrote: >> >>>On November 08, 1998 at 14:34:00, Robert Hyatt wrote: >>> >>>>On November 08, 1998 at 12:51:19, Christophe Theron wrote: >>>> >>>>>On November 08, 1998 at 08:13:10, John Wentworth wrote: >>>>> >>>>>>10 to 20 years ago there were a lot of chess programs on mainframes and these >>>>>>were competing in the ACM tournaments. All of a sudden they disappeared, cost >>>>>>and the advances in PCs I'am assuming. However, seems like there must be someone >>>>>>out there programming on a mainframe, I mean they are so superior in speed over >>>>>>the PC's it's laughable. Last I heard Deep Blue no longer existed, and someone >>>>>>was working on Socrates, but you never hear of that anymore either. >>>>> >>>>>Socrates is Don Dailey's baby, who post here often. As far as I know Don is >>>>>still working on an "all platform" multiprocessing chess program called >>>>>"CilkChess". Cilk is a parallel oriented language. CilkChess is written in Cilk, >>>>>and so has the ability to be compiled for uni- or multiprocessor platforms. >>>>> >>>>>Unless I am wrong, Don can produce a PC or mainframe version of his program when >>>>>he wants. >>>>> >>>>>Don, please correct me if I missed something. >>>>> >>>>> >>>>> >>>>>> Also, like to know where all the older programs are now, like Belle, Cray >>>>>>Blitz, Nuchess, BEBE etc. Probably been erased or sitting on a shelf somewhere, >>>>>>just curious if anyone knows. >>>>> >>>>>Cray Blitz was written by Bob Hyatt, who post here more than often. I've heard >>>>>that Cray Blitz has not been erased at all, and has even run some long test >>>>>suite recently (less than one year). >>>>> >>>>>Cray Blitz' successor is the well known freeware program Crafty, which is >>>>>discussed here very often. >>>>> >>>>>Bob, your turn to correct me. :) >>>>> >>>>> >>>>> >>>>> Christophe >>>> >>>>Pretty much correct. Cray blitz still exists... but is not being modified >>>>since Crafty was started. I started the "crafty" project after the 1994 ACM >>>>event in Cape May... machine time is *very* difficult to get, there is little >>>>room for the "unexpected" (such as a weather delay or whatever) since the >>>>machines are so tightly scheduled... >>>> >>>>I wearied of the process of setting up machine time every year, dozens of >>>>phone calls, emails, begging, borrowing, etc... >>>> >>>>The Cray's will still blow off any collection of microprocessor-based machines >>>>you'd care to use, but at $60,000,000 they are expensive and difficult to get >>>>hold of. >>>> >>>>I gave up not because the micros were catching up in speed (which they weren't, >>>>not even close) but because the micros are so much easier to get access to... >>> >>>Come on Bob let's Cray Blitz going again. I would love to see it kick some butt. >>>Let's see, if each CCC member took out a $100,000.00 mortgage maybe we buy our >>>own Cray :) >> >>It would be nice, wouldn't it? But I think I might fall behind on the >>repayments, and I would not want my building society to re-posses a piece of the >>communal Cray! On a more realistic note, If Hsu ever gets permission to market >>his chess chip boards (as used in Deeper Blue), they could boost the speed of a >>chess program immensely. I don't know how a rack of several of these boards, >>driven by a host PC, would compare to a Cray in terms of raw speed, but I am >>sure the price would be more manageable than $60,000,000. The point is that this >>technology was designed for chess from the outset, unlike a conventional >>mainframe that was designed for multi-purpose use and probably has many >>excellent (and expensive) features that are not very useful for chess. I think >>for maybe a few thousand dollars it might be possible to put together a pretty >>awesome chess playing machine using Hsu's technology - it's such a pity that IBM >>hold the rights to the chip, since they have a vested interest in never letting >>it be used. >> >>Best wishes, >>Roberto > >I take it then that the chips are not dependant on the original hardware? I ask >because Bob has often explained that Cray Blitz could only exist on a Cray, >being designed to make use of it's unique architecture. I thought the chips >might be similar except that they are directly hardware instead of software. >Bob? Do you know? > > Albert Silver The DB chess processors are machine independent... they could be interfaced via a PCI card, for example, as well as the VME bus they used in the IBM machine... But they wouldn't help a program until it was rewritten to communicate with the chip and it would be a *complete* redesign of the program from the ground-up to do that... It would fly, of course... since each chip is supposed to run at 24mhz (very low by todays speeds) and search about 2.4M nodes per second per chip... Of course, for those that don't think NPS is important, this is all a moot point. :)
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.