Slashdot Mirror
Codename Brutus: Chess-Playing FPGA PCI Card
rockville writes "Brutus, a FPGA add-in PCI card developed by ChessBase and Dr. Christian Donnegar, just dominated a strong field of human players at a tournament in Germany. It's the first serious chess-playing FPGA architecture since Deep Blue was disassembled after its victory over Kasparov in 1997. Pictures of the card and a short description are here."
Computers are still very weak at this Asian board game. And despite many people trying to make substantial progress with that. The best open-source one, GNU Go, is btw not very far away from the best commercial ones.
There is a Go Wiki with, among other things, a short introdcution.
It's in the Smithsonian and the hardware is more or less intact. It's in the typical condition of a decommissioned computer, i.e. you can't just flip a switch and start using it, but there's some chance that the folks who built it could get it working again sometime. This is described in the book "Behind Deep Blue: Building the Computer That Defeated the World Chess Champion" by Deep Blue's designer F.-H. Hsu. Hsu later got interested in building a Shogi (Japanese chess) machine using FPGA's. He says with today's custom VLSI, the equivalent of Deep Thought could be built on one chip and mounted in a compact flash card. You'd put the card into your Zaurus or Ipaq PDA and have a grandmaster-strength pocket chess machine. He put some effort into commercializing such a device but couldn't get enough backing so he went off to greener pastures.
VHDL the language may not be specific, but to properly use/take advantage of an FPGA architecure it must be specifically targeted. Writing VHDL for implementation ( and not just simulation) requires targeting the specific final implentation (FPGA, CPLD, VLSI) and even the specific brand/type/etc.
FPGA designs require dynamic processes (so that they can be reprogrammed) and highly regular, repetative, and predicatable implementation. That does NOT make them easier to change process with. Unlike memory/processors that can be reclocked or relabeled, the FPGA needs to be as exact as possible to get proper functionality out of it.
If you want a good "Getting started with Hardware Design" I suggest attending a university for a Computer or Electrical Engineering degree.
Coding VHDL for HW implementation is NOT easy, its not just VHDL->Synthesis->DONE. There is tons of testing and retesting to determine if it synthesized right, if its timed right, if its functional under all inputs/circumstances. Getting a properly simulated and funciontal VHDL design (in the synthesizable VHDL subset that is) is only step ONE of a design. You then have to get a design that is still functional that will synthesize. Then you have to get a design that is still functional that synthesizes that performs correctly.
Its not the kind of thing that you can learn in 21 days from a Sams publishing book.
FunOne
Chess grandmasters do not tune their skills by playing lots of games, how could the best player ever get better? The interesting part only occurs in the middle game. Most GMs learn by reading books and replaying of important games. Its about seeing the pattern and knowing how to counter it. This is exactly the kind of thing computers are good at. They don't have to be taught how to understand the game or how to put to getter some new stratagy, just want to do when something happens. This is why Gary Kasparov was so upset at loosing, the computer had been programmed to recognise all know plays and knew how to counter them. So it simply waited for Gary to make a mistake.
James
You mean the full board positions cannot be encoded in about 32 bytes (12 different figures=4bit * 64 squares = 32 bytes)? And the resulting move cannot be communicated back in less than 3 bytes? (from x=3bit + y=3bit + new_x=3bit + new_y=3bit)
Code poet, espresso fiend, starter upper.
tarquin $> xboard -size small
That will do the trick
No one knows for sure why Deep Blue was disassembled. IBM also fired the entire team that worked on Deep Blue after it won the series against Kasparov. There are two theories:
1. Deep Blue was useless, because it was only a chess-playing machine, and had defeated the world champion of chess (at the time), Kasparov.
2. There is heavy speculation that IBM was cheating with Deep Blue. Some people believe that the machine was also being helped by GMs, behind the scenes, and it was dismantled to destroy the evidence. The reason for this suspicion was the notorious time when Deep Blue hesitated to play a move, and then suddenly changed the move. The Deep Blue team admitted it was not programmed to do this.
You need more than 12 bits to store a move-- en passant captures, castling and pawn promotions have to be handled seperately.
Oh my god, do you ever suck
No, and stop coming on to me you fag
You can't take the sky from me...
Deep blue WAS helped by some very good chess players both as coaches and as members of the Deep Blue team. Basically the match was Kasparov vs a bunch of way above average chess players and one really honkin fast chess computer as an aid. And Deep Blue was never dismantled, half of it is at the Smithsonian intact, the reason they don't have both towers is they simply didn't have room for it all.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.