What got you into computer chess?

[Leo]

It was always nice to play chess whenever you wanted and at different skill levels. You could have hours of entertainment. I got my first set in 1987.

Oh yea. I think it was 1981 my friend got Atari chess which was a joke even back then.

[phhnguyen]

I was teaching about AI in a university. I knew some basic computer chess knowledge such as AlphaBeta algorithm as well as the basic design of a chess program and could transfer smoothly those knowledge to my students. The problem I have never written any chess program. Thus I worked with a talent student group to build our first chess (Xiangqi) program (about end of 1990s). The program worked well even it was so weak. After that those students moved to other projects and then graduated. But I have stuck by chess programming from that time

[pilgrimdan]

Chess Challenger 7 …

[Angrim]

Back in college, my CS teacher made an Othello playing engine and left it running in the computer lab. I don't remember if it had alpha-beta, but at least it had minmax. That motivated me to write my own, and a few weeks later mine could beat his reliably, then I made a version that played checkers, then ported it to C(the original was in Turbo Pascal) and then I added Chess and Suicide Chess to it's collection of games that it could play. I never really polished the chess version, but the ones that played variants got to be pretty good.

[MikeB]

Not this one , but a similar black and white ad in the Wall Street Journal, 1977. Note the letters to label the rows.

It took me about a year to beat it regularly, I then tried the chess club and promptly got my head handed to me game after game. After a few years l (1981) I ventured to New Paltz, New York, for the World Open, watched a skinny 17 year old kid tie for 1st in the open section. Joel Benjamin would later earned his GM title.

[mclane]

I learned to play chess when I saw two other people playing an interesting looking game but yet unknown to me in 1969.
They said they were playing chess and one of them said he couldn't teach me because I was too dumb to learn chess.
We were in the elementary school library, so I read the chess sections in Compton's Encyclopedia, Collier's Encyclopedia and Encyclopedia Britanica.
After that I played chess weekly from about a month or two until the teachers took the chess sets away. They didn't like us playing chess.

In the early 1980s in undergrad, I had an AI class which was fascinating. Normal programming and programming problems were boringly easy but AI
required you to really turn your head upside down to get an elegant solution. At that time, I was reading the game code of several games that
came with the universities PDP-11/34 minicomputer. That stuff was cool. Also, I got into playing chess again after undergrad in the late 1980's
and humans could be the best computers - CXG Super Enterprise - I bought one then and it still works, Fidelity Excellence....
Also, BattleChess and ChessMaster for the Nintendo along with Sargon for the PC. That put the icing on the cake. Why could humans
beat computers? Were humans really searching that deep? So, I started reading chess books on strategy and positional chess. I did this for
a decade and read dozens of books.

While in grad school, I studied all the required classes OS, Compilers, Computability Theory/Algorithms and CPU design. Also, I took classes in other areas like Neural Networks. I immediately thought of Neural Nets as a position eval and other things for chess programs. I did some
experimental work and it was a bust. My 25 MHz 386 SX with a Cyrix math coprocessor was not fast enough to ever be effective. So that was abandoned.

When I wrote Noonian in the 1990s it had a huge position evaluator which caused it to be very slow. It played the opening and early middlegame
very well but would loose to speed after that.

Since then, I still spend most of my computer chess programming time working on the position eval. That is where the real AI is. That is where you
give the program a personality and style.

Yes but the big progress in hardware has changed the world of computer chess. Shannon a, b, c made sense in early days. And to be honest, I never liked a, but was fascinated by b or c.
Only a few programs tried b. Most programs were a mixture of a and b and with the hardware progress they suddenly were capable to beat humans. But do they really play chess ? I would say NO.
They calculate chess.

The b Strategy programs were only a few.
Mainly the engine in sci sys mkv (6502) and successors of it.
Isn't cxg sphinx (68000 CPU) one of them ??

But those programs were beaten by the ab programs.
Only a few programmers tried Shannon c.
I would see Mephisto III and it's 68000 version III S
in this category. An engine that computes 1-3 NPS ( the 68000 version maybe 3-6 NPS).

With the fast pcs everything changed.

Suddenly knowledge was not important anymore but depth.
Nullmove came. And the slow knowledged programs were hopelessly beaten.


Today we have Komodo, stockfish, Houdini on top. Even
The new approach of mcs and neural networks.

I wonder why nobody tries to build b and c programs. Just for fun. Just for the evidence to proof that
It is possible.
That there is a difference between calculating a move like a machine is doing and the way humans do it, and create software that emulates how humans think.


The hardware and ram is amazingly fast today.

So the resources are there.

The b program should only follow the few moves "that make sense". While the c strategy programs should really create a PLAN !
This is how humans do it.

And why is nobody trying out those things??

There is nothing to lose.
The race against stockfish or Komodo or Houdini or lczero or humans is over.
Now it is about AI.

[Vinvin]

What got you into computer chess? Battle Chess, a love of programming, and no chess set.

My elder brother bought a "Tandy Computerized Chess Game" around 1980 : https://www.youtube.com/watch?v=nZvGSv9awZM . I played a lot of games against it. Rising the level again and again.
Then "Video Chess" for the Atari 2600 console around 1982 : https://www.youtube.com/watch?v=wdpRDYQgm84
Then "Colossus 4.0" for the Commodore 64 around 1985 : https://www.youtube.com/watch?v=0ZTbjXMghso
Then "Chessmaster 2000" on the Amiga : https://www.youtube.com/watch?v=u_KzZIu2dqs
Still on the Amiga, "Chess Champion 2175" : https://www.youtube.com/watch?v=rIRk8JqwXG4
Next computer, a 486 PC, first chess program : Mchess Pro 3.x : https://www.youtube.com/watch?v=Qw2WyKuK6ck
Still on PC for a long time : ChessGenius, HIARCS, Rebel and Fritz.

[tsoj]

I wanted to learn programming and i thought a chess AI would bei a nice project. Just some kind of minimax and some ASCII figures to move around. Yesterday i took a look at the code again and the spaghetti factor is 10000.

[Ajedrecista]

Hello:

I would not know exactly. I think that my first chess software was ChessMaster 9000 back in 2005 or 2006. I started to read some chess forums like Rybka Forum, Immortal223 and TalkChess, following news about Ippolit, Robbolito, Ivanhoe, Houdini, Critter...

I discovered perft by chance around 2011 or so. I really liked estimation of perft values (see Perft estimation for higher plies and Perft ratios in this web). Then I discovered a fast perft counter (JetChess), then Steven Edwards' pool of guessing Perft(13) while running his perft programme on that value, so I registered myself in TalkChess to write my estimate and stayed there until now.

So I guess that ChessMaster personalities were my start in computer chess and perft was my kick-off in TalkChess.

Read David Levy’s book on Computers and Chess in 1977 and was inspired to write an engine in Fortran, typed onto punched cards, to run on an IBM 370, 5 seconds of compute per move.

(OFF TOPIC). Fortran... I have a long time question but I do not want to hijack the thread: do someone know which is the current strongest engine written in Fortran, no matter of what Fortran version? Maybe Cray Blitz? (I read that it was not fully written in Fortran: 'Cray Blitz was written in Fortran, time-critical parts in the Cray Assembly Language CAL'). Thanks in advance.

Regards from Spain.

Ajedrecista.

[Leo]

252045-video-chess-atari-8-bit-screenshot-checkmate-i-lose.tif

Atari Chess. 1979. Checkmate. You lose.