Old CP/M programs vs Dedicated Chess Computers

Designed for posting all types of tournaments and Games (e.g. Man vs. Machine, Computer vs. Computer and basement matches.)

Moderators: Harvey Williamson, Watchman

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Old CP/M programs vs Dedicated Chess Computers

Post by TracySMiller »

OK, so I decided to pit some of these old chess programs running on a CP/M emulator against my dedicated computer units. BenRedic was kind enough to make several of these programs available to us. I'm going to start with Dave Kittinger's MyChess (1979 version).

This is going to be pretty informal, because I still have my Affordable Chess Computer Tournament going on. I'm going to start with pairing MyChess against my lower-rated units, two games each (so each program will play White once). Once I get to units that seem to be on close to equal footing, I'll play more than two games to get a better idea of the comparative strength.

Mychess requires some patience and lots of practice to play with. You have to type in the moves accurately, because backspacing or deleting input will quite often screw things up. Also, occasionally the program will just stop accepting legal moves, leading to an abandoned game. But I forge ahead, because I guess I'm geeky like that.
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Mychess (1979) vs. Radio Shack's Companion Chess

Post by TracySMiller »

So first up in the MyChess (1979) challenge is a match vs. the awful Radio Shack Companion Chess. I can't stress enough how horrible this unit is. I often have to restrain myself from throwing it out the door and onto the street, but I don't want to put such horrible trash out there, lest it corrupt the youth of my neighborhood. It has no opening book, the pieces are not magnetic, and the nicest thing I can say about it is it makes a decent paperweight.

I cannot find any reliable documentation of the levels on the Radio Shack unit. It claims to have 64 levels, one for each square. The best I've been able to do is level B8. Some moves it makes in less than 15 seconds, and some moves take considerlly more than 5 minutes. Level C1 seems to take much longer on average to make its moves.

I am using Mychess Level 5. At this level is often moves in just a few seconds, but 4 or 5 times per game it'll hunker down for a 10+ minute think, which is frustrating. But Level 4 moves almost instantly on each move. Also, if you try to choose a Time Control, no matter what you choose, it moves almost instantly.

[fen]5r2/pp2b1Qp/3kp3/8/8/1q2P3/5PPP/2R3K1 w - - 1 30[/fen]

In this first game White (RS Chess Companion) is down a Bishop in this position, but surprisingly may have some chances if he plays 30.Qd4+, but after snatching the h-pawn, it's all over. Game One goes to MyChess (1979).

Here's the first game: https://lichess.org/XT99XM8z
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Mychess defeats RS Chess Companion 2.0-0.0

Post by TracySMiller »

This second game was an old-fashioned butt-whoopin'. In the following position, the lack of opening book and lack of any decent programming allows it to play the horrible ...Nb4. Mychess wins easily 2.0-0.0.

[fen]rnbqkb1r/ppp1pppp/8/3n4/3PP3/8/PP3PPP/RNBQKBNR w KQkq - 0 4[/fen]

I suppose it didn't consider White's response, Qa4+.

Here's the game: https://lichess.org/LHf3HnSF
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Mychess vs. Radio Shack 1650 Fast Response

Post by TracySMiller »

Next up to face Mychess's wrath is the RS 1650 Fast Response. In Game One, the Radio Shack unit unwisely chooses the King's Gambit and is quickly overrun tactically.

Can you quickly spot's Black's best move here?


[fen]r1bqr1k1/pp3ppp/1bp5/5nP1/1Pp2Q2/2P4N/P2P3P/RNBK1R2 w - - 1 15[/fen]

Answer below:
.
.
.
.
.
.
.
.
.
.
.
.
15...Ne3+!


Here's the bloodbath: https://lichess.org/I9P2v5Qc
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Mychess (1979) defeats RS 1650 Fast Response 2.0-0.0

Post by TracySMiller »

Radio Shack 1650 Fast Response plays much better in Game Two, but a few key mistakes cost it dearly. In the following position, although its game would still be tough, Black has to play ...Be8, to be prepared to intervene on the upcoming Queen check.

[fen]3r2k1/3b2pp/2pQp3/8/5P1q/4P3/P2p2PP/1R3RK1 w - - 1 29[/fen]

But after 29...e5, White mops up nicely and defeats RS Fast Response 2.0-0.0.

Here's the game: https://lichess.org/wcipsRFu
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Mychess (1979) vs. RS 1680L

Post by TracySMiller »

On to the next challenge for Mychess (1979), the Radio Shack 1680L, the younger and slightly stronger brother to the 1650 Fast Response. This first game didn't last long though, as Mychess makes it pay for its mistake.

[fen]q3rrk1/B1p1Rppp/p2p4/8/1Q6/1PN4b/P1P2P1P/4R1K1 w - - 1 22[/fen]

Here the RS 1680L plays the unforgivable Nd5??, leading to mate four moves later.

Here's the game: https://lichess.org/NLVTI0qU
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Mychess (1979) defeats RS 1680L 2.0-0.0

Post by TracySMiller »

The RS 1680L makes 9 mistakes and one blunder, but does last 77 moves against the mighty Mychess (1979).

Here's the game: https://lichess.org/QaZBjpiY
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Post by TracySMiller »

I'm going to have to move on to the next old chess program to test, because this 1979 Mychess version is just too buggy. For every game I've successfully completed, I have at least two games (some that went as long as 25-30 moves) where a bug made me abandon the game. It's just too frustrating. I'm going to move on to the 1983 Mychess version (also using the CP/M emulator). Perhaps by 1983 they had ironed out these bugs.

But to summarize, the 1979 Mychess (running at Level 5), defeated the RS Companion 2-0, the RS 1650 Fast Response 2-0, the RS 1680L 2-0, and the RS Partner 1680X 2-0. Then things got interesting. The two games with the RS 1750L Portable ended in a 1-1 tie, so I played two more games, which were both draws, so still tied 2-2. I decided to move on to the RS 1850, which Mychess defeated pretty easily 2-0. Go figure. I'm not sure if the 1750L is stronger than the 1850, or if Mychess just had a tougher time with its particular style. Or too few games to draw a conclusion (the real answer!).

With the 1983 Mychess version I'm going to pursue a smarter strategy and start somewhere in the middle of my dedicated computers strength-wise, and scale up or down as needed. And I'll make things a little more thorough by doing best of 5 matches.
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Post by TracySMiller »

Well, I was able to get in one game with the 1983 CP/M version of Mychess. It was defeated by the Radio Shack 1900L. The next game was about 20 moves into a fairly even game when it bugged out again, putting a piece on the board for no reason. I'm done playing around with the two MyChess CP/M versions. It was fun for a while.

Next I tried Chess.com, which is also called Fortran Chess. It was written in 1979 and only has two levels. In the documentation, the author claims level 0 takes about 90 seconds per move, and level 1 takes several minutes per move. My experience playing it using the CP/M emulator was that Level 0 made moves almost instantly, and Level 1 took 1-2 seconds per move.

I matched it against the Radio Shack 1850. The 1850 trounced it on both levels, easily, with the 1850 set to level 4 (30 seconds/move). I played 3 games like that. Then I put the 1850 down to level 1 (3 seconds per move), and it still won both games it played against Fortran Chess. This Fortran Chess ended abruptly in the first game when checkmated. The window just vanished into thin air. I thought that was funny. But in the next two games, it was facing a mate in 2 or 3 when it ended. I thought maybe the program had some kind of evaluation function that, when faced with certain defeat, would shut down the program.

But then I matched it up with the Radio Shack 1650 Fast Response, which I set down at level 1 (3 seconds per move). In the first game, Fortran Chess was winning with a nice attack when it abruptly ended. I started the game again, and again, in this game was several pawns ahead when it ended itself.

ARRGHHHH, these bugs!
TracySMiller
Full Member
Posts: 865
Joined: Sun Mar 13, 2016 2:24 am
Location: Kingsport, TN

Post by TracySMiller »

So I played a few games with Microchess using the CP/M emulator. It was interesting having to enter moves using that octal code, but not terribly difficult, especially with a board display right there in front of you. And quite often, if you're watching closely, you can see which changes occur on the board when Microchess makes it moves, and tell the appropriate move without having to look at the octal code.

In any case, I played two games with it against the Radio Shack 1650 Fast Response set at Level 0, where it makes a move in one second, and set Microchess at Level N (the strongest of its three levels). Radio Shack won both games easily. Microchess played really weakly, similar to what some programs that have a Handicap mode. In one game, it left its Queen en prise, and in another it overlooked Mate-in-One. Still, it made legal moves, and did not display any bugs in the two games I played against it.

I found another Microchess version online that I'm going to try to see if it's any stronger. This other version you can run straight from the executable, and you can enter your moves in using normal coordinates, although it still displays its own moves using the octal code.
Post Reply