The future of computer chess

[Uri Blass]

The point was: is the huge progress in computerchess part of hardware or software development.

With the 16 bit cpus the adress range was much bigger.
Suddenly not only 32 or with bank switching 96 kb was usable. But lot of memory more.

Suddenly we were in megabyte range.
That made hash tables possibe.
The engine could grow and use heavier evaluations and lot of new stuff.

So was this a feature of software or more a feature of hardware change from 8 bit adress room to 16 bit addresses ?

With the new 68000 cpu richard lang was capable to program a very strong engine in assembler.
And with the use of 68020 or 68030 and the increase of the mhz from 12 to 33 or 66 or even 133 mhz
the software was made faster.


The next big step was that the pcs made progress from a few mhz into much more mhz.
We remember 8086, 286, 386, 486-33, 486-66 and more. Then came pentiums much more speed.
And one day we had cpus with 2 cores, then with 4 cores and the race continues,

Today we have very fast cpus with lot memory, high speed and several cores,


So how big was the software progress that was not due to big hardware progress ?

Can somebody create a 6502 engine that beats ed schroeder , dave kittinger ?

People have no motivation to do something that nobody or almost nobody is going to use.

If there is a big prize money for the best 6502 engine in one year from the announcement (let say some rich person donate a prize of a million dollar for it) and nobody get something that is significantly stronger than what we have now then you have some evidence that it is really hard to get a significant improvement with the knowledge of today but today we have no evidence for this claim.

It is possible that only changing weights in the evaluation of what Ed did at that time may give 100 elo improvement and at that time it was hard to test but today I guess that with some work we can emulate Ed's engine in new hardware except being something like 1000 times faster that mean that with new hardware we can test changing weights 1000 times faster relative to what Ed could.

[Viz]

I mean pure search tuning in stockfish - aka just constants changes without any other code changes was 15 elo or so when first implemented somewhat properly.
And this is at times of no less than smth like stockfish 14 IIRC, which was tuned and retuned at smaller sets of data, yet it was so proficient.
Heck, you can even look there - https://github.com/official-stockfish/S ... XInTheDark this is more or less constant gains with nothing but numbers optimization for longer time controls (while regressing at shorter ones).
Sure w/e was present in 1960 had much fewer parameters but let me tell you if they were anywhere near close to optimum it would've been a miracle. Modern optimization techniques can squeeze a lot of elo without even adding any code more or less.

[towforce]

People have no motivation to do something that nobody or almost nobody is going to use.

This.

I am sure that if the resource were invested in it, better engines could be made for 1970s 8-bit CPUs (even if just by tuning) - but nobody is going to make that investment today.

Regarding address space: I don't have expert knowledge on the subject, but here's some simple maths:

2^8 = 256 (so 1 byte can address 256 different memory locations). That's not enough for a good chess program (you'd do well to run ANY chess program in 256 bytes for both program and memory!).

2^16 = 65536 (= "64k")

So 16 bit addressing would have been used. So fetching a memory address from memory (which, in the 1970s and early 1980s, was on a different chip!) would have entailed locking the computer bus for the CPU's usage, then taking multiple clock/bus cycles to fetch the contents of the memory at the 2-byte address!

If your computer had a 16-bit CPU and a 16 bit bus, I imagine that the process of fetching data (or an instruction) from a 16 bit address would take far fewer clock cycles. CPUs improved in many other ways as well.

The 8-bit CPUs were an absolute revolution: they enabled "ordinary people" to have their own computers - which was phenomenal! But these "ordinary people" weren't getting supercomputers.

[mclane]

Thats why the old 8 bit homecomputers have 64 kb memory.
2 to the power of 16 was also in our CD players.

All cds you have use only 65536 levels for the sound.
From the most silent situation in a concert to the loudest in a rock n roll concert. Only 65536 levels of how loud or silent something is.

The human ear is much better then 65536 . Thats why todays music lovers use flac24 files.

The the most silent to the most loudest sound has 16.777.216 levels and comes closer to the human ear.
2 working human ears can find out if music is 65536 or 16.7 million levels .

I can identify the difference because when it is flac24 the hairs in my neck go upwards and the emotion in the music is big, like in a live concert while flac16 sound is technically perfect but almost without emotion.


The 16 bit homecomputers like amiga500 or atari ST had only 10 mhz cpu but megabytes of ram instead of 64 kb.
This made it possible to use different data structures in chess engines.

[jefk]

with sufficiently long time controls, balanced openings will only lead to draws for the strongest
engines. Even unbalanced (but not biased) openings will lead more and more to draws.

That's why in the higher leagues of TCEC some 'advantageous' (biased) lines are chosen;
then ofcourse a specific engine should play such line with both colors.

https://wiki.chessdom.org/Openings_FAQ

[mclane]

I did a little tournament between very old engines.
Mainly home computer engines or early dedicated machines . The PC engines run on IBM XT with 350 cycles.

There is even a 4 bit engine.

The only sad thing so far is that the TI 99/4A engine Video Chess is so weak.
Considering that the TI is a 16 Bit CPU, the results are lousy.

Code: Select all

 Code:
    Motor                                                       Punkte CoCyMeGaSfSaChFiPhGrMaCoLoAtMyScScPsCXVi    S-B
01: Conchess 2 mhz {Brettcomputer}                              4,5/5  ·     =   1 1       1         1           11,75
02: Cyrus Dragon Chess {Brettcomputer}                          4,5/5    ·           1 1 =         1     1        9,75
03: Mephisto ESB II {Brettcomputer}                             4,0/5      ·   1   1   0         1       1       10,00
04: Gambiet81 TRS80 {Brettcomputer}                             3,5/5  =     ·         =   =     1     1     ?    8,25
05: Sfinks4 TRS-80 {Brettcomputer}                              3,5/5      0   · =           1     1   1          6,50
06: Sargon 2.5 Applied Concepts / Chafitz {Brettcomputer}       3,0/5  0       = ·     1 =         1              7,00
07: CheckCheck 1.75 XT 350 cycles                               3,0/5  0   0       ·     1   1             1      6,00
08: Fidelity Chess Challenger Grandmaster Voice {Brettcomputer} 3,0/5    0           ·             0   1   1 1    3,00
09: PhilidorChess XT {Brettcomputer}                            2,5/5    0 1 =   0     ·   1                      8,25
10: Grandmaster Chess C64 {Brettcomputer}                       2,5/5    =       = 0     ·     1     =            6,50
11: Mattel Chess Aquarius {Brettcomputer}                       2,5/5  0     =         0   · 1               1    3,75
12: Conic Computer Chess Modell 7012 {Brettcomputer}            2,0/5          0   0       0 ·       1     1      3,00
12: Logichess2.2 Lynx {Brettcomputer}                           2,0/5                    0     ·     0 = 1 =      3,00
14: Atari Video Chess VCS {Brettcomputer}                       2,0/5      0 0                   · =     =   1    1,50
15: Mychess DOS IBM XT 350 cycles {Brettcomputer}               1,5/5    0     0 0   1           = ·              4,00
16: Scisys Sensator Chess {Brettcomputer}                       1,5/5  0                 =   0 1     · 0          3,25
17: SciSys Chess Champion MKIV {Brettcomputer}                  1,5/5        0 0     0         =     1 ·          2,50
18: Psi Chess Sinclair ZX Spectrum {Brettcomputer}              1,5/5    0 0                   0 =       ·   1    1,00
18: CXG Sphinx Seville CXG-807 {Brettcomputer}                  1,5/5              0 0       0 =           · 1    1,00
20: Video Chess TI99/4A {Brettcomputer}                         0,0/5        ?       0     0     0       0 0 ·    0,00

50 Partien von 190 gespielt
Name des Turniers: Zep tepi summer2024
Ort/ Land: LAPTOP-B04J1QBI, Germany
Spielstufe: Turnier 40/120 20/60

You can follow the games in the Schachcomputer.info forum. Even replay them online in the browser.
the tournament begins at page 5:
https://www.schachcomputer.info/forum/s ... 890&page=5

i have also made many pictures of the machines in action.


The programs/programmers used in this tournament are:

Code: Select all

01: Conchess 2 mhz Ulf Rathsmann
02: Cyrus Dragon32 Chess Richard Lang
03: Mephisto ESB II Thomas Nitsche and Elmar Henne
04: Gambiet81 TRS80 Wim Rens
05: Sfinks4 TRS80 William Fink
06: Sargon 2.5 Applied Concepts Dan+Kathe Spracklen
07: CheckCheck 1.75 XT 350 cycles Wolfgang Delmare
08: Fidelity Chess Challenger Grandmaster Voice Ron Nelson
09: PhilidorChess XT 350 cycles David Broughton
10: Grandmaster Chess C64 Fritz Schäfer
11: Mattel Chess Aquarius Copyright 1983 Heuristic Software Corporation
12: Conic Computer Chess Stephen Chen
13: Logichess2.2 Lynx Camputers Kaare Danielsen
14: Atari Video Chess VCS Julio Kaplan / Larry Wagner
15: Mychess IBM XT 350 cycles Dave Kittinger
16: Scisys Senator Chess Mike Johnson or Julio Kaplan ??
17: SciSys Chess Champion MKIV Mike Johnson
18: Psi Chess Sinclair ZX Spectrum Steven L. Watson and Jack Wilkes
19: CXG Sphinx Seville (4Bit!!) Levy+Taylor
20: Video Chess TI99/4A David Levy

How can we help the TI99/4A ??
Is the engine so weak that it cannot compete although it used an early 16 bit machine ??
Was Video Chess on the TI 99/4A the only ever written chess engine on that machine ?

If you want to change the future, you have to learn about the past.

[Viz]

People have no motivation to do something that nobody or almost nobody is going to use.

If there is a big prize money for the best 6502 engine in one year from the announcement (let say some rich person donate a prize of a million dollar for it) and nobody get something that is significantly stronger than what we have now then you have some evidence that it is really hard to get a significant improvement with the knowledge of today but today we have no evidence for this claim.

It is possible that only changing weights in the evaluation of what Ed did at that time may give 100 elo improvement and at that time it was hard to test but today I guess that with some work we can emulate Ed's engine in new hardware except being something like 1000 times faster that mean that with new hardware we can test changing weights 1000 times faster relative to what Ed could.

We actually have quite the opposite evidence.
When sebastian lague (I think it's spelled like this) announced small bots chess challenge it dragged some reasonably big names of modern chess engine development landscape - https://github.com/SebLague/Tiny-Chess- ... gsList.txt
Same goes for 4k tournament at TCEC which resulted in 4k engines reaching 3000 http://computerchess.org.uk/ccrl/4040/c ... +opponents
CCRL elo.
Just that no one wants to do stuff for really ancient hardware for reason of... what exactly? Proving smth on talkchess?

[mclane]

Ed Schroeders 8 Bit engines are very strong compared to the 6502-5 mhz CPU they run.
I guess if he would implement a few endgame knowledge things that do not really interfere with opening/middlegame behaviour
but rescue and increase MM5/academy/Monte Carlo IV/Polgar/Nigel Shorts strength in games that reach endgame.

I mean easy things like Kqk can mate and Krk can mate but not knk or kbk. wrong colored etc.

Coming back to the TI99/4A behaviour in my tournament, i really want to know why the engine is so weak. I guess it has to do with the fact that it came out VERY EARLY in 2nd quarter of 1979 (!!) for TI 99/4 by Tom Merrow as programmer it is said.

TI99/4A came out 31st may of 1981 with a 16 bit CPU TMS 9900 3 mhz, 16 KB RAM, 26 KB ROM .
The TI99/4 came out june 1979.
This early date, june 1979, is the main reason the cartridge Video Chess has such an early production time.
In year 1979 computerchess engines were not that strong. And even David Levy as advisor could not help much.
He maybe helped to create a convenient I/O routine for the engine, it can even play simultaneously against humans, like the later SciSys MKV dedicated unit (also part of Levys environment team). But the strength of the engine is very lousy.

[mclane]

Video chess „was originally released during the 2nd quarter of 1979 on Solid State SoftwareTM Command Module with part number PHM 3008“.
Tom Merrow worked together with David Levy at that time.

„ Texas Instruments was planning to launch the TI 99/4 personal computer and wanted consultancy support in the design of a chess program. Levy had expertise in chess algorithms and programming opening books and was well qualified to carry out the task. He wrote a detailed specification and over an extended period, involving many visits to Texas, he provided advice to their development group. The program had many original features and the resulting Chess cartridge is believed to be the most successful sold for the TI99/4.“

But merrow left TI in december 1978:
„ I was a software engineer (nominally a project leader) and I personally wrote much of the system software and the byte-code interpreter. The first screen you see on power-up (greenish background with little colored blocks)--I wrote that code. I also worked with David Levy and wrote a lot of the chess program (another engineer took over the project and completed it after I left).

I left TI in December 1978 (if I recall correctly), primarily because I couldn't stand Lubbock. I moved to Tektronix Computer Research Lab in Portland Oregon--a very nice place to work and a great place to live.“

So who worked on the game then ?

[smatovic]

Video chess ...

Just for the files, "Video Chess" was also called the Atari VCS/2600 cartridge by Atari from 1979:

https://en.wikipedia.org/wiki/Video_Chess
https://www.chessprogramming.org/Video_Chess

Idk how they stored even a move list with 128 bytes of RAM/registers

--
Srdja