Chris Whittington's Chess System Tal!

[chrisw]

So, to answer your question, there is nothing contractual now stopping me from writing another chess program. Instead I got interested in neural networks, wrote one in C that I can control all parameters of several years ago (did various things with that including a mega-strong Backgammon, also tried RRKK but it was not much good) but am now playing around with historical stockmarket data to see what transpires.

Hello Chris,

I vaguely remember that you also talked about writing a shogi program many years back. Did you ever do that, or were the plans abandoned?

I've been wanting to work seriously on computer shogi for a long time, but unfortunately people will never let me stop working on my chess program.

Tord

Hi Tord,

Yes, I produced a very agressive Shogi engine in about three months. I think it was built on the back of the routines in CSTal3 (not the bitboard ones!) in 1998/99 or so.

What has happened was that OS had licenced Jeff Rollason's Shogi to its Japanese distributor, the revenues were large and important (PC and Play Station). There was strong Shogi competition and Jeff was struggling a bit at one stage. OS had very little control over what he did since nobody understood, and there was always the threat of losing the Japanese deal to competitors. So I attacked the Shogi problem for two reasons, one to put a rocket up Jeff's backside and the other to try and understand the basis of Shogi and Shogi engines in order to manage the project better.

The result was immensley agressive, the underlying routines of movgen, move, unmove are fairly simple to change, so really all the work for a chess programmer is evaluation. I just went for massive king attack and used standard piece values for the different Shogi pieces.

The Japanese distributor came over and watched it play a couple fo games and liked it. Jeff had got everything back together again by then, so I didn't do any more with mine and the source probably sits around somewhere.

Any, the moral is that Shogi is not very difficult for a chess programmer who already has a working chess program.

Best wishes,

Chris

[chrisw]

thats not much. what we would need would be the
64 bit bitboard UCI CSTAL III engine

do you mean the one I wrote that has never been published?

Very interesting that you rewrote CSTAL with bitboards. I remember reading the very interesting debates you had with Bob Hyatt over the usefulness of bitboards in eval. How do you approach the eval with bitboards? Do you still use attack tables? I would be very interested in a version to play against...

Hi Zach,

I'll need to get Thorsten to remind me what I did (I would guess he has a copy of the source). I think it played chess (it was auto-testable against the CSTal2 version), but I don't think I'ld done much with the evaluation. If I remember right it was mixed 0x88 and bitboard. I think attack tables (and everything actually) was doubled up as the old bitwise patterns (16 bits for each sides peices) and also 64-bit board based attack patterns. Thinking back, I was probably intending to discover which representation worked best for any one function.

Best wishes,

Chris

[chrisw]

Oh, let them try....First they should know. Then they should come here. Then, to probe I did sent the stuff over there. Etc.
Sore maybe they are, but I doubt they are so dumb people to try that gamble.
In any case no one took my offer, so the shop is closed.

To anyone of the badly harmed guys regards
Fernando

Hi Fernando,

You are pretty safe down in there Tierra del Fuego, you and I could just hide ourselves amongst the alpacas

Best wishes,

Chris

[chrisw]

Chris:
If you feel it is no proper to do what I said I could do, just tell me and I will not do it.
Or perhaps the receiver of the copied stuff could deliver to you the book you choose? That would be another way...
I do not pretend to get a profit of this. I have given many things for not even the cheap book as a compliment, but simply these days I cannot do it anymore due to financial problems. A $5 book just cover PART of my post expenses, the package, etc. In terms of my motivations, you know well I have been a fan of Cstal since lot of time, even writing some stuff about it as you surely remember. So when I offer to do this, with the investment of time of copying a disk, going to the stationary to get an envelope, then to the post office, etc, I do it to keep the flame .
But, again, if you feel different, just let me know.
Truly yours
Your friend Fernando.

Hi Fernando,

No problem for me - this stuff is not in any way my property any more. Not really very sure who owns it now, maybe the liquidator - for sure somebody owns it, ownership doesn't just vaporise into thin air.

I do know there are some very sore venture capitalists out there who lost around £16mn of investment, and they might just love the opportunity to sue someone (via the liquidator) to recoup some few shekels.

Beware of sharks is my advice.

Your old friend,

Chris

Hi Chris,

Did you sign a non-competition agreement? If so, when does it expire?

If you are not (now) under an agreement to not write another chess program then what is stopping you from writing one?

Hi Michael,

It was a lot more complicated than a non-competition agreement. Those venture capitalists make it way more certain than that. Part of the way it was done was to pay me a large yearly consultancy fee so that anything I produced would belong to them. Therefore I produced nothing and became a farmer instead.

When they went bust the consultancy fee (paid monthly) ceased.

So, to answer your question, there is nothing contractual now stopping me from writing another chess program. Instead I got interested in neural networks, wrote one in C that I can control all parameters of several years ago (did various things with that including a mega-strong Backgammon, also tried RRKK but it was not much good) but am now playing around with historical stockmarket data to see what transpires.

Best wishes,

Chris

That should read KRK, btw. I also tried a few other simple endgame databases by ANN. Not exactly successful. There is one ANN scheme that works well with chess engines. I just might reveal it one day

[mclane]

yes. but it is a pity that the "project" or the nearly relased version 3
lies arround with nobody trying to finish or publishing the almost finished
stuff.

IMO it would be nice to get the bitboard version running...

or that somebody would continue the project.
if it is not you, we all understand that you have no interest anymore, it should IMO be possible that somebody else could try to finish it.

the search of cstal2.03 was a weak point.
transforming the ideas of CSTAL into a modern program of today with a
modern search would be an interesting project.

why don't you "direct" this. you said you are older now and would like to direct and organise things instead of writing yourself.

you could try to help or manage the others who try the job.

[mclane]

i remember i bought those strange pieces with strange symbols on it and tried to learn these chess variation.

i still have the pieces but never found somebody else who also knows the rules.
so i gave up learning.

[chrisw]

i remember i bought those strange pieces with strange symbols on it and tried to learn these chess variation.

i still have the pieces but never found somebody else who also knows the rules.
so i gave up learning.

Shogi is a very nice game. More interesting than chess, imo. You can learn by playing against a Shogi program.

Apart for the boardeing larger (9x9), captured pieces cna be parachuted back onto the board, so the game becomes more complex and exciting as moves are played out. Unlike chess which simplifies down, Shogi complicates up.

[chrisw]

yes. but it is a pity that the "project" or the nearly relased version 3
lies arround with nobody trying to finish or publishing the almost finished
stuff.

IMO it would be nice to get the bitboard version running...

or that somebody would continue the project.
if it is not you, we all understand that you have no interest anymore, it should IMO be possible that somebody else could try to finish it.

the search of cstal2.03 was a weak point.
transforming the ideas of CSTAL into a modern program of today with a
modern search would be an interesting project.

why don't you "direct" this. you said you are older now and would like to direct and organise things instead of writing yourself.

you could try to help or manage the others who try the job.

But 'we' don't own the CSTal3 source, so 'we' can't use it.

[Tord Romstad]

Hi Tord,

Yes, I produced a very agressive Shogi engine in about three months. I think it was built on the back of the routines in CSTal3 (not the bitboard ones!) in 1998/99 or so.

What has happened was that OS had licenced Jeff Rollason's Shogi to its Japanese distributor, the revenues were large and important (PC and Play Station). There was strong Shogi competition and Jeff was struggling a bit at one stage. OS had very little control over what he did since nobody understood, and there was always the threat of losing the Japanese deal to competitors. So I attacked the Shogi problem for two reasons, one to put a rocket up Jeff's backside and the other to try and understand the basis of Shogi and Shogi engines in order to manage the project better.

The result was immensley agressive, the underlying routines of movgen, move, unmove are fairly simple to change, so really all the work for a chess programmer is evaluation. I just went for massive king attack and used standard piece values for the different Shogi pieces.

The Japanese distributor came over and watched it play a couple fo games and liked it. Jeff had got everything back together again by then, so I didn't do any more with mine and the source probably sits around somewhere.

That's a shame. I guess somebody else owns this code too, and that there is little or no hope of ever seeing it released in any form?

Any, the moral is that Shogi is not very difficult for a chess programmer who already has a working chess program.

If it were only the engine, it wouldn't be too bad. The problem is that I have to write a GUI as well. The GUI for my chess program is messy and poorly written, and I think modifying it to handle shogi would be about as much work as writing a completely new GUI from scratch. I would also have to do some work to port the low-level bitboard stuff to shogi, which isn't really technically very hard, but awfully boring.

Tord

[Tord Romstad]

i remember i bought those strange pieces with strange symbols on it and tried to learn these chess variation.

i still have the pieces but never found somebody else who also knows the rules.
so i gave up learning.

How about an e-mail game, Thorsten? At rook handicap, perhaps? A rook handicap is far less than it sounds like for a chess player, but I am a very weak shogi player myself, so I guess a rook handicap would be about right in a slow game against a beginner with chess experience.

Shogi is a very nice game. More interesting than chess, imo. You can learn by playing against a Shogi program.

Apart for the boardeing larger (9x9), captured pieces cna be parachuted back onto the board, so the game becomes more complex and exciting as moves are played out. Unlike chess which simplifies down, Shogi complicates up.

I agree, it is much more interesting than chess, both for playing and for programming.

Tord