31 bit hash values. How often will it fail?

Discussion of chess software programming and technical issues.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Aleks Peshkov
Posts: 887
Joined: Sun Nov 19, 2006 8:16 pm
Location: Russia

Re: 31 bit hash values. How often will it fail?

Post by Aleks Peshkov » Mon Sep 08, 2008 5:43 am

I doubt that whatever old program written in C/Pascal have special data restrictions to natural word size.

Carey
Posts: 313
Joined: Wed Mar 08, 2006 7:18 pm
Contact:

Re: 31 bit hash values. How often will it fail?

Post by Carey » Mon Sep 08, 2008 2:35 pm

bob wrote:you haven't mentioned the "era" of the program, which would give insight into the search techniques used. Pre-1989 would mean no null-move or singular extensions, no reductions. Probably basic alpha/beta with a couple of types of extensions....

It is definetly pre-1989. If it comes after the mid 80's, the odds are good I don't care about it. (With the exceptions of novelty programs, such as Javascript chess, micromax, etc.)


The program is a little convoluted and it's hard for me to follow the basic structure. And the author did experiment with a few ideas. So things aren't always done the traditional way. And there might even be a few bugs...

Hopefully today I'll get time to expand things to 60 bits of hash. (I'm just going to use the numbers already there, just in a different order.)

We may still have other issues, of course. We probably do have other issues to deal with. It's entirely possible expanding the hash size wont help at all. But it's worth a shot.


Really wish I could say more, but if I start ignoring authors wishes about releasing or talking about the programs before they want, then not too many would trust me with their old programs. (Not that I'm likely to find many more... :cry: )

bob
Posts: 20930
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: 31 bit hash values. How often will it fail?

Post by bob » Mon Sep 08, 2008 3:13 pm

Carey wrote:
bob wrote:you haven't mentioned the "era" of the program, which would give insight into the search techniques used. Pre-1989 would mean no null-move or singular extensions, no reductions. Probably basic alpha/beta with a couple of types of extensions....

It is definetly pre-1989. If it comes after the mid 80's, the odds are good I don't care about it. (With the exceptions of novelty programs, such as Javascript chess, micromax, etc.)


The program is a little convoluted and it's hard for me to follow the basic structure. And the author did experiment with a few ideas. So things aren't always done the traditional way. And there might even be a few bugs...

Hopefully today I'll get time to expand things to 60 bits of hash. (I'm just going to use the numbers already there, just in a different order.)

We may still have other issues, of course. We probably do have other issues to deal with. It's entirely possible expanding the hash size wont help at all. But it's worth a shot.


Really wish I could say more, but if I start ignoring authors wishes about releasing or talking about the programs before they want, then not too many would trust me with their old programs. (Not that I'm likely to find many more... :cry: )
On that topic, I received an email today from someone that says he has a source _listing_ of chess 4.6. I told him to hang on to it although I am not sure the OCR approach is worth anything, and it is a compass (CDC assembly language) program.

Carey
Posts: 313
Joined: Wed Mar 08, 2006 7:18 pm
Contact:

Re: 31 bit hash values. How often will it fail?

Post by Carey » Mon Sep 08, 2008 3:28 pm

bob wrote:On that topic, I received an email today from someone that says he has a source _listing_ of chess 4.6. I told him to hang on to it although I am not sure the OCR approach is worth anything, and it is a compass (CDC assembly language) program.
An electronic copy of 4.6 is widely available. Slate (I think) gave a PDF version to the ComputerHistory.org people, and I got it from them before they posted it. (It's on my website. Posted with Slate's & Atkin's permission, of course.)

It's a pdf, but it's just plain text in there. It's likely the same as the printed version your guy has.

It'd be interesting to do a C copy of the program, just to see how it performed on modern hardware. But doing that kind of port would be far too much trouble.


I wish I could get hold of Atkin again to ask him about Chess 3.6. (The email address I had is no longer valid and I don't see anything newer.)

Supposedly he either included a copy in his thesis or at least thoroughly described it.


Not much left on the classic selective search programs. I know the reasons they have problems, but I still think the idea is better / more elegant than brute force.

Although, with the more agressive prunings modern brute force programs do, they are starting to look a little more like their selective search ancestors.


Speaking of Chess, Atking & Slate.... back when I asked Slate about NuChess, he wasn't in the mood to release it, but he did say to be sure and let him know when your CrayBlitz source became publicly available.... Sounds like he's wanting to look at his old competitor.... But maybe I can barter with him and he'll offer up more versions of Chess and/or NuChess.


Carey

bob
Posts: 20930
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: 31 bit hash values. How often will it fail?

Post by bob » Mon Sep 08, 2008 5:11 pm

Carey wrote:
bob wrote:On that topic, I received an email today from someone that says he has a source _listing_ of chess 4.6. I told him to hang on to it although I am not sure the OCR approach is worth anything, and it is a compass (CDC assembly language) program.
An electronic copy of 4.6 is widely available. Slate (I think) gave a PDF version to the ComputerHistory.org people, and I got it from them before they posted it. (It's on my website. Posted with Slate's & Atkin's permission, of course.)

It's a pdf, but it's just plain text in there. It's likely the same as the printed version your guy has.

It'd be interesting to do a C copy of the program, just to see how it performed on modern hardware. But doing that kind of port would be far too much trouble.


I wish I could get hold of Atkin again to ask him about Chess 3.6. (The email address I had is no longer valid and I don't see anything newer.)

Supposedly he either included a copy in his thesis or at least thoroughly described it.


Not much left on the classic selective search programs. I know the reasons they have problems, but I still think the idea is better / more elegant than brute force.

Although, with the more agressive prunings modern brute force programs do, they are starting to look a little more like their selective search ancestors.


Speaking of Chess, Atking & Slate.... back when I asked Slate about NuChess, he wasn't in the mood to release it, but he did say to be sure and let him know when your CrayBlitz source became publicly available.... Sounds like he's wanting to look at his old competitor.... But maybe I can barter with him and he'll offer up more versions of Chess and/or NuChess.


Carey
He's more than welcome to a copy. Time has been so tight I have not made progress on the book stuff. I could try to put together something that will compile and run with no book, just for fun, although not having a book is a bummer...

bob
Posts: 20930
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: 31 bit hash values. How often will it fail?

Post by bob » Mon Sep 08, 2008 5:24 pm

woohoo... a breakthrough. Someone had emailed me a question about an old position from some Cray Blitz game and I was looking thru my chess problem file (a paper file, not disk) and lo ahd behold, I found a print-out of the book code. I will work on getting this entered and maybe we can release a complete functional version of 1991-era Cray Blitz. I will have to figure out if "define file(...)" is in the unix fortran, and if not, make some changes, but it ought to be workable with a little poking around...

More later...

User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 6:43 pm

Re: 31 bit hash values. How often will it fail?

Post by sje » Mon Sep 08, 2008 6:45 pm

Carey wrote:Speaking of Chess, Atking & Slate.... back when I asked Slate about NuChess, he wasn't in the mood to release it, but he did say to be sure and let him know when your CrayBlitz source became publicly available.... Sounds like he's wanting to look at his old competitor.... But maybe I can barter with him and he'll offer up more versions of Chess and/or NuChess.
If I can recall correctly, Slate wrote a commercial chess program in the early 1980s for the Apple II that used the eight bit 6502 CPU. The program was called "Chess 7.0".

Carey
Posts: 313
Joined: Wed Mar 08, 2006 7:18 pm
Contact:

Re: 31 bit hash values. How often will it fail?

Post by Carey » Mon Sep 08, 2008 6:45 pm

bob wrote:woohoo... a breakthrough. Someone had emailed me a question about an old position from some Cray Blitz game and I was looking thru my chess problem file (a paper file, not disk) and lo ahd behold, I found a print-out of the book code. I will work on getting this entered and maybe we can release a complete functional version of 1991-era Cray Blitz. I will have to figure out if "define file(...)" is in the unix fortran, and if not, make some changes, but it ought to be workable with a little poking around...

More later...
EXCELLENT!!

If this version isn't too different from the other two (paper copies) you sent me, we might even be able to recreate those as well.

If the differences are too great, we can at least release the OCR'ed text and the jpg scans.

Carey


P.S.... Maybe you ought to poke around your paper archives a little more....? :)

Before, you managed to find a copy of Blitz and two copies of CrayBlitz that you thought had been lost. Now another part of CrayBlitz... Who knows, you might still have more stuff that you forgot about or thought was lost....

Maybe hidden in some box in your office or basement might even be the last surviving copy of the source for MacHack VI, Chess 3.6, Kotok-McCarthy and even the location of where Elvis & Jimmy Hoffa are hiding... :lol:

Carey
Posts: 313
Joined: Wed Mar 08, 2006 7:18 pm
Contact:

Re: 31 bit hash values. How often will it fail?

Post by Carey » Mon Sep 08, 2008 6:50 pm

sje wrote:
Carey wrote:Speaking of Chess, Atking & Slate.... back when I asked Slate about NuChess, he wasn't in the mood to release it, but he did say to be sure and let him know when your CrayBlitz source became publicly available.... Sounds like he's wanting to look at his old competitor.... But maybe I can barter with him and he'll offer up more versions of Chess and/or NuChess.
If I can recall correctly, Slate wrote a commercial chess program in the early 1980s for the Apple II that used the eight bit 6502 CPU. The program was called "Chess 7.0".
That was Atkin.

He mentioned it once
I did write Capablanca for the Destiny Systems Great Game Machine, but that's not going to do anyone any good. It was written in 6502 assembly language, and interfaced directly to an 8-character 16- segment display, and a 12-key keypad. That was later converted to be CHESS 7.0 for the Apple ][, Atari 800, and Commodore 64. Even later it became "How About a Nice Game of Chess" for the Apple ][e. If you have some way of reading 25-year-old Apple 5 1/4 inch floppy disks, I might be able to find them. They are pretty dusty, and probably unreadable.
As you notice, he did offer to look for the disks for those games...

Unfortunately I haven't been able to get in touch with him since. I do know a couple guys with a "cat weasel" (I think it's called) which is a card that plugs into a PC that let it read just about every floppy disk ever made.

User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 6:43 pm

Re: 31 bit hash values. How often will it fail?

Post by sje » Mon Sep 08, 2008 6:53 pm

bob wrote:woohoo... a breakthrough. Someone had emailed me a question about an old position from some Cray Blitz game and I was looking thru my chess problem file (a paper file, not disk) and lo ahd behold, I found a print-out of the book code. I will work on getting this entered and maybe we can release a complete functional version of 1991-era Cray Blitz. I will have to figure out if "define file(...)" is in the unix fortran, and if not, make some changes, but it ought to be workable with a little poking around...
From time to time I've seen complete Cray machines with software up for sale on ebay. The prices are unbelievably low, perhaps a thousandth of the original cost.

The circular chair model sees to be popular. But where would one get the Freon needed for cooling?

There is (or was) at least one commercial Cray production class emulator available for the Apple Macintosh.

Post Reply