Hello all

[mclane]

hi dave,
i am very happy that you are still reading and learning about computerchess.

i remember the good old years very good.
i still have, like many others, many dedicated chess computers.

in an interview you once said 2009, that the last year you produced a machine for novag was long time ago, 10 years. that would have been
1999. of course we are interested to know, how novag was capable to produce different kind of machines without you and your help AFTER this.

e.g. Citrine comes to my mind.



it came out 2006

cpu H8/3687
type 16 Bit
20 MHz
RAM 3 KB
ROM 56 KB
book 24.000 half-moves

did you sell them the rights to continue the development of your sources, or how was that possible ?


Another interesting question to me is, what has happened with the
Super-Constellation program.

There was the Constellation-Expert machine, right after the Superconny in wood. but it played a little different IMO than the cuperconstellation.
not so exciting.

later i often wondered if you used the superconstellation core in one of the 6301 or even in the H8 machines too.

can you explain how much the development of those constellation and super-constellation programs changed over the years into the 6301 and H8 machines that followed later ?

i once wrote an article about the differences in Super-Forte A/B/C.
the super-forte C was really a very strong playing 8 bit program.

i also thought the 6800 hardware was a very good platform and that the software on this machine played some excellent chess.

i did not really liked the H8 programs with those little RAM amount.
they often had lack of understanding. maybe a reason was the small amount of memory that was used ?!

best wishes,
Thorsten Czub

[Dave K]

Like you, I haven't pursued developing in decades, but for several years have been running mini tournaments in the background testing current engines.

Still just Dave K

[ThatsIt]

Hi Dave !

We've included WChess 2000 v1.2 (MCS) in our lists,
http://www.husvankempen.de/nunn/40_4_Ra ... liste.html
and
http://www.husvankempen.de/nunn/40_4_Ra ... e/957.html

Not bad.

Best wishes,
G.S.
(CEGT-Team)

[DaveKitt]

Thanks for all the nice comments! As for the '5 labors of Hercules', not likely. Unfortunately, through the years and moves, I have thrown out a lot of the old stuff. I think I only have the sources for some versions of WChess (Harvard cup version, Universal hardware version, my production version as well as a few of the old H8 versions.

I used to be a big pack rat, kept copies of everything both sources and paper trails (contracts, royalty statements etc). However, after so many years, divorce, moves etc, I slowly started whittling down the large cache of materials (heck, I even had the old 8" floppies from MyChess). This makes it very difficult for me to clear up absolutely some of the questions regarding versions of CM. However, as I recall, my program was used in the original CM for Apple and PCs. Then the Spracklen version was used for CM 2000, but they had problems getting PC version ready, so only used for Apple. Mine was used for CM2000 and 2100. However, it may have been for CM 3000. I do know one of the versions used my engine for PC and Dan and Kathy's for Apple. The first CM for Apple was based loosely on the old Mychess that was sold by Datamost. The PC version was rewrite in C based on an assembly version (6502 probably).

As for all the Novag versions, the early Robot Adversary and Savant were created from the z80 Mychess assembly program - meaning there was a high reliance on a swap-off evaluation rather than quiescence search. The Constellation was a complete new program written in 6502 as I recall, and did away with swap-off in favor of quiescent search. While working for Novag, there was generally a 'high-end' product and then some 'low end' products. The high ends went from z80 to 6502 to 68k as I recall. The low ends were such devices as 6301, 6501, TLCS47 and other odd birds. Each year would be a new CPU which would have more ram and more rom for same price, but generally completely different architecture. So each program had to be designed for the various indexing and register schemes to try and maximize speed/strength. Note, these started with something like 8k rom, maybe 1k ram. One low end had 2k bytes ROM, 196 nibble (4 bit) ram. I think the high end robot had 48k ROM and 8k ram. Constellation was either 8 or 16k rom, don't recall exactly, with probably 4k ram.

Goal with WChess was to have a program in high level language (C) whilch would facilitate testing new search and eval ideas. This was a continual process over probably 10 years. A funny thing, I had tried NULL move but sort of messed up the implementation and hence did not find it better than the static forward pruning I had been using, so never switched over. WChess also used piece value tables and in fact used a high level PVT generator written under contract by Don Dailey and Larry Kaufman. Larry also contributed some opening books, although this was also an ongoing process in my 'lab'. Experimented and adopted Singular Extensions, One reply to check. Spent lots of time looking at things like recapture extensions, k safety extensions, p to 8th extenstions and all manner of pruning ideas.

Major drawbacks of the programs:
1) dependence on PVTs for bulk of chess knowledge. Led to weaker play at deeper levels as PVT knowledge generated from root position became less relevant/accurate as depth increased.
2) insufficient eval of passed pawn threats - I saw in Stockfish that they tested control of queening sqs, I guess easier to do w/bitboards.
3) pawn structure eval was never very strong.

Just for the record, I did have some collaborators over the years besides Don and Larry as mentioned above. I think Hal Bogner was first 'chess contributor', he helped with testing and input for evaluation ideas back when I worked in Van Nuys. After Hal, Scott McDonald contributed chess knowledge and reviewed literally hundreds of autotest games looking for weaknesses and improvements. When I moved to Mobile, Al, James Parker worked with me off and on for several years. James is s very bright fellow and wrote a Shogi program for Novag. He also contributed to the chess efforts and the Chinese Chess program as I recall. There were also a number of ideas adopted as result of conversations with various programmers at the computer chess tournaments I attended.

Dave

[JuLieN]

Wow, thanks a lot Dave, that was a lot of information! I bet the various wiki contributors we have here got their birthday present already!

Regarding null move, it took a lot of time to get it right anyway (and we all have to find the settings that fit best to our own engines).

Let's hope you can put your hands back onto some WChess sources, then. No need to improve it at first: a UCI version for a start would make everyone happy.

@Fern
So, now you know, about CM3000: nobody knows, even its creators. Not sure you'll get your appetite back.

[F.Huber]

I think I only have the sources for some versions of WChess (Harvard cup version, Universal hardware version, my production version as well as a few of the old H8 versions.

Hi Dave,

would you give the permission to publish your WChess 1.05 (PC version)?

I'm sure that many members here still have this program on floppy disks, but can't use it anymore because the installation count had reached its limit (3 if I remember correctly). So they could still use WChess e.g. in DOS-Box.

Franz

[Dr.Wael Deeb]

I think I only have the sources for some versions of WChess (Harvard cup version, Universal hardware version, my production version as well as a few of the old H8 versions.

Hi Dave,

would you give the permission to publish your WChess 1.05 (PC version)?

I'm sure that many members here still have this program on floppy disks, but can't use it anymore because the installation count had reached its limit (3 if I remember correctly). So they could still use WChess e.g. in DOS-Box.

Franz

Would be a great idea and an amazing present for us,the computer chess geekazoids

[fern]

Good Joke....

Let us see:
which is the mathematical probability that two David K persons, both dedicated to chess programming and same age and coming to same weird site in Internet are NOT the same person?

Fern

[DaveKitt]

I do have some of the old WChess floppies - with 3 installs

I may be able to find a non-protected executable, but would need to find a machine w/3.5" floppy. Easier than finding a zip drive I think...

Source code is way too ugly to ever put out there. As I said, I was a pack rat, meaning I never saw a line of code that could be deleted when it was just as easy to put a conditional around it....

I'll try and look around this weekend and see if I can get an executable. Most the C chess programs I wrote used other UIs such as ChessMaster and PowerChess and as such were proprietary (ie - unique) and not compatible w/winboard or UCI. The version I played on chess.com for a few games just had text board output, which was basically a debugging version.
Dave

[fern]

That is great new.
You see, here, added to some degree of general madness, you will find inexcusable acquisitive people not happy if they does not have 450 engines to play or test and always the last one or the older one, the strongest and the weakest, the pretty and the ugly, the one with historical value AND strong as yours and others without any value whatsoever.
It does not matter.
They are all nuts as asses.
Well, David, when will get it....?

fern