gooner:

The one where he "dared me" to play the Double Muzio against him? He was taunting me on the ChessTalk forum, and it spilled over to here. I should have just ignored his challenge, but it is not like me to back down from a fight.

I did not know he had the Chessbase CD with a refutation of the line he asked me to play. I gave up a Bishop and a Knight for just 2 pawns, very early in that game, and it is still going on.

To prove to him that I could really sacrifice a Knight and Bishop against him, I took him out of published play in our next game and I won that in 26 moves. And his rating was 2700 at the time, before the celing was removed. Since that time, he was rated 2706, and I won a second game against him in our "3 wins" match.

You should comment on this game, since it shows what a player can do "without help" against the sum of all of the accumulated chess knowledge in the world.

http://brainking.com/en/ArchivedGame?g=532067

Now you know the full story.

Pedro, if you do not think it is a losing move, challenge me to a rated game, and play it. I will play the other side.

The piece that would have moved from 24 to 28 is now off the board, and cannot influence the play whatsoever.

That subgame, by defintion, cannot possibly occur in this game, since it is impossible to create any such position from the parent game with an extra piece on the board.

ustica:

Take a chance? Translation. Lie about the move and see if your opponent believes you.

Sorry pal, one look at "24-28" (by the way, why did you say that instead of refer to the letters on the board on BrainKing? Oh, that's right, computer programs have numbers on them, don't they?) and I knew that was a losing move.

http://brainking.com/en/ShowGame?g=952957&i=16

I direct everyone's attention to the above position. After I made my move, my opponent, Ustica, made this remark to me in our game:

ustica: ustica: (7. August 2005, 01:48:33)
if i was white i do 24-28 not 1-6 you like publesh moves. oh well just my opion.

On our board, 1-6 is g1-f2 and 24-28 is b6-a7. The move the "great" ustica is trying to tell me to play will lose.

I am wondering, if he is so good, why doesn't he know this? Does he actually think he can persuade me to make a losing move.

What a great champion he is.

Kingsrow has the 10-piece database now.

http://pages.prodigy.net/eyg/Checkers/10-pieceBuild.htm

Ladders are suppose to work as follows:

1. The "Top N" players are "sacred" and only accept requests from within the group. In most cases, it is the top 10 players.

2. The top level MUST BE VERY ACTIVE and cannot turn down a game request.

3. If any player on a lower rung (i.e. a higher number) wins over a higher rung (lower number), the winner takes over the rung and the loser drops one place.

Example, Rung 7 defeats Rung 4, then 7 is "teleported" to rung 4, old rung 4 gets demoted to 5, and 5 goes to 6.

So you can get bumped to a worse ladder rung (higher number) even if you yourself have not lost. Every time someone over top of you loses, there is a chance you will go down too.

4. If anyone outside the "sacred" circle wins, like rungs 11 to 100, then the winner moves "half the distance" closer to the loser if starting below the opponent. The loser drops one slot.

Example: Player 80 defeats Player 20. The new spot for Player 80 is 20 + (80 - 20)/2 = 50. Player 20 is then bumped down to 21, 21 to 22, etc.

This is so that someone down the bottom can't just "get lucky" and take over a high spot with one win. You have to "claw your way up" by moving "half the distance" repeatedly.

The problem with ladders is that the initial seeding gives the #1 spot a position that is almost impossible to wrest. A drawn match does not swap the ladder position, so the first player to reach #1 almost always keeps it forever.

Not easy to spot this one since there is still 8 pieces on the board, but this will distill into Second Position. Test your own precognition

http://brainking.com/en/ShowGame?g=846243&i=44

I have played against three World Checkers Champions. Alex Moiseyev, Ron King, and Don Lafferty, winning one game from each of them.

I could give you the other dates as well, but "treasure" is a strange word to apply to something that is just a simple matter of fact.

And if Alex Moiseyev is listening, what would he think, given that he drove hundreds of miles to play some checkers with me on December 18, 2004?

Next Thurday night at 9 PM Eastern I will be the guest speaker on chess.fm radio.

If you go to http://www.chess.fm just turn on your computer's speakers, and you can hear the broadcast.

There is a question-and-answer segment on the show, and for 20 minutes, I will be answering the questions tossed my way.

Over the years, starting with the MS Gaming Zone in 1996-1996, Case's Ladder in 1997-1998-1999, Playsite.com over the same tenure, there has been one common denominator.

Some virtually unknown player has issued remarks claiming "I cheat". I was 456-0 on Case's Ladder with about 20 draws. I was undefeated on Playsite.com, winning a game from (then) World Champion Ron King. Even George Miller and I first "met" on Gamezone, and we played many an interesting game.

I took my lumps, like everyone else, but once I got good, I changed my ID from "HoodedClaw" to "Marion Tinsley", and boy, did I get yelled at! It was like a sacriledge (I still did not know he had recently become deceased, about 8-9 months prior) so I switched it to "Ed Trice" very soon thereafter.

I believe this "sacriledge" in their minds made me a marked man first. So be it, the big guns knew who I was now, and we would duke it out and the shouts would come flying also.

That was the net then... and some things do not change. Rather than issue far-flung statements, let me call to the attention of everyone some unbiased facts:

1. I beat Chinook with its 6-piece database of 2.5 billion positions several times. I was the first person to defeat it twice in one day. It you go to http://www.cs.ualberta.ca/~chinook/WallofHonor.php you can see this.

Novice level
286. Ed Trice, October 20, 1996

Amateur
52. Ed Trice, October 20, 1996

Then again the next day, on its highest level:
Intermediate
86. Ed Trice, October 21, 1996

You have to understand something. This list is built from the bottom up. You will see there are VERY FEW people underneath.

I was only the 6th person to win against Chinook on its highest setting, and it had been online a little over a year at that point. To win 3 game in 2 days against it was unheard of at that point.

So, people might say "You were using a program". You have to remember, back then, there were no programs! Dr. Schaeffer did not make his databases available to the public until June of 1997. In fact, Gil Dodgen and I were the first to use his 6-piece database probed in RAM when we released World Championship Checkers (WCC) on August 13, 1997.

Chinook was the undisputed strongest program out there at the time, plus it was running on a 150 MhZ SGI box in Canada. And, at the time of my victory, 100 MHz Pentiums were top of the line, with 133 MHz comming out later that fall.

Gil's previous program, Cornell Checkers, was a DOS program that had no databases in 1990 and a 4-piece database in 1992. He discontinued it in 1994 when an operating system change at Microsoft undermined his 32-bit DOS extender program. His software ran for the last time in 1994.

After I played a phone game against his Cornell Checkers program, and beat it soundly, Gil "hired me" on the spot to serve as the evaluation function consultant for our new checkers program. I coined the name WCC, published it, and we were a team.

2. I beat the World Computer Chess Champion, Deep Thought, in 1989. I know, not checkers related, but I gave Deep Thought its quickest loss ever, 20 moves.

http://www.chessgames.com/perl/chessgamegid=1272214

In fact, I am the only person who may make the claim to have defeated programs that have held World Machine Champions in the game of checkers and chess.

Basically, I know how to find positions that might be problematic for software programs, because I am software programmer myself. I know how hard it is to get a program to "understand" a piece of knowledge such as a tailhook, single corner cramp, double corner cramp, diamond cramp, 2-holding-1, when to break a bridge, when a doghole man-up scenario will lose and draw, etc.

I am uniquely quialified to beat up software, since I know what is usually outside of its domain of understanding.

On a side note, I also wrote the first software program to eclipse the 2200 mark (delineating the Master Class) when "The Sniper" won its 3rd round game against Mike Tempkin in 1987 at St. Joseph's Prep High School in Philadelphia.

This was that many years after Ken Thompson's "hardware master", Belle, eclipsed the 2200 mark.

He had a $600,000 system funded by Belle Labs. I had my 512K Macintosh with a 7 Megahertz clock.

That's Megahertz, not Gigahertz.

3. I have already showcases two checkers games where I should have lost. Any "majors" player could have won the game where I missed the Andrew Jackson Defence in The Switcher. I would like to beleive a Master Class player would be needed to complete the difficult ending to win against me in the other game.

Again, programs have books to avoid such losses, and endgame databases can drive you the rest of the way home.

4. I play dangerous, sacrificing lines that are far beyond the horizon of computers. Look at these games:

http://brainking.com/en/ArchivedGame?g=325125&i=17

In this game, I sacrificed a Chancellor for Archbishop, then I throw away my Knight, all for position play, against a very strong player. There is a mate in 59 with optimal play.

http://brainking.com/en/ShowGame?g=821982&i=31

In this position, I played Ri3 with my Queen hanging. If my opponent takes the Queen, I have a mate in 47. He avoided it. I wonder if anyone here could resist?

http://brainking.com/en/ArchivedGame?g=347378&i=67

Here I sacrifice an Archbishop to begin a very lengthy mating sequence in Janus Chess, which I am sure came as a surprise for Caissus, a very strong Janus player.

http://brainking.com/en/ArchivedGame?g=532067&i=31

And of course, here is my 26 move win against 2700 rated Alex, with a two piece sacrifice at the end to wrap it up.

The point of all of this: To have any form of software duplicate these moves, you would need, in some cases, over 80 plies of searching. This is BEYOND what a program can do tactically, and it is in the domain of what we call strategy.

Strategy is, essentially, long range tactics.

Why do you first try to double rooks in an open file, then seize the 7th rank with one, then the other? Because you have done it 1,000,000 times and you know it is a formula for success.

Sometimes moves such as Rad1, Rd7, Rf7, R1d7, Rxg7+, Rxh7+ are intersperesed between pawn pushes, recaptures, some checks or check evasions, or some other minor piece exchanges. This can push the tactic beyond the horizon of the program's ability to search. So, while we humans say "oh, double the rooks, seize the 7th, and you win", programs cannot do this.

Such is the case with my games on here. I do well because I bring the sum of my experience at beating software programs with me. I am unique in that I program them, have written some World Class software myself, and I do not give up thinking of ways to destroy my opponent until every last plausible attempt involving any amount of sustained attacks have been exhausted.

I will kill anyone who wanders onto my board and does not play with equal aggression.

Do I need to prove that too?

As for the earlier comments about my endgames.

Look at these two games.

Game 1

http://brainking.com/en/ArchivedGame?g=647831&i=47

My opponent could win on the next move with d2-e1 but he played d2-c1 instead. If I was using a computer program, I would never have gotten into this psotion to begin with. I will allow someone like George Miller to confirm I am in a loss. I can't be responsible for my opponents not winning.

Here is a second example, when I forgot the Andrew Jackson Defense to the Switcher...

http://brainking.com/en/ArchivedGame?g=208684&i=6

From here I should lose. f8-g7 was the way to draw that line, and I played it enough to know better, I just merely forgot it.

So, again, Jake and his far-flung assessments are demonstrated false. Every prorgam has moves in the opening book to avoid the early loss, and endgame databases can salvage draws from a great distance, which could have saved the first game.

Oh, I get it. Jake is Pedro's pawn account, now it all makes sense. Jake is Pedro. Got it. Bye bye (hides Pedro.)

I see Pedro needs a new hobby. Gee whiz Pedro. Don't you have a life, or are you still trying to derail the Gothic Chess Federation, lol.

Jake Lopez's Checkers record:

4 wins, 0 draws, 2 losses

Why am I even wasting my time with you?

You are now on hide.

Have a great life in your little imagined world of greatness. I am sure you will be the World Champion in no time.

(laughs so hard I almost fell off the chair.)

Say what you want, you can't make a 3000 member organization that was based in Canada disappear. The PGN of the games exist, I have seen them, I went over the games from my tournament, the move list is accurate.

I have computed the 10-piece database. Not only that, it has perfect play information, not just "win-loss-draw" like other programs. It also has an Aggressive Draw Hueristic that will put you into the position with the fewest legal moves to draw, move after move after move, until your own 8-piece database program's draws are beyond its horizon, meaning you will spiral into a loss before you software realizes it.

It sits on a box down at the University of Pennsylvania with 256 Gigabytes of RAM and 32 separate 2-terrabyte hard drives to store the databases. Its search speed is about 8 million positions per second.

I can't remember if I sent Purple an early copy of the paper I was writing on it, or not. He can reply if he wants.

If I used that program against people on here, I would have 0 draws, not 16.

You should really do some more research before you say fairly unintelligent things.

You can start your research here:

(a book I am in...)
http://ostg.pricegrabber.com/search_getprod.php/isbn=1402077092/nrd=1/

By the way, I helped to correct the Chinook 8-piece databases, which had errors in them until late 2001. It is mentioned in that book, page 210. Write to the publisher, ask them to email you the "Acknowledgements" paragraph from that page. Or send an email to Dr. Jonathan Schaeffer himself, he will tell you.

http://www.gothicchess.org/7_piece_perfect_play.pdf

As for my abilities, articles I had written for Draughts Razoo, and English magazine, were very well received. And World Champion Alex Moiseyev said my annotation of one of HIS games was the best quality of analysis he has seen, and it will be given the most real estate in the book he is publishing. I guess a computer program I wrote distilled the essential strategic concepts and was able to put them into laymen's terms as well.

And I don't have 2 computers, I have about 175 or so in my company.

So you criticize my play, yet I have played in a live tournament, while you have not.

I see no logic in your statements.

If I am such a "nobody", and, using your own metric of "nobodiness", you are less of a nobody, I don't think anyone would care about the result of of "nobody" vs. "less than nobody" contest.

So, why don't you go play with yourself.

Still waiting for you to tell us where you have played live tournaments. All games from the 22 NACA tournaments are in the Wyllie Games Archive from what I have been told.

That's because Alex asked me not to discuss the result

I beat Kingsrow 1-0 with 19 draws, get your facts straight. That version of Kingsrow had the 2.9 trillion positions in the 9-piece database and an opening book of 800,000 moves.

How well would you do against it?

Present at the NACA tournament were two past and present Canadian champions, some mail players, and some people who started playing checkers about 20 years before I was born. While the Canadian Checker Federation was not as sizeable as the ACF, it has master members who have taken points off of ACF masters, and I took points off of them.

Say what you want. I played 32 games in 3 days, lost none, and won 18.

I played in one tournament, and, from what I saw, checkers tournaments don't have a lot going for them.

Let me hire you as the President of my company. You are now making $675 an hour.

Here are your choices.

Work Friday and half of Saturday, and earn about $8100 in straight time pay.

Or, take off Friday, earn none of the aforementioned money, travel about 600 miles to a tournament playing site to breath smoke for hours at a time, win maybe $50 - $75 in a checkers tournament, drive another 600 miles back, then listen to more buttheads like yourself say "Gee you only won 1 tournament... you couldn't win if condition X-Y-Z happened."

Now, Mr Lopez, care to disclose your checkers tournament career?

I wonder why World Champion Alex Moiseyev came to my house on December 18, 2004?

Hmmmm... I wonder.

And what tournaments have you played in?

By the way, I won a NACA tournament in 1997 (an Open invitation held in Nyak, NY), 28 points out of 32, with no losses. It was 11-man ballot.

Did you forget about that one tournament result?

I don't know if all checkers tournaments are like this one, but there were 3 people there under 60 in my section (myself and two others) and it was not played in the best of locations. The cafeteria featured creaky wooden chairs that were very uncomfortable and not well balanced. There was some Bridge event going on not far from us, which allowed smoking, so the place stunk like old cigars (though the cigarette smokers outnumbered them probably 10 to 1, the cigar smoke was what I remembered most.)

The mens room was crowded at times, and in the summer the pungent odor of urine wafted you when you were still 50 feet away. It was enough to make you wish for a third kidney.

There were only about 40 players in the so-called Masters Section, with a surprising 60 or more in what was labeled the "Mixed Amateur" section. There were only 2 ladies present in the whole event.

NACA ran events from 1996-1999 and was relatively short lived, based somewhere in Canada. They claimed to have 3000 members, but offered no publication (just a BBS online) and only organized 11-man ballot checkers, bridge, and some other non-poker card game that I cannot remember.

Ahh yes, that's it, have been reviewing some lassie games since i am playing a few now, i know i saw one somewhere, thanks

How many different legal jumps can 1 player have on one turn, stemming from an actual game (not a constructed problem)?

I have seen, on more than one occasion, positions where I had 3 jumps choices (which is usually a little surprising since jumps are forced.)

This usually comes about in a turbulent position where double jumps have occured in pairs.

Does anyone have a game runup where there is more than 3 jumps available for the side to move?

I usually go by the name of Ed, but maybe cousin Al is pretty good too.

If anyone tunes in to Channel 6 tomorrow at 10 AM, they can see me at the Franklin-Wyndham Hotel in Philadelphia in the Main Exhbition Room. In front of me is a checkerboard, and a member of the hotel staff that will be making the moves Chinook sends via instant messenger to his laptop.

That version of Chinook has the 6-piece database only and the 1990 version of its opening book.

I will not be playing that one.

The machine I will be playing against has a boatload of RAM (32 GB) and either 4 or 8 processors (I forget which). This will not be done via the web.

Dr. Jonathan Schaeffer of the University of Alberta has agreed to let me cross swords with the strongest version of the Chinook checkers program. Once it solves the next slice of databases it is working on, it will have over 39 trillion positions solved (39,000,000,000,000).

Stay tuned for more info.

my tournament is open for all, non-exclusive. it doesn't have a fancy title, and that's the one I will be playing in.

http://brainking.com/en/Tournaments?trg=8873&trnst=0

And take a guess who this is:

http://brainking.com/en/Profile?u=15519

(yawn)

Jake Lopez: If you don't know the answer to that, you don't get the American Checker Federation bulletin. You are now on hide Jake, I will no longer respond to you. Have a great life.

I did not write the Kingsrow program, and everyone knows I beat the 9-piece database version of Kingsrow 1-0 with 19 draws last November.

I am a candidate to attend the next a.i. gaming conference in Taiwan this coming September. As such, I have been asked to perform a "peer review" of some of the papers that are being submitted for publishing. One of the papers I will be grading is being written by Dr. Jonathan Schaeffer of the Chinook team.

Dr. Schaeffer has already solved the 10-piece database for the game of checkers. The drawback is, you have to start with 5 kings against 5 kings, then do 5 kings vs. 4 kings + 1 checker, 4 kings + 1 checker vs. 4 kings + 1 checkers, etc., until you get to the most interesing database: 5 checkers vs. 5 checkers.

So, to solve the 11-piece datbase, you need to do 6 kings vs. 5 kings... and the whole mess takes forever.

I proposed a means to "jump right to" 6 checkers vs. 5 checkers, knocking off about 12 years of computation (the databases are really that large) in a casual conversation to his team in 2003. It turns out, I was partially incorrect, but they came up with a means to bridge the gap.

Using a new technique, they are tackling 6 checkers vs. 5 checkers without having to do the tremendously wasteful computation starting with the kings.

It is a fascinating paper.

A program's weakness is that it is will not play like a strong human player in the endgame.

I thought that my cited example was obvious.

EVERY program that probes endgame databases will prefer a KNOWN db result to a score backed up from a leaf node that has imperfect information associated with it.

A prorgam would throw away half its army if it could recover it with a draw.

A program would throw away a very mobile king for no reason when it is practically running you out of moves, all for the sole purpose to get into its endgame databases.

This has been demonstrated.

Greg Murray, aka Usurper, is the player named Bullet.

It is up to everyone to decide on their own. I showed a position where there is no reason to throw away a king, yet software programs do it all the time.

I am a researcher of games programming, so I am fairly knowledgeable in this area.

http://www.GothicChess.org/papers.html is one source of technical papers I have written.

I am just stating the facts after I have been given a tip.

Well, someone tipped me off that Bullet might be using a computer program, and they asked me to try and uncover proof of this. It is no easy task to deal with computer programs, and the only way to get at the root of it all is to somehow force the opponent into a situation where he leaves his fingerprints behind.

I would like everyone to take a look at "Exhibit A"

http://brainking.com/en/ShowGame?g=813374&i=42

In this position, please notice, my opponent has just thrown away his king, for no reason. My side is seemingly hard pressed to find the draw (the only way to expose a computer user is to get yourself into a scrape that you can just barely get out of, while demonstrating "inhuman" play leading to the draw from their side) but you can see how to achieve it easy enough.

I need to get 3 checkers along one diagonal (f6-g5-h4) and just trade off a piece to break the secondary bridge (f6-e7, d8xf6, g5xe7, with an easy draw thereafter).

So why did my opponent throw away his strongest resourcce, his freely mobile king, while I am "squirming"?

The answer: software always plays for the first opportunity to enter into a database position that is "known". Throwing away the king then winning back one of my pieces virtually assures this in short order.

Every other "human" on the planet would move the piece on d6, perhaps d6-e5 forcing f2-g3 or the like.

And that concludes the investigation.

[IP tracer 3.4.1]
[(c) 2005 Antimicrosoft Hackers, Inc.]
[define target layer = URL]
[URL = ?]
[http://brainking.com/en/ShowGame?g=813374]


[Trace Mode = ? {! for options}]
[__active__userhit__names@ip.reg__]
[
;;;; trace started pass 001 ;;;;]
[;;;; trace started pass 002 ;;;;]
[;;;; trace started pass 003 ;;;;]
[;;;; trace started pass 004 ;;;;]
[;;;; trace started pass 005 ;;;;]
[__domain_user_1__ "Ed Trice" ]
[;;;; trace started pass 006 ;;;;]
[;;;; trace started pass 007 ;;;;]
[;;;; trace started pass 008 ;;;;]
[;;;; trace started pass 009 ;;;;]
[;;;; trace started pass 010 ;;;;]
[;;;; trace started pass 011 ;;;;]
[;;;; trace started pass 012 ;;;;]
[;;;; trace started pass 013 ;;;;]
[;;;; trace started pass 014 ;;;;]
[;;;; trace started pass 015 ;;;;]
[;;;; trace started pass 016 ;;;;]
[__domain_user_2__ "Greg Murray" ]
[+++ trace end @ 2 users found +++]

Live checkers play is available for limited testing (until the full site is built).

Come on in and play in real time.

http://www.GothicChessLive.com


This is a shakedown to get the bugs out, sign in, play, and give feedback (look for the "Checkers Test" room under "The Gothic Room" when you sign on.)

Everyone who has had the World Checkers Champion come to your house to play a game of checkers with you, raise your hand.

(puts his hand up)