Home » Forum Home » General

Topic: rollie tesh is browsing pente.org?
Replies: 166   Views: 128,960   Pages: 12   Last Post: Mar 17, 2020, 12:19 AM by: watsu

Search Forum

Back to Topic List Topics: [ Previous | Next ]
Replies: 166   Views: 128,960   Pages: 12   [ Previous | 1 2 3 4 5 6 7 8 9 10 11 12 | Next ]
rollietesh

Posts: 110
Registered: Apr 17, 2010
Age: 51
Re: rollie tesh is browsing pente.org?
Posted: Apr 25, 2010, 11:30 PM

Thanks watsu, that was quite an effort, and much appreciated. I was able to to also dowload these games, thanks to up2ng's link to Mark Mammel's site, but these diagrams are so much easier. I wish though that it was possible to "take over" a diagram use it as a normal board to play over other moves, and then be able to return to the original game. One can do that something like that at chessgames.com.

Looking over these games is like looking over someone else's games. I vaguely remember some of the key moments and positions, like when I was on the ropes again Dan Allen, or how Dave Wallace shockingly played straight into the hammer, and then I went on to lose to him just as quickly with the white pieces. In some of the games however, I'm surprised by my opening choice, such as my final game against Hawkins. "I played that!?" I do remember that a lot of my opening choices were due to having knowledge of my opponent's history. I'd find an old game were I felt they played inaccurately and then I would play into that opening hoping for them to repeat their error. This was an especially potent weapon if my opponent had ended up winning their prior game despite their error. This increased the chance of them playing right back into it, as happened in a key game against Bill Webber.

Anyway, up2ng's comment -- "I think we pretty much agree with your claim that P1 has a forced win starting from the first move with perfect play. This may have been audacious when you first claimed it, but we've come to accept it here for a while. However, humans are not capable of playing perfectly in every match in timed, over-the-board play without use of outside resources, so Pente is still a fun game even though in the abstract it has been basically solved." -- this comment has taken the wind out of my sails, in a good way. You explained the situation very well, and I agree completely.

Therefore, I guess that the only reason to try and systematically prove forced wins in various openings is if there was some major project to write a book. But that would take too much time and the market is way too small now. Back in 1984 I briefly toyed with the idea of undertaking such a project and write a book called "Pente: Busted". When Gary Gabrel got wind of this he called me a "Judas for even thinking about it." That too took the wind out of my sails, and that was pretty much the end of my full-time focus on Pente -- exactly half a lifetime ago.

continued ...


rollietesh

Posts: 110
Registered: Apr 17, 2010
Age: 51
Re: rollie tesh is browsing pente.org?
Posted: Apr 25, 2010, 11:55 PM

... continued.

zoeyk, every one of your proposed starting positions is interesting, though in a couple I would deviate a little earlier. Any one of these positions would be fun to play. I appreciate your diagramming efforts also! That must be a lot of work, and I need to learn how to do that. I'm a little slow when first learning things, and would need a specific list of the item-by-item steps to take. Especially since right now I am unusually busy.

zoeyk also raises another interesting question about who is the best player ever, especially with such players now as you say Nososky, Barklin and richardiii. At Chessmetrics they attempt to answer that question regarding past chess players, using an intricate formula. But there are so few games of Pente to judge by, the only way to know for sure is for each of these players to play full matches while in their primes, which is impossible.

What would happen in a match between me in my prime and Nososky in his prime? All I can say is that I would be confident of winning, though he would certainly and perhaps more justifiably say the same thing. The only player that I ever thought I'm might lose to in a lengthy match was Scott Justice, though I still expected to win in a squeaker. If you all conclude that Novosky is better than Scott Justice ever was, then perhaps Nosovsky is indeed the best player.

There are some things I realize that I am actually better at now, while playing Pente again, such as systematic discipline. But there are other things I am weaker at, such as memory. Still, I wonder how I might fair against Novosky, even today, given a relaxed time control, and time to prepare?

Anyway, what do we do here and now? I would be interested in a playing a collaborative game, at our mutual leisures, though only one game at a time. It can be any opening. One of zoeyk's, or up2ng's suggestion of 1) ... N8. Is that still called the "Allen Opening"? That opening for example might be a fun one to demonstrate a forced win against.

You guys decide what to do, and thanks again.

PS - Page 6 I think is too buried with diagrams I cannot get most of watsu's diagrams to load. zoeyk, I wonder if you can delete the diagram series of all possible opening moves? We can focus on one game at a time anyway. Thanks.

Rollie

zoeyk

Posts: 1,964
Registered: Mar 4, 2007
From: San Francisco
Age: 41
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 25, 2010, 11:58 PM

me and u posted at the same time, i think u didnt see all the posts on page 6.

any how, writing a book, yes ive been slowly trying to work on one with the help of the community. there is a private forum youll get a invitation too soon. i think at this point, the book is not to solve the game and make it obsolete, but more over to enhance the game with a mixture of perfect play and human error, physiology, logic, and other things. you dont want to hand them a life time supply of cryo-packed frozen fish and call it a day, it be much better to give em a spear and teach them the art of spear fishing, or something like that. give them tools to find the answers, or give them tools to make better tools with by them selves through there own imagination, but dont give all the answers, give hints, with out plainly showing how the magician pulls a rabbit, and all his other tricks to boot.

define solved rollie, means knowing which side has sure win from the start in perfect play? i would think that was obvious when the game was first created, since white was ahead one move, and its literally impossible to fill the board with a tie with out the board imploding with captures. does it mean a systematic formulated check list of geometrical laws P1 simply follows mindlessly to find victory? we have yet to create such a check list. does it mean P2's options are limited to a finite small number that can be contained and memorized in ones mind or written in a book known as all answers youll ever need to know? impossible i think, P2's options are so vast a 6 inch thick book could not contain all this data.


and, again, if K10,N8, then your second move is ?

thanks Rollie,

z

Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
zoeyk

Posts: 1,964
Registered: Mar 4, 2007
From: San Francisco
Age: 41
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 25, 2010, 11:59 PM

ah boy we keep posting at the same time LOL.

Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
zoeyk

Posts: 1,964
Registered: Mar 4, 2007
From: San Francisco
Age: 41
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 12:11 AM

if you want to play nosovs in a relaxed time frame, you will be required to play him Turn Based at brainking.com.
he does live play here, and turn based over there.

if your interested in that, we can give you instruction for how to set it up. turn based is where you can for example make one move every 30 days, or every 1-2 days, or even several moves in a day. turn based is for relaxed time control. but since your familiar with chess sites i must assume you aware of what turn based is.

in order to be known as the best player, in my opinion, you must master live play too. a turn based master does not mean best player in my opinion. many low level players hit masters status in turn based because they use resources. where in live play this factor is much less.

i would like to see you do both, when you are ready and up to it, and when you find it enjoyable to do both.
nosovs in live play will get your heart pumping i tell ya! hehe

Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
zoeyk

Posts: 1,964
Registered: Mar 4, 2007
From: San Francisco
Age: 41
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 12:31 AM

ok, i have edited the game post on page 6, K10,N8, your move...

Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
rollietesh

Posts: 110
Registered: Apr 17, 2010
Age: 51
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 1:59 AM

Hi zoeyk and all,

Thanks again for all your efforts.

If Novosky is as good as he appears to be, and any games between us were to mean something, then I would need time to prepare. For one thing, he is likely going to play error-free. Therefore, everything will come down to innovations targeted as kinks in each others armor. I would need time to review his previous games, and he mine (I presume). Perhaps sometime later, if he is up for it, but only when I'm between design project and can study for awhile.

I'm still not getting Page 6 to load right, so let's start fresh here.

After 1) K10, N8, there are a few moves I might try. I would not play 2) N10, as in the old days. I sense that an immediate win here is to play away from black's stone, or to play something unusual like 2) N11 and work towards a scissor. Then I see in your database that 2) N11 has the highest winning percentage. Not wanting to copy these kinds of database stats, I'm instead going to play the somewhat more normal move:

1) K10, N8
2) G10

BTW, I just noticed something. Is there no "I" file on the board?

Rollie


Message was edited by: rollietesh at Apr 25, 2010 8:00 PM


watsu

Posts: 988
Registered: Dec 16, 2001
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 2:20 AM

Correct, Rollie no I. At some point in the alphanumeric development, it was considered to be to close to a 1, so left out. Who would confuse a move in the middle of the board with one on the edge?
Ah, well. We're stuck with it.

I don't want to play non tournament rules games. If you take one of my unrated invites, play tourney
rollietesh

Posts: 110
Registered: Apr 17, 2010
Age: 51
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 3:26 AM

zoeyk, I tried to accept your game invite. Bare with me as I must be doing something wrong. I'm expecting to see a position with 1) K10, N8 already played. If so, my move will be 2) G10

watsu, thanks for the info on the missing "I" file. I'm still not sure why you guys prefer this notation over the old-school R,L,U,D system, but I'll try to adapt

watsu

Posts: 988
Registered: Dec 16, 2001
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 3:33 AM

Rollie, on turn based Pente games (which I assume was what the invite was) there is no way to start a game from any position except for the 1.K10 opening. You and Zoeyk will have to proceed from that point, and until Z makes the N8 move, you will just see a game with opponent's turn board- with just the K10 stone placed.
As for the preference, I think most of us prefer not to have to count intersections out from the center in both directions. Looking out to the edge of the board in two directions to see the intersection I at least find easier than going ok that's up 5 and now how many right? maybe it's simply a matter of being used to it, though i never found it easy to use in the old Pente books either.

I don't want to play non tournament rules games. If you take one of my unrated invites, play tourney
up2ng

Posts: 542
Registered: May 9, 2002
From: Northeast USA
Age: 38
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 9:28 AM

I'm glad this discussion is still going strong! I hope that my comments didn't take the wind out of your sails too much Rollie! Knowing that P1 has a winning advantage in Pente should not stop it from being fun to play. I think playing the game, especially live time-constrained games is an excellent test of skill and continues to be fun. In addition, analysis can be fun also -- but not from the point of view of "who should win", so much as "we know white should win, so if white did not win, where was the error and what improvement will guarantee victory?" Typically this analysis involves working backwards from the end of the game and finding the first place where white appeared to make a game changing mistake. Then, an improvement is tried and then black has many ways to try to thwart that improvement. Many times it doesn't work out, and black finds a sure win. So, we must go back even further in the game, repeating this process -- sometimes all the way back to the opening before white's losing move is found and corrected. I still find this type of analysis interesting.

Zoey, I doubt very much that it was known that the game had a built in win for Player 1 when it was created. That is an extremely advanced concept that players do not come to the true realization of until they are playing at a very high level and have played for a long time. Many relatively good players will argue up and down that it is not true still -- until they get just a little bit better. You didn't believe this to be true when you first started playing and neither did I. Not only that, but the tournament rule was created to "eliminate the P1 advantage" when it was finally universally accepted that P1 has a winning advantage with no opening restrictions. Of course, we now know this was not a true solution, it was only effective in creating a more complex game, which humans would have a harder time playing flawlessly. Other variations, such as keryo-pente, d-pente and others were also invented with the goal of creating an even game -- and yet none of them can be. The only solution to creating an even Pente match was FINALLY adopted here at this site very recently, which is to consider a match to be a sequence of two games, where the players switch sides to play the second game (a set). The expectation in such a match is a draw and so that entity has no built in advantage for either player. In fact, it is amusing that a world championship of pente back in the game's prime would be designed so that a coin flip determined which side each player would sit at for the 3rd and final game of the match! Of course you cannot have live, over the board tournament matches lasting forever either so it's a tough situation. Plus, after all, the NFL still decides who gets the ball in a sudden death overtime with a coin flip...

-------------

Ok, so this current game with Rollie -- instead of trying to create a turn-based game, maybe it would be better to just continue showing the position on the board here in the forums. Maybe zoey can create a board here and then just continue to edit it as we go (instead of creating more and more boards in this thread -- as Rollie said, after a while older computers will have trouble loading the page, I was having that issue on page 6 before also). If someone else, like zoey, creates the board, I will also be able to edit it if I'm around -- but if I create it and then I'm away for a while, no one else can edit my posts so I'll leave it to someone else to start it off.

Rollie, very interesting of you to purposely go away from the database results. It's one of the reasons that I pushed for N8, because I had a strong feeling that you would not respond with N11. (I actually do not personally prefer N11 either for what it's worth)

After all, database results can be misleading. For example, just punching in the position and seeing that N11 has the highest winning percentage -- does this mean that N11 is the best move? Perhaps N11 appears to be the strongest simply because only the best players play it, and not the other way around. (I'm interested to see if people will misread what I've just typed there, you really need to read it closely and think about it to catch what I'm saying.)



Ok, after 1. K10, N8; 2. G10

My own personal style is to strongly consider 2 ... M9 although there are at least a half dozen reasonable tries here. (Note, I tend to not use the database either, even in "turn-based" games like this -- so I honestly don't know how proven my suggestion is statistically, but I know why and how I would play it out.)

Looking for lots of ideas on this one -- after all, the first moves of the game are the most important.

zoeyk

Posts: 1,964
Registered: Mar 4, 2007
From: San Francisco
Age: 41
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 26, 2010, 11:10 AM

to up2ng, i had a feeling you would go with a draw attack for the opening. im more inclined to lean in the direction of N6 for P2's second move. those are both good, just 2 different styles. i have a feeling the draw move is easier to read than N6 is. i hesitate to go into details further in front of rollie.

we should meet up in the lobby and go over game plans and come to a vote on the approach. since this is a community team effort to beat rollie.

i will make a new thread for this game. since it will surely become lost in this now vast growing thread.


to rollie;
yes the N8 is known as the allen's opening.
but N8 is easier to type

and the missing "i' line. is because I and l look too similar. (note that l is actually a lower case L ), and not a number one,..1. if i say I9 it can look like l9. this I9 L9 confusion is the reason as far as i know.

and, one more reason, there is no I in pente, its a game of 2 or more people, hehe. ok bad joke.


Message was edited by: zoeyk at Apr 26, 2010 11:54 AM


Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
up2ng

Posts: 542
Registered: May 9, 2002
From: Northeast USA
Age: 38
Re: rollie tesh is browsing pente.org?
Posted: Apr 27, 2010, 12:05 AM

Hey zoey,

I just read my email and it looked like you were a bit disappointed that Rollie had checked out the database, but by the time I saw your edited post here it seems like maybe you changed your mind about that.

I agree that using the database during a game makes it even harder for black to win. To be fair, Rollie has always been a player who plays with a high amount of "preparation" for his games, one of the reasons he became a champion. He often used all of the resources available to him, including known previous games played by his upcoming oppponents, etc, and that would literally change his approach to the line that he would play for that game, even with the white stones. That's not generally how we play white around here, since we are so confident that white has a sure win and that we are destined to always find it that we don't bother to adapt our white lines to our opponent, but that is how he has always done it. So, using the database is a logical step.

This is generally why I am such a huge fan of live, over-the-board play with a timer (or live, online play with a timer). Research and analysis are not allowed, and if someone tries to do some research anyway while playing on line, the timer limits the benefits that can be gained. It is virtually impossible to beat a strong player's white in a turn-based game. (Although it does happen)

We should not take this so seriously anyway. This is just an excercise of playing out an "exhibition" game with a great champion.

Hopefully some other players will chime in soon on their thoughts for a 2nd move. If not, I will try to catch you in the lobby and see if we can work something out. If we cannot coordinate that, just communicate with PM to discuss strategy.

I'll leave it to you to lock in the final moves. Hopefully we will have a 2nd move within a day or two.

watsu

Posts: 988
Registered: Dec 16, 2001
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 27, 2010, 12:31 AM

Going back to your earlier post about relative ratings of the all time best players for a second Rollie...
I had the opportunity to chat several years back about that question with Gary Barnes (his profile here is progambler) and his reply at the time was basically it depends on how you define the time setting. His pick for strongest at timed thirty minute games was Istvan Virag (profile virag) and for two days per move games it was Dmitri Krasnonosov- who didn't play much here (as far as I know, only a few games as student) but who does have some games in the database here from another site, IYT. Unfortunately, most of his games- along with everyone else's were lost to the public when IYT's database crashed several years ago.

The only games between nosovs and sjustice I can find right now are turn based, and in the limited selection available I see 2 wins apiece. Since Scott has said that he's a slow player who needs time to analyse things fully and these are tournament games during an active period for scott, we'll assume he was more or less in his prime for them and playing in a time setting in which he could perform well. No way to tell by those results who was better between them at that time. Scott won 5 of 8 in the turn based games I can pull up against virag with 1 player 2 win, nosovs 6 of 9 against virag turn based with 3 p2 wins. Which doesn't necessarily mean nosovs is/was better than Scott, of course.
Dmitri Krasnonosov, in the turn based games I can pull up against virag was 9 for 11 with 3 p2 wins agains virag, playing as p1 in six games, so call it 8 for 10, since his p1 losses were miniscule.

FYI- richardiii had not yet reached his prime when I had this conversation with Gary, so I've confined myself to trying to work within a fairly limited reference frame for ease of comparison. Other players who have reached or approached the top of the pile over a long period will probably have a better first hand knowledge of the relative strengths of the top players, but my main point is this- there is/was a select group of say top 10 all time online players in both real time and turn based games, Gary Barnes was IMO definitely in that group and played a lot of top players from sjustice's prime era. I think he'd have said that nosovs is at least as good as scott in his prime was.

I don't want to play non tournament rules games. If you take one of my unrated invites, play tourney
zoeyk

Posts: 1,964
Registered: Mar 4, 2007
From: San Francisco
Age: 41
Home page
Re: rollie tesh is browsing pente.org?
Posted: Apr 27, 2010, 12:37 AM

well, i wasnt concerned to beat rollie's white in turn based. that wasn't the main point of my goal here. i just wanted to see how a champion 1984 style player with no modern influence would approach modern day openings. once the pandora's data base box is opened its hard to forget what you saw sometimes. as watsu put it "it kind of defeats the purpose of the exercise."

and, i kind of wanted to keep this in private channels. too late. any how, i plan to use moves that might have the solution shown in the data base, so if it is viewed, then little point. the N8, G10 is overly documented. its only fun if he doesnt have that resource i suspect. if we do K9 instead, i do have some new novelties how ever..

any how, no biggy, all is ok and cool. im just going to go with the flo here.

Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Replies: 166   Views: 128,960   Pages: 12   [ Previous | 1 2 3 4 5 6 7 8 9 10 11 12 | Next ]
Back to Topic List
Topics: [ Previous | Next ]


Powered by Jive Software