Re: rollie tesh is browsing pente.org?
Posted:
Apr 24, 2010, 3:00 AM
up2ng,
Your 11) M9 suggestion, along the same lines as the actual game, may very well be the most direct win. Was this analysed elsewhere? Sorry if I overlook anything posted before. I guess the idea is that after 11) M9, the move O9 can be answered by 12) O11, L9* 13) M9, P12 14) P11 is that it?
I'm also seeing a forced win with 11)H11*, but you are right that 11) M9 could be faster.
So, are you all now claiming that P2 loses demonstrably after the 2nd move
Posts:
2,220
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Re: rollie tesh is browsing pente.org?
Posted:
Apr 24, 2010, 7:15 AM
i will say this, im almost 100% convinced that P2's 4th move is a losing one, and P2's 1st and 2nd moves are losing ones we know, so now i just wonder if whites 3rd is sure win or not. P2's 3rd is forced how ever, this means im suggesting a alternate P2 4th. and the one that i currently find interesting is the capture. im 99% sure this loses, but its worth a look perhaps.
here's a game me and watsu did recently;
and here's an alternate line i looked at;
in these games i already see whites first error, i'll let you try and see it. i will further say that P2 has many choices for the 5th move that are worth looking into.
with all that aside tho, i look forward to playing the other half of our set, and starting the next set as well, i have many openings i would like to test on you. you have been away for 26 years and im fascinated to play you, because it really feels like im reaching through a window to 1984 and testing 2010 openings against the pente awareness of 1984. my curiosity is to see "have we really come very far from back then?" i mean sure, we have really perfected our new P1 paths, and found countless P2 lines that are amazing, but we don't know how they would had played out vs you. Rollie, you are a window into a preserved past, a bridge into a time very few can reach for answers without the help of people like your self. a 1980s world champ that has never seen the pente.org data base before,...priceless.
let me know when you are available for playing, and hopefully you will find time to play some other players, there is a long line of players just itching to play you hehe.
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Posts:
542
Registered:
May 9, 2002
From:
Northeast USA
Re: rollie tesh is browsing pente.org?
Posted:
Apr 24, 2010, 8:12 AM
Watsu, yes in that other line 20. H13 attacking the pair looks like it works out, black has no way to force a 5th capture and cannot defend its own capture trouble.
Rollie, yes a few of us took a look at a lot of possibilities together in the game room and not all of those results were posted here. 11. M9 after 10 ... N13 was found to be good and so 10 ... N13 was dismissed. I think your idea works also for white. So yes, we think we've shown now that white wins this line all the way through, going all the way back to the 4th move.
In my opinion, looking at a different fourth move is pretty much changing the opening and is really no longer an analysis on the current game in question. Plus, looking at things so early on means that there are many many possible paths to explore to see if the moves to that point were solid. Too many really.
In the variation watsu and zoey looked at, a new fourth for black is chosen, to capture. Even here, maybe there were even more 4th moves black could try such as oddball moves like K11 or H11. Black is probably behind at this early stage no matter what so he just has to pick something and see what happens, and these all lead to completely different games.
In this variation, we assume white's response on the 5th is to capture back ... are we sure? If so, I really don't like black's 5th. This is too passive, nothing lines up for black from there and he is also spread out. I assume that the 6th for white was not the best move and even that could probably be made to win some other way.
I guess I'm saying before going too far down that rabbit hole, just know that there are endless possibilities when you start looking this far back -- which is what makes pente still interesting to play. Once the opening is well established and we're into the midgame, now black's moves are generally forced with very limited choices, and white has enough structure so that it should be straightforward to see which one or two options should be best, so a relatively small number of scenarios can be analyzed to determine that white is on the right track.
For what it's worth, I'm confident that white is still winning after black's new 4th, and very confident that white is winning after black's 5th. I won't suggest any improvements just yet.
Posts:
2,220
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Re: rollie tesh is browsing pente.org?
Posted:
Apr 24, 2010, 10:46 AM
to up2ng; i admit it was indeed a passive looking position. there are better ones, but i thought it good to let others discover and point out the stronger ones that perhaps have something more in the way of,.. conductivity? i pretty much agree with every thing you said there basically. and yes, other than a new 4th, i think we have beaten this opening to death now. at some point ill try to recreate my solution you had requested, showing how i use the (3 x split 3) to force the win.
_________________________________________________________ added- to Rollie; ahh ok, just re-read the previous page again, i think your asking to take the actual game (moves we did) and do a step by step summed up version of analysis. i'll work on writing the P2 side of it for each move i did and what i was thinking at the time, and ill leave it to you to take care of the P1 side of it. then we can just single shuffle them like a deck of cards if you know what i mean.
or perhaps you just meant to try out a new line someone was suggesting, that up2ng has since stated is now moot by way of alternate solution found.
Message was edited by: zoeyk at Apr 24, 2010 5:17 AM
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Re: rollie tesh is browsing pente.org?
Posted:
Apr 25, 2010, 4:06 AM
Comparing old-school opening theory with current opening theory is indeed intriguing. For one thing there is a lot more of it today. I'm curious about what has since been discovered in some of the key lines left hanging after the 1984 collapse of interest in the game. Thank you for resurrecting that interest.
There are, however, some opening conclusions reached during the tense energy of the "golden age" of Pente that I suspect have been forgotten. Then again I'm not sure because I've forgotten most of it myself.
Is there a way in your excellent database that I can research old games? For example, can I bring up all my old games somehow? Somewhere around my house I have old Pente newsletters, but I'd have to dig
At this time I'm mostly interested defending my claim that P1 has a forced win in all lines. In fact I'll go one step further and make the audacious statement: "I cannot be beaten as P1" This is perhaps a silly statement considering that I lost as many games as P1 as I did with P2 during the 1983 cycle. Nonetheless, it is a fun claim for you to try and disprove. I would enjoy that humiliation
If everyone is up for it, then my first move is 1) K10. See if you can beat me. Let's focus on one game at a time, one move at a time, with a 'comfortable' time control. Some days I'll can make a few moves in a row, other times I might need to pause for a few days.
If you are all up for it, then collaborate on finding what you think is a most challenging defense. I believe that I can demonstrate a forced win against all defenses. One caveat: if I do get beaten then I claim that I can quickly find my error and replay the correction to a win. If I still cannot demonstrate a quick win with my second effort, then I will eat my words, and you guys have saved Pente
My first move is 1) K10, what's your collaborative best shot?
Rollie
PS- Thanks up2ng for the attempt to teach me how to create these cool diagrams on my own. I took a look at the page you linked me too, but I'm not seeing any buttons to click to create anything. Does one need to be a member to be able to do that? Or can you point me to the exact place that explains things? Thanks.
Posts:
542
Registered:
May 9, 2002
From:
Northeast USA
Re: rollie tesh is browsing pente.org?
Posted:
Apr 25, 2010, 8:40 AM
Fun variations zoey,
In the first one, the obvious choice for white that comes to mind is to cover up the double cap at N8 (maybe white should have already played an early extension to K7? Not sure...) This move protects white's tria from immediate capture and now black has to deal with that tria. Meanwhile, white threatens 3 extensions along the seven line for a likely victory. Black probably responds with L11 and N12, but then what? Black probably defends by attacking the keystone pair by playing O9, and I believe white can then get away with defending that pair with L12 and still be in a winning position, although it is certainly not a direct forced win from there. Nice one.
Second variation at first glance looked identical to the one we've already looked at and was illustrated by your match with karlw, but the high block on the 5th changes black's defensive position slightly. Unfortunately black still has nothing happening and has to play defensive from there since white does have the L-line and 11-line working. Defending at J11 just doesn't help black very much. Black could try M11 to be a bit more aggressive but white still seems in control after a bit more jockying for position.
Even better in this one, again, seems to be to not play J9 at all and have white move directly to H11*. If black attacks the pair with J9, white has pretty decent options from L9, and if black does something else, white now has the attack at H8 and then H9 available.
The third one is tricky but I think white's in pretty good shape after 6. N9
Rollie - Talking opening theory with you would be a lot of fun. I hope that you can remember what some of those "key lines" were at the time, and I am sure there have been discoveries and solutions determined for those lines since many of us studied those games from the old newsletters as part of our learning process, and extensively played some of the openings that were common, such as "the wedge".
I'm not sure if any of those old games ever made it into the database or not. And I don't think documentation for ALL of your games during that period has been seen. But a pretty good collection of those games can still be reviewed from some of the old newsletters that were preserved. I am constantly losing the link to that foreign renju site which has links to a few of the tournament newsletters such as the '83 Championship. Perhaps watsu can post it here again. (Note, I hope that someone takes it upon themselves to save a copy of those files -- perhaps Dweebo could host them and create an area at this site where they can be viewed...)
In addition, if you go to Mark Mammel's site, here:
... you can find a link to the data files for most of the games that were played during the '83 championship. If you then download his WPente program (the AI for this program is the computer AI used here by the way), and open these files, you can scroll through the moves from those games as they were recorded.
Side note, if you want to see another place to get nostalgic about old pente materials, check out a couple of your OWN puzzles at this page!
Anyways, 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. I personally wish that some of the variations offered at this site, such as D-Pente which involves swapping in the opening, would gain some momentum since we'd see much more variety in play and more difficult openings to analyze, which should help balance the percentage of wins for each side, at least more than the current standard of Pente with the tournament rule. Even in this case one side or the other always has a sure win, but the higher the complexity the greater the chance for humans to play a flawed game.
Side Note -- As far as creating these diagrams in the forums, unless the link I posted is broken for you for some reason, it should explain the instructions for creating them. You'll note at the end that Dweebo has plans for creating a more user-friendly way of doing it, like point-and-click to place the stones on a board, etc. But, for now, you actually have to type out the moves according to the "code" shown. For example, using the [ game ] and [ /game ] tags (without those spaces) and listing the parameters and moves exactly as shown. In your message it will look like text, but when you hit the Preview button, if you've done it correctly, it should show the board. If it does, click Post Message and it should work.
------------
Your challenge of seeing if you can be beaten while playing white will be fun. I vote for 1 ... N8.
Posts:
2,220
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Re: rollie tesh is browsing pente.org?
Posted:
Apr 25, 2010, 10:27 AM
well on one hand, ide say K9 is strongest, and N8 is next in line of the 12 total moves...
L9 comes in 3rd, or tied with N8, although some would disagree,.. any how,
heres what i would like to know, i want to know what are your second move answers to all 12 first P2 moves. after we know this, it will become more clear possibly as to what first move of black will work best on you.
so,
if;
K10,K9,N10 (<---we know this now)
K10,L9,?
K10,M9,?
K10,N9,?
K10,O9,?
K10,M10,?
K10,N10,?
K10,M8,?
K10,N8,?
K10,O8,?
K10,O7,?
We await your answers.
_______________________________________________________ we might not get you in the first attempt, with enough games i think we can find a error to exploit tho, but you know with hours or even days to make a move it becomes much easier. if you played 10 games per player vs 10 of the worlds top players in rated 20 minute live sets, ide be absolutely blown away if your P1 didn't fall at least once.
you have yet to play Alexander Nosovsky from Russia and Viktor Barykin from Russia and Richardiii from USA these are the worlds modern day Grand Masters of Pente that are currently active.
Players like me and karlw and up2ng as just a few examples are a slight step below their level. but we are gaining on them
the depth they can see in a minute might take us a week to analysis. they are faster(although i will mention karlw is a speed pente champion), and often times a bit more creative. any how, at some point i will write you a list of top active players that you simply must play at some point.
Message was edited by: zoeyk at Apr 25, 2010 5:22 AM
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Rollie, here's the link up2ng mentioned which has some old newsletter pictures. http://renju.se/rif/pente/pentgifs.htm We were (a few of us here) in contact with Tom Braunlich a few years back and he said he had copies of all of the old newsletters which we could digitalize, but to the best of my knowledge that project never came about, and so the link I've posted above remains about the best us online folk can do to see some of the old theory, analysis and games. That, and as up2ng mentioned Mark Mammel's 131 games from the '83 Championship, which probably the closest thing to what you are looking for. The database here primarily consists of games which were actually played online, so unless someone has replayed a game of yours online- which does happen- it won't be in the database but likely if it has been replayed it won't have your name on it so you won't be able to find it in the database, unfortunately. In my forum post here, http://pente.org/gameServer/forums/thread.jspa?forumID=27&threadID=4556&messageID=19625#19625 I transribed part or all of three of your old games using the link to post boards which up2ng gave, but it's a laborious process involving playing the game through on a board with alphanumerical coordinates and then putting those moves into the [ game ] [ /game ] brackets. Then finding and correcting any errors you may have made the first time, so that you actually get what you intended...
Message was edited by: watsu at Apr 25, 2010 1:23 PM
Retired from TB Pente, but still playing live games & exploring variants like D, poof and boat
I was just thinking that I rarely saw 1 ... N8 used in the old newsletters and so his game might not be as fully developed against that move yet. We know now that N8 is one of the strongest (most complex) choices. I think that K9 or L9, although more standard, and maybe stronger in the strictest sense, probably just plays into his strengths. Plus, he has already played a game with you against K9. He might not be comfortable giving away all of his 2nd moves, although at some point that will make for a great discussion. I'll leave the final decision up to you.
Okay, I hope I've done this all correctly... I used Mark Mammel's records of the '83 Championship games to make boards for all of your games played in Boston in '83, Rollie. Hope this helps some.
Round 1
Round 2
Round 3
Round 4
Round 5
Quarterfinal
Semifinal
Final
Message was edited by: watsu at Apr 25, 2010 5:40 PM
Retired from TB Pente, but still playing live games & exploring variants like D, poof and boat
I found the newsletters again and reviewed these games using WPente and you have posted them all correctly. The only minor thing -- in the 2nd Quarterfinal game (of 3) against Allen, Tesh was playing the black stones.
The original newsletters, containing analysis of some of these games can be found from a series of links from this website:
EDIT: Watsu, I hope you didn't go through the pain of translating from the other style coordinates for all of these games since the txt files used for Mark's WPente collection of '83 games contains the alphanumeric coordiates already and so it's a MUCH faster editing process to get it into the format to post here on the forums...
----------
By the way, I guess I was wrong, Rollie did in fact face 1 ... N8 in a couple of games in that tournament. In light of this ... I STILL vote for 1 ... N8!
Message was edited by: up2ng at Apr 25, 2010 4:46 PM Note to Watsu...
well im not so sure, i bet there are some P2 second moves after K9 that hes never seen before, for example i'll throw out karlw's famous K9 M8 terror of a opening as an example. that opening in his hay day was a beast!
but, im flexible, since this is a community vs rollie tesh project. ide be interested to hear some others opinions on it. watsu? richard? nosovs? brf? s3v3n? jayhawk? karlw? ext ext...too many top players to list
ok rollie, out of curiosity, if K10, N8, then your 2nd is ?
or should we walk into this blindly like a roll of the dice hehe. ill say that we are used to being able to look at a players P1 moves to select a weakness to attack. i havnt much to go one here as far as targeting P1 moves in advance. but i suppose thats a part of the challenge?
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
