Posts:
2,395
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Smoke and mirrors vs....
Posted:
Oct 3, 2026, 2:02 AM
Ok so karl, i brought up with owen the concept of p2 being smoke and mirrors. And, he detests that thinking. Prefers another. And I thought it would be a very interesting conversation. Because I feel like you would both agree and disagree on stuff. Like usual lol. Yeah im stirring the pot maybe. Let's see.
Owen, tell Karl what i was saying about smoke and mirrors essentially, and what you said.
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Re: Smoke and mirrors vs....
Posted:
Oct 15, 2026, 2:22 PM
Three blocks. 1. My reply in PM was: > Hate this approach in general, why not to play approximately equal positions where you need skill and calculation instead of smoke and mirrors? --- 2. More.
Imagine a chess game with black player being handicapped to a rook. Imagine there is a community around it. These chess players just don't know black can put two rooks on the field so they play with one. Full set of white pieces vs black pieces without a rook. Also they have a rule: bare king loses immediately so draw is impossible. They call it chess-pro. They want to make the game fair so if two players meet, they play 2 games: player1 white and player2 black, without a rook, and then vice versa. And they consider that a strong player can win even without a rook. Well, it is true to a certain extent.
Then a conventional chess player comes. And they say: show us your power, win us without a rook! Learn smoke and mirrors to distract your opponent, yes, you are one rook down, but true master can win it! Go on, kid! How do you think, what a reaction may be? Maybe he will try to tell them that playing without a rook makes the game imbalanced ? Maybe he will try to show them at least a step further to balancing the game? If bare king loses and there is nothing to do with it, well, we can at least give black some more material, a bishop or a knight (if we can't give another rook). Still it is better than chess-pro. Not sufficient but definitely better towards balance. And swap2 analogue may be even better alternative since now player 1 is interested in making the position as balanced as possible when playing the opening.
Hope the idea of this comparison is clear.
--- 3. And finally. The last 10+ real-world tournaments in pente were played using swap2 rule or some variations of Tarannikov rule. When was the last tournament by pente pro rule? How many players were there?
Gomoku encountered the same situation. The last tournament by gomoku pro was played in 90s (WC was in 91 then some minor regional tournaments occured), part of gomoku elite was resistive to any changes but finally swap2 won the race: WCs are resurrected since 2009, people play gomoku swap2 a lot IRL and on numerous resources - playok, vint, gomokuquest, renjucaffe, logicgames, ..., ..., ... And there is one community, namely Gambler (seemingly dead by now but very active in 00s), which still uses gomoku pro rules.
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 12:54 PM
As someone who has played both a lot of pro Pente and swap2 Pente, I think where the analogy breaks is that there's no good way to make the bare king not lose in Pente. But, I'd like to hear from someone who sees further than I do (if he's willing to chime in here) on how far swap2 might go towards replacing the missing rook for pro Pente. Pente_gon?
ETA: I'm not contending that there's no way to introduce draws into Pente - there are some. What I'm saying is more to the effect of there's no known way to introduce the amount of drawn in principle with best play by both sides from the opening needed to reach a swap2 gomoku/renju/checkers/chess level of balanced positions. One might be able to add a bishop or a knight to one side or take a queen from the other side, or even manage to play a game so well from the disadvantaged side that neither side has a win if draws are introduced (unlikely but theoretically possible). The less imbalanced swap2 Pente rules might help, but so far I've seen very few top swap2 gomoku take the game seriously due to lack of balance.
Message was edited by: watsu at Oct 24, 2026, 11:31 AM
Message was edited by: watsu at Oct 24, 2026, 11:40 AM
Retired from rated TB Pente, but still playing live and TB variants
Posts:
62
Registered:
Sep 25, 2021
From:
Boston
Age:
57
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 1:58 PM
Owen I think the chess-pro analogy is really helpful.
Building on it, let me propose an additional thought experiment. Try to imagine for the moment that Pente had never been invented. And then someone -- let's call her Suzy -- comes along and proposes a game with Pente's basic mechanics: Two people take turns placing stones on a board; you win if you get five-in-a-row; you can capture pairs by flanking. But they haven't yet fleshed out those mechanics into a fully specified game: no rules (yet) for how to start; open question whether you can win by capture and if so how; open question how exactly capture works (e.g. whether you can play into a capture); open question whether there should be additional rules constraining move choices for either player; etc. Suzy comes to a community of Very Smart People who are invested in these kinds of games with a request for help: "Assuming the basic mechanics can make for a deep, rich, interesting game, how should I flesh them out in order to do so?"
Owen I take it that you think: Clearly *not* in the way Pente (even with pro rule) works. And Karl et. al., I take it that you *agree*. Assuming so, it's a good and interesting question what the best advice to give to Suzy is.
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 2:27 PM
"Well, Suzy, we've now tried pie rule, capture two + capture 3, captures don't win, restrictions on the first player's second move, exactly five needed to win, playing into captures, playing into captures forbidden, smaller boards, unbreakable five needed to win, and various combinations of those. So far, there seems to be no real consensus on the best way to flesh this idea out, because as far as we can tell a bare king still loses. We can keep thinking about it, run some AI tests for ideas and comparisons, etc. but..."
Message was edited by: watsu at Oct 24, 2026, 12:28 PM
Retired from rated TB Pente, but still playing live and TB variants
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 6:38 PM
Regarding the debate "pro Pente vs swap2 Pente" or "how to make pente more balanced".
I think a first important question is the following. Are human capabilities sufficient so a well-trained player can, from any given position: 1) play perfectly as both White and Black? 2) evaluate accurately which side is winning?
Obviously the answer depends whether we are talking about turn-based or live games. I am definitively not qualify enough to answer these questions myself, and I would be very interested to hear what top players think.
If I had to guess, I would say that most top players think that both 1) and 2) are doable in turn-based games, whereas only 1) is doable in live games. Am I right?
@Owen, based on previous posts, I get the impression that your opinion differs somewhat from this. Correct me if I'm wrong, but if I remember correctly you once wrote something similar to "1) and 2) are possible for human when playing pro Pente because there are too few effectively playable branches in the game tree that the human brain can remember and play all of them accurately". Obviously, if this is your belief, then the swap2 version appears to be way better than the pro one. But I suspect this is where your opinion diverges from that of most other top players on this site (in the sense that they may think that 1) and 2) are doable, but for different reasons).
--------
If my guess about the answers to the question above is correct, then I think that most top players would agree that, in turn-based games, the swap2 version does not really solve any of the well-known problems of the pro version, apart of course from allowing a greater variety of playable positions.
Moreover, in live games, if 1) is doable but not 2), then the swap2 version, while solving one problem, might introduce another one: the opening phase might be way too important compared to the rest of the game.
After P1 plays the first three stones, unless P1's opening is bad, P2 cannot decide accurately if it is preferable to play as White or Black. So P2 must play two additional stones very wisely, probably spending a substantial portion of his available time on this decision. Then P1 must decide carefully which color to play, probably consuming a large amount of his time as well. From there the game could then be played relatively quickly, since both players are able to play accurately with their limited remaining time.
I might be wrong but I think this could be a real drawback of the swap2 version at the highest level. For instance, I've already heard of people who do not really like to play swap2 gomoku for this reason. In case there are around top gomoku players used to play live swap2 games, can this become a real problem, or am I misunderstanding how these games are played?
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 6:56 PM
About the "smoke and mirror" concept.
Although no definition has been given, the name is sufficiently explicit that I think I understand roughly what it means.
If the answer to 1) from my previous post is yes, then I understand why Owen does not like it. In that case it would just be a technique used by P2 to trick a P1 player who is not yet capable of playing perfectly by deliberately creating a more complicated or confusing position in which P1 might fail to find the winning line.
Although mastering such a technique might require considerable creativity and skill, and might also allow a top player to obtain a better win rate against non-top players than other top players, it seems irrelevent in games between two players capable to play perfectly.
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 7:32 PM
Finally, about the comparision between pro Pente and Chess with one rook for P2. I understand your point Owen, but your hypothetical scenario is exaggerated.
To be more accurate, chess players with one rock for P2 should have known for a long time that chess, chess960 and other variant exist. They should even have implemented these variants on their dedicated website for chess with one rook for P2, with regular turn-based tournament being organized, even if chess variant tournaments attract far fewer players than those played with one rook for P2.
There is an interesting question here: Why do they still prefer to play chess with one rook for P2 rather than one of these apparently more balanced alternatives?
From my own experience with Pente, pro rules seems to do the job reasonably for casual players. I just checked the database to test this belief, and among 541,085 turn-based games in which both players were rated below 2000, P1 won ~51% of the time (while among 26,738 turn-based games in which player were rated above 2000, P1 won ~58.5% of the time).
@top players, although you all seem to agree that pro rules make Pente highly unbalanced, you still seem to prefer pro Pente over other variants. Am I right about this? And if so, why? Do you think other variants suffer from similar problems? Even swap2 and Keryo Pente?
Posts:
2,395
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 7:34 PM
It is no mystery that P1 has sure win in perfect play. It is the games design. There isn't draw, and P1 moves first, and tournament rule still allows P1 to maintain said sure win. I think we can all agree on that.
Smoke and mirrors was originally karls concept. Which I later adopted. And it is also a very straight forward concept, that is very complex to master.
Essentially, human vs human, P2 can only win if P1 makes error. Provoking said error, can come about different ways, and at different stages of the game. Smoke and mirrors can come in many forms.
spook move that looks bad for P1 even though its not.
bait move, where a move that looks good for P1 is a trap.
multiple doors position. Where P1 gets overloaded with choices.
Deep Dark Forrest. A trecherous complex and confusing position. Where only one player can escape the forest.
Adzi face off. A unintuitive position of slow tempos, where the player with greater understanding of adzi navigates and wins.
DataBase tactics. Utilizing False wins and False loses to exploit players who rely on database to navigate.
Chat warfare. Using text chat to get into your opponents head to either probe for data, or psych them out, to gain advantages towards winning. And Owen's immoral favorite, asking a player to play a weak move as P1, and then in other game dont return favor, play strong P1. This one is a good way to get blocked by players, as it is without honor. Not recommended. *not a ad hominem, just trying to be thorough, and give credit to who invents new tactics.*
Without any smoke and mirrors, pente becomes more pointless. And mastering it, takes skill.
But yes, essentially you are correct.
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 8:57 PM
If I were to advice Suzy, I would start by decomposing the problem into two parts. The first would be changing the mechanics in such a way that some form of a draw zone emerges. The second would be eliminating the first-player advantage.
Conveniently, the world of abstract strategy games already has a technique for solving the second problem: the pie rule. Exactly how to apply it, be it swap2, swap, T10, or something specific to Pente, it's a technical question that should not be too difficult to resolve. For example, in the world of Renju, tournaments using a wide variety of opening rules were held in many countries over several years, allowing players to experience the pros and cons of different opening rules for themselves and make a well-informed choice.
The first problem, however, is extremely difficult for me.
Over the past few months, I have been thinking about possible alternatives and have come up with the following ideas.
First of all, boat, capturing threes, poof, and other rule changes that affect both players symmetrically will not work (I have played these variants a bit to get a better feel for them). These approaches lead to a dead end. Or, more precisely, they will not balance the odds, but they may give rise to new games. The history of Gomoku provides a useful analogy: attempts to balance the game eventually gave rise to Renju. The forbidden moves did not ultimately balance the game; the opening rules did. But the search for balance in 19th-20th centuries gave birth to a new game with new mechanics.
The same applies to chess variants without a rook: introducing a random starting position (Chess 960) will not help us find a balance, nor will Atomic Chess, Crazyhouse, or other variants, as long as a bare king still loses and the two sides do not have equal material.
So finding a draw zone in Pente looks very difficult.
But I have had several ideas, namely:
a) In case of a capture victory, award a win only if the winner has captured at least two more pairs than the opponent, while still requiring at least five pairs. A table tennis (volleyball, tennis, ...) approach.
b) In case of a five-in-a-row win, award a draw instead if it is the first player who completes the five-in-a-row and the second player can complete his own five-in-a-row in reply by the very next move.
c) Introduce a mechanic that allows captured stones to be placed back on the board as one's own stones. This effectively allows two stones to be placed in a single turn, so some strict restrictions would be necessary. Nevertheless, this might create a draw zone.
d) Introduce a limited pool of stones and allow stones to be moved: each player starts with a limited supply of stones (I think the optimal number would be somewhere between 15 and 20). On each turn, a player may either place a stone from his supply on the board or move a stone which is already on the board to another intersection. To preserve existing attacking techniques as much as possible, it would probably be worth prohibiting the movement of a stone that forms a solid pair with another stone. If a player cannot make a move (supply is empty and all stones on the board are part of pairs), he passes. If both players pass consecutively, the game is a draw. Repetition of a position (the same position with the same player to move at two different moments of the game) also results in a draw.
e) Captured stones remain on the board as "neutral" or "dead" stones: their intersections cannot be played on, but their former owner can no longer use them to create a five-in-a-row. Whether their former owner can still use them to capture his opp's stones is an open question. I would try both versions. Each seems to have its own pros and cons.
Option (a) does not seem like a solution. I think the first player would win the current version of Pente easily even if capture victories were dismissed totally. Even a single extra pair of stones gives too big gomoku-style advantage.
Option (b) seems likely to lead to a catastrophic change in gameplay: the defending player would now make his own three rather than block the opponent's three. On a board with an even number of intersections along each dimension, the second player would even have a trivial drawing strategy. Would the game resulting from this change still resemble the Pente we know today? Unclear. Would it even be playable, or would every game become a trivial draw? That's why I don't like this approach.
Options (c)–(e), on the other hand, may be worth discussing. Option (d) introduces repetition of the position and even "infinite checks" option to chess pro; (c) and/or (e) may produce some variants which can fill the board thus ending the game in a draw. Combined with any pie rule implementation, of course.
Posts:
2,395
Registered:
Mar 4, 2007
From:
San Francisco
Age:
45 Home page
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 9:31 PM
Hey Karl, Bej is asking why do we prefer regular pente over other varients. You should take a crack at answering it maybe.
For me, its the original pure version. And it ain't dead yet, so im still at it.
Keep in mind i grew up playing boat pente vs a master player. So, I do like a varient. Getting into a new varient takes time. And varients have less players. Easier to find strong players in regular pente. Not sure why you call it pro pente. Maybe for tourney rule? And we all know pente. No long learning curve. Im already good at it.
I like pentes simplicity. And finding the complex in that simplicity. It is the version for the purist. And given that so much has been mapped, I enjoy the challenge of finding the needle in hay stack.
New varients, to actually master them. Too time consuming. Im not that young now. My time is important to me. So. But maybe ill play varient sometime. But as a new mainstream professionally taken serious competitive varient, I doubt it. Probably more as a casual thing. Its a time issue really. As opposed to, cant teach an old dog new tricks thing.
Scire hostis animum - Intelligere ludum - Nosce te ipsum - Prima moventur conciliat - Nolite errare
Posts:
1,095
Registered:
Mar 7, 2006
From:
Philadelphia, PA
Age:
38
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 9:42 PM
A lot to respond to here, and I'm still on vacation, so I have to keep it brief.
I agree that Owen's comparison of pro-pente balancing with rook-odds chess is a hilarious exaggeration. It's yet another piece of evidence supporting the claim that his attack on pente is in bad faith, misinformed, or both.
Having played both games extensively, I believe the closest analogy is pawn odds: enough to tilt the balance strongly in favor of p1, but only at the highest level of human play. At everything below master level, a single mistake (which is almost guaranteed to occur) wipes out the advantage.
It is also interesting that you use chess as an example, Owen, because chess has its own balancing issues that are becoming more serious with each generation. Engine-based openings that almost guarantee a risk-free draw for White and plenty of chances to push for a win against an opponent who isn't fully booked up, etc.
Finally, I love Zoey's classification of smoke and mirror strategies. I am inspired to try and find an example of each from master games. Don't reveal too much, Zoey! Or we'll never win a p2 game again, and Owen will have triumphed.
Edit: didn't see Zoey's most recent post. I've ranted at length about my rules preferences in other threads, but to summarize: pro-pente is optimized for a quick, decisive, tactical battle where the first three to five moves are a beautifully complex skirmish for initiative, and then after some brilliant tactical displays, one side grabs a VCT and the game ends, usually bloodlessly.
Captures to me are absolutely essential to the game's tactical dynamism. Without them, it feels too much like tic tac toe, and I get bored of the static nature of the position.
I'm also generally averse to swaps, but I totally see their merit in balancing inherently unbalanced games. I prefer my moves to be my own, thank you.
I also don't like variants that add complexity or variety without any tangible benefit (subjective, I know, but indulge me). For example, playing into captures or starting on the edge of the board.
That pretty much leaves my three favorite variants: pro pente, keryo, and boat.
It's no use going back to yesterday, because I was a different person then.
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 10:48 PM
> Why do they still prefer to play chess with one rook for P2 rather than one of these apparently more balanced alternatives?
You have an answer from Zoey: > For me, its the original pure version. And it ain't dead yet, so im still at it.
People don't like changes. Chess has a small but significant first player advantage but they will never change it. They experience problems with draw death, they say that a dedicated 2500 may force a draw out of 2700 but they do nothing with it. That's the truth. If you try to change anything, you will experience a heavy resistance from top players and a conservative part of the community. Even if the game is dying... but not dead.
I've witnessed the same process with gomoku 20-25 years ago.
> I might be wrong but I think this could be a real drawback of the swap2 version at the highest level. For instance, I've already heard of people who do not really like to play swap2 gomoku for this reason.
You are absolutely correct describing the gomoku swap2 game process on high level. Yes, it is a drawback, at least it may be considered as a drawback.
Any rule has drawbacks, and any rule change introduces those who decline to play with new rules. =( Well, do these people suggest an alternative to swap2?
Re: Smoke and mirrors vs....
Posted:
Oct 24, 2026, 11:57 PM
> Are human capabilities sufficient so a well-trained player can, from any given position: > 1) play perfectly as both White and Black? > 2) evaluate accurately which side is winning?
My answers for any given position (whenever the game tree is sufficiently wide or deep) would be: 1) No. 2) No.
Both depend heavily on skill and, to some extent, luck. A decent player can easily play a fully consistent game. But even a top player may make a mistake in an unfamiliar position. Therefore, even top players cannot reliably do either (1) or (2) in arbitrary, unfamiliar positions.
But there is a catch.
In pro Pente, assuming P1 plays optimally, the game tree is not that wide. For (1), you may need to memorize only around 100 diagrams, provided your calculation skills are good enough. For (2), you need to know the evaluations of perhaps several dozens of opening positions.
So, to me, it is obvious that any dedicated player can learn pro Pente surewins.
So for pro Pente and dedicated players the answer should be: 1) Yes. 2) Yes.
There is, however, one small obstacle: the lack of an accessible opening base and means to build it easily. You can't find these diagrams on Internet or in books. In Renju, Gomoku, Chess, other mind games there are programs that help with analysis. You can get the ground truth of a given position in minutes or even seconds. In Pente, however, there are no programs that are stronger than human players. As a result, most players are not sufficiently motivated to build their own theoretical knowledge base since it requires a lot of time. And if you rely on a database to find the best plan, well, you can get caught.
So we are still in the Stone Age of Pente opening theory. If you want to know the true evaluation of a particular opening position, you may have to spend hours analyzing it yourself (and still you have no guaranteed result if the position is complicated enough).
Back to your question.
For (1), in well-studied surewin openings, the answer is yes for P1 (not for P2 since there is no meaningful concept of optimal play for P2, only smoke and mirrors). For (2), in pro Pente surewin openings, accurate evaluation is definitely known or derivable for most P2 defenses.
But in Swap2 Pente, my answer to (2) is no. And since the answer is also no for Swap2 Gomoku, I believe Swap2 Pente is playable and reasonably well balanced, even at the highest level. So for swap2 Pente: 1) No. 2) No.
