Posts:
542
Registered:
May 9, 2002
From:
Northeast USA
D-Pente 101: Rules, Tips, Tactics and Why You Should Play It
Posted:
Mar 6, 2004, 5:03 AM
Greetings,
I'd like to start a healthy discussion of D-pente here to spark some interest in this exciting new variation of our beloved classic game of Pente. There are several reasons why this variation is so exciting and why I am pushing for everyone to give it a try. First, it goes a long way towards "solving" the "player 1 advantage" crisis that has been hurting the highest level of the game for quite some time now. Secondly, it breathes some fresh life into a game that tends to become a little stale after playing out thousands of iterations. The reason for this is the shear volume of variety of games that become plausible under this new system. It is no longer enough to scour the database and "memorize" the best responses to the small number of game variations that are played over and over again under regular pente (pro-pente) rules to become one of the best players. A much greater degree of fundamental understanding combined with creative innovation will now be required to conquer the game and playing vastly different games each time you play should keep individuals much more entertained in the process. It is already considered by many to be the future of competitive pente. Most importantly, you should all give D-Pente a try because it's FUN!!!
RULES OF THE GAME:
First of all, just what the heck am I talking about anyway? What is this D-pente and how do you play?
Let me digress briefly here and point out to the site developers (as I'm sure they are already aware) that the rules of D-pente are not yet posted anywhere at the time of writing this (except in the forums). It may be beneficial now that we have several variations supported at this site to update the main page, the FAQ pages and the tutorials of this site to include all of these variations, including D-pente, so that at least people will know how to play and will not be intimidated to at least give these variations a try. In addition, at this point I would strongly suggest some sort of link, directly from the game room, that yields a small pop up window which precisely explains the rules of the game for each variation. This link might be placed directly next to where it says "Game:", with something like "(help)" or "(?)" or something. (Incidentally, In Mark's WPente program, in the "Help" menu, he has two help topics (links) called "Pente Rules" and "Pente Variants", which perhaps if he gave his permission these could be reused here with very minimal development effort.)
Ok, here are the rules for D-pente. First, you decide who will sit in the "Player 1" seat and who will sit in the "Player 2" seat. The person who sits in the Player 1 seat will be responsible for "creating an opening". Once both players click "Play", a white stone will be automatically played at K10 (this may or may not change in the future -- I vote for this to change, but I understand that this may cause substantial changes behind the scenes to make this happen). Now, PLAYER 1 will be playing ALL of the next three stones. Player 1 will play one black stone, then Player 1 will play one white stone, and finally Player 1 will play another black stone. At this point, there will have been two white stones played and two black stones played and the third white move will be pending (a capture is allowed at this stage, although it is unlikely that Player 1 will opt to do this). The second white stone is unrestricted in D-Pente -- there is no "tournament rule".
Now, with the third white move pending, PLAYER 2 now has the OPTION to play the rest of the game as white (swap to player 1) or as black (decline the option to swap to player 1). After this occurs, whoever is now sitting in the Player 1 seat will play the third white stone, and play continues as normal. At all times before and after this D-pente SWAP TIME, the rules of the game are exactly the same as in pente -- it takes 5 in-a-row or capturing 5 pairs of stones to win the game.
RULES SUMMARY: 1) 5-in-a-row (or more) wins 2) 5 pair captures (or more) wins 3) Player 1 plays two white stones and two black stones (the first white stone is actually forced to K10, but there is no "tournament rule" restriction placed on the second white stone), then Player 2 gets to OPTION to swap seats with Player 1, then play continues as normal.
CONVENTIONS OF ETTIQUETTE:
Just like in Pente and every other variation of Pente played here at DSG, you should ALWAYS play D-Pente games in sets. In the case of D-Pente, this means that one player STARTS one game as Player 1, then the other player STARTS the next game as Player 1. Keep in mind, that unlike any other variation, you MAY not have to actually swap seats after the first game in order to correctly play a set. If Player 2 excercised his option to swap to the Player 1 seat in the first game, then the two players would simply stay seated and click "PLAY" to start game two. This will require the players to keep track of who started the last game as Player 1, but I think you can all handle that. There are two reasons why you should be playing D-Pente in sets. First, despite popular belief, the swap rule DOES NOT eliminate all inherant advantage that one player has over the other which exists in Pente (and Pro-Pente). In fact, instead of Player 1 having the advantage like in all other variants, in D-Pente PLAYER 2 actually has the advantage!!! Wierd but true. The second reason to always play in sets is that it takes a great degree of skill to create the opening at the beginning of a D-Pente game, and it is only fair that both players be required to share this burden equally.
SOME D-PENTE THEORY AND BACKGROUND:
Before diving into tips and tactics for playing D-Pente, first, if you will humor me, I will entertain you with some Pente theory. The crux of understanding Pente is to realize that one player ALWAYS has the advantage in any given game at any given time. Always. There is no such thing as a tied position -- and a competitively played game will never end in a tie. This differs slightly from other stone games such as renju where the edges of the board will actually prevent either player from achieving a win. In Pente, someone will always win. It follows then that someone is always winning.
Next, as I'm sure you've all heard at some point at DSG, Player 1 has the inherant advantage in Pente. Consider regular Pente (no tournament rule, no move restrictions). After white makes his first move, white has an infinitely stronger position than black. Black has no stones on the board and white has one. Next, when white plays his second stone, he now has 100% more stones on the board than black. This is a massive advantage. If you sent two armies to battle each other in a field armed only with swords and you give one army twice as many men as the other, those outnumbered soldiers had better be pretty darn good swordsmen if they are going to win. Being outnumbered 2 to 1 is an overwhelming advantage. Thus, at the instant just before white plays his second stone, white is winning. Always. Now, when white plays his third stone, he has 50% more stones on the board than black, and if he has chosen wisely, he has several more paths to victory available than black (if he has put 3 stones into a triangle -- he has created "potential" paths to victory on a horizontal, vertical and diagonal direction at the same time -- black with only two stones has at most one "potential" path to victory in only one direction. Again, white has the advantage.
Notice here that white's numerical advantage has decreased significantly from infinite, to 100%, to 50% in just three moves. In addition, just before playing white's second move white is ALWAYS winning. This is not the case for the third move. There are some legal combinations of moves where after white's third move, black's advantage is overwhelming -- but this is not a numerical advantage, it's a positional advantage. If you put two armies of archers on top of a huge hill behind a massive fortress's walls and pit them against three armies of archers that must trudge up this open hill through quicksand in order to get in range -- well, I'll put my money on the two fortified armies of archers. But it didn't have to be that way. If those three armies were equally well fortified they would have the advantage.
The point here is that the only way for black to ever gain the advantage as more and more moves are played is for white to make an error. This conclusion may be a leap of faith for some but if you think of the logic it must be true, and in fact it has been proven to be so. In fact, in regular pente, if white does not make an error in just the first few moves, then this advantage becomes so decisive that it quickly becomes crushing. The tournament rule in Pro-Pente was then invoked and intended to slow down this steamroll effect. However, even with the tournament rule, following the same logic as above, white is still ALWAYS winning until he makes an error.
Next, understand that, from black's perspective, the moves played at the beginning of the game are VASTLY more important than those played near the end. Look at the progression above again -- white's numerical advantage goes from infinite, to 100%, to 50% and lessens further with every move played. However, if BLACK plays his second move in error, for example, the result is much more devastating than playing his 6th move in error, for example. Let's say that black plays his second move in error. For the purposes of making my point here, we'll treat this blunder as if black failed to play a stone at all. Now, when white plays his next stone, instead of having a 50% numerical advantage, he will have whopping 200% advantage!! This advantage is so crushing that white cannot help but to win the game quickly and easily -- the momentum is far too great for anything else to occur. Taking it one step further, white would then have to make TWO such blunders for black to gain the advantage instead of just one! Let's say that after black's blunder, white plays an equally terrible blunder for his third move and black comes back with a good third move of his own. Well, now, after white plays stone 4, he will still have a 50% advantage! (Treat one white stone and one black stone as if they didn't exist and you are left with a 3 to 2 advantage). Now black makes another good move, move 4, and white blunders again! Finally, if black's 5th move is sound, he will finally gain the advantage -- and this will be a comparitively measley 33% advantage. Instead, suppose black plays all stones perfectly until move 6. By making a blunder at move 6 he is only giving up a 7 to 5, or 40% advantage to white after white makes his next move, instead of the incredible 200% advantage after move 3. Plus, black making the first 5 moves correctly gives white ample time to make his own error.
Now, it should be obvious to you that this model is oversimplified, but it does accurately show that in unrestricted pente, white has an inherant advantage and is always winning and this advantage cannot be overcome unless white makes an error.
To improve this model slightly, let's also consider "complexity", "momentum", and "degree of error". First, momentum and complexity often go, at least partially, hand-in-hand. Notice that, after white makes his second move, while his numerical advantage is 100%, his "momentum" generated is close to 0. There is some momentum, but it is some number close to 0. This means that black's move must respond at least in some indirect way to white's positional (not numerical, but positional) advantage. He isn't "forced" to respond directly, in the sense that white has not yet even made an open tria that must be blocked. Conversely, consider a position where white has made a triangle and there are no black stones in sight. Once white then makes his next move, he has generated incredibly high "momentum". This is because white makes a vertical tria, black must block, white makes a horizontal tria, black must block, white makes a diagonal tria, black must block, white makes another vertical tria, black must block ... you can see that the train is rolling and cannot be stopped -- trains have lots of momentum. Next, on the move prior to creating this triangle, white does not have high momentum, there are no "easy" moves to make. He has made a vertical "potential" but if he creates the tria at this time it does not lead anywhere on his next move -- there is no momentum -- since there are no "easy" moves, this position is more "complex". When a position is more "complex" there is a vastly larger number of options available that are all closer to being "optimal" than when a position is less complex. One important side effect of this observation is that highly complex positions are MORE LIKELY to induce an error than less complex positions. In the above example, at the moment that white has decided to create a triangle, he has played a complex move that generates lots more momentum than white had previously enjoyed. It was not as easy or obvious as the subsequent moves to make trias were but it was the correct one in this case. He had the opportunity to make this complex move because he was given the "initiative" to do so by the fact that he has had the numerical advantage and has not yet made any errors.
So the question arises -- if the position was so complex, how do we really know whether making this triangle was indeed the optimal move? Perhaps placing the stone in another position which also makes a triangle was another good alternative? Or perhaps another move altogether was better than making the triangle. Since the position is complex, there are no "easy", "obvious" moves to be made. Nevertheless, there is exactly one "correct" or "optimal" move in every situation. This is precisely why the best players spend so much of their time thinking about the first few moves of the game. During the first few moves, the position is generally highly complex and momentum is low. It is also easier to make a small error since the correct move is less obvious -- so they spend more time trying to get such moves right. In general, since white already has the "initiative" from his inherant numerical advantage, white's efforts are to try to simplify the position and create more momentum of attack. Black, starting at a disadvantage, plays more defensively, trying to create more complex positions (to invite errors) and disrupt white's momentum until such time that white makes an error that black can take advantage of by "taking the initiative" and building his own momentum.
To follow from our previous method of calculating white's advantage, if white makes an absolute blunder (such as playing his second move to T1), then we can indeed continue on by considering that this blunder stone does not exist at all (after this second move, white still has one stone, and black has one stone). However, what if white plays a move such as 2. K15? Is this a complete and total blunder? Is there no possible way that this K15 stone will come into play? Or perhaps is this more of an 80% blunder? Maybe we should more accurately say that white now has 1.2 stones and black has 1 stone. So, instead of white having a 100% numerical advantage, he now has only a 20% numerical advantage. And in highly complex positions, maybe white makes the second best move -- and thus has made a 10% blunder. Well, if this occurs after the second move, white will still have 1.9 stones to black's 1 stone. At the same time, white has made his position more complex and has given himself less momentum -- all of these factors can be included into some overarching definition of "advantage" -- or "initiative". Whoever has it is winning, whoever keeps it will win, and black can only win if white has a large enough "degree of error".
D-PENTE TIPS
Enter D-Pente!!! By now you should understand how important the opening moves are to the outcome of a pente game. The idea of D-Pente is to create a "more even" opening that gives a nearly equal chance of winning to both players. Here are some tips for how to create d-pente openings and how to decide whether or not to swap:
General rule of thumb: To counteract the massive 50% numerical advantage that white will enjoy as soon as white plays the third stone, try to create openings that appear to be greatly in black's favor. The most common mistake to beginning D-Pente players is to not tip the scales far enough into black's favor. In general, if you are the player deciding whether to swap -- when in doubt, swap. It is likely that white still has the advantage.
Avoid creating openings that are common in regular or pro-pente. These openings are known to be good for white in nearly every case and you will find yourself getting swapped into the player 2 seat and put at a disadvantage quite often. In general, this includes any white second move to K13, K14 and similar. BE ADVISED! There are a few cases where positions that involve such white second moves are known to be good for black -- but again if they are "known", they are likely to be known by your opponent and you will be put into the seat which has the disadvantage.
Be very careful when creating openings that are common "alternative" openings in pro-pente (IE, g-pente style openings). In general, this includes white second moves to L13, M13 or similar. Very often these positions are good for white -- however, if you are the one deciding whether to swap, look carefully since black's stones are more important in these cases and may, in a few cases, tip the scales into black's favor.
Avoid openings that appear symmetrical. Consider 1. K10, K9; 2. K12, K7; While this appears to have the same paths to victory for each player at first glance, since white gets the first (and next) move, white will have a massive advantage in such positions.
Avoid any opening that gives white a vertical or horizontal (and probably even a diagonal) potential. There are really no two places to put black stones that adequately defends this strong white position.
It is terrible to put two black stones into a position to be captured. It is a viable option to put two white stones into a position to be captured.
It is viable to put black stones into a position that will create a keystone attack on white's stones -- do not give white this advantage.
In most cases, put white stones "farther apart" than black stones.
It is not necessary, and is often incorrect, to allow white's third stone to be placed in such a way that allows for an open tria with both of the other white stones.
------------------------------ I'd love to hear any and all D-Pente responses at this forum. Specifically, more tips for opening play and some specific opening line suggestions would be appreciated! ------------------------------
Be creative and innovative! Experiment! Practice! Practice! Practice!
Posts:
1,032
Registered:
Dec 16, 2001
From:
Powell, OH
Age:
37 Home page
Re: D-Pente 101: Rules, Tips, Tactics and Why You Should Play It
Posted:
Mar 6, 2004, 4:32 PM
Great post up2ng! I think your tips are great!
I appreciate your tutorial here, it would go well in the DSG tutorial format perhaps. One more thing I don't really have time to do but it's a good idea.
I agree about the rules being posted and having a link from the game room, that is another thing I've thought about for a long time, it does need to be done to push D-Pente!
Posts:
55
Registered:
Jan 21, 2002
From:
ohio
Age:
65
Re: D-Pente 101: Rules, Tips, Tactics and Why You Should Play It
Posted:
Mar 7, 2004, 10:12 PM
everyone should try "D", its certainly become one of my favorites, it evens the field between white and black, and makes it more reasonable from black position.
Posts:
1,032
Registered:
Dec 16, 2001
From:
Powell, OH
Age:
37 Home page
Re: D-Pente 101: Rules, Tips, Tactics and Why You Should Play It
Posted:
Mar 9, 2004, 8:28 PM
I just read part of this again (who can read all of up2ng's posted in one sitting ) and noticed his comment about the first move at K10.
I originally coded D-Pente to not require K10 as the first move (not a major coding change actually) but later decided against it. I understand the desire to not have that restriction, it allows player 1 to start the game off in a corner or side of the board and use the sides of the board as part of the game. Ordinarily the sides don't come into play in regular Pente.
I changed my mind 1/2 way through because I remembered reading or hearing somewhere that Pente is played on a 19x19 board BECAUSE you aren't supposed to ever reach the end of the board. The idea was to have an infinite playing field. For this reason I decided to keep the K10 restriction, just to try to keep the variation as close as possible to the original game of Pente.
Maybe we need a variation to D-Pente for this rule Just kidding...
Posts:
260
Registered:
Dec 16, 2001
From:
Maryland
Age:
53 Home page
Re: D-Pente 101: Rules, Tips, Tactics and Why You Should Play It
Posted:
Mar 9, 2004, 9:25 PM
Great post! The one additional note I have is that opening for D-Pente was suggested by Don Banks of Canada. (The 'D' is for Don).
Note that my implementation of the variant in my program and at PBEM has the first player move to the center K10, the _second_ player makes the next three moves, and then the first player has the option to swap. It works out exactly the same of course, (except that now the "first" player has the advantage of deciding whether to swap) but it was easier to program it that way at PBEM.
We'll have to have a D-pente tournament here I will start the 2004 pente tourn at PBEM shortly, and I am going to use the d-pente variant there.
Re: D-Pente 101: Rules, Tips, Tactics and Why You Should Play It
Posted:
Jul 17, 2004, 10:42 PM
Wonderful post!
For what it's worth, I would like to add that this idea called here "D-Pente" actually goes back much further to the old glory days of Pente in the early 1980s. As I recall, R. Wayne Schmittberger, who was then the editor of GAMES magazine and a fine Pente player, suggested this rule back around 1982 as his suggestion for handling the edge of the first player.
He called it the "Pie Cutter Rule", because the exact same method has been used in many other games of skill to try to neutralize theoretical advantages for one player or the other. The Pie Cutter name comes from an even more ancient use for this concept -- two kids want to share the remaining apple pie in a pan. How can they do it fairly? Use the Pie Cutter Rule -- Have one kid cut the pie, and the other kid may choose which piece of pie he wants. The same thing is going on here with "D-Pente".
Schmittberger's original suggestion, as I recall, had the first player making the first 3 moves of the game and the 2nd player choosing sides, rather than the first 4 moves as in D-Pente.
Personally, I like the idea although it seems a little awkward. One thing I like about it is that it not only tends to equalize the first move advantage, but also eliminates the requirement for two-game matches in tournament play. In tournaments, the director pairs the two players and names one of them as the pie cutter and the other as the pie chooser in each game. In subsequent rounds players are assigned the pie-cutter role and the pie-chooser role an equal number of times, but against different opponents. This would allow over-the-board tournaments to have more rounds in the same amount of time.
The other nice thing about the rule, of course, is that there would be a lot of creativity in preparing, in advance, your "Pie Cutter" strategies. For example, players could seek to find formations that are not well known and which are "tricky", such as a position that seems to obviously favor black when in fact it favors white (if you see a "trick").
I'd like to see more experience with this variation! Perhaps it is the wave of the future, though I still prefer Keryo!
) and noticed his comment about the first move at K10.