to add
Game number: Main Page
White: shyryan    Black: bob440 (LG) | This game (LG) | Download JTwixt file
On 2020-07-06 at 21:29, shyryan (info) said:
Not sure where the fatal error was. Maybe 17.l9 was better. I was caught off-guard by 20.o6, but now that I look at it, 16.j6 was pretty well-supported.

On 2020-07-07 at 11:25, Peyrol (info) said:
|23.e10 24.g11 looks good for black. One line is 25.h11 26.h13 27.f10 28.j14 So I suppose black J6 wins the left side, in large part due to the D14 peg.

Yes |17.l9 is best. White threatens J5 which constrains black's response. White also threatens O6. There is nothing black can do. 18.n6 19.m7 for example.

On 2020-07-07 at 21:55, MisterCat (info) said:
Crafty, there, Twixter; That L9 move, threatening J5 or O6 appears to win the game for white! You are playing a move that sets up future threats, which are not at all apparent - at least to moi. K8, loses, L9 wins ...
Actually, pretty cool. I like that. I hope to be able to think that way. Maybe someday.

On 2020-07-08 at 08:02, Peyrol (info) said:
Well it was shyryan's idea in post mortem. You could call the L9 link a shallow glide (gliding "down" to the top border,) as opposed to K8 which is a steep glide. All that room for white on the right of J6 is what makes L9 work. For example |17.l9 18.o5 stops both J5 and O6 but then 19.l5 20.k5 21.n6 22.m4 23.p5 works for white.

On 2020-07-09 at 20:06, bob440 (info) said:
anybody look at |17.l9 18.o6 ?

On 2020-07-09 at 20:48, Alan Hensel (info) said:
|17.l9 18.o6 - how about 19.j5 20.i4 21.k7 22.k5 23.l5 or 22.k3 23.n4. Typically the 0-5 gap would be attacked in the middle, so 19.l6, for example, but then after 20.m7 21.j5 it looks pretty similar to just going straight for 19.j5. 19.m6 might also work, but looks messier - 20.l7 21.o7 22.p8 23.p9 24.m8 25.m7 26.m5 27.r6 28.q6 29.t7

On 2020-07-10 at 05:40, MisterCat (info) said:
So I am trying to generalize:

with the rest of the game being irrelevant, after 1. J10 2. J6 3. L9 then it would appear that White has already achieved reaching the top border, and Black can do nothing to prevent this!? White has Peyrol's threats of J5 and O6, and so far, Black's attempts with 4. O5 , 4. N6 , and 4. O6 have been busted. Is there nothing that Black can do? Maybe 4. M5 ?

If not, it just seems to me that the Twixt board has been shrunk - like you can be guaranteed to reach the border with a peg in the square formed by 1. I9 , 2. P9 , 3. P16 , and 4. I16 !

On 2020-07-10 at 22:57, Peyrol (info) said:
Without the other pegs on the board, I'm not really sure what you are asking. Yes L9 wins the top if you ignore the bottom. So white can win a 13 row handicap game in this way.

The opening moves are crucial. There are three or maybe four moves you make in a good game, usually in the opening, which involve more intuition than calculation, and the rest of the game you try to tactically justify the plan that you are now stuck with. More and more these days, opening intuition is being replaced with specific opening knowledge. This is why I like 30x30, not necessarily more than standard, I just like it.

Oh and thank you for the compliment earlier. I often don't take compliments well.

On 2020-07-11 at 18:43, MisterCat (info) said:
I noted that L9 was shyryan's original idea, and failed to give him credit; credit given where credit is due!

My 'compliment' to you was to the follow up moves you cited, which explain the underlying threats made by L9; these are moves that I would normally have not spotted, and I found them to be 'crafty', and thus my comment.

To further explain what I meant in my last post about 'shrinking the board', perhaps some polygons:

1.i9 2.k8 3.l8 4.m8 5.n8 6.p9 7.q11 8.q12 9.q13 10.q14 11.p16 12.n17 13.m17 14.l17 15.k17 16.i16 17.h14 18.h13 19.h12 20.h11

One used to think that it was necessary to have a peg on, or past this polygon in order to be guaranteed of reaching the nearest edge.

However, the discussion above with 1. J10 2. J6 3. L9

seems to imply that a peg on or past THIS polygon (with L9 on it) will do the trick: 1.j10 2.l9 3.k9 4.m9 5.n9 6.p11 7.o10 8.p12 9.p13 10.p14 11.n16 12.o15 13.l16 14.k16 15.m16 16.j15 17.i14 18.i12 19.i13 20.i11

The second polygon is SMALLER, which is what I meant. I am trying to see if I can generalize a rule here as to how far from the edge one can be in order to GUARANTEE reaching it, past any defense. Of course, in this specific game,

1. J10 2. J6 3. L9 , blacks specific defender of J6 may be the problem here - maybe if black had chosen a different move, then the J10 - L9 link would NOT be guaranteed to reach the top.

What do you think? Have I at least made my concept and question clear? (meow)

On 2020-07-13 at 17:54, shyryan (info) said:
David, are you aware of a repository of opening knowledge that is coming to replace opening intuition? All I've really seen is charts that show whether most players swap on certain opening moves.

On 2020-07-14 at 20:49, Peyrol (info) said:
Mr. Cat: The issue of whether J10 is unstoppably connected to the top or not sounds like you are talking about a template, which is a term used in Hex. Certainly templates can be useful in Twixt, but IMO it's not as easy as it is in Hex to draw a conclusion about the full board position from template knowledge. So if you agree with me about that, then I guess we are in agreement.

shyryan: I'm just referring to the growing archive of games which are available to study both on this site and with Jtwixt.

You must be logged in to add comments.