Arimaa Forum (http://arimaa.com/arimaa/forum/cgi/YaBB.cgi)
Team Games >> 2009 One vs TheMob >> 2015 Move 12g
(Message started by: deep_blue on Dec 19th, 2015, 5:25pm)

Title: 2015 Move 12g
Post by deep_blue on Dec 19th, 2015, 5:25pm
Sharp made its move. Ideas?

Title: Re: 2015 Move 12g
Post by aaaa on Dec 19th, 2015, 11:39pm
Seems like the choice is between the horse frame and the horse flip.

Title: Re: 2015 Move 12g
Post by Lion on Dec 19th, 2015, 11:40pm
We need to try to work out whether framing the horse wins outright or not--and I think we ought to be able to calculate this one essentially to the end. I see this as follows.

1. We permanently frame the S horse.
2. Comp permanently hostages our G Horse (cc7e rb7e hb6n mc6w), but his own S horse will need many moves to get back into the game.
3. In the meantime, we might attempt to rotate out our G Elephant (I need some more time to figure out if we have enough time for this, but this looks probable at first glance.
4. Alternatively, we may consider blockading the S camel on b6 and rotating out our G camel (we might have enough time for this, too).

In other words, framing the S horse looks winning.

If this turns out not to be the case, we may consider getting our G horse away from a6 (Mb5e Ra5e Ha6s Dd2n). However, then both of our b5 and g5 G Rabbits will face imminent danger.

Title: Re: 2015 Move 12g
Post by Lion on Dec 19th, 2015, 11:41pm
That's right, there is also the horse flip to consider! :D

I imagine then the comp will play mc6w ee3w ed3n ed4w. Next, if we capture the g5 horse, the comp will likely capture the c2 cat while threatening both the a6 Horse and d2 Dog (hb4s ec4s ec3e Cc2nX). This looks to be losing. However, there are also other possibilities for us instead of capturing the g5 horse, but we should probably calculate the horse frame first.

Title: Re: 2015 Move 12g
Post by SilverMitt on Dec 20th, 2015, 12:09am
It seems like the long-term evaluation after 12g f3:H 12s Cd7 bc7 Hb7 Mb6 is complicated, though it makes me uncomfortable at first glance, as silver need not even try to force anything immediately.

What I don't like about the horse flip is that it sort of saves time for silver in terms of going after the a6 horse.  But if it tactically works out as a trade, maybe it's our best option.

I'm curious about a retreat along the lines of 12g Mb4 ab5 Ha5:R now that the silver camel has swung west.

Title: Re: 2015 Move 12g
Post by aaaa on Dec 20th, 2015, 12:29am

on 12/20/15 at 00:09:47, SilverMitt wrote:
I'm curious about a retreat along the lines of 12g Mb4 ab5 Ha5:R now that the silver camel has swung west.

Silver could simply defend against the frame in the east and all I see then is Gold being behind in the rabbit-pulling contest with nothing to show for it.

Title: Re: 2015 Move 12g
Post by browni3141 on Dec 20th, 2015, 3:56am
I think any 12g horse retreat concedes clear advantage to silver per aaaa's reasoning.

Title: Re: 2015 Move 12g
Post by Hufflepup on Dec 20th, 2015, 7:06am
Flipping the horse could lead to 13g: mc6w ee3n ee4w ed4w.
We cannot take the eastern horse next turn because of 13s: ec4n Mb5e ec5s hb4n (We cannot prevent camel capture next turn). There is no other 13g in this scenario which will stop us losing at least a cat with no material compensation. Silver also has decent follow up threats around both C3 and C6, and we don't have any. For this reason I don't think the horse flip is a viable option.


If we framed the horse then:
12s cc7e rb7e hb6n mc6w (Taking horse hostage)
If we don't want to lose the horse next turn or break the frame we have to play:
13g Mb5e X X X
The trouble is there are no other steps that would stop something like:
13s: ee3n ee4w ed4w df6w (Sacrificing the framed horse)
There is now no 13g/14g that we can play which will stop us losing at least a horse. This still looks a lot better than the situation after the horse flip though.


Title: Re: 2015 Move 12g
Post by aaaa on Dec 20th, 2015, 11:27am
Framing the horse and then immediately giving it up with E->c5 after m->b6 might work out OK, but things would immediately get desperate as Silver is bound to take over the southeast trap while also being ahead in any goal race. The camel helping with the defense would either be futile or an extremely complicated affair.

12g Mb5s hb6s Mb4s hb5s might be one way to try for a horse trade.

Title: Re: 2015 Move 12g
Post by Lion on Dec 20th, 2015, 8:59pm

on 12/20/15 at 07:06:03, Hufflepup wrote:
Flipping the horse could lead to 13g: mc6w ee3n ee4w ed4w.

If we framed the horse then:
...
13s: ee3n ee4w ed4w df6w (Sacrificing the framed horse)
There is now no 13g/14g that we can play which will stop us losing at least a horse. This still looks a lot better than the situation after the horse flip though.


I am not convinced this is the case.

13g. Mb5e Dd2n Rg5e X
13s. ee3n ee4w ed4w df6w
14g. Ef4w Ee4w Ed4w Mc6w

Next, silver can't capture the horse as then we could capture the silver camel.

Title: Re: 2015 Move 12g
Post by aaaa on Dec 20th, 2015, 10:07pm
Lion, if Gold were to defend the horse by putting the camel on c5, it needs to be followed up with a rabbit step to b5. Otherwise, Silver could win the camel by fencing it with his own camel on b5 and his elephant on c4 or d5 (occupying the vacant b6 with the gold horse then would only create a false-protection pattern).

Title: Re: 2015 Move 12g
Post by Lion on Dec 20th, 2015, 11:41pm
That's a good point, but maybe we can get away with it.

13g. Mb5e Dd2n Rg5e Ra5e
13s. ee3n ee4w ed4w df6w
14g. Mc5e Md5n dd6w Ef4w

It gets complicated after this, but perhaps we could end up losing only a dog in what follows.

Edit: never mind, we would lose the camel here. Hmm, I guess the computer really did find a good move.

Title: Re: 2015 Move 12g
Post by SilverMitt on Dec 21st, 2015, 12:15pm
I actually don't think that silver has a great tactical option after 12g f3:H 12s Cd7 bc7 Hb7 Mb6 13g Mc5 ab5 gh5 Dd3, as something like 13s Ec4 Dfe6 Hx gives us a great attack in the east after, say, 14g Mf5 Hg4.

It may become a maneuvering battle, with silver having a long-term plan of getting the b7 horse active, while we would need an objective of our own, perhaps trying to rotate out the g3 horse start a fight around c6.  We'd have to be alert for potential tactics on every move, and I'm still not sure on the long-term evaluation, but I think there is a lot of upside here.

Title: Re: 2015 Move 12g
Post by half_integer on Dec 21st, 2015, 6:08pm
Now that I've finally had some time to look at this move, I want to be optimistic.  Sharpvector not continuing to threaten the rabbit directly may mean that we have exposed a greater threat in the west.  For those who were worried about the silver camel coming west, it appears that silver sees the situation the same way.

If we frame the horse, I don't think we can expect to rotate out the elephant.  To do so would require two more pieces than we have at the f3 trap and I don't see us taking the number of full moves it would take to set up the rotation.  I count at least 16 steps to carry out a rotation.  So I think that a horse frame gives us an elephant deadlock at f3, with the important note that we are freer to leave than silver is.

I'm not fond of the horse flip; I agree that it just seems to save steps for silver.  And having the horse out of play for a while (on b7) is worth almost as much (or more, depending on tactics).  If moving the b6 horse is a possibility, then I think a double pull to Mc4 should be considered as an alternate.

I don't like Mc5 Rb5 Ha5 X because silver pulling the rabbit to b6 appears to be indefensible resulting in a pinned gold camel.  

I was going to suggest looking at Mc5:H Hb6:R (pulling the rabbit is optional), but 12s Ed5 Dd5e5 seems to lead to unavoidable gold camel loss plus horse hostage or horse trade - so not a good idea :) .

So having worked through these moves, the horse frame seems like the only acceptable move to me, and I also think we should not leave f3 with our elephant before silver does, sacrificing the horse.  If it goes as expected with a horse hostage at c6, I still think we can protect the camel with Mc5 Rb5 and the result will perhaps be a horse trade; I don't know who will come out positionally at the end.  Maybe we can start another attack in the east at the end, while silver's strongest three pieces are all grouped in the west.  (The a5 rabbit would presumably also be lost after a horse trade and gold camel fleeing the west.)

Title: Re: 2015 Move 12g
Post by Hippo on Dec 22nd, 2015, 11:25am
Yes I agree we would not have time to capture east horse after the horse flip. And horse frame is good enough compensation for the horse hostage with hb7. The situation with Rb5 Mc5 looks very safe and may be the stalemate situation could allow slow rotation of the elephant out of the frame.

I like the frame, especially with the camel west and horse going to burry itself. Even when the silver horse activates, I cannot see it beeing good strongest free piece.

OK I cannot see fast enough rotation, but our position would be solid.


Title: Re: 2015 Move 12g
Post by half_integer on Dec 22nd, 2015, 5:56pm
The last three posts have been in support of the horse frame as the best move, and I don't think any of the other posts have shown a great outcome for any other move.  Therefore, should we conduct a voice vote for this move?  It could allow us to gain reserve, and also puts the clock back on lightvector during the holiday (when the Mob may be preoccupied anyway).

Additional search for silver replies to this move could be fruitful.  I just looked for some and all I came up with is something like 12s Mc5 Dd7 Hd6 to not bury the horse as much.  But silver didn't want to leave the camel on c5 last move so he probably won't want to do it here either.  He could also try to start a goal attack in the east, but he doesn't have any strong pieces and we could block the h-file with a single step.

Title: Re: 2015 Move 12g
Post by Hippo on Dec 23rd, 2015, 12:23am
+1

Title: Re: 2015 Move 12g
Post by Lion on Dec 23rd, 2015, 1:39am
Yes, let's put this up for a vote.

I think the only real alternative to the horse frame is the horse pull, but we probably just lose a rabbit in this case. I'm thinking of something like this:

12g. Mb5s hb6s Mb4s hb5s
12s. mc6w mb6s Ha6e hb4w
13g. Mb3n ha4s Mb4w Dd2n
13s. rh6s rh5s Hb6e Hc6x mb5n
14g. ha3e Ma4s hb3e hc3x Ma3e
14s. Rg5w hg4n df6w Rf5n Rf6x

Title: Re: 2015 Move 12g
Post by Hufflepup on Dec 23rd, 2015, 2:47am
I also think we should put this up for an early vote.

Title: Re: 2015 Move 12g
Post by deep_blue on Dec 23rd, 2015, 5:46am
I am about to go to sleep (haven't had some for like 2 days now :-/) so I won't read the whole discussion. But since there seemed to be consensus on a voice vote I can quickly start one.

Title: Re: 2015 Move 12g
Post by clyring on Dec 23rd, 2015, 9:26am
---Analysis tree---

- precedes a silver move. > precedes a gold move. Pardon any SM notation errors, and my lack of evaluations on many lines.

f4w g4ws e4e/::Hf3 (1)
- c7e b7e b6n c6w/Cd7 Rc7 Hb7 Mb6 (1a)
- > b5e a5e g5e dd2n/Mc5 ab5 gh6 Dd3 (1a.1)
- > - d3nww (f3x) f6w/Ec4 (Hx) Dfe6 (1a.1a)
- > - > c5eee g3n/Mf5 Hg4
- > - d3nnw (f3x) f6w/Ed5 (Hx) Dfe6 (1a.1b, NEW)
- > - > //Prevents an f6 attack a little more effectively.
- > f4nwww/Ec5 oo (1a.2)
- > - f6ss g5w h6s/Df4:R h5 (1a.2a, NEW)
- > - > //Threatening e3n e2n f2w f3s/Ee4:D Hf2:C, but it's surprisingly hard for silver to actually threaten captures in f3. We should definitely take time while pondering to try to better evaluate this type of position, if we play the horse frame.
- c6s d7n b6ee/Mc5 Ddd7 Hd6 (1b)
- > //I don't think this is very threatening.

b5s a5e a6s */Mb4 ab5 Ha5 * (2) -=
- //This seems to covert to a rabbit pulling game, where we are behind.

b5ss b6ss/Mb3::Hb4 (3)
- c6ws a6e b4w/Mb5:H Ha4
- > b3nw a4s d2n/Ma4:Ha3 Dd3
- > //Probably there is a better 13g available, but like Lion I don't think we can avoid material loss or get more than small compensation.
- > - b6e b5n h6ss/Mb6:Hx h4
- > - > a4se a3x/Mb3:Hx
- > - > - g5wnx g4n f6w/(Hg5 De6)::Rx

b5e b6ss c5w/::Hb4 (-1)
- e3nww c6w/Ec4 Mb6 -/+
- > //We seem to be in tactical trouble due to the threat of c4ns b5e b6s/Ec4:Mc5 Mb6

b5e b6s a6e */Mc5:H Hb6 * (-2)
- //Loses big material.
- d3wnn c7e/Ed5 Cd7

b5sw b6ss/Ma4::Hb4 (-3, NEW)
- c6ws a6e e3w/Mb5:H Ed3
- > //Our life seems pretty rough here tactically.

So far of the discussed options, the horse frame seems very likely best. I add my voice to the voting chorus.

Title: Re: 2015 Move 12g
Post by SilverMitt on Dec 23rd, 2015, 12:56pm
So, I think we agree that we are tactically fine initially after the frame.  But, both the frame and the hostage are persistent and hard to break and free, respectively.  Silver may definitely want to hold off on giving up the framed horse to force the issue around c6, and it's not clear to me whether we can afford to leave c5 with the camel, as that may allow the silver horse to become very strong in the southwest.

The other place to improve is around f3 in the east.  We would need to bring additional forces from the west, and it's not clear whether we can really achieve anything, although silver may not be able to just clog the area with rabbits, as that might allow elephant rotation if we can get our horse to g4.

In any case, in order to follow up the frame on 12g with a move including Mc5 on 13g, we need to have a long-term improvement strategy.

Title: Re: 2015 Move 12g
Post by aaaa on Dec 23rd, 2015, 1:56pm

on 12/23/15 at 01:39:16, Lion wrote:
Yes, let's put this up for a vote.

I think the only real alternative to the horse frame is the horse pull, but we probably just lose a rabbit in this case. I'm thinking of something like this:

12g. Mb5s hb6s Mb4s hb5s
12s. mc6w mb6s Ha6e hb4w
13g. Mb3n ha4s Mb4w Dd2n
13s. rh6s rh5s Hb6e Hc6x mb5n
14g. ha3e Ma4s hb3e hc3x Ma3e
14s. Rg5w hg4n df6w Rf5n Rf6x

I don't see why Gold can't play the more obvious 13g Mb3w Ma3e ha4s xxxx here, after which the given 13s looks like a bad idea, as instead of recapturing immediately, Gold could frame the horse first (with a goal threat).

I'm not quite sold on the appeal of the horse frame as it's not clear what Gold's strategy afterward should be. Not only can Silver slowly activate his horse to become the strongest free piece (as long as the frame stays in place), the camel being on c5 and then possibly only being able to escape through the center is a dangerous proposition no matter how you slice it. It's not that far-fetched to imagine Silver abandoning the framed horse at the right moment after having congested the center, not only winning back the horse but also taking the camel hostage in the center.

Title: Re: 2015 Move 12g
Post by arimaa_master on Dec 23rd, 2015, 6:52pm
What about simple 12g E to c5?

For instance:
12g Ef4n Ef5w Ee5w Ed5w
12s Rg5w hg4n df6w Rf5n Rf6x
13g Ec5s mc6s hb6e Mb5n

and Gold seems to be fine.


Title: Re: 2015 Move 12g
Post by half_integer on Dec 23rd, 2015, 6:56pm
After 1a.1 I think we also need to beware of
 13s Rb5w mb6s hb7s rc7w / M:Ra5 Hb6 Rb7
since this allows silver to get the camel for the framed horse, with 14s Ec4 or Ed5 if we don't respond.  Even
 14g Ef4wwww / Eb4 doesn't help:
 14s Hf3n Ee3nww / Hf4 Ec4
 15g Eb4s mb5s Ra5e Mc5e / Eb3:M Rb5 Md5
 15s Rb5e mb4n Rc5nx ec4n / M:Rc5 E:Rc6x

I'm not yet sure how we avoid or respond to this line.  The camel can leave after 13s but then silver can reestablish the hostage.  I'm not sure who benefits from this back-and-forth.

A wild idea I just considered would be to not defend the hostaged horse at c6 but bring the camel east and plan to rotate out the elephant next move.  Could a (somewhat) free elephant be worth a whole horse (and rabbit)?

If we go the more expected route, I think we clearly need to figure out how to flood the c6 trap with small pieces to free our camel, before silver can set up a horse-for-camel trade and abandon f3.

Title: Re: 2015 Move 12g
Post by half_integer on Dec 23rd, 2015, 7:11pm
Maybe instead of figuring out how to avoid a camel-for-horse trade we should look further ahead to set up more compensation.  If we can do something like Ee5:D with our rabbit still at g5 then we can get camel-for-(horse and dog).  Even if silver can avoid risking the dog, Ee6 with the g-file horse advance could possibly provide positional compensation.

Title: Re: 2015 Move 12g
Post by SilverMitt on Dec 24th, 2015, 3:03am
It does seem like any 13g with Mc5 might be close to refuted now.  If silver can force us to address threats while also activating the horse, we are probably behind.

Title: Re: 2015 Move 12g
Post by Lion on Dec 24th, 2015, 8:06am

on 12/23/15 at 13:56:07, aaaa wrote:
I don't see why Gold can't play the more obvious 13g Mb3w Ma3e ha4s xxxx here, after which the given 13s looks like a bad idea, as instead of recapturing immediately, Gold could frame the horse first (with a goal threat).


Won't silver respond 13s rh6s rh5s Rg5w hg4n, winning the rabbit soon thereafter?

Title: Re: 2015 Move 12g
Post by Lion on Dec 24th, 2015, 8:16am

on 12/23/15 at 18:52:18, arimaa_master wrote:
What about simple 12g E to c5?

For instance:
12g Ef4n Ef5w Ee5w Ed5w
12s Rg5w hg4n df6w Rf5n Rf6x
13g Ec5s mc6s hb6e Mb5n

and Gold seems to be fine.


Gold is just down a rabbit in this line, or am I missing something?

Title: Re: 2015 Move 12g
Post by clyring on Dec 24th, 2015, 10:01am

on 12/23/15 at 13:56:07, aaaa wrote:
I don't see why Gold can't play the more obvious [3a] 13g Mb3w Ma3e ha4s xxxx here, after which the given 13s looks like a bad idea, as instead of recapturing immediately, Gold could frame the horse first (with a goal threat).

Yeah, looks like I did an un-tactic. The line still feels dangerous to me; I'll have to look at it some more later.


on 12/23/15 at 18:52:18, arimaa_master wrote:
What about simple 12g E to c5?

For instance:
12g Ef4n Ef5w Ee5w Ed5w
12s Rg5w hg4n df6w Rf5n Rf6x
13g Ec5s mc6s hb6e Mb5n

and Gold seems to be fine.


We have some compensation, but I guess 13s c5eee e3n/Mf5 Ee4 is incoming with an eastern counterattack to follow. Is it enough? And like the 13g Ec5 line considered above, f6ss g5w h6s/Df4:R h5 is another possible answer, now threatening to take g3 with a goal-in-one threat. Maybe playable, but I definitely want to see a lot of discussion before we make a move like this one.


on 12/23/15 at 18:56:34, half_integer wrote:
After 1a.1 I think we also need to beware of
 13s Rb5w mb6s hb7s rc7w / M:Ra5 Hb6 Rb7
since this allows silver to get the camel for the framed horse, with 14s Ec4 or Ed5 if we don't respond.  Even
 14g Ef4wwww / Eb4 doesn't help:
 14s Hf3n Ee3nww / Hf4 Ec4
 15g Eb4s mb5s Ra5e Mc5e / Eb3:M Rb5 Md5
 15s Rb5e mb4n Rc5nx ec4n / M:Rc5 E:Rc6x

I'm not yet sure how we avoid or respond to this line.  The camel can leave after 13s but then silver can reestablish the hostage.  I'm not sure who benefits from this back-and-forth.

A wild idea I just considered would be to not defend the hostaged horse at c6 but bring the camel east and plan to rotate out the elephant next move.  Could a (somewhat) free elephant be worth a whole horse (and rabbit)?

If we go the more expected route, I think we clearly need to figure out how to flood the c6 trap with small pieces to free our camel, before silver can set up a horse-for-camel trade and abandon f3.

Yeah, when silver stops threatening anything on 13s, my first instinct is to immediately bring the camel east with [1a.1] 13s b5w b6s b7s c7w/Mb5:Ra5 Hb6 cb7 14g c5eee a2n/Mf5 a3 14s b6en b5n d6e/Hc7 Mb6 De6 15g e4wwn f5s/Ed5 Mf4 15s b6sn a6x/Mb6::Hx 16g d5wwe b6s/Ec5:M, which looks okay for us.

Another 12g idea that occurred to me and deserves some serious consideration is 12g g4ww f4ns/Ef4:He4.

My confidence in the frame as the best move has eroded somewhat. Maybe I was distracted by the thought of getting some of that juicy reserve back... :salivate:

Title: Re: 2015 Move 12g
Post by Lion on Dec 24th, 2015, 10:23am

on 12/23/15 at 13:56:07, aaaa wrote:
I don't see why Gold can't play the more obvious 13g Mb3w Ma3e ha4s xxxx here, after which the given 13s looks like a bad idea, as instead of recapturing immediately, Gold could frame the horse first (with a goal threat).


Upon further consideration, this might be a good option. It's a rabbit sacrifice in exchange for a camel hostage.

12g. Mb5s hb6s Mb4s hb5s
12s. mc6w mb6s Ha6e hb4w
13g. Mb3e ha4s Ma3w Dd2n
13s. rh6s rh5s Rg5w hg4n
14g. Ef4w Ee4w Ed4w Ec4n
14s. df6w Rf5n Rf6x Dd3w ee3w
15g. mb5s Ec5w mb4w Eb5s

Next, if 15s. ed3n ed4w ec4n ec5w then 16g. Cc2 Dc3s Rb1w Rg2e. In the end, the likely outcome is that a pair of horses gets traded, we lose the a5 rabbit but silver also loses a rabbit on h4, and silver camel gets stuck in the west while ours has free reign in the east. If this is correct, then this is good for us.

Title: Re: 2015 Move 12g
Post by Lion on Dec 24th, 2015, 10:49am

on 12/23/15 at 18:56:34, half_integer wrote:
After 1a.1 I think we also need to beware of
 13s Rb5w mb6s hb7s rc7w


Yeah, this is pretty good. It lets silver get his horse in the game immediately. However, this does not necessarily mean this is a bad position for us. If the silver horse goes after our rabbits in the east (to prevent elephant rotation), then we are likely to get control over the c6 trap, freeing our camel. On the other hand, the silver horse cannot easily penetrate through the center to attack c3, so perhaps it has to stay put in the center. If this is the case then maybe we'll have time to rotate the elephant. Therefore, I am comfortable choosing the horse frame for 12g.

Having said that, maybe the horse pull could turn out to be a better option, after all, as I noted in the post above. Gotta calculate some more ... :)

Title: Re: 2015 Move 12g
Post by half_integer on Dec 24th, 2015, 3:06pm
I also am not so sure that the frame is the best move anymore.  The double pull to b3 seems to work out OK, and the Ec5 also seems to threaten the camel enough to avoid silver attacking the f3 trap.  I'm not sure whether the tactics actually lead to a capture though.  Still, it moves the fight into silver's half of the board instead of around f3.

The frame seems to lead to silver setting up a camel capture for horse loss - however I think we can set up an additional capture or positional advantage which might still leave us ahead.  I recall that I think sharp is weaker at playing after asymmetric trades too.

Time to make an expanded tree I think.

Title: Re: 2015 Move 12g
Post by half_integer on Dec 26th, 2015, 8:42am
OK, the move (horse frame) has been made.  In our discussion, we only identified one silver reply to this move (12s Cd7 Rc7 Hb7 Mb6) so I think we can work on the next Gold reply.

The only piece that can protect the a6 horse is our camel, put on c5 or d6.  d6 doesn't appear desirable as it quickly leads to camel loss or various frames.  Mc5 alone is not good as it allows silver Mb5 Ed5 in one move which leads to immediate camel loss (Hb6 is false protection and the silver camel pull cannot be blocked).  To prevent that 13s I see two measures: pull the d6 dog with Mc5:Dd5 which allows Ed6 in 4 steps, or put a piece on b5, where Ra5b5 is the only one that can reach.  This latter move leaves us two free steps which we can debate below.

The wild-card move would be to not defend the horse and play for a rapid elephant rotation, with Mf5 or Me4.  There may be some moves with the same goal which do not use all four steps for the camel this move (though Me5 is not one of them: Dd5 E:Me6 is immediate camel-for-horse trade).

It is important to note that the camel-for-horse trade may be inevitable, at least in the lines where we defend c6, so rather than always reject lines with camel loss for horse sacrifice, we should determine if we can protect the a6 horse or get greater compensation.  Conversely, if we cannot protect the a6 horse long term then making progress now by not defending it becomes attractive.

So, an updated tree (SM notation only to save typing):

12g ::Hf3
12s Cd7 Rc7 Hb7 Mb6
13g
   1 Md6:Dc6
   - 1.1 Dc5 Ed5 (Hf3x) - indefensible camel loss with horse hostage

   2 Md6:De6
   - 2.1 Ed5 (Hf3x) Mb5
   - - 2.1.1 Ec5
   - - - 2.1.1.1 De7 Cg7 Df7 Hb6 - f6 cannot be defended but maybe a camel trade can be threatened?
   - - - - 2.1.1.1.1 E::Mb3
   - - - - - 2.1.1.1.1.1 Ec4::M
   - - - - - - 2.1.1.1.1.1.1 (E/R) Md5 Dd3 - and we seem to have freed our camel and gotten a camel hostage after two more moves?
   - - - - 2.1.1.1.2 E:Mb4 (Dd3 Hg4)
   - - - - - 2.1.1.1.2.1 Ec4::M - and our camel seems to have escaped
   - - - - - 2.1.1.1.2.2 Ec4 Ma4 X - our camel escapes here too
   - 2.2 Ed5 (Hf3x) X
   - - 2.2.1 Dc5
   - - - 2.2.1.1 E:Mc6 X X - camel loss is unavoidable
   - - 2.2.2 Ec5
   - - - 2.2.2.1 E:Mc6 X X - this frame looks like an easy elephant rotation for silver, though we are up a horse in the short term
   - - - 2.2.2.2 De7 Cg7 Df7 X - f6 cannot be defended, but this time it is harder to make camel threats in response

   3 Mc5 X X X
   - 3.1 Ed5 (Hf3x) Mb5 - camel is indefensible and we still have a horse hostaged

   4 Mc5:Dd5 - note I didn't find any really bad responses to this yet
   - 4.1 Ed4 (Hf3x) Dd6 Ed5
   - - 4.1.1 Ec4 Mb5 - doesn't appear profitable for silver, he gets a horse trade at best
   - 4.2 Ed4 (Hf3x) De5 Ed5
   - - 4.2.1 Ec4 Mb5 - still just a horse trade for silver
   - 4.3 Ec4 (Hf3x) Mb5
   - - 4.3.1 Ed6
   - - - 4.3.1.1 E:Mc6 Mb6 X - this frame looks difficult but not impossible to rotate, so this position might be OK for gold (up a horse for a camel frame)
   - - - 4.3.1.2 E:Mc6 Hb6 Ra7b7
   - - - - 4.3.1.2.1 Ha7 M:Ha6 X - a strange little position which might leave the silver camel as strongest free piece but also gives gold access to the back ranks in the west
   - - - 4.3.1.3 is there an even better silver move?
   - 4.4 Ec4 (Hf3x) Cd6
   - - 4.4.1 Rb5 M:De5 X - silver seems to get the horse and rabbit at c6 but it looks like gold's camel can always escape

This has taken a while so I'm going to edit in more later.
   5 Mc5 Rb5 X X - to be expanded
   6 Mf5 / Me4 / Md4 X / Md5 X X / Mc4 X X - lots of work to do here

I noticed that I'm almost exclusively focused on the silver moves which sacrifice the framed horse; there is the possibility of silver responses which defend but don't try for a trade.



Arimaa Forum » Powered by YaBB 1 Gold - SP 1.3.1!
YaBB © 2000-2003. All Rights Reserved.