Could the Conflict Of Heroes solo system work for ASL?

[peterk1]

Hi Guys,

I haven't posted here (or played ASL) for quite a while but I pop by every now and then to see what is going on. Something happened recently that has got me thinking about solo ASL again. Have any of you tried out the solo system just published for the Conflict Of Heroes game? It's a card driven AI, where the deck is prepared ahead of the scenario. It also has a very nice way to generate unit placements. And I've been thinking whether something like this could be used to finally have a good way to play some of the great standard ASL scenarios solitaire.

Things I like about the CoH system:

o At the beginning of the scenario, you drop all the counters for the AI's OOB into a cup. Suspect markers are placed in a bunch of good setup locations. When one is revealed, you draw a unit counter from the cup. What's nice about this is that you can easily and quickly get a reasonable set-up for a standard scenario. In a custom solo scenario you can limit the possible OOB for a more consistent experience between different sessions/users than is possible with standard SASL. If you want, you can even have more units in the cup than Suspect markers, to add a tiny bit of variability to what will actually show up. You can also have different cups (a cup for heavy weapons in upper floors; cups for different buildings/set-up areas; locations for leaders). If you want the possibility of dummy Suspect markers, a dr before pulling a counter from the cup does the trick.

o The order in which AI actions are resolved is driven by cards and not the fixed left/right north/south in SASL and it leads to a much more intelligent ordering of actions. Important actions tend to get performed first very consistently. As an example, if the AI has a large killer stack meant for Prep fire, then it will almost always lead the turn, before a sniper can wreck the stack or kill the leader.

o The action that a unit attempts is context sensitive and not driven purely by a random DR. Each card has 5-6 possibilities on it and you read from top/bottom until you find an action that applies.

o AI movement is categorized into low risk / normal risk.

o The deck is prepared ahead of the scenario and can be massaged as the scenario progresses in response to how things are going. In CoH, if you inflict a large number of enemy casualties, they will lose Command cards and the AI side will progressively become more predictable and less aggressive. In ASL, for Guards Counterattack for example, you could remove a lot of the "prep fire" cards on the last turn to get more movement if the AI still has to take a building for the win. If the AI has the win, you could take out all or most of the movement cards to make it stay put and defend.

o You don't need to give the AI a huge numerical advantage for a challenging game. About half of my plays have had the German/Russian fairly evenly matched.

I've had a ton of fun with this thing in the last 2 weeks and was just wondering if anyone else has played it and also thinks there are possibilities in the ASL world for such a system. I like CoH, but I miss playing ASL, but I don't think I can play the normal solitaire module any more.

 

[Michael Dorosh]

Hi Guys,

I haven't posted here (or played ASL) for quite a while but I pop by every now and then to see what is going on. Something happened recently that has got me thinking about solo ASL again. Have any of you tried out the solo system just published for the Conflict Of Heroes game? It's a card driven AI, where the deck is prepared ahead of the scenario. It also has a very nice way to generate unit placements. And I've been thinking whether something like this could be used to finally have a good way to play some of the great standard ASL scenarios solitaire.

Things I like about the CoH system:

o At the beginning of the scenario, you drop all the counters for the AI's OOB into a cup. Suspect markers are placed in a bunch of good setup locations. When one is revealed, you draw a unit counter from the cup. What's nice about this is that you can easily and quickly get a reasonable set-up for a standard scenario. In a custom solo scenario you can limit the possible OOB for a more consistent experience between different sessions/users than is possible with standard SASL. If you want, you can even have more units in the cup than Suspect markers, to add a tiny bit of variability to what will actually show up. You can also have different cups (a cup for heavy weapons in upper floors; cups for different buildings/set-up areas; locations for leaders). If you want the possibility of dummy Suspect markers, a dr before pulling a counter from the cup does the trick.

o The order in which AI actions are resolved is driven by cards and not the fixed left/right north/south in SASL and it leads to a much more intelligent ordering of actions. Important actions tend to get performed first very consistently. As an example, if the AI has a large killer stack meant for Prep fire, then it will almost always lead the turn, before a sniper can wreck the stack or kill the leader.

o The action that a unit attempts is context sensitive and not driven purely by a random DR. Each card has 5-6 possibilities on it and you read from top/bottom until you find an action that applies.

o AI movement is categorized into low risk / normal risk.

o The deck is prepared ahead of the scenario and can be massaged as the scenario progresses in response to how things are going. In CoH, if you inflict a large number of enemy casualties, they will lose Command cards and the AI side will progressively become more predictable and less aggressive. In ASL, for Guards Counterattack for example, you could remove a lot of the "prep fire" cards on the last turn to get more movement if the AI still has to take a building for the win. If the AI has the win, you could take out all or most of the movement cards to make it stay put and defend.

o You don't need to give the AI a huge numerical advantage for a challenging game. About half of my plays have had the German/Russian fairly evenly matched.

I've had a ton of fun with this thing in the last 2 weeks and was just wondering if anyone else has played it and also thinks there are possibilities in the ASL world for such a system. I like CoH, but I miss playing ASL, but I don't think I can play the normal solitaire module any more.

Sounds interesting.

 

[pcfrederick]

I know that there are some "homegrown" solo versions of doing ASL outside of SASL. I'd be interested in a system that can take the 7,000+ scenarios and make them more or less solo with some sort of an AI. Can someone provide some thoughts on Mike O'Leary's version of solo ASL?

 

[peterk1]

I know that there are some "homegrown" solo versions of doing ASL outside of SASL. I'd be interested in a system that can take the 7,000+ scenarios and make them more or less solo with some sort of an AI. Can someone provide some thoughts on Mike O'Leary's version of solo ASL?

I haven't looked at it for a long while, but it was super, super basic. Not at all what I'm after. IIRC the main idea was randomizing the defensive setup by using concealment counters. And that was basically it. Otherwise, it was mainly the old boring "play both sides to the best of your ability". It was meant as a way for players to play both sides of scenarios having concealed units in the OOB. Very narrow scope.

In CoH, the player has to make very, very few decisions for the system. There haven't been any deck of cards systems done yet, unless you count one of Wild Bill Wilder's things.

I'm easier to please than you. I'd be happy with something that lets me play just a few of the better, less complicated scenarios without having to play both sides to the best of my ability.

Did a little brainstorming tonight using ASL SK1 and the first scenario and I came up with some ideas that work well - but it's far too mathy/thinky to be smooth playing.

 

[Akashi]

Hi Guys,

I haven't posted here (or played ASL) for quite a while but I pop by every now and then to see what is going on. Something happened recently that has got me thinking about solo ASL again. Have any of you tried out the solo system just published for the Conflict Of Heroes game? It's a card driven AI, where the deck is prepared ahead of the scenario. It also has a very nice way to generate unit placements. And I've been thinking whether something like this could be used to finally have a good way to play some of the great standard ASL scenarios solitaire.

Things I like about the CoH system:

o At the beginning of the scenario, you drop all the counters for the AI's OOB into a cup. Suspect markers are placed in a bunch of good setup locations. When one is revealed, you draw a unit counter from the cup. What's nice about this is that you can easily and quickly get a reasonable set-up for a standard scenario. In a custom solo scenario you can limit the possible OOB for a more consistent experience between different sessions/users than is possible with standard SASL. If you want, you can even have more units in the cup than Suspect markers, to add a tiny bit of variability to what will actually show up. You can also have different cups (a cup for heavy weapons in upper floors; cups for different buildings/set-up areas; locations for leaders). If you want the possibility of dummy Suspect markers, a dr before pulling a counter from the cup does the trick.

o The order in which AI actions are resolved is driven by cards and not the fixed left/right north/south in SASL and it leads to a much more intelligent ordering of actions. Important actions tend to get performed first very consistently. As an example, if the AI has a large killer stack meant for Prep fire, then it will almost always lead the turn, before a sniper can wreck the stack or kill the leader.

o The action that a unit attempts is context sensitive and not driven purely by a random DR. Each card has 5-6 possibilities on it and you read from top/bottom until you find an action that applies.

o AI movement is categorized into low risk / normal risk.

o The deck is prepared ahead of the scenario and can be massaged as the scenario progresses in response to how things are going. In CoH, if you inflict a large number of enemy casualties, they will lose Command cards and the AI side will progressively become more predictable and less aggressive. In ASL, for Guards Counterattack for example, you could remove a lot of the "prep fire" cards on the last turn to get more movement if the AI still has to take a building for the win. If the AI has the win, you could take out all or most of the movement cards to make it stay put and defend.

o You don't need to give the AI a huge numerical advantage for a challenging game. About half of my plays have had the German/Russian fairly evenly matched.

I've had a ton of fun with this thing in the last 2 weeks and was just wondering if anyone else has played it and also thinks there are possibilities in the ASL world for such a system. I like CoH, but I miss playing ASL, but I don't think I can play the normal solitaire module any more.

The problem is activating enemies. COH is a I go you go system, whereas ASL is "I do all my stuff then you do all your stuff". The COH system do not work for that, you can not identify an enemy unit based on your last unit that just moved or fired. You can also not determine if the AI is going to OP-fire (or defensive fire) nor will the AI system be able to determine if the enemy wants to move or do prep-fire. When you think of it, you need either to play ASL as I go you go and that will lead into a whole new game (no phases anymore) or rewrite the cards...

 

[peterk1]

or rewrite the cards...

It would definitely be with brand new cards. I wasn't really thinking about using the CoH cards themselves - it was more the technique of using cards to drive the game and make decisions as as a whole. You could have things like "largest group of units farthest from VP Hex or Target <do something>". You'd be able to get behavior like "units in contact with the enemy will prep fire and approach VP hexes cautiously and units farther away will move to contact quickly/aggresively" fairly easily. The concept still works and it's faster and more fun than rolling dice for everything.

You could certainly flip cards to determine whether an enemy will Def Fire on you in a given hex as you move (using strength of shot, TEM/modifiers or distance from the shooter) which would bring a lot of suspense to your movement phase. If I remember correctly, SASL is fairly deterministic with it's Defensive Fire choices.

You're correct, though, it's easier to do in CoH because it is an impulse game and ASL is a "plan/execute a move for your entire side". It's harder to make a bad/stupid move in CoH. It definitely changes things a little. I've heard rumors that the Lock n' Load system will be getting a version of the CoH system for all of its games, but I think that's another impulse type game, so it's a better fit.

 

[pcfrederick]

Texas ASL Banzai magazine had an article on a hybrid between ASL and SASL, although I think it is more about adding FOW for 2-player game.

 

[Michael Dorosh]

It seems like apples are being asked to mix with oranges - to wit, a random action generator on the one hand, but a fixed order of battle on the other.

The easiest thing to do would be to surrender the idea of playing pre-made scenarios in a solitaire format since the two are, if not incompatible, then at the least cumbersome.

I say this because I get the feeling the "new style of scenario" - say, 5.5 turns, CVP cap plus exit/control of this or that - is so heavily designed/"balanced"/tested as to be virtually scripted. I played a Journal scenario just today - one board wide, 5.5 turns or so, 15 German squads attacking down this single board. Compare to the old school ASL scenarios of 10 or 12 turns. There was less likelihood of a set "solution" for the scen (not to mention you actually had time, if desired, to do clearance attempts, entrenching, etc.)

With that being said, if it true that some scenarios have "solutions", then coming up with a random system of actions that will replicate that solution would seem to be difficult, unless your solitaire system also had a "script" for each specific scenario.

Which isn't a bad idea - look at Ambush, for example - there were different Conditions (usually 4 or 6 per scenario) which guided the overall flow of the game. You could possibly do such a thing for an ASL system - say a German company (-) is defending a village. Condition 1 would be the advance to contact, Condition 2 the outposts defending the line, Condition 3 the outposts fall back, Condition 4, reinforcements counter-attack (all this according to standard German defensive doctrine).

But I think it would be difficult to have generic rules with a fixed order of battle.

Which means another way to go would be to get rid of the fixed order of battle from the scenario card for the AI side.

 

[Michael Dorosh]

And thinking further on this, you probably could create a system just like Ambush! for "scripting" one side of the other in ASL. It would not be a true solitaire on the fly system, as it would require someone to spend a lot of time writing, or 'programming' the AI, for each scenario. But if you really wanted to retain the fixed order of battle, you could have the same kind of hex-by-hex "operations" model from Ambush! where each movement might trigger a specific "sighting", i.e. an enemy defensive position which is taken from the fixed order of battle, and set up by whoever programs the AI.

Who knows? Such a thing could get popular if it was popular and enough people wanted to contribute such things to the community. Programming AIs could become just as popular as designing scenarios. Would take at least as much skill.

 

[peterk1]

Hmm, not all that enthusiastic about the idea here. Perhaps an example would help? I spent a few hours today thinking about it again and fooling around and made up some prototype cards and rules to take for a test drive. This is Retaking Vierville and we're coming in at US turn 2 of 5. Germans currently occupy 2 of the 4 victory buildings and would win if the scenario ended now.

I've actually been running both sides using my prototype to give it a better workout. I've had to hold myself back from jumping in and taking a side.

===================

US turn 2. Rally phase has just occurred and we're beginning PFPh.
3 reinforcements are arriving from the north who obviously won't be firing. Let's check the others.

Starting with strongest groups and working down.
The strongest group is the 9-1 and 2 squads in building hex M4. There is another -1 leader and 2 squads further up north in the orchards, but this one has an adjacent 747 that can be pulled in to a firegroup if need be so this group is stronger.

These guys have a few targets. The obvious shot to M6 right across the street is probably what will come up, but there is also a long range shot to a 548 in K8.
Lets see what the cards say!

View attachment 52781

The part of the card that we're interested in is the Def Fire >=8 right at the top. Initially I was only using it for Defensive fire decisions but I've started using it for Prep Fire decisions as well. I'll be changing that label when I make my next set of cards. As an aside, currently the deck has 20 cards in it, and I haven't exhausted the deck yet in a single half-turn, so you don't need to draw all that many cards.

The card is saying that the group of units in M4 will prep fire if they have a shot available that is the equivalent of a 8(0). Since the 2 possible targets are both in stone buildings, that means I would need to be able to muster a 20(+3) or a 16(+2). All of these 3 shots have the same probability of results on the IIFT. All I'm doing is converting the +3 TEM and the -1 leader bonus into column shifts to get a single number telling me how strong the shot should be. It makes it easy to compare choices that way. The 2 squads the leader is stacked with would only produce a 12(+2) shot so I would need to fire group with the neighboring squad to get a 20(+3) shot on M6. I lose the leader bonus but no big deal.

Normally I would draw a second card here to determine the target, but there is only one possibility with the card I drew for firepower so I don't have to. The other potential target is at long range and I wouldn't have enough firepower if I took that shot.

US rolls 9, bumped up to 12 and it's a PTC. 8-1 German leader fails and is pinned, but 467 squad passes.

Nice consequence of all that is that the leader hasn't prepped and there happens to be a broken squad not too far away. He will probably try to move over there to help with the upcoming rally.

Next Prep Fire choice will be the 8-1 and two 747s up in the orchard to the north. They have some long range targets in the grainfields. Let's draw a card.

View attachment 52782

It's an 8 again, so a 8(0) equivalent is needed. If all of those guys fire together at long range with the leader directing it would be a 6(-1) which is equivalent to 8(0), so all of those guys will prep on O8. N8 is not a target because of the extra grain hindrance. It's not a bad decision. US need to gain back 2 victory locations, but it's still early on and there's no great pressure to move out now. And once again, no need to draw for a target. Bit of a shame, I would have liked to demonstrate that.

The cards list FP values from 1-8, so both of these draws were on the high side. With lower numbers in those 2 cards, we would have had weaker shots but a squad or 2 left free for movement. An 8 FP card results in a shot that has roughly a 50% chance of getting a result on the attack. That doesn't mean that the AI will never take shots using more than 8FP. As we've already seen, if the terrain is unfavorable, he can bump that up 3 columns and go for a 20FP attack. Hindrances such as smoke will lead him to trying to bring even more force to bear. The AI will try to get a good shot with as few resources as possible. We may see a better example of how that works in the DFPh.

Stack fires to the grainfield at 6(-1). Roll is 7 and thats a 1MC. 9-1 leader rolls snake eyes and is wounded and broken. The 548 squad likewise breaks.

Theres only one more US unit on board to resolve; a CXed 747 right across the road from the group that just got hit! A chance to draw blood. Should be interesting. Let's draw the card.

View attachment 52783

Another 8 and unfortunately that unit can only produce a 6FP shot to open ground, he has no negative modifiers - he has a +1 due to CX, so he does not prep fire due to lack of firepower and will move. I would have needed a 4 or lower card for this shot to happen - there was a 60% chance.

There are 4 8FP cards in the deck and we've just drawn 3 of them one after the other.

End PFPh.

=================

So what's different really between this and SASL? Well in SASL I would be processing the units geographically in a fixed scan of the board - here I'm processing the units/stacks in order from strongest to weakest. It's a small detail, but doing this will result in a stronger game. When moving units, I will likewise process those from smallest/weakest to largest/strongest and that will also result in a stronger AI game.

Another difference is that in SASL I would just be rolling on a table for whether a unit Preps or not. The actual position on the board and the quality of the shot available don't matter. Here it's dynamic and the available offensive units and defensive terrain of the defender all matter. When I get to turns 4 and 5 the rules will change slightly and the player which is winning or losing the game may also have an impact on the choice. It mimics a human player quite well. The AI got results on both of it's shots. SASL would never have formed that firegroup losing the leader bonus and SASL would never have 1 unit in a stack fire while another moves later on. I can get both of those with the cards.

Last difference, I didn't have to consult any rules/charts doing this whole procedure. It was natural and quick - and fun.

I'll be doing some movement next. The card based system really shines during movement and defensive fire and IMO makes it as exciting as in a 2 player game.

 

[peterk1]

The easiest thing to do would be to surrender the idea of playing pre-made scenarios in a solitaire format since the two are, if not incompatible, then at the least cumbersome.

Not sure about this. The magic of ASL really comes from its library of great scenarios. If you throw those out - you might as well build a brand new game from the ground up that is 100% solo friendly.

It's interesting really - if you're willing to just play both sides, ASL is far better than CoH. In CoH, you flip flop between the two sides so quickly, it's difficult to identify with either side. But attaching the solo system onto CoH, I imagine was quite smooth, and once that is in, the problem goes away entirely and you completely identify with your own side and want to crush the AI.

Didn't the ASL "clone" game from Critical hit (forget the name now - ATS?) have some sort of solitaire robot? Has anyone ever played that?

 

[telengard]

This is pretty interesting. I've played this expansion with CoH and think its brilliant, never thought of applying it to other systems. When you have something to share I'd like to give it a spin...

 

[peterk1]

Thanks for the vote of confidence! If things go well, I'll try to put together something to share. Might take a while though.
I've been playing CoH non-stop the past 2-3 weeks. While I had a lot of fun with it, I was a little disappointed by the scenarios that were provided. For the most part, they were way too small. Even something smallish like retaking Vierville has more pieces for the player to control than the biggest scenario in CoH Solo. I have to confess that all the while, I was just dreaming of having something like it for ASL, and I'm too impatient to wait and see what they do for Lock 'n Load solo.

 

[peterk1]

US turn 2 movement phase.


Americans have 2 units on board who can move and 3 more entering play from the north. I won't do anything fancy here - just a simple dr to determine whether I enter an offboard unit now or not.


The dr indicates that a reinforcement will enter, so I toss them into a cup and draw one. There are 2 747 squads and an 8-0 leader, so all I am really randomizing is whether the leader will enter play now or not.


Since these guys will be out of the action when they enter (not within range of a VP hex; not in an enemy's normal range, regardless of LOS), if I draw the leader first, then the entire group will enter and move together.


View attachment 52793


And it's a squad that enters on row V. I don't need the cup any more since the next reinforcement to enter will definitely be the leader/squad pair.


Now to do his move. This is one of the trickiest things to devise an automated system for but I'll try to describe what I have so far.


As I mentioned earlier, these units are out of the fight when they enter. They are not in the normal range of any enemy and they have no enemy in their normal range. The first task is to get closer to the battle as quickly as possible. Rather than micromanaging the move hex by hex, we'll try just to pick a good destination hex.


The first thing on my priority list to look for is a hex offering cover within normal range +1 of an enemy unit. The closest enemy unit are the broken guys in O8, so I look for a hex offering cover within 4 hexes of them, that my 747 can enter or get adjacent to in this movement phase.


There's a few good possibilities. We'll break ties using the TEM offered by the cover and now there are only 2 possibilities. The building in Q6 or the building in R7 which can both be reached by using CX.


Now we need to check if the moves would be safe. Basically, can the unit get to the destination without incurring a FFMO? Both of those buildings can be reached safely, but Q6 can be reached without entering enemy LOS at all, so that is the winner.


The squad enters play on V4.

View attachment 52787


and moves all the way to R5 for 4MP.


Next unit. Random selection for on board or reinforcement returns a reinforcement, so now the squad and 8-0 leader enter play. We've already done some of the analysis for this types of move for the last unit, but this group can move a few hexes further. Does that give us any new possibilities?


Yes, it does! If this group goes CX, they can even reach N5 or O6 which is adjacent to two VP hexes, so the orchard in O5 becomes the movement target.


The group enters play on the westernmost road, on hex V4, as before.

View attachment 52789

They can move all the way to Q4 for 6MP completely out of view of the enemy.

View attachment 52790

And then they use the 2MP gained by CX to move to the orchard in O5. At the very last hex of the move, they are spotted by the German 467 in N6 who may decide to fire. If that last hex was open ground rather than Orchard, then this move would have qualified as being risky, and I would have had to have drawn a card to approve the move before starting it. Let's draw a card and see if the German unit will First fire.

View attachment 52794

 

[peterk1]

The 467 will fire on a shot of 1(0) or equivalent and they have a 4(-1). That shot would have occurred on any card draw except an 8 and there was only a 1/17 chance of that.

German rolls 6 and that is a 1MC.

Leader and squad both pass.

Now there are only 2 more units left to move. A 747 which is about to lose a CX marker and the 9-1 leader. These we will do in a priority order and the squad in P6 will go before the leader. It's an obvious move. Assault move adjacent to the VP hex in N6.


A whole bunch of units can fire at this guy - the unit which just fired can Final Fire. A 548 in the grainfield across the road can shoot and there is also a 467 in the other VP building which hasn't acted yet. As always, let's draw a card.

View attachment 52795


A shot will happen on a 4(0) or equivalent. The 747 is in a stone building, so that means I need a 12(+3) shot. I don't have it, so no fire. It's worth noting that if the German 8-1 leader would not have pinned in the PFPh, then there would have been a shot here.


Lastly, the leader in M4 will try to move to K5 in an effort to help the half squad in K6 rally. This is a safe move. I'm not sure, if the leader should really be doing this for best play, but SASL always gave priority to leaders moving to help with rallying, so I'm going to test with that being an automatic move and will see what sort of results I get in repeated plays. It's worth noting that if the leaders modifier would have been required in the PFPh, then he wouldn't have made this move, so variation is still very much present and it comes out of the card draws.


To L3 for 2 MP. We need to check if M6 will shoot.


View attachment 52796


FP 8. As long as the leader sticks to the stone buildings, the best shot from M6 is a 4(+2) which is equivalent to a 1(0). L3 to K4 for 4/6 MPs. In this hex, we are also in long range of the 548 in the grainfield. Another card.


View attachment 52797


FP2 on the card is still not good enough and the leader completes his move by entering the most dangerous hex in his voyage. K5 for 5MP. There will probably be a shot here.


View attachment 52798


We need a 2(0) shot or equivalent. The 467 firing across the road would be 4(-1) and the 548 shooting long range from the grainfield would be 2(-1). So they both qualify. Currently my priority lists say to use the unit with the least FP, so that would be the 467. The logic is to achieve the firepower objective from the card while preserving as much FP as possible to deal with future moves. The AI doesn't know that this is the last unit to move and I want to keep things simple and not have special rules for that. The 467 also has the best shot, in this instance, so it also happens to be the best play anyways.

He rolls a 3,3. The attack cowers to 2FP because of the pinned leader and that's an NMC and the leader passes.


I'm currently testing the system with only one Defensive Fire attack per hex entered, so that's all, even though it would make sense for the 548 to shoot now rather than in the Defensive Fire phase.


View attachment 52799


End MPh

===========================

And that's a small example of what's possible using a little deck of cards to drive the action while playing a standard scenario solo. I've been impressed enough to continue fooling around with the system to see if it has some legs.

What I like better using the cards opposed to traditional SASL rules is that rather than have hardcoded rules guaranteeing that a shot will occur, it's dynamic and there's some suspense every time you move to a new hex and draw a card. In this experiment, the normal SASL rules probably would have behaved the same way, but I have had some instances where a totally unexpected Defensive Fire shot has occurred.

...but Michael is correct in an earlier comment, in saying that, comparing this to SASL is apples and oranges. One is a mission generator and this is a way to play standard scenarios. Still, I'm shocked so far as to how well it does work.

 

[peterk1]

Not sure if anyone is interested, but I have been elaborating on the idea I posted here and have been experimenting with a card driven ASL AI that can play normal scenarios for almost a year. Just recorded a few videos showing how things work:

Rules have been pretty stable for about 6 months. I've been focusing on SK1 to get the fundamentals down well and keep things simple, but I've also played a bunch of stuff from Beyond Valor and just finished up a play of the Decision At Elst CG. I've gotten perhaps 40-50 games out of my bot last year, and for the most part they have all been interesting and a great way to get familiar with a scenario to prepare for HtH play.

 

[Michael Dorosh]

Just noticed the discussion of this on Facebook. Not much into watching videos - what is the intent for distribution of the rules (if you've thought that far ahead)? Would love to try it for myself. Will you be self-publishing, publishing through a TPP, or will you be releasing into the wild?

 

[peterk1]

Hi Michael,

The rules and basic cards are here
https://www.dropbox.com/sh/ssr1rray8ktwp7g/AACVTa_wDgeXdfSqYIh5609La?dl=0

For the nicer cards that I use in the video, visit the discussion group on Boardgame Geek and go towards the end of the thread to get a dropbox link. They should also be in the BGG files section.

https://www.boardgamegeek.com/thread/1546763/solitaire-asl-bot/page/1

 

[GerryM]

I recently played an ASLSK scenario using Peter's BOT system - even has a card plug-in for VASL.

I played a little SASL years ago. I must say Peter's system feels much more natural and engaging. I even found myself rooting for the BOT at times. And it plays quickly once you get the hang of it.