Recommend
6 
 Thumb up
 Hide
40 Posts
1 , 2  Next »   | 

Pax Porfiriana» Forums » Variants

Subject: New solo variant rss

Your Tags: Add tags
Popular Tags: [View All]
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Hi everyone,

I just wanted to mention that my new solo variant, the Pax-o-matic, has been approved by the Admins, and the rules file is now available for download here:

The Pax-o-matic - A diceless solitaire variant for Pax Porfiriana

Any feedback would be most welcome. The idea was to remove the use of a dice roll, and instead use some basic logic to drive the AI actions. This does make the AI more time consuming to use, but hopefully it gives the feeling of a (semi-)realistic opponent. It won't always do the most optimal move, as that would require an extremely cumbersome instruction set, but the combination of 2 opponents that try to beat you should make for a decently tough opponent.

Thanks,

Colin
18 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Bryan Felsher
United States
Rancho Palos Verdes
CA
flag msg tools
mbmbmb
Wow Colin! This is excellent. Thanks for your effort. It seems to me that this should make a good tough game, as the bots play quite defensive (which is probably the best idea for scripting the bot as uncomplicated as possible.). The chart is very easy to follow, and is simpler than the AI in some of the wargames I play. Bravo, and I look forward to trying it out. I may get a chance on Wednesday. I have think this could also be a good training aid for newer players looking to learn the strategies of the game, as the flowchart seems to follow good principles of common sense play, which will require cunning, backstabbing, and underhanded play to beat. That's Pax Porfiriana!
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Thanks, Bryan. To be honest, I'm quite a novice at the game, having only played it once before embarking on this variant project. I would be keen to hear what people think, as this is realistically still a work in progress. I am willing to consider suggestions, and update the bots as needed. However, I am trying to stick to my semi-artificial constraint of 2 sides of paper. That limit helps me stick to a reasonable number of rules, and not get too complex. But the areas of interest to me are:

1. Win conditions - I played around with various options, but ended up at just the normal rules. But, as you said, the bots play quite defensive, so they are not hugely likely to win the game early. Therefore, I had considered the condition that the human must win a Topple, with no Gold Victory possible. That way, the bots play more similarly to a COIN bot, i.e. they try to win if possible, but really, they are essentially hoping to win the long game (in COIN, the human player cannot win until the last/4th victory check). I also played around with the idea of using the senility scoring variant, but felt that the 4th Topple would be too easy for the human.

2. Command Victories - I am a little concerned that a Command Victory may be the best path for a human to take. With 2 Command points on offer from the Public Cards, and the bots unlikely to ever buy them, it may turn out to be too easy. Of course, there are ways around that, but I wanted to start with the bots using every normal rule, to ease the feeling that the bots are cheating in some way. We'll see how the feedback goes.

3. Difficulty options - I have some ideas on how players can modify the difficulty, but would be up for suggestions. My current options are:

i. Start the Pax-o-matics off with more money
ii. Use 3 Pax-o-matics (probably gets too cumbersome)
iii. Increase Diaz prestige for the human, but not the bots

Thanks,

Colin
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
Thanks for this Colin - just gave it a run through after seeing it earlier today. Much better than the Diaz dice variant - it really gives a sense of the whole intrigue of the game, but it's still different to just playing separate hands as well, which I have done when I was learning the game. The three players really makes the decisions a bit more interesting and of course largely the bots are less random.

Don't be alarmed, but I won on the first topple. I think it was just the luck of the thing, but see what you think.

I drew Orozco (Revolution and Command), and the Bots were Reyes (Loyalty and Command) and Terrazas (Loyalty and Command). So revolution was always going to be a good shout and that's the way it worked out.

Terrazas seemed to get stuck picking up mines and banks in the 0 column that he then couldn't play because he couldn't afford it. Then he'd get a relatively useless partner and play that instead just as he was building up a few coins. He did nationalise an enterprise of mine though at one point but that was about as far as he could get.

Reyes was doing much better for himself. At one point he picked up a plantation with slaves so I liberated that as quickly as I could.


Anyway, we were in anarchy and the right topple came out. I only needed 4 - Reyes had a Revolution point and so did Terrazas- and on top of the plantation, and another revolution partner, I had the cash to get the public card, buy the topple and flip my hacendado.

Where I'm not sure is, there were no opportunities to change the regime, and there were cheaper options than the topple for the bots to buy still. A human player would probably have seen it unfolding, I think - seen that I had maybe the chance to pick up the 2 extra points, even if I didn't have them yet - and just bought the topple early. But your rules say, the cheapest option that satisfies one of the criteria, so I did it to the letter and won (please let me know if I'm interpreting that right!)

I'll give it another go at some point for sure - like I said, I think it was just everything falling into place. But the senility variant would fix it I think. I'll try at least one of each and get back to you on how they go.

1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Hi Thomas,

Thanks for the feedback. It sounds like you did everything correct, so I wouldn't worry on that front.

To be honest, the Public cards are what worry me most about the variant. I wanted to keep maintenance of the bots to a minimum, so rather than have the human player having to look across the entire market each turn, I have the bots prioritizing the cheapest cards first. Of course, this means that they will rarely get the chance to buy a Public card, as there are so many other cards to get through first. This could lead to a human player manipulating the game through those cards. I was particularly worried about the Command cards, seeing as they are so rare in the deck, but have 2pts on offer through Public cards.

I do have the first card purchase as Topple, so it can opt to remove it, if the human is losing. But, of course, taking a Topple early is not likely, as there will be many other cards to consider in the market first. I could alter the rules such that it considers a Topple anywhere first. I don't generally like exceptions, but this would be a minor one, as the Topple is already the first option. I could just reword to say any Topple.

I had thought about saying the game uses the senility variant, which would make an early win less likely. But I was worried about the ease of a 4th Topple win instead. Maybe that is unfounded, and it's worth changing. If this seems viable, I'd prefer that over the Topple option above.

Just a quick question, were there no Headline cards that would have altered the Regime? Remember that it will take a Headline, even if it hurts them discard-wise, as long as it changes the Regime away from a human win.

Luckily, we have many ways to alter the difficulty, officially, and unofficially. You could:

- set your own Diaz settings, say 3,2,2,2
- give the Hacendados slightly more cash to start
- make it impossible for you to win on a Gold Victory (after the 4th Topple)
- allow the bots to play an Orange/Black card for Strawman points. Right now, it will only use Orange cards for Scapegoat plays, but allowing Strawman actions, it may help. Do you happen to remember if this would have done anything? I'm assuming not in this case, as you didn't lead until the turn you took the Topple.

I think, for now, I will leave it as is, and as we get more feedback, we can consider tweaks, like those listed above. I don't, by any stretch, that this variant is finished, so this is great input. This is my first attempt at a solo variant of any kind, so I'm convinced there are loopholes and strategies to winning that I didn't consider. BTW, someone else played the bots over on the 1 Player Guild, and he lost, so I know that it's not a certainty.

BTW, ignoring the 1st Topple win, how difficult did you find the rules to implement? I know that removing the dice and replacing with a flowchart is more complex, but did it ever feel burdensome? During testing, it's hard to tell, as my games are very slow, due to me analyzing every move, to see if the rules need to be altered. I haven't really ever just played it straight up.

Thanks,

Colin
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Oh, and one more thing. Seeing as the bots like Partners, I just wanted to check that neither of them picked up a "During Toppling" card? In my games, they always seem to have at least one of those, making it tough to Topple on the first card.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
There were no headline cards with regime changes - again, the luck factor! I'd also killed the one partner card out there that had "during topple discard for 1 revolution point".

I also must have misread the rules because I did have terrazzo do a strawman early on. I'd forgotten this. He did get an enterprise out but then nationalized it because it was the only card he could play (I thought). It was a bit odd and then he got stuck as I mentioned.

As for ease, I'm not in front of the document. It was mostly okay, I might have some suggestions when I get a chance.

I think the fact that someone else has played it out and lost is a really good sign for it given my experience. Just like real pax, you never know...
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
OK, so you didn't miss a Headline, so that's good. The Pax-o-matics will not use an Orange/Black card for Strawman actions, only Scapegoats, but this is something I'm considering adding, if the Strawman takes the lead away from the human.

As for rules clarity, I bet after another try or 2, you'll have the rules down. I try to keep them to 2 sides max, otherwise they get too weighty. I'm also working on similar projects for Panamax and Concordia.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
One question - some rules require a prestige point comparison. I played that I had to be solely in front of the bots before they'd take any actions that depend on that. For much of the game I was level with one of the bots so they didn't do anything. Was this correct?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
ColintheFlea wrote:
OK, so you didn't miss a Headline, so that's good. The Pax-o-matics will not use an Orange/Black card for Strawman actions, only Scapegoats, but this is something I'm considering adding, if the Strawman takes the lead away from the human.

As for rules clarity, I bet after another try or 2, you'll have the rules down. I try to keep them to 2 sides max, otherwise they get too weighty. I'm also working on similar projects for Panamax and Concordia.

Thanks,

Colin


Yes I'm sure the rules will clear up. From memory what I just posted and also trying to work out order of actions as opposed to order of buying (which i realized didn't matter) were the initial problems and then it flowed pretty nicely.

It would be nice to include strawman but yes perhaps only as a catchup mechanism rather than being forced into it by the flow chart (which it wasn't, i must have just mistook scapegoat).
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
casualcasual wrote:
One question - some rules require a prestige point comparison. I played that I had to be solely in front of the bots before they'd take any actions that depend on that. For much of the game I was level with one of the bots so they didn't do anything. Was this correct?


I think it depends on the action in question. For example, the Pax-o-matic will only play an Orange/Black card if the human is tied or behind the bots, and will stay that way after getting the victim-awarded Prestige. Initially, I had that as the human must be behind, and stay behind, but another user said that was too restrictive, and the bots never used that option. I agreed, and altered it to make it a little broader for the bots. However, they still won't give the human a Prestige point in an area that the human is already leading in, just to use a card.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
Sorry, yes, that's what I meant I think (I didn't have the rules in front of me).
I think the wording that I was unsure of was:

"Only if the human player does not lead the Tripartate in the associated Prestige, and would continue to do so after receiving the victim-award"

So - "does not lead" - if I have 2 prestige points, and bot B has 2 prestige points - I do not lead, however, gaining a prestige point would put me in the lead, so Bot A wouldn't play it.

However, if Bot A had 3 points in that area, he would play it, even though it would give me 3 to tie with him (because I still would not lead).

I might change the wording - "and would still not lead after receiving the victim-award", but maybe that's just me.




 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
Two last things (sorry - but it is a reflection of the fact I really like this!)

- I presume that when you say buying an enterprise "would provide income" means that the bot will happily purchase enterprises it can't immediately play and have them in his hand? (This happened with one of my bots and caused him to get stuck). But he wouldn't purchase a mine or a bank in a depression, or an enterprise in Chihuahua when no Chihuahuan enterprises could be played as a result of a General card.

- And, finally, I'm trying to think of a circumstance where the bot has already purchased the 1st card and therefore skips buying the 1st card.

("Buy 1st Market/Public card
Skip this Action if the 1st card has already been purchased.")

When would this be the case? Later in the flowchart there's a separate "Buy 2nd card" option. Maybe I'm just missing something obvious!

Thanks again - as I said, I'll have another go when I get a chance and then maybe try with some of the difficulty options too.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
casualcasual wrote:


I might change the wording - "and would still not lead after receiving the victim-award", but maybe that's just me.


Your interpretation is correct, except that you don't just consider 1 of the bots, but the Tripartate, i.e. both bots + 2 from Diaz. And you are probably right, saying "must not be leading, and would continue to not lead" is a little odd wording. I'll maybe change it, if I end up with other edits to do.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
casualcasual wrote:
Two last things (sorry - but it is a reflection of the fact I really like this!)

- I presume that when you say buying an enterprise "would provide income" means that the bot will happily purchase enterprises it can't immediately play and have them in his hand? (This happened with one of my bots and caused him to get stuck). But he wouldn't purchase a mine or a bank in a depression, or an enterprise in Chihuahua when no Chihuahuan enterprises could be played as a result of a General card.


Yep, it will stockpile Enterprises that if it COULD (not will) play, would immediately gain income cubes. However, it does not have to be able to play them right away (maybe due to lack of gold). There are several reasons it may pass on an Enterprise:

- Won't always take a Smelter or Water Works etc... if it doesn't have the relevant coupled Enterprise to get income.

- Doesn't have any income cubes to place

- Is a bank/mine and Depression

I'm sure there are other times, as well.

casualcasual wrote:
- And, finally, I'm trying to think of a circumstance where the bot has already purchased the 1st card and therefore skips buying the 1st card.

("Buy 1st Market/Public card
Skip this Action if the 1st card has already been purchased.")

When would this be the case? Later in the flowchart there's a separate "Buy 2nd card" option. Maybe I'm just missing something obvious!


Right, so it will always try to buy a card first. Then, for its second action, it will try to play a card. Then, its preferred third action is to play another card, rather than buy. The way I wrote it, it allows the bot to switch the priority from buying to playing. Does that make sense? If I just had Buy-Play, then if it bought the first card, it may just buy a second card, and seeing as that takes 2 Actions, that's all it would do. My way, it will Buy, then try and Play. If no Play, then it may Buy again. But if it does Play, then it can't Buy twice, as it runs out of Actions, so moves to other things.

casualcasual wrote:
Thanks again - as I said, I'll have another go when I get a chance and then maybe try with some of the difficulty options too.


Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
casualcasual wrote:


- And, finally, I'm trying to think of a circumstance where the bot has already purchased the 1st card and therefore skips buying the 1st card.


Actually, are you interpreting "Buy 1st card" as meaning the 1st card in the Market? What I mean by "1st card" is whether it used an Action on buying a card already. Therefore, seeing as it frequently buys a card with its first Action, it also frequently will skip that instruction for its 2nd/3rd Actions. The only real reasons it doesn't buy a card with the first Action is:

- It doesn't have any money, and neither 0 cost cards are desirable

- It already has a full hand

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
Yep, I see (and no I hadn't interpreted it that way)

Potentially, with a full hand, the bot skips buying, then plays a card in its hand, and now can go back and buy a card. But then it can't buy a second obviously so it needs to skip "buy 2nd card".


I'm trying to think of a neater way to put that into the flow chart. I think i was just confused by the first sentence in the chart talking about skipping the step when its not immediately clear why it would, especially at the beginning - my bot was playing partners or oranges so didn't quite get to 5 cards at the beginning of a turn. It's probably okay though - I just kept going and it was fine obviously, in a turn or two I most likely would have come across it and figured it out.

OK I'll definitely leave it there and let you know after a few more playthhroughs if there's anything else i notice with trends or difficulty.

Good luck vs PNE

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
I suppose I could call the first section just "Buy Card". It will generally do that, unless its hand is full, or none of the cards are desirable/affordable. If it doesn't buy, it will try and play. If the hand was full, but it does play one, it will go back to step 1 in Action 2, and will now try and bring the hand back up.

And we need all the luck we can get. Not a good start to the season so far.

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
I had a few more plays through the week. Overall, I've enjoyed it, there's a level of strategy there missing in the dice versions whereby you can work out what the bots will take and plan accordingly which is actually enjoyable - it does mean you are kind of "working" them though, but it's just the game so that's not an issue.

So the results were

Revolution Victory, 2nd topple – Orozco (vs Reyes and Carranza)
Outrage Victory, 2nd topple – Boss Shephard (vs Terrazas and Orozco)
Command victory, 1st topple – Reyes (vs carranza and madero)
Command Victory 4th Topple – Orozco (vs Carranza and madero)

They were all with the straight rules except the last one which I added to the Diaz prestige for the first topple although I don't think it made a difference.


Some interesting things came up out of the plays.

Bots collecting prestige - Those newspapers present some interesting choices for the bots. At least a couple of times they've made the "wrong" choice. I.e., they've oriented the partner card because they are beating the other bot or because it matches their hacendado, but it might have been the smarter move to consolidate things. In one instance I had collected three revolution points which I still held in my hand. I knew those were going to come out on my next couple of turns and that I'd be in an excellent position then to go for the win - but the bot still chose to go "loyalty" instead of "revolution" at that point. I thought about having him go "revolution" but just played it by the book.

Headlines - I also had a similar situation where one bot bought a headline and caused strife because it resulted in me getting rid of two cards to his one. However this flipped the regime to the one that I wanted for the topple.

Strife for bots - I had a few times where there was no way to split the cards that they had to get rid of, so I went with working backwards up the "play partner" section.

Topples
- I'm undecided on this, but I had a thought that maybe the bots should purchase topples as soon as they can afford them if they will prevent you winning. Then I thought that might make it next to impossible and result in most games going to cash. Then I realised there is a way to mitigate it which is to speculate on the topple card and prevent the bots from doing this while you work towards it. I might try that on my next play.


Bot money - very often, I've noticed the bots get a bit tied up with their hands. They'll have a whole heap of enterprises in hand, then pick up a partner, which are often cheaper to play out, and play it out immediately. So they don't get to play out their more expensive cards. Slowly they'll get rid of them by selling them, but it's like a drip, it only happens every few turns because they'll have played a partner or something else and so can't sell. I actually used this to my advantage on at least one or two occasions; eg., I was worried about one of those nationalise everything in x cards but I let the bot have it because I knew it would be forever before it actually played it.
Having said that, two of those victories were super tight (as in, I spent all my money achieving the topple).

Orange and Black - these are fairly restrictive for the bots and you can work around this pretty effectively. If you can keep your prestige levels equal with one of the bots, then basically you are going to be free of trouble because the bots won't play them against you. I was also unsure of whether or not the bots would play them on each other - I think you intended not? Having them straw man doesn't exactly make it easy or neat either - in what cases might they do this? Only when they are trailing prestige perhaps?
I also had a situation in the last game where there was a black card in the market in each row in the zero gold space. I didn't actually do this, but because neither bot could use either card, they just sat there, and I realised I could speculate on something and know that the bots would never buy that card because it would never reach the zero column. I could have speculated on a topple and then just let it sit there till I was ready (I didn't need to as it turned out).


Anyway - I've enjoyed all the games - these aren't criticisms at all, just observations. Like I said, it becomes a slightly different game which is enjoyable in and of itself, and it's much better in that way then playing different hands, and is quite different from the dice solo variant which is of course a bit more random.








1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Hey Thomas,

Thanks a lot for posting more thoughts. I don't have my game handy right now, to follow all your points, but I wanted to respond to let you know I'd seen them. You've probably played it more than I have now, so are in a better than me to critique and improve, so let me know if you have any ideas. I'll try to think about your points more over the next day or so.

You never know, if we can get it balanced right, and be a fulfilling game, maybe we can get in the Living Rules!

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Alright, time for a more detailed response.

casualcasual wrote:
So the results were

Revolution Victory, 2nd topple – Orozco (vs Reyes and Carranza)
Outrage Victory, 2nd topple – Boss Shephard (vs Terrazas and Orozco)
Command victory, 1st topple – Reyes (vs carranza and madero)
Command Victory 4th Topple – Orozco (vs Carranza and madero)

They were all with the straight rules except the last one which I added to the Diaz prestige for the first topple although I don't think it made a difference.


Oh dear, no losses yet. I'm guessing you've played the game a lot, and are of a much higher level than myself. It's good that you are pushing the bots, as I can only test them so far. At this point, I think the best solution is to start altering the difficulty through win conditions, or increasing the starting gold, as in general, I'm happy with the bot logic. But let's see if there are any opportunities below.


casualcasual wrote:
Some interesting things came up out of the plays.

Bots collecting prestige - Those newspapers present some interesting choices for the bots. At least a couple of times they've made the "wrong" choice. I.e., they've oriented the partner card because they are beating the other bot or because it matches their hacendado, but it might have been the smarter move to consolidate things. In one instance I had collected three revolution points which I still held in my hand. I knew those were going to come out on my next couple of turns and that I'd be in an excellent position then to go for the win - but the bot still chose to go "loyalty" instead of "revolution" at that point. I thought about having him go "revolution" but just played it by the book.


OK, I've never played the MP version of the game, so don't really know the dynamics that it brings. So, my question is, would a human opponent have a good idea that you had 3 Revolution points in your hand? Even if they did, would they still be working towards their own win? I guess what I'm asking is, what is the overriding aim, to progress your own position, or block others? We do have to be careful saying the bot made the wrong move. It may not make the move you know to be right, knowing your hand, and what you are going to do. Rather, did it make a different move to what a human would have done?

If the answer is yes, it went for the wrong prestige, and there is no way the human wouldn't have gone Revolution, then perhaps a tweak is necessary. Remember, I made the bots defensive, if they KNOW you lead the Tripartate, just not if they suspect you COULD lead the Tripartate. If it doesn't see you lead, it tries to improve its own position, by playing the prestige it leads the other bot in (this tries to reduce inter-bot competition where possible). The solution would be to change the definition of leading to include all cards in your hand as well. That way, it would have taken Revolution instead. But I'm not sure i like that option. First, it increases the complexity of the move. Second, it makes the bot super-human, having full knowledge of your hand. I suppose the counter to that is that you know his hand, so why doesn't it know yours? (Hmmm, I wonder if the bot hands should only be seen during their turns, making it a little trickier to remember what they have.)

casualcasual wrote:
Headlines - I also had a similar situation where one bot bought a headline and caused strife because it resulted in me getting rid of two cards to his one. However this flipped the regime to the one that I wanted for the topple.


In your opinion, what do human players generally use Headlines for? To cause a loss of cards for an opponent? Regime change? Or both? I initially made the bots use Headlines for the former, but maybe it should be for the latter. I could alter it to say it only takes a Headline if the regime change goes to one it has the most prestige for. But then you run the risk of it losing too many cards, as card color is not something that comes into its thinking.


casualcasual wrote:
Strife for bots - I had a few times where there was no way to split the cards that they had to get rid of, so I went with working backwards up the "play partner" section.


I'm assuming you applied the tiebreaker of protecting the least number of cubes? If so, then I use the tried and tested method of just selecting randomly. But either way works.

casualcasual wrote:
Topples - I'm undecided on this, but I had a thought that maybe the bots should purchase topples as soon as they can afford them if they will prevent you winning. Then I thought that might make it next to impossible and result in most games going to cash. Then I realised there is a way to mitigate it which is to speculate on the topple card and prevent the bots from doing this while you work towards it. I might try that on my next play.


Sounds reasonable.


casualcasual wrote:
Bot money - very often, I've noticed the bots get a bit tied up with their hands. They'll have a whole heap of enterprises in hand, then pick up a partner, which are often cheaper to play out, and play it out immediately. So they don't get to play out their more expensive cards. Slowly they'll get rid of them by selling them, but it's like a drip, it only happens every few turns because they'll have played a partner or something else and so can't sell. I actually used this to my advantage on at least one or two occasions; eg., I was worried about one of those nationalise everything in x cards but I let the bot have it because I knew it would be forever before it actually played it.
Having said that, two of those victories were super tight (as in, I spent all my money achieving the topple).


Hmmm, something doesn't sound quite right there. Maybe in my games, we've had more Enterprises, as the bots are generally not that hard up for cash. Therefore, frequently, they'll take a Partner, play a Partner, play an Enterprise as their 3 Actions. The reason it likes Partners over Enterprises is twofold. First, the Partners usually provide prestige, so getting something down to start making it more difficult for the human is important. Second, the Partners frequently have some game benefit that is more useful to the bot, such as a "During Toppling" card, or making card buys cheaper. Personally, I can't say I've seen a huge problem playing Enterprises. Maybe it decides to sell a more expensive one occasionally, but getting stuck is not something I've experienced. Maybe I should try a few more games.


casualcasual wrote:
Orange and Black - these are fairly restrictive for the bots and you can work around this pretty effectively. If you can keep your prestige levels equal with one of the bots, then basically you are going to be free of trouble because the bots won't play them against you. I was also unsure of whether or not the bots would play them on each other - I think you intended not? Having them straw man doesn't exactly make it easy or neat either - in what cases might they do this? Only when they are trailing prestige perhaps?


OK, I want to make sure you are playing this right. The bot will play an Orange/Black card on you if you don't lead the Tripartate, not the bot. From the way you wrote it above, it sounds like you are being a little too restrictive, so I thought I'd check. I wouldn't say they use those cards a lot, but they will use them, so beware. Also, yes, you are right, I consider the 2 bots to be cautious friends of each other. They both want to win, but they have a pact not to play Orange/Black cards on each other. If they did, it would only serve to help the human.

casualcasual wrote:
I also had a situation in the last game where there was a black card in the market in each row in the zero gold space. I didn't actually do this, but because neither bot could use either card, they just sat there, and I realised I could speculate on something and know that the bots would never buy that card because it would never reach the zero column. I could have speculated on a topple and then just let it sit there till I was ready (I didn't need to as it turned out).


Well, there is a release clause in the rules for this situation. Perhaps the bots just never triggered it. Note the "Buy Non-Headline, 0 Cost Card" step. This is designed to help the bot a. keep the line moving, and b. set up a card sale, as it's buying a card it doesn't really want. I've seen it do that Action several times, so I know it does trigger. Don't be sure it won't do it in your games!


casualcasual wrote:
Anyway - I've enjoyed all the games - these aren't criticisms at all, just observations. Like I said, it becomes a slightly different game which is enjoyable in and of itself, and it's much better in that way then playing different hands, and is quite different from the dice solo variant which is of course a bit more random.


Don't worry, I didn't take your feedback as a criticism. The tone was certainly positive. I am concerned that expert players may find it too easy. But I like to keep my rules streamlined, with options to vary difficulty outside of them. That's why I would suggest that unless a rule is broken, they all stay as is, and people increase/decrease the starting Diaz prestige/gold/cards as necessary. I know the bots don't always do what you would do, but they have other advantages (here, it's the effect of essentially 2 against 1). We could make the bots do the perfect move every time, but then the rules would be pages and pages long. I'd rather give people a framework to go with, and tweak themselves from there.

But do let me know if you have any specific suggestions on changes.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
I think canny players will have a pretty good idea of what their opponents are picking up, especially when it correlates to the hacendado prestige. But I agree with your distinction about making the wrong move vs not making a human move. I actually enjoyed that part - hoarding prestige points from the bots as a strategy. I don't think a change is required.

The headlines thing is a similar issue really. A human player might avoid the headline - or even purchase it for the status quo to prevent a regime change that would help an opponent. But as you say, they're bots. I can't remember what it says specifically but maybe if the bots would not purchase a headline if it would change to a regime that the human player leads the tripartite in.

The bot money situation - yes I think I have been playing that wrong. The flowchart to me suggested that having played a card the bot wouldn't play another card on the same turn. Mostly they were speculating. I probably just completely missed something obvious there.

The orange/black cards - the way i read it was, if I'm tying in prestige, then the bot won't play either because I would be given the lead. Are you saying the bot will play the orange/black card in that situation?

Okay, I missed the trigger as I got stuck on the bots only buying black cards they could use. It only happened once anyway, quite late in the game - so I'll be sure to remember that next time.

Completely agree with your assessment - I think it works well as a framework. I'll see if my misinterpretations make much of a difference first and then try the topple thing too.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Hi Thomas,

More answers below:

casualcasual wrote:
I think canny players will have a pretty good idea of what their opponents are picking up, especially when it correlates to the hacendado prestige. But I agree with your distinction about making the wrong move vs not making a human move. I actually enjoyed that part - hoarding prestige points from the bots as a strategy. I don't think a change is required.


OK, but I am going to think a little about the open/closed hands questions. Should we make the bots hands hidden until their turn to move, to mimic a real game? Or conversely, we could pretend the bot knows the human hand, and that prestige there would count in its calculations? Something to mull over.

casualcasual wrote:
The headlines thing is a similar issue really. A human player might avoid the headline - or even purchase it for the status quo to prevent a regime change that would help an opponent. But as you say, they're bots. I can't remember what it says specifically but maybe if the bots would not purchase a headline if it would change to a regime that the human player leads the tripartite in.


Well, it does, sort of. It will want to alter the Regime if:

- the human leads the Tripartate in the current Regime, and not in the changed one
- if it leads the Tripartate in the changed Regime, but not in the current one.

However, regardless of that, it will take the Headline if the human loses more cards. So it could unsuspectingly change it to something you prefer (based on cards in your hand, or bad luck, i.e. you lead right now in the changed Regime). However, I suspect that is rare. To be honest, in my games, the bots never seem to take the Headline, as it rarely loses less cards than the human. Most of the time, they just fall off the Market.


casualcasual wrote:
The bot money situation - yes I think I have been playing that wrong. The flowchart to me suggested that having played a card the bot wouldn't play another card on the same turn. Mostly they were speculating. I probably just completely missed something obvious there.


OK, then that's actually good (well, and bad, seeing as I need to find a way to emphasize that more). The "Flowchart Guidelines" section states "For each Action, start at the beginning of the flowchart", meaning that if it doesn't buy a card the first Action, but does play a card, for the second Action, it starts over at Buy A Card again. So, if it had a full hand of 5 cards, it will skip buying, and try to play. If it does, it now has a space, and the next Action, it will attempt to buy. And so on.

That is why I put in 2 separate Buy A Card actions. If I only put in one, then as it started over again, it would just buy a second card. What I want it to do is try to Buy A Card. Then Play as many cards as it can, then return to buying a second card if necessary. If it Buys/Plays for the first 2, and can't play again (can't buy, due to it needing 2 Actions), then it enters the later flowchart items.

casualcasual wrote:
The orange/black cards - the way i read it was, if I'm tying in prestige, then the bot won't play either because I would be given the lead. Are you saying the bot will play the orange/black card in that situation?


It's your wording I need you to clarify. When you say tying in Prestige, I want to make sure you are comparing to the Tripartate, and not just the bot with the Orange/Black card. So, let's say you have 2 Revolution points, the non-playing bot has 0 Revolution points, and the bot with the card has 1 Revolution point. Seeing as you trail the Tripartate 3-2, it will play a Revolution Orange/Black card on you, even though you lead that particular bot 2-1. Does that make sense?

casualcasual wrote:
Okay, I missed the trigger as I got stuck on the bots only buying black cards they could use. It only happened once anyway, quite late in the game - so I'll be sure to remember that next time.


Yep, it can sometimes buy a card it cannot use, but only as a last resort. It usually speculates first, but then rather than speculating again, it will take one of the 0 cost cards (with the aim of selling it in the future).

casualcasual wrote:
Completely agree with your assessment - I think it works well as a framework. I'll see if my misinterpretations make much of a difference first and then try the topple thing too.


OK, let me know how it goes. I think the possibility of a double play round will make a difference. As will the larger chance of using Orange/Black cards.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Australia
NSW
flag msg tools
Avatar
mbmbmbmbmb
That all makes sense Colin - and I feel like a bit of an idiot to be honest! I think those clarifications with regards to the flowchart and then the tripartite will correct things and make those orange and black cards more viable options for the bots in my games. I don't think on reflection I had missed the trigger to buy those 0-spot cards, but wasn't getting down there because of the way I was playing.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Colin Taylor
United States
California
flag msg tools
Avatar
mbmbmbmbmb
Hah, no worries. To be honest, I'm just glad someone has taken an interest, and if it makes me alter the wording in places, to make it easier to play, then that's all worth it. For example, bolding the bit about starting down the flowchart fresh for each Action may be worth doing, seeing as that is key.

Thanks,

Colin
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
1 , 2  Next »   | 
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.