AI? (Full Version)

All Forums >> [Current Games From Matrix.] >> [Age of Muskets] >> Horse and Musket: Volume I, Frederick the Great



Message

michaelincol -> AI? (6/17/2009 1:36:54 PM)

I play games solo - ready to buy this one but I have seen nothing mentioned yet as to how AI is? Thanks for any info.

Michael



sol_invictus -> RE: AI? (6/17/2009 2:14:57 PM)

I guess the AI is difficult to evaluate with an engine like H&M. The Activation rules can make the best plan fall apart for both the AI and Player. This is the strength of the engine. I almost always play email games because I enjoy the human interaction but several years ago I was a dedicated solo gamer. I guess I would feel safe in saying that if you are a relatively competent Players you will win the majority of your games against the AI. I view the AI of this and all games as simply a mechanism to keep me honest and engaged. If you haven't played the H&M series before then I think you will be satisfied with the AI. It is also best to play the attacker which is usually the Prussians. If you enjoy the era and tactical wargames I highly recommend the game and encourage you to try some email games. The tension is so much greater when going against another Player.



staryhrac30 -> RE: AI? (6/17/2009 9:53:26 PM)

Hi,
I'm just thinking about the purchase and was just about to ask the same question as OP. I only play single also. As I like the era and scope of the game, AI is my only concern now. I've found some sparse info on the previous H&M titles, and AI appears to be a quality one (according to Wargamer review). But still, I would like to hear a few more opinions about this/previous title longetivity in SP.



Zaratoughda -> RE: AI? (6/18/2009 12:11:44 AM)

Guys... I posted this in a separate thread before this one but guess you missed it...

As far as the AI is concerned... looks like it hasn't changed materially from PWM. In other words, it is woeful to the point of the game being unplayable (for me at least) against the computer.

If you watch what the computer is doing it is marching units this way and then back again like a total idiot... and then it will do things like change the facing of it's units leaving them blatantly exposed to a flanking attack. Also, in the original HnM the AI tried mostly frontal attacks... but in PWM.... it will almost always try to outflank. The problem with this.... is leaves it's center wide open and it is easy to anniliate the AI.

I beat the computer Austrians in Mollwitz 71 to 4 which is a total anniliation. I tried Chotositz anyways but got totally turned off watching the computer move it's units back and forth from one side to the other.

I won't be purchasing this update until the AI is drastically improved and am not gonna hold my breath on that.

Too bad because from what I have seen of what Magnus had done with the artwork, there is considerable improvement there. But, if the game is essentially unplayable against the AI it isn't worth it.

Zaratoughda



michaelincol -> RE: AI? (6/18/2009 1:04:17 AM)

thanks for update - guess I will go with buying COG:EE - I am really enjoying FoF and it's AI is more than adequate. This note on AI, coupled with developer's note in forum somewhere I read today downplaying the game's AI with comment to the effect that anyone can beat AI with usual plug for PBEM - IMO a lame disclaimer that is usually case when AI is not decent - especially given that most polls show most play against AI. I'll stick with FoF and CoG and WBTS (I haven't even figured out how to play that yet let alone go up against the AI!

No disrespect intended and I am sure the game here is good for PBEM - I just give my dollars to developers that take in to consideration the need for credible AI.

Michael



Tim Coakley -> RE: AI? (6/18/2009 1:12:31 AM)

Gamerincol,
No disrespect taken, but I do ask you to consider two points. My comments in the other thread were in response to Zaratoughda who had posted over a year ago under another thread and was completely discounting the game because of his result against the AI. The AI is certainly important to me...it is a significant item for future improvement, but this game is an update and combination of the previous games.

Regards,
Tim



michaelincol -> RE: AI? (6/18/2009 1:15:46 AM)

I didn't feel right as soon as I posted my diss reply, so I looked up post from Tim and he did not say what I said he did. My apologies - he was actually more encouraging somewhat about AI and did not mention the often used disclaimer to use PBEM. I guess I was hoping for a decent AI as I love this type of game, jumped to cynical and untruthful conclusions, and instead of keeping them in my head went and typed 'em out and posted 'em here. My bad, and my apologies.

Michael



Tim Coakley -> RE: AI? (6/18/2009 1:27:35 AM)

No problem, a good discussion all around. I had been a solo player for years before getting into PBEM and online play. The quest for a fantastic AI is always an ongoing battle.

If you do find HnM is not to your liking and and you are in the mood for a new game, I recommend any of the Pather Games WW II games. The best AI I have ever seen...the whole system rivals the professional military simulations I have used in training.

Regards,
Tim



Zaratoughda -> RE: AI? (6/18/2009 5:21:45 AM)

quote:

ORIGINAL: gamerincol

No disrespect intended and I am sure the game here is good for PBEM



Well, actually the game is not that good for PBEM.... and I believe this was what Inaki H. alluded to in a previous post.

The problem.... is the game has each leader (and this started with the original FtG) move his forces individually and it is semi-random as to when leaders, on both sides, move. So, if you exchanged a save file each time a leader from the other side was supposed to move, and you had a bunch of leaders on each side, you would have to exchange a lot more than two save files per turn.

But, you can delay when a leader moves so, what you have to do is decide before hand in what sequence leaders move, typically all the leaders from a given side move first, then the other side's leaders, and when the first player is moving he just delays all the other side's leader's moves until he is done moving his leaders. If you want to, you can have some of the one side's leaders move first, then all the leaders of the other side, then the rest of the leaders from the first side.... and still have just two save file exchanges per turn.

This is a bit of a pain to do but is doable and the only way you can reasonably play PBEM.

Yeah, don't expect this to change anytime soon.

Zaratoughda



Iņaki Harrizabalagatar -> RE: AI? (6/18/2009 8:00:17 AM)

I tried another thing with the old H&M2, I modded the scenarios clasifying leaders initiative in a rigorous order, so that first one side, then the other side, moved all their leaders. It ruined the activation system uncertainty, but allowed PBEM.



IronWarrior -> RE: AI? (6/18/2009 11:46:12 AM)

Ironically these criticisms are making me even more interested in this game. [:D] I only play against the AI to learn a game and I don't mind the slower pace of pbem, nor do I want absolute and total control over everything. The C&C rules really sound interesting.

With the Tiller's games I set Defensive fire on and didn't mind the extra turns. Maybe this game can fill the void for a good tactical game that was left when I quit playing Tiller's games. The straw that broke the back for me in those titles was artillery consistently firing to no effect at hex range of one, or a turn where 5 battalions in line formation charging 5 routing battalions in the rear lost all but one engagement, cavalry worth too many VP's and cannot be used effectively nor historically, square formation too effective against line in melee... etc etc.

Hopefully HnM is more realistic and not too much like the Tiller's games.



Tim Coakley -> RE: AI? (6/18/2009 1:39:22 PM)

Inaki,
this is one of the solutions I am considering. No final decisions for the future, but this method would allow for faster PBEM play. I will also be looking at the bombardment and command phases for integration.

Great feedback.

Tim



sol_invictus -> RE: AI? (6/18/2009 2:54:39 PM)

I completely disagree with the comments about the game not being good in pbem. I think it works beautifully in pbem. The only problem or possible criticism I can see is if someone wants a quick game over several days. What's the rush? The player can always choose a small scenario or a portion of a battle if there is some time pressure. PBEM is where the game comes into its own imo.



Iņaki Harrizabalagatar -> RE: AI? (6/18/2009 3:26:20 PM)

Well, it is a matter of preferences I suppose, but I fail to see the fun in exchanging files just to see how your leader fails activation, to me PBEM have to keep some pace to keep it interesting and fun.



Zaratoughda -> RE: AI? (6/18/2009 3:29:15 PM)

quote:

ORIGINAL: IronWarrior

Ironically these criticisms are making me even more interested in this game. [:D] I only play against the AI to learn a game and I don't mind the slower pace of pbem, nor do I want absolute and total control over everything. The C&C rules really sound interesting.

With the Tiller's games I set Defensive fire on and didn't mind the extra turns. Maybe this game can fill the void for a good tactical game that was left when I quit playing Tiller's games. The straw that broke the back for me in those titles was artillery consistently firing to no effect at hex range of one, or a turn where 5 battalions in line formation charging 5 routing battalions in the rear lost all but one engagement, cavalry worth too many VP's and cannot be used effectively nor historically, square formation too effective against line in melee... etc etc.

Hopefully HnM is more realistic and not too much like the Tiller's games.


I bought a couple of Tiller's games.... and got really turned off. It seemed things weren't going like they should so I did some testing and found if you broke a unit down into like 4 units they did 2.5x the amount of damage and took around 2.5x LESS morale and readiness loss. In other words, the game system was totally unbalanced in favor of smaller units (yeah, posted some nasty words on that here in this forum and got yelled at <g>).

The thing that is most impressive about Dave Erickson's design is that it does a very good job about differentiating between line infantry and other types. Though there are improvements that can be made, very good and much better than the HPS games in this regard.

However... as far as the AI is concerned.... Bill Sherman did a full set of HnM scenarios for the FIW and the ARW, and I wrote a program that converted them to HnM2:PWM. But, a very basic test of games along these lines is the battle of Guildford Courthouse and the North Carolina militia. These troops.... lol..... were some of the most ill-trained in the history of warfare. What Greene got out of them were two volleys and then they ran.... never to be seen again in that battle, in that campaign, and in fact never again in the war <g>.

So, if as the British you can't get by the two lines of NC militia in Guildford.... you are doing real, REAL (!) bad and.... in HnM2..... the AI got by the first line but got stopped by the second. :-(

But, in the HPS game.... didn't even get by the first line (sheesh). Imagining Cornwallis getting captured by the NC militia is almost unbelievable.

So, the AI in these games is not very good. Bad and worse.

Zaratoughda

P.S. Oh, I modded the HPS Fort Necessity scenario so that it played out reasoanably by combining small units into larger ones so that all units were within 2x, and was able to WIN the Battle of Fort Necessity as George Washington (!). This only made sense given the AI's tendency to throw it's units in piecemeal attacks but, at least it was reasonable to that degree.

P.S.2. Oh, one thing I should say about the HPS games... they seem to have done a good job on the OOBs. Maybe use as reference for scenarios in other games.



Zaratoughda -> RE: AI? (6/18/2009 3:37:21 PM)

Oh, I should say that, the major criticism on the HnM2 C&C system is that, when a leader fails his activation, you can't do ANYTHING with him or his troops (possibly your higher level commander can but lots of times that is not possiblle).

So, in these cases your units just sit there doing nothing. Sheesh..... can't even fire when engaged in combat.

This is an example of an interesting system that.... needs improvement.

Zaratoughda



Tim Coakley -> RE: AI? (6/18/2009 3:52:38 PM)

Zaratoughda,
remember that while you can not initiate fire with a unit under a commander that failed activation, you can return fire. This models the other leader having the initiative and still gives you some reaction.

Tim



Sumter -> RE: AI? (6/19/2009 3:42:08 AM)

I only play against the AI, but it seems that the game editor provides a very simple solution if you find the leader activation process inhibiting PBEM.  Just eliminate all subordinate commanders, increase the range of command and control for each army commander to cover the entire map and play a true turn-based format.  Just be sure to save the new version of the scenario under a different file name.



lancerunolfsson -> RE: AI? (6/19/2009 5:32:48 AM)

Excellent PBEM fix Sumter!

The activation system though seems like it makes solo play better. Yet I got to wonder if this fix helps the AI in solo play? By removing a layer of complexity.



Sumter -> RE: AI? (6/19/2009 6:28:51 AM)

I don't actually know that.  I like the complexity of the command structure.  I do wish there was a way to switch back and forth between playing solo and hot seat.  The original HNM permitted this and made it possible to correct some of the AI's more errant behavior -- like unsupported artillery units moving to attack infantry or cavalry units.  Still, I like PMW and plan to buy HNM Vol I in the next few days.



Page: [1]

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
1.079102