AI/Realism (Full Version)

All Forums >> [New Releases from Matrix Games] >> WarPlan



Message

FAA -> AI/Realism (8/5/2020 9:27:37 AM)

I think about buying this game but I have two questions:

1. How’s the AI? Particularly, how well can it retreat to avoid excessive losses, can it pull something like Dunkirk in 1940, how good are naval landings? Most games really struggle with these.
2. How realistic/plausible is it? War in the East and War in the West for example are so detailed that they’re quite plausible and realistic. Classic wargames usually are not, and from what I can see WarPlan doesn’t have a detailed economy or resource management. Does it stand to reason that it doesn’t model the strategic aspect of war properly, particularly with critical shortages for the axis like fuel, rubber, metal etc.



kennonlightfoot -> RE: AI/Realism (8/5/2020 2:39:06 PM)

My opinion is:

1. The AI is like all other games of this type. Good for leaning the basic functions of the game but can't handle the complexities of a long war. It does alright if you play only scenarios where the major strategic decisions are already made for it. No one has the deep pockets to fund developing an AI for these type games like they did for Chess. If you plan to only play the AI, the game is going to be of limited use past the learning stage.

2. This depends on what you think is possible as alternatives. The game doesn't force you to follow historic choices although I suspect on the main choices like Sealion, France, and Russia the player deviates at considerable risk if playing an opponent. For the scale it uses I think it is one of the best I have come across for handling WW II. It has resources, trade, research, and production in it but presented in a simple enough way that it doesn't become tedious. The combat scale is such that having specialized units (infantry, mechanized, armor, cavalry, para, various air) works without seeming artificial. I would place the game mechanics wise between Strategic Command (very high level) and War in the West (very detailed at operational level).



Cigar King -> RE: AI/Realism (8/6/2020 1:56:05 AM)

1) the Ai is not bad. It does give ground, and generally does well on the defense. The one drawback is amphibious invasions. The current air rules, if they intercept an invading unit, will destroy it. I have never seen the Allies launch D-Day successfully. They do build up a good invasion force, but, they invade on too wide a front, and the air attacks destroy most of the invading troops before they can get to land.

It's nothing like WiE/WiW. That being said, the strategic model, while it has some quirks, actually works well. The Germans have to keep a close eye on their manpower and oil. Fortunately, the Russian winter effects are so devastating that it's a good time to build up reserves to make summer attacks.

I think it's perhaps the best WW2 game out there, and there is excellent support from the designer. Buy it.



pzgndr -> RE: AI/Realism (8/6/2020 3:10:47 PM)

quote:

ORIGINAL: kennonlightfoot
The AI is like all other games of this type. Good for leaning the basic functions of the game but can't handle the complexities of a long war. It does alright if you play only scenarios where the major strategic decisions are already made for it. No one has the deep pockets to fund developing an AI for these type games like they did for Chess. If you plan to only play the AI, the game is going to be of limited use past the learning stage.


So, what would you say are the insurmountable complexities of a long war, specifically the entire WWII ETO campaign 1939-1945? I created the Advanced Third Reich mod for Strategic Command. IMHO, clever use of event scripting and AI scripting that allows for variable triggers and results is fully capable of providing a challenging computer opponent for both Axis and Allies for the entire campaign.

Examples:
- Variable national strategies for research and production (default, air, naval, or ground)
- Historical or early Axis attacks in the West in 1939-1940
- BEF to France or not
- Germany creation of Vichy France or not, and variable status for French colonies after surrender
- Sea Lion, or Spain, or Turkey options
- Variable Axis or Allied coups in Norway and Yugoslavia
- Axis and Allied support for North Africa operations, and transports for reinforcements
- Historical or early Barbarossa in Russia 1941
- Variable Axis and Allied offensive emphasis in Russia (North, Center, or South)
- Allied invasions/transports for North Africa/Sicily/Italy 1942-43, depending on French colony status
- Early D-Day in France 1943 or not
- German "Fuhrer Bunker" transition to defensive in 1944-1945
- etc.

I don't exactly have deep pockets for my efforts, but I had motivation to develop my A3R mod and plenty of patience over the past 10-15 YEARS as the Strategic Command series evolved and I was able to keep getting closer to what I wanted. And I will say that it does take a lot of time and effort for both the developer to keep improving capabilities (something Al will keep working on) and the modder to keep up and continually improve. Anyways, FWIW, my A3R mod has all of those things I listed above, which makes playing an entire campaign game challenging and different each time, for either side. My A3R mod v1.3 for SCWWII:WIE is in pretty good shape these days. I may tweak it some more for v1.4, when I get a chance.

Bottom Line: When I hear criticisms of "all other games of this type" blah blah blah, I politely disagree. For WarPlan, Al has a pretty good start on things based on his similar experiences with SC mods and I expect the WarPlan series will evolve nicely.



Omnius -> RE: AI/Realism (8/6/2020 6:01:39 PM)

The Artificial Ignorance is rather Pathetic. Look at how poorly it does on the historical level in the 1939 scenario taking down Poland, takes 4 turns to do a 2 turn job. On turn 1 I like to invade Norway since the British navy is in port. The British AI does absolutely nothing about that, just cedes Norway without a fight. As others have said it's only good for learning the ropes without playing both sides but better off learning both sides playing both sides hotseat.

I'm glad Alvaro is fixing the Yugoslavia cheat, that was definitely an ahistorical cheat that too many players abused. Otherwise it's fair at depicting history, vastly superior to Paradox's garbage HOI 4 game.



kennonlightfoot -> RE: AI/Realism (8/6/2020 9:23:51 PM)

Scripted AI even with some random event is still very predictable and easily countered unless the game parameters are fudged to compensate. Something like giving the AI side 20% more supply, experience, etc.

What we are calling AI in these games isn't really AI. True AI analyzes millions of possible outcomes before acting. That can be done in Chess because the board is only 64 squares. Poland has more hexes in its western half than this. Some things can be scripted in a more predictable manner because enough things are fixed to do it. Like what to do to defend Poland. But as the game gets past these restricted situations like Poland and France, the possibilities become overwhelming for anything short of using Watson as your opponent.

Some games get around this by letting the computer control the units for both sides. The player is limited to high level decisions whose outcome is decided by the game. Examples are Sid Meir's Gettysburg, Total War, and Scourge of War games.

But board style games like WarPlan, War in the East/West/Pacific, etc. have to large of a map area for a PC to analyze. So they generally follow a script of some kind with a little tactical logic thrown in so they don't make stupid unit attacks.



pzgndr -> RE: AI/Realism (8/6/2020 10:36:54 PM)

quote:

ORIGINAL: kennonlightfoot
Some things can be scripted in a more predictable manner because enough things are fixed to do it. Like what to do to defend Poland. But as the game gets past these restricted situations like Poland and France, the possibilities become overwhelming for anything short of using Watson as your opponent.


Like what, exactly? I'm serious, what are you really expecting when you sit down to play a WWII ETO campaign game, whether against a computer opponent or a human opponent? Certainly some human players will try to get away with bizarre strategies that may or may not work, like massive armor builds or massive tac air builds or whatever, and these ahistorical surprises may appeal to some players but not others.

Methinks your criticism of what is actually possible with the variability of event scripting and AI scripting, as Al and I are used to from the SC series and here in WarPlan, is based on you not fully understanding what I'm talking about. No matter, I've been arguing with guys like you for a very long time now. It was my motivation to work on my A3R mod to prove you guys wrong. And I believe I have. Even my "very predictable" scripting that I wrote has surprised me more than once, because I deliberately scripted many things to be as unpredictable as possible, within reason.

So again, what exactly are these insurmountable complexities of a long war of which you speak, that your trustworthy human opponents can master but the AI scripting cannot? Give us some ideas, so Al and I can consider improvements.



baloo7777 -> RE: AI/Realism (8/7/2020 12:07:39 AM)

I play Ai and pbem a little. The AI I give 10% bonuses to in the start screen (10 and 12% I think). It makes the AI much tougher. I can beat the AI sometimes, and have yet to beat a pbem opponent (only on 4th game). Sometimes, often even, I need an AI game to play. I appreciate anyone who works to make a better AI game whether scripted or with bonuses. I think there are many more players that play this and other wargames vs the AI. Probably more than play PvP. The game is different played using the same rules against the AI as against a person (for all games). That said, a developer can only play-balance the best he can so that both types of players will like the gameplay, or they won't be selling as many games. We are a very vocal group of PvP players here, myself included. I think that Alvaro has been one of the best developers for listening on this site and making changes he feels makes the game better. I don't always agree with him, but that is just fine. If a modder is willing to try making the game better as they see it, I'm all for giving their mod a try... like PanzerMikes excellent Map and Unit mods!



Edorf -> RE: AI/Realism (8/7/2020 2:17:42 AM)

I just want to say that pzgndr have some very good points regarding helping the AI in games like this. Thumbs up.



pzgndr -> RE: AI/Realism (8/7/2020 11:44:53 AM)

quote:

ORIGINAL: baloo7777
The AI I give 10% bonuses to in the start screen (10 and 12% I think). It makes the AI much tougher.


Yes. In fairness, you cannot play against the AI on default settings and expect a challenging game. Even for my A3R mod, I balanced it for human play but the AI still needs some handicaps. Better players may want tougher handicaps. It all depends on what you want out of a solitaire game.

The reason some bonus is necessary is because, frankly, the generic AI making moves and attacks to execute either AI scripting or generic routines is not as efficient as a human player. Sometimes it can be surprisingly efficient, but often over-extends itself and takes unnecessary losses. So some bonuses for extra production or extra strength points or increased spotting or whatever helps balance the game. Also, I've scripted additional units and other enhancements for both Axis and Allies, for throughout the campaign, to provide some help. So just to clarify, I'm not suggesting the AI is challenging all by itself on default settings; the computer opponent still needs help.

That said, the solitaire game versus computer opponent can be as challenging and as unpredictable as playing against a person. I grew up playing 3R/A3R so I know the game's strengths and weaknesses. I also played SC many many years and started as a playtester early on, so I knew that game's strengths and weaknesses, as well as the AI's. When I started scripting for my A3R mod, it was always to implement my own various strategies so that I would be playing a challenging game. All of this is non-trivial, and takes a LOT of time and patience. It's no surprise that most games fail to reach this level. For WarPlan, I believe Al also has this passion to take his extensive gaming experience and make it happen. WarPlan is off to a good start and will improve with age.



AlvaroSousa -> RE: AI/Realism (8/7/2020 2:05:51 PM)

A game like Unity of Command it is fairly easy to do an A.I.
#1 Single Front
#2 Land only
#3 Abstract Air
#4 The operational strategy is predetermined. In the original all the USSR had to do is defend
#5 Which is genius on their part.. they programmed in ~10 moves the A.I. can do depending on the map situation vs the human.
The WarPlan A.I. would be almost as competent as the UoC A.I. and I could make it better.

But in a game of this scope it is simple impossible at default historic levels. I have 11 levels for a reason + 3 supply levels. I put in various strategies for the A.I. that it randomly picks. I also put in some special code to take advantage of situations and defending. One of the largest challenges was large encirclement of A.I. forces. That took a long time for me to formulate out a reasonable code.

I wrote Brute Force for SC2. It took forever to write an A.I. for that. I must have spend 6 months on it at least.



malkarma -> RE: AI/Realism (8/7/2020 2:31:10 PM)

One thing that can be improved Alvaro, is the garrisonign of ports by the Italian. At moment Rhodes and Misurata are always left undefended. Can we make something to fix this?
Another oddity that I managed to achieve is killing the whole invasion force of the US in 1942. If I´m not wrong, the US are scrtipted to invade Northern Africa as soons as the have enought landing ships to disembarkeits units. The main isssue was that I managed to capture Gibraltar, so as the US transports didn´t had a valid pat to enter the Mediterranean, they stayed iddle in the Atlantic (the script aims to a quick capture of the Algerian ports). In this situation, the AI should have aim for tyhe Morocco ports at least, but instead they let them die thenmselves in the waters of the Atlantic.

This said, this example shows how difficult/complex is to pregram an reliable AI able of interact and counter its human opponent actions.



Omnius -> RE: AI/Realism (8/7/2020 2:47:00 PM)

Adding a script to begin each scenario would help get an AI off to a better start. Scripting can't be done for the long haul but scripting a beginning turn or two would make the AI better. Like in the 1939 scenario make the German AI take down Poland in two turns. It's pathetic to watch the German AI move the armor unit in the south stupidly around the cavalry unit and never make a single turn 1 attack. Same with the other armored unit in the north. At least with a script Al could make the German AI smart enough to capture Warsaw and the other important Polish cities in two turns. Invade Luxemburg on the first turn since there's no penalty for doing so like in other games where any attack on any Benelux country would trigger all of them becoming active.

When considering AI's I like to think of the old George Carlin joke about television. "I turn the brightness knob up all the way but the tv doesn't get any smarter"! Sadly when we turn up the difficulty on AI's in war games all that does is give the same old crappy AI more stuff to throw at us in the same lame way, absolutely no improvement in smartness. Sure would love to see Microshaft do a Deep Blue project for a war game, that would really be awesome.

I just don't play against AI's, there's no challenge to it. I prefer to play hotseat solitaire even if that does ruin the element of surprise.



pzgndr -> RE: AI/Realism (8/7/2020 3:03:48 PM)

quote:

ORIGINAL: malkarma
The main isssue was that I managed to capture Gibraltar, so as the US transports didn´t had a valid pat to enter the Mediterranean, they stayed iddle in the Atlantic (the script aims to a quick capture of the Algerian ports). In this situation, the AI should have aim for tyhe Morocco ports at least, but instead they let them die thenmselves in the waters of the Atlantic.

This said, this example shows how difficult/complex is to pregram an reliable AI able of interact and counter its human opponent actions.


Excellent example! And it is indeed very complex. But you know what your options are and act accordingly. I scripted my AI the same way, first to consider the situation:
- Status of Morocco and Algeria (Allied, Vichy, neutral?)
- Status of Gibraltar (Axis or Allied?)

Based on the status, US invasion force will ideally invade Morocco and Algeria, but maybe just Morocco if Axis controls Gibraltar. Instead of invasion, maybe send fast transports if Allies control Morocco and/or Algeria. Then the invasion needs to be reinforced, again either through Morocco or Algeria. How much and for how long, so US doesn't send its entire force pool to North Africa? Status checks for max units, and cancel scripts if not needed. Keep a stream of reinforcements coming to invade Sicily and then Italy, until Italy surrenders. Then some reinforcements can go directly from US to Naples.

The point is that with enough patience and some cleverness, you can indeed script/program the AI to do what you yourself would do. If this, then do that; else do something else. Difficult yes; impossible no. I've done it. It can be done. Others can do it too. If they want to.



AlvaroSousa -> RE: AI/Realism (8/7/2020 5:44:49 PM)


quote:

ORIGINAL: malkarma

One thing that can be improved Alvaro, is the garrisonign of ports by the Italian. At moment Rhodes and Misurata are always left undefended. Can we make something to fix this?
Another oddity that I managed to achieve is killing the whole invasion force of the US in 1942. If I´m not wrong, the US are scrtipted to invade Northern Africa as soons as the have enought landing ships to disembarkeits units. The main isssue was that I managed to capture Gibraltar, so as the US transports didn´t had a valid pat to enter the Mediterranean, they stayed iddle in the Atlantic (the script aims to a quick capture of the Algerian ports). In this situation, the AI should have aim for tyhe Morocco ports at least, but instead they let them die thenmselves in the waters of the Atlantic.

This said, this example shows how difficult/complex is to pregram an reliable AI able of interact and counter its human opponent actions.



I'll take a look.



AlvaroSousa -> RE: AI/Realism (8/7/2020 5:46:55 PM)

As for Poland a human maximizes what he can do. The A.I. can't. There is no script I can write that will overcome this. I would have to change the Pole's setup.



kennonlightfoot -> RE: AI/Realism (8/8/2020 1:20:51 AM)

Just to remind myself of how the AI is doing I played it using Historic settings and using version 8U6.

First I don't really expect the AI to be all that great. These games with thousands of hexes and hundreds of units moving in the same turn with supporting actions is far more complex than say Chess which only one piece moves per turn on a maximum of 64 squares.

Second, I play to test strategies and tactics against the AI so would prefer that it react to those moves in a reasonable way. I use the historic setting because I will learn nothing from an overpowered opponent using bad tactics.

So I started a the 1939 Scenario with me Allies and AI Germans.

The AI immediately ran amuck in Poland. It spent all of 1939 trying to take Warsaw. Finally did on on the last Dec turn.

It then (probably following script) declared war on Denmark. It took it four turns to take it. Mostly because of bad weather. And, I didn't bother to contest it.

Next item probably on the script is to take Norway. It at least didn't declare war but worse moved the German fleet over in preparation but couldn't land because of weather. The first attempt was on 3/1. The last attempt was on 4/26. This initiated about about 4 or 5 turns of fleet warfare where I hit anything that came out with French when possible and UK to finish them off. Eventually something triggered the German AI to stop trying.

3/15/40 - German AI declares war on the Netherlands. And fails to take it until 8/2/40.

5/24/40 - The AI declares war on Belgium and attacks it along with the unconquered Netherlands and parts of France. By then I figured it couldn't take France so moved all by UK combat units into France.

The spring and summer is spent driving the Germans back into Germany. I was one turn sort of saving the Netherlands. With winter in October the two sides stalemated on a line with the Allies holding part of the Rhine river. Don't think there is going to be any Barbarossa in Spring of 41. I doubt the UK can generate a large enough force to break the Rhine border. The final positions are shown below.

Would giving the AI bonuses on Experience and Morale change things? Yes, but not the tactics used.
As I said earlier, the AI is good for learning the mechanics of the game but not the tactics. Throwing all my forces into France like I did would get me destroyed by a player opponent. It's fun in the old "beer and pretzels" game type that Avalon Hill and SPI use to turn out. But if you plan to play an opponent it will be very misleading.




[image]local://upfiles/22005/28FF86E1F7204EC2AB9193B222857BFB.jpg[/image]



baloo7777 -> RE: AI/Realism (8/8/2020 2:32:28 PM)


quote:

ORIGINAL: kennonlightfoot



quote:

Would giving the AI bonuses on Experience and Morale change things? Yes, but not the tactics used.


Giving the AI a hefty bonus will lead to it blasting through Poland and Denmark, and later the Netherlands and Holland. Norway is another matter. Belgium and France, even with Brits and armor, will fall by September even by being bludgeoned with lots of German Infantry. I don't think the AI uses supply trucks, instead regaining its fighting ability much more quickly. The tactics the AI uses are preprogrammed by the developer.
My real question is the economics...that is, on Steam, or on Matrix, are the largest amount of people buying games to play solitaire vs AI, or to play PvP. The server in this game has been rumored to have no more than 200 games at a time on it (worldwide I think). Hopefully, WarPlan has sold substantially more than 400 units, which means that purchasers are mostly AI players. That said, my point is, yes you can say the AI is not as good as a person (obviously), but if you state that it is only good for learning the game, it relegates it to an uninterested vast majority of purchasers. So the real question is, do developers plan on selling to only PVP/pbem people?



baloo7777 -> RE: AI/Realism (8/8/2020 2:41:31 PM)

On the AI scripting, I wonder if it is feasible to have an on/off type button that would allow a player to switch between AI or Hotseat as the game is played? It would have to be a choice in the options that is set before a game is started. That way a player could turn it off (much like Diplomacy) before creating an online/pbem game. This would allow an AI player to prevent the US from invading Algeria if Gibraltar was Axis, etc. You turn on the hotseat button, play the turn and end it, then as the Axis turn you turn it back off. Ai players would gain some FOW info on the computer forces, but could solve problems with the AI scripting and still play a reasonable solitaire game.



AlvaroSousa -> RE: AI/Realism (8/8/2020 4:34:15 PM)

To put things in perspective I can stop the SC3 Axis A.I. in 1940 as the Allies. I am not a very good player at SC as I have not played another person in a decade almost.

I will look at the Norway scripts.



malkarma -> RE: AI/Realism (8/8/2020 6:58:35 PM)

Another thing that you can look is the behaviour of the computer in Poland. Nearly in all games the AI leaves the Poland corps close to Danzing without zocs near. That means that you can move it in suicidal mission that use to end in the capture or Berlin or Stettin by turn 3...or the overrun of a cpuple of air units. I best case situation that move make nearle all the Lutwaffe redeply, hurting his chances of conquering Poland in time.
Another thing that I realized is that the AI moves too much trying to get the best odd possible, but due the efficiency loss, she enda fdoing a waltz with his units instead of attacking. The funny thing is that after that, she finishes making attacks with a single unit with odds lower than 2:1.
I have the feeling that this is the difficult thing to solve, but maybe a scriot in wich if the computer manages to get a 4:1 or more she attacks instead of keep moving his units? Also an script to not commit suicidal attacks under 2:1 or less with a single unit? I think that this would make that the AI would be able to do ore damage to the player without crippling his units in the process.

This said, since I´m not programmer, I don't have any idea if this could be implemented, just a random thought.

ps: Also noticed that the German AI dont take care of his subs, so she eventually lose them in the convoy attacks, Maybe some script to stop moving subs with 1 or 2 steps remaining, and repairing them instead.



kennonlightfoot -> RE: AI/Realism (8/8/2020 7:53:48 PM)

Poland probably is the second easiest situation to script for the AI (easiest would be Denmark if it wasn't for weather). But even it will give the computer AI considerable problems. The initial position is good for scripting because you can tell the AI exactly what it needs to do because you know where every unit and objective city is. But unless you give the AI enough bonuses to destroy every unit it can attack, turn 2 requires analysis that only a higher level AI can handle. Units have moved and objectives are screened. The AI has to plan how to break that screen and take Warsaw. This requires more than a script although if you have enough bonuses it can just power its way through them.

Main trouble I saw in my game is the AI seems to not know it needs to put the subs on a convoy route. Interestingly, as the game progressed and the Germans got more than three sub groups the AI got much better at it. In late 41 they started doing significant damage to the UK shipping. About the only thing in needed to do was know how to use Oilers. But I think that was due to it not expecting to have to fight without French bases.

But the kind of AI that could "look" at the the layers required to actually plan and execute even a defense only plan for a game this complex are far beyond what a PC can do even if backed with deep pockets and lots of AI developers.



pzgndr -> RE: AI/Realism (8/9/2020 1:35:02 PM)

quote:

ORIGINAL: kennonlightfoot
But the kind of AI that could "look" at the the layers required to actually plan and execute even a defense only plan for a game this complex are far beyond what a PC can do even if backed with deep pockets and lots of AI developers.


I guess I never got the memo that said I couldn't do what I in fact did with my Advanced Third Reich mod. [8|]

Have you ever even looked into the scripting capabilities of WarPlan or Strategic Command, or do you enjoy making bogus assertions and expect anybody to take you seriously? As I mentioned above, it takes a LOT of time and patience, combined with a LOT of experience with the game and the AI performance, to create a challenging computer opponent. It is difficult; it is not impossible. The real problem is that developers stop developing a particular game before moving on to their next game, never quite reaching AI nirvana and leaving players eternally disappointed. Third Reich PC computer game is one such example; close but never got there.

I would also argue that a WWII ETO game is not as complex as you make it out to be. There are two sides, Axis and Allies, with a fairly well-defined default strategy for both sides plus several major alternatives (Sea Lion, Spain, Turkey) and other variations here and there. It's not rocket science. Napoleonic Empires in Arms, on the other hand, with seven major powers and changing alliances over the period 1805-1815, is a more complex game in itself and the C++ coding is more difficult than text scripting. I do both; I know.



AlvaroSousa -> RE: AI/Realism (8/9/2020 3:17:45 PM)

The thing is that we can work the system till the cow comes home. Sometimes a developed A.I. system has limitations and you have to restart basically. Coding for scripts takes time, coding for A.I. even more time. Then testing even MORE time. What developers will not tell you is that there is a cost analysis that goes into design over the longevity of the game.

For example as well as WP has done I have effectively only made minimum wage for my hours put in over profit minus costs for 3 years of full time work.
Since the sales of WarPlan are going down should I invest another 1000 hours, ~6 months, in vastly improving the A.I. around all the flaws I currently see in my first attempt at a game?
The answer is that I would be shooting myself in my financial foot if I did. Needless to say I still work on A.I. for the game because I enjoy it. But somethings will require a lot of work around with many hours put in. This is what developers go through. So if you have a developer living off of games like WP then this is the struggle they have to make.

I literally make more an hour playing poker by far than I have profit in WP. But I enjoy this more.

So what is done then? The next version is improved and that is how things are done. Developers take what works, improve it, and pass it on to the next version of the game. Now some developers, and this is just my opinion, just reskin their game, with minor improvements. I don't do that and I know Hubert, from SC3, doesn't. We email each other discussing things.

I have a specific sequence of events for WP. WP2 will be what I really wanted to create. So why didn't I do it in the first place? It would be too difficult to create and design. Just like making a car you need to start small before you make a race car.

The ETO isn't complex to a human. But the small details in operational actions is for an A.I.. For example in my recent game my opponent pulled off a massive operation involving 3 paras and 5 tanks which cut off part of a front. I had my Stalingrad. A.I. can't do this without a Google Deep mind server farm. Certainly not one man.

Next it also time. Do you want to wait 10m while the A.I. does massive calculations? And this is my i7. If someone has an i3 or i5 CPU that could be 20m. It's a scale of ratio. WP wouldn't sell at all if I did that.

I do little changes on the engine I don't post in the logs because they are often unnoticeable to the players. There is a new one I am considering currently about the A.I.'s aggression.



kennonlightfoot -> RE: AI/Realism (8/9/2020 4:20:05 PM)

Third Right isn't as complex a game as this one. I can't really comment on it since I never played the computer version just the old board game and that was many years ago.

The problem with script driven AI is it can't embed the kind of logic needed to handle things when they don't go as planned. Also, in these large games it is very difficult for the AI to handle the logic to maintain continuous lines that are strong enough not to be broken. A player can tell at a glance at the map where their opponent is building up enough force to break his line. The AI doesn't do "glance at the map" very well. A good example is my game against the AI. I played it out into 1942. When the Russians join the war the AI wasn't scripted for handling two fronts in this way. It continued to maintain a double line against France and actually push them back some. But on the Eastern front it did almost nothing and Russia took Germany about as fast as it could move its units. Berlin was taken in about five turns.

As the number of hexes that must be controlled increase relative to the number of units available to control them the decisions the AI must make start getting into the trillions. Building the logic to make the AI capable of handling this is far beyond a PC's capacity.

What these games have is "Weak AI". Which more or less follows plain "IF THEN" rules and is implemented in normal programming languages. It can't do anything that the programmer didn't explicitly set it up for. This is my level of expertise in AI. I am trained in Real Time control systems using high level languages like Prolog, Gensym G2, Ladder Logic in Control Systems, etc. Games like WarPlan are much to complex for this level of AI.

And, I doubt any game companies can afford Strong AI. Maybe in the future when AI becomes more generally available in low cost systems.



Cigar King -> RE: AI/Realism (8/9/2020 6:28:47 PM)

The AI works as well as can be expected. Tweeks are fine, but, I doubt any amount of reasonable work would allow it to compete with a skilled human. Consider how long they worked on computer chess before it was able to beat a proficient human.

Give the AI some bonuses if it is not challenging enough for you. Solving some of the game play issues would, IMHO, provide more bang for the buck than focusing on the AI.



pzgndr -> RE: AI/Realism (8/9/2020 6:44:50 PM)

quote:

ORIGINAL: kennonlightfoot
What these games have is "Weak AI". Which more or less follows plain "IF THEN" rules and is implemented in normal programming languages. It can't do anything that the programmer didn't explicitly set it up for. This is my level of expertise in AI.


Alright, one more response. I invite you to study the scripting capabilities in WarPlan and Strategic Command. I am not convinced that you understand.

A very simple example of scripting variability is IF 2-4 enemy units approach within range 3-5 of an objective, THEN do something (deploy extra unit(s), initiate an attack, set up a defense line, etc.) So when 3 enemy units are within 4 hexes, what happens? You don't know. It might trigger the response, or it might not. You scripted this, but you cannot know explicitly what will happen. Almost like playing another person. You don't know what your opponent is going to do, something brilliant or something stupid. In this sense, the scripted computer opponent behaves in the same way. Also, there is "fuzzy logic" in the generic AI routines, at least for Strategic Command and I think Al has implemented this also.

Anyways, my final comment on all this and Al just confirmed above what I've been saying, is the time and effort necessary to create an exceptionally challenging computer opponent is generally more than a developer can afford to invest. I focused on Third Reich at the grand strategic level because it is more manageable. WarPlan at a more grand operational level is indeed more complex. That makes it more difficult, but again not impossible.



kennonlightfoot -> RE: AI/Realism (8/9/2020 7:58:00 PM)

My opinion is the AI can be used to learn the basics of the game but the game is to complex for it to handle campaigns. You can crank up the bonuses so it won't be a cake walk but it will still use very basic tactics. It's just the nature of the beast.

However, I do think it could handle more restricted scenarios that don't require it to do strategic planning.

That said, I will try and write a scenario for Poland just to see if I can do it. Hopefully, with such a limited scope I can make it work at a reasonable level for both sides.



AlvaroSousa -> RE: AI/Realism (8/9/2020 9:43:07 PM)

Interestingly A.I. is my favorite part of coding.



pzgndr -> RE: AI/Realism (8/10/2020 11:01:52 AM)

quote:

ORIGINAL: AlvaroSousa
Interestingly A.I. is my favorite part of coding.


+1

After working on game bugs past several weeks, it was refreshing to do some AI work this weekend.



Page: [1] 2   next >   >>

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
1.21875