Matrix Games Forums

AI only.   I tried once to let the AI control the SW Pacific and South Pacific so I could control
the CBI theater(re),   which I did.  But the AI performed so poorly in SW Pacific and S. Pacific I couldn't continue.  I am slowly getting a WITP urge back now though.    I'll have to control the SW Pac and S. Pac area again and let CBI go to h**l.    

ORIGINAL: Rainer

Peter,
you should set Control Zone S also to Human Control.
There is a confirmed bug which makes AI interfere with your actions in Northern Australia, Papua and New Guinea.
Again, ONLY when Control Zone S is left under Computer Control.

During the first couple of weeks it doesn't really matter, but when you start to operate in the Port Moresby Area the bug hits. Examples: Air Units reset given targets, LCUs reset given orders. Ghost Units appear in several places.

Having said that: I am a convinced Allied Human against Japanese AI player (currently April 4, 1945), and I use the same settings as you do (British, Chinese handled by AI), except Dutch East India (Sumatra, Java, Borneo) because of the bug.
Cheers
Rainer

Thanks! Didn't know about this bug, would be nice to see this fixed!

ORIGINAL: Sonny II

I think the fact that folks are still working on it and the code has not been released to the public (there was talk of this a year or so ago) would indicate that there is still some market for the game.

Never been any serious talk of releasing the code. That's only in the fevered dreams of a few forumites...

Not sure if this matters but I would be happy to buy an add-on pack ; having the focus just for cleaning up things with the code/including interface as well as making the AI a bit better player....though I guess financially wouldn't make much sense.

ORIGINAL: DrewBlack

its rubblish!!!

Woudl never call it an Ai!!!! that implies intelligence, which it DOES NOT HAVE!!!

Joel ive said it before, I would pay good money ( lot of good money) for any sort of Ai in this game... i love it but the Ai is none existant... even something on the lines ofGBWII Ai would be great. Something please[:(][:(][:(]

Drew

i agree AI for me means Amazing Incompetence

ive sen the AI do astoundingly stupid things

1. bringing ryujo and her TF into Port Moresby, where she was promptly sunk

2. bringing tankers in range of wake, where they are promptly sunk

3. allowing me to cut off and destroy 80000 nips during the early burma campaign

4. following that failing to even truly resist my subsequent rampage back thru burma, thru malaya, indochina, and up the coast of china. to shanghai, where i finally decided to stop advancing

[:D]

Hi all,

221 entries in the poll so far... so that's (more or less [:D]) how many of us visit this place regularly...


Leo "Apollo11"

ORIGINAL: marky
i agree AI for me means Amazing Incompetence
[:D]

I believe the real defination of AI is Animated Ignorance. Note how no matter what game you chose, the AI's actions seem primarily motivated by the need to "look busy" and give the player something to react to. This one is both better and worse than many. Better in that it does reasonably well in following it's "scripted moves" in the initial Japanese conquest..., and worse in that there really dosen't seem to be any AI at all after that.

Well I have only played as the Japanese against the AI and I have to say I have been given a decent contest and it is enjoyable.
 
I play fairly historically and to enjoy the historical aspects of the game rather than trying to win at all costs or to find exploits. And I play on the very hard setting.
 
To qualify my experience with the AI : I have yet to get beyond December 1942 ( due to other demands on my time and numerous restarts) and I usually leave the China front to be controlled by the AI. So maybe the AI performs worse after 1942 or in China ?
 
Sure the AI has faults, but nothing that breaks the game for me.
 
One reason I haven't tried PBEM is that it seems you need to agree on a long list of house rules to ensure a balanced game.
 
Whilst acknowledging the flaws in the game identified by some in this forum I do think this is a wonderful game capable of delivering years of enjoyment and I thank the developers and modders responsible producing such an addictive game. 

 

While those of us who play against the AI can in time predict an outcome, or action of the AI, maybe we should have a seperate way (or method)of calculating victory.
For me, I play to limit losses.
An end game screen similar to that of SPWAW showing detailed losses would be a nice touch/utility in time?
To date, I have managed to not lose an Allied carrier till approx mid '43,(without cheating by keeping them locked up in port.)
In fact, playing against the AI might have an extra criteria of points for the following:
1. Ability to hold the Phillipines
2.The ability to hold New Guinea
3.(Java,etc.)

Yeah, I know each side already gets points for these things, but maybe the ability to alter the points (to favor the AI) might be considered as an option on the preferances screen instead of just the editor?
I mean a detailed preferances capability which might show maybe a sliding guage with an arrow, and a list of all the important bases with "value" boxes which could be changed prior to the game, based on possible game strategies?

Just read through this thread, trying to decide wether or not I should buy WITP, never been much of an MP player, simply don't have the devotion for it, particulary not when it comes to PBEM etc, would you guys still say its worth buying the game?
 
I really enjoyed Pacific War from Grigsby even against the AI!

ORIGINAL: Gargantou

... I really enjoyed Pacific War from Grigsby even against the AI!

I've played that game as well; it's very manageable, enjoyable and doesn't take years to finish, but the AI isn't impressive, and how things are done -- including final victory conditions --are sometimes confusing.

I think the biggest AI issue in Grigsby games -- WitP, UV, PacWar -- is that AI TFs act like samurai and repeatedly bonzai charge islands and airfields, often w/predictable results. Although I would -- and should -- expect the IJA to act like this, the navy shouldnt be so suicidal!

No chance. When I reported this bug a long time ago (3 years?) it was acknowledged, however the developers said it is too complicated (too many side effects) to be fixed. So we have to live with that.

ORIGINAL: Joe D.

ORIGINAL: Gargantou

... I really enjoyed Pacific War from Grigsby even against the AI!

I've played that game as well; it's very manageable, enjoyable and doesn't take years to finish, but the AI isn't impressive, and how things are done -- including final victory conditions --are sometimes confusing.

I think the biggest AI issue in Grigsby games -- WitP, UV, PacWar -- is that AI TFs act like samurai and repeatedly bonzai charge islands and airfields, often w/predictable results. Although I would -- and should -- expect the IJA to act like this, the navy shouldnt be so suicidal!

Heh yeah, you know what would be interesting? If someone attempted to do a game like Pacific War/UV/WiTP etc, with a learning AI, basically, the AI would 'record' what actions led to the most deaths on his side, and learn from this to avoid those actions.

Just a dream of mine.

I play with a group that uses WiTP a lot. Our scenarios lead into a big CPX h-2-h lasting over a weekend. We design our own scenarios and play WiTP out to the point we switch to the invasion scen based on TacOps, with specific maps etc .. Almost all of us use an AI based scen to devise alternatives and gauge the risk of a particular maneuver. Then, we use the results as a basis for risk evaluation and go forward, with a realistic appreciation of the probability of success/failure.

An AI is a tool. It is not an opponent. It does not have the ability to recognize strategic lines of advance, it does not have the ability to recognize flanking operations. An AI is a reactive piece of software code that will do what the programmers tell it to do, without regard as to what you are doing. The longer you play against an AI, the more out-of-sync the AI opponent will be; that’s the way computer AI things are.

I don't agree completely with you JWE, I have seen some games that have had quite good AIs, such as the Silent Storm series.
 
I think the hardest AIs truly are the ones for these grand-scale strategy games, they become far more predictable than in other games.
 
I do believe we're moving closer and closer to truly good AIs though, perhaps in some 5-10 years we'll atleast have AIs that can mimmic a human opponent, i.e. 'record' their tactics and use them as a basic formula from which to work with.
 
But IMO, in most of these strategy games like War in Russia, Pacific War, UV etc, the term AI is very unfair, because it feels more like a "Programmed Opponent" rather than an AI.
 
The biggest weakness when I play strategy games in particular, is the ability for the AI to adopt.
 
Oh well, me being an optimist I will just keep hoping that we'll eventually reach an AI that atleast is close to an average humans capability to think logically.

ORIGINAL: Gargantou

I don't agree completely with you JWE, I have seen some games that have had quite good AIs, such as the Silent Storm series.

I think the hardest AIs truly are the ones for these grand-scale strategy games, they become far more predictable than in other games.

I do believe we're moving closer and closer to truly good AIs though, perhaps in some 5-10 years we'll atleast have AIs that can mimmic a human opponent, i.e. 'record' their tactics and use them as a basic formula from which to work with.

But IMO, in most of these strategy games like War in Russia, Pacific War, UV etc, the term AI is very unfair, because it feels more like a "Programmed Opponent" rather than an AI.

The biggest weakness when I play strategy games in particular, is the ability for the AI to adopt.

Oh well, me being an optimist I will just keep hoping that we'll eventually reach an AI that atleast is close to an average humans capability to think logically.

I absolutely feel what you say will happen. Computers are more powerful everyday and eventually someone with some "muther" of a computer will solve the problems. And then, it will be driven down to be usable to other less massive computers and eventually to the public.

It won't happen it my life time, but, failing a giant meteor, super volcano, or other massive catastrophe will happen. And I suspect it won't be that far in the future.

Edit forgot to spell check.

Same here, though I'm hoping it will be before the end of my life-time(which I am estimating to end around 2060-2070 if by 'natural causes'), it's pretty insane if you look at how far society has come in merely the past 60-70 years or so!

Oh well, wandering off-topic, sorry mods![:)]

ORIGINAL: Gargantou

Same here, though I'm hoping it will be before the end of my life-time(which I am estimating to end around 2060-2070 if by 'natural causes'), it's pretty insane if you look at how far society has come in merely the past 60-70 years or so!

Oh well, wandering off-topic, sorry mods![:)]

In the 2060-2070 timeframe, I will be over 100 years old.

Not likely, so they better damn well hurry up!

Regards,
Feltan

ORIGINAL: Feltan

ORIGINAL: Gargantou

Same here, though I'm hoping it will be before the end of my life-time(which I am estimating to end around 2060-2070 if by 'natural causes'), it's pretty insane if you look at how far society has come in merely the past 60-70 years or so!

Oh well, wandering off-topic, sorry mods![:)]

In the 2060-2070 timeframe, I will be over 100 years old.

Not likely, so they better damn well hurry up!

Regards,
Feltan

I was born 1990, so I'd be 70-80 years, anyway, I saw an interesting article with some guy who was a geneticist(spelling?) about a few months ago, he said that he believed the first person that will reach the age of 160+ years old is already born, and I don't find it all that unbelievable, I mean compare the average life lengths of people from the end of the 19th century to the end of the 20th, there's a substantial difference, atleast over here in Sweden.

In the end, isn't the biggest 'cause' fo death when it comes to old age basically that the bodies own immune defense becomes very weared down which makes you far more weakened and thus far more sensitive to even things such as flu?

ORIGINAL: Gargantou
...In the end, isn't the biggest 'cause' fo death when it comes to old age basically that the bodies own immune defense becomes very weared down which makes you far more weakened and thus far more sensitive to even things such as flu?

Maybe for some. As for me, a lifetime of poor habits and immodest living will likely be the end of me!

Regards,
Feltan