Do we need an UNHISTORICAL LEVEL?

[fbastos]

I hear that the AI can be soundly beaten on the VERY HARD level. Do we need something like a REALLY VERY HARD level with unhistorical constraints to the player, to keep the challenge going?

Like, give the AI a hadicap on replacements level and quantity, advance the AI's reinforcements queue, boot its production?

Or give the AI combat advantage, so the player would need to get an overwhelming force to win combats and would lose handsomely if fought uno-on-uno?

Or even yet set more strict goals for the player, like must invade and conquer Japan (or Australia) in order to achieve a Decisive Victory?

Somehow I feel I'm going to get bashed on this thread.. but let it come!

F.

[steveh11Matrix]

Probably a mod, something like HOI's "Stony Road" one?

Personally, I'd have thought that simply trying as the Japanese ought to have been challenge enough...

Steve.

[52nd Lowland]

After playing against the AI and having started campaigns by PBEM i think that PBEM wins hands down.Simply because the quirks ofplaying against a unpredictable human player means you have to think carefully and think out your strategies.I found that playing the AI made me a lazier player as i could generally predict the AIs general moves in theatre.
However two turns into one campaign the PBEM player had thrown a couple of unexpected moves that meant i was playing catchup from the start.[:D]

[drw61]

Yes! I would like start screens were I can pick extra replacements, extra pilots, ship land and aircraft arrival dates are 30/90/365 days in advance or late. This would allow for more of a challenge from the AI. On the same screen would be an option for historical so that hopefully everyone would be happy.

[2Stepper]

I know over the years to come that there'll be an effort to improve the AI further... I mean what game doesn't see SOME form of issue with an AI? I can't think of any, though some are better'n others.

I've said this before, and I'll say it again... The Mod'er community will take hold totally and before long we'll have our own mods to suppliment this gem. The overall strength or weakness of the AI doesn't worry me in the least... [8D]

[Captain Cruft]

VERY HARD already gives the AI combat advantages. It won't make much difference though.

It will still send out penny packet unescorted transport TFs to be slaughtered by the dozen.
It will still attempt to supply empty bases.
It will still disband major fleets in size 3 ports near to your LBA.
It will still send TFs consisting solely of PT boats or barges across the entire Pacific without an accompanying fuel source.
It will still set half of its front-line fighter squadrons to "Night Ops - Long Range Cap" and the other half to "Training"
It will still move all its air groups around every turn for no good reason.
It will still deposit useless slivers of engineering units on every empty base from Anchorage to Karachi.
It will still not react to anything you do.
It will still, in a word, be utterly useless

The only challenge playing against it as either side is to how to avoid auto-victory on 1st Jan 1943.

JMHO of course, no offence intended ...

[52nd Lowland]

ORIGINAL: 2Stepper

I know over the years to come that there'll be an effort to improve the AI further... I mean what game doesn't see SOME form of issue with an AI? I can't think of any, though some are better'n others.

I've said this before, and I'll say it again... The Mod'er community will take hold totally and before long we'll have our own mods to suppliment this gem. The overall strength or weakness of the AI doesn't worry me in the least... [8D]

Excellent point..look at the work already started on things like scenarios and graphics mods.IMHO that will just build on what is an excellent game so far that can be tweaked and modded into something v special.

[Sneer]

[:(]there is no use from Ai
I'm breaking my head how to avoid auto victory as for March /42 as Japan
I have almost 6:1 advantage and it seems I can put it even further

[SiTheSly]

I'm breaking my head how to avoid auto victory as for March /42 as Japan
I have almost 6:1 advantage and it seems I can put it even further

One thing that you can do is change the value to the Allies of SF and LA so it is the same as it is for Japan.

So in other words SF and LA are worth 3000 each. This gives a bonus to the Allies of 5940 which means that you need an extra 24000 points to get the victory.

I'm playing Japan at the moment and will get the auto victory (but no where near your level) which is annoying since I wanted to see what it would be like facing the Allies when they receive their Essex carriers and better planes. I know I could play a later scenario but I like playing from the begining.

I'm going to try it for my next game and see what it's like.

[brisd]

I should clarify my comments on another thread that I have not played at Very Hard AI level (as Japan vs Allied AI), so I am using player comments and my own handling of AI at the Hard setting to state that most players will do well vs Allied AI as Japan in scenario 15, no matter what the setting. Not on first play, but after that yes. We have all the advantages, we know their OOB, their weaknesses, what to do and not do. AI can cheat and that will help but as Capt Cruft pointed out, it is not going to matter in long run. Having said that, I am curious how people playing Allies vs Japan AI are doing? The game seems designed with the idea that most campaigns will be played vs Japan AI, curious if it does any better. Players will come up with edited uber-AI scenarios eventually to fill the need.

[Mr.Frag]

Just so you guys clearly understand, there is a HUGE difference between HARD and VERY HARD.

Don't try and extrapolate between the levels.

Historic = AI plays completely blind

Hard = AI can peek at where you are going

Very Hard = all sorts of bonuses to logistics, combat mods, peeking, etc

[mongo]

The AI peeking..

Are we talking on the order of our SIGINT? or are we talking SIGINT on steroids?

[Mr.Frag]

The AI peeking..

Not sure exactly of the level of peeking, but you can probably consider it similar to loading the game and looking where things are for the other side now and then.

[mongo]

Either or would have sufficed.

You answered it though - steroids [:)]

thankies,

[fbastos]

there is no use from Ai
I'm breaking my head how to avoid auto victory as for March /42 as Japan

Hmm... camp your boats at Pearl Harbor's bay?

[fbastos]

Just so you guys clearly understand, there is a HUGE difference between HARD and VERY HARD.

Don't try and extrapolate between the levels.

Do you think that VERY HARD is hard enough, Frag?

[Mr.Frag]

Do you think that VERY HARD is hard enough, Frag?

When people start posting that they have auto-victory Jan 1st, 1943 for both sides, I'll get worried about it. Until then, I'm not too worried.

[Grotius]

If you impose a few modest constraints on yourself, the AI can give you a good game even on Historical or Hard level. I am on Hard, playing as the Allies, June 1942. The AI has the Phillipines, DEI, Malaya, Burma. I didn't put the US CVs in the DEI, or move a USMC division into Batavia, or strip all of India to defend Singapore, or anything like that. I limit replacements in areas that seem to me isolated from replacements. And I've been trying to follow a course parallel to that followed by Allied commanders in the war. So instead of piling all my marines into an attack on Eniwetok or other CentPac bases, I'm building up in the South Pacific, moving my CVs there, etc.

I also never -- EVER -- save and reload. That's a huge "cheat" in favor of humans. If part of a ground unit doesn't load, I plan for that and send a follow-up transport TF with the remaining bits and pieces to follow it. If I lose a CV battle because I forget to put up CAP, so be it.

Also, I'd like to try playing against the Allied AI on Very Hard in a 1944 or 1945 scenario.

Sure, PBEM will always give you a more intelligent opponent. But AI has its advantages. Because it tends to play within a historical framework, you you can get a war more like the real war. It's fast; you can actually play the whole war. I can realistically hope to finish my current game in the current calendar year. AI doesn't quit. And it doesn't care if you want to play 10 turns a night or take a few days off.

[Charles2222]

ORIGINAL: Grotius

If you impose a few modest constraints on yourself, the AI can give you a good game even on Historical or Hard level. I am on Hard, playing as the Allies, June 1942. The AI has the Phillipines, DEI, Malaya, Burma. I didn't put the US CVs in the DEI, or move a USMC division into Batavia, or strip all of India to defend Singapore, or anything like that. I limit replacements in areas that seem to me isolated from replacements. And I've been trying to follow a course parallel to that followed by Allied commanders in the war. So instead of piling all my marines into an attack on Eniwetok or other CentPac bases, I'm building up in the South Pacific, moving my CVs there, etc.

I also never -- EVER -- save and reload. That's a huge "cheat" in favor of humans. If part of a ground unit doesn't load, I plan for that and send a follow-up transport TF with the remaining bits and pieces to follow it. If I lose a CV battle because I forget to put up CAP, so be it.

Also, I'd like to try playing against the Allied AI on Very Hard in a 1944 or 1945 scenario.

Sure, PBEM will always give you a more intelligent opponent. But AI has its advantages. Because it tends to play within a historical framework, you you can get a war more like the real war. It's fast; you can actually play the whole war. I can realistically hope to finish my current game in the current calendar year. AI doesn't quit. And it doesn't care if you want to play 10 turns a night or take a few days off.

There's nothing wrong with saving/loading if you use some discipline. I had thios rule in CIV3 that whatever the turn turned out as, I could only change the deficiency, such as forgettingto put up CAP in the WITP case. If I didn't expect trouble and thereby didn't put it up, then I suffer. I mean, a commander doesn't expect enemy activity and forget to put up air cover. If you wish to continue doing that sort of thing, a realistic caveat you ought to impose on yourself too, and that is the firing of that TF commander, even though it costs you PP's (assuming he lives through the fight).

[Grotius]

Hehe, I also play Civ 3 "Ironman" style -- I never save and reload. It's just the way I prefer to play; it's neither better nor worse than anyone else's way. In WITP, if I forget CAP, it's my tough luck. It gives me incentive to be more careful about remembering it.