I'm not sure whether to continue much longer

Matrix Games Forums

Forums Register Login Photo Gallery Member List Search Calendars FAQ

My Profile Inbox Address Book My Subscription My Forums Log Out

I'm not sure whether to continue much longer - AI Game

View related threads: (in this forum | in all forums)

Logged in as: Guest

Users viewing this topic: none

Printable Version

All Forums >> [New Releases from Matrix Games] >> War in the Pacific: Admiral's Edition >> I'm not sure whether to continue much longer - AI Game

Page: [1]

Message

<< Older Topic Newer Topic >>

I'm not sure whether to continue much longer - AI Game - 3/1/2011 6:17:33 PM

brian800000

Posts: 225
Joined: 9/15/2010
Status: offline

First, I want to say that the AI team did a great job. A credible performance by the AI is a great achievement considering the complexity of the game.

However, I am now in late April 1942 and starting to assert dominance as the allies. I know I've sunk 1 CV and 2 BBs, and it looks like 2 CVLs are about to go under (if they haven't already). Several CAs are probably also sunk (I'm guessing 3-4). I've only lost 1 CA (no BB or CVs) since December 7. I think I can hold Singapore, I know I'll hold Java and Sumatra (I sank an invasion force in which the AI took 50,000+ casualties).

I'm not sure whether I should continue. On the one hand, I want to learn the game for a future PBEM, and only going to April isn't much of a learning experience. On the other hand, I want to also do an AI game as the Japanese before I start a PBEM, and I would probably learn more as Japan rather than waiting around for several months until I have enough units to start an offensive.

What do the more experienced players suggest?

Post #: 1

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 6:30:01 PM

Xxzard

Posts: 440
Joined: 9/28/2008
From: Arizona
Status: offline

The AI will continue to pour resources into hopeless endeavors. For as cleverly scripted as it is, it is very "target fixated" and very easy to attrit by predicting where it will try to attack next. I recall in my longest game vs the AI, it just kept attacking the Solomons and Java, until all it's resources were depleted. I went all the way into mid '43 with that game, but the Japanese resistance eventually grew so weak due to depletion of fighting units that it just wasn't much fun anymore.

A PBEM game will probably diverge so far from your experience with the AI that it may not even be all that comparable, especially in April 1942 (That's when things can get very interesting

).

I suggest moving on to a new game, either another as the Allies at the start to work on your opening moves, or a new game as Japan for a change of pace.

_____________________________

(in reply to brian800000)

Post #: 2

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 6:49:44 PM

m10bob

Posts: 8622
Joined: 11/3/2002
From: Dismal Seepage Indiana
Status: offline

OTOH, I play against the AI as I did in WITP and find the AE version to be robust and original later on in the game.

I have seen it stage new endeavors at least in late 1943 which had nothing to do with any (then) campaigns.

My work hours have never allowed me to play another human and compared to WITP, this is a vast improvement.

Don't forget, the editor allows modders to program their own AI in depth for later stages of the game, and this in itself is reflected in the stock AE game I am playing.

It can be a challenge.

_____________________________

(in reply to Xxzard)

Post #: 3

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 8:09:06 PM

HansBolter

Posts: 7704
Joined: 7/6/2006
From: United States
Status: offline

I've played extensively against the AI and can tell you that if you are aggressive early as the Allies the attrition you cause the AI will cause the game to skew heavily to the Allied side by the fall of '42.

There are two ways to counter this.

Play passively early, pretend you don't have the hindsight of history to know where the AI will be going and when.

Try the Iron Man scenario. Even with aggresive early Allied efforts the game won't be as skewed in the Allied favor unitl early '43.

Combine the two and play Iron man passively as the Alles early and your game will stay viable for a much longer time period, giving you a chance to experience more of the campaign without feeling like the game isn't worth playing on.

_____________________________

Hans

(in reply to m10bob)

Post #: 4

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 8:14:10 PM

JeffroK

Posts: 6391
Joined: 1/26/2005
Status: offline

quote:

ORIGINAL: HansBolter

I've played extensively against the AI and can tell you that if you are aggressive early as the Allies the attrition you cause the AI will cause the game to skew heavily to the Allied side by the fall of '42.

There are two ways to counter this.

Play passively early, pretend you don't have the hindsight of history to know where the AI will be going and when.

Play historically, build up bases and garrisons just as you would have to IRL.

Try a different line of approach and see what happens.
Maybe leave your forces with 2nd line aircraft and see what happens.

The AI is a tool to play with, not smash which then leads you into boredom or PBEM (much the same)

_____________________________

Interdum feror cupidine partium magnarum Europae vincendarum

(in reply to HansBolter)

Post #: 5

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 8:37:37 PM

brian800000

Posts: 225
Joined: 9/15/2010
Status: offline

I tried playing historically, without the benefit of hindsight. The problem is that this was my first game and especially at first I was figuring out how to set up convoys as much as any grand strategy.

I'm somewhat attached to my game--I've had fun and a part of me wants to play it out. But considering the time commitment to finish the game, I'm not sure I want to finish smashing an opponent that is on the verge of being beaten this early.

(in reply to JeffroK)

Post #: 6

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 9:28:15 PM

vettim89

Posts: 3615
Joined: 7/14/2007
From: Toledo, Ohio
Status: offline

I agree with other posters in that playing an AI game requires a little bit of self control as the Allied player. You need to pull back to historical lines of defense. A fighting withdraw is fine but you need to let the AI have the SRA with only historical forces resisting the advance.

The AI is much better in AE than WiTP. There are actual multiple pathways that can be followed as far as the AI scripts. So the AI will vary from game to game. That said, any attempt by either player to break the AI will likely be successful. Some one invaded Hokkaido in 1942 a while back and the AI didn't even try to defend it. There are only so many hours in the day and space in the computere as far as writing and storing the AI scripts. Not every possibility can be accounted for by the AI; so if you go off the reservation so to speak you can expect teh AI to collapse

_____________________________

"We have met the enemy and they are ours" - Commodore O.H. Perry

(in reply to brian800000)

Post #: 7

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 9:36:07 PM

brian800000

Posts: 225
Joined: 9/15/2010
Status: offline

quote:

ORIGINAL: vettim89

I agree with other posters in that playing an AI game requires a little bit of self control as the Allied player. You need to pull back to historical lines of defense. A fighting withdraw is fine but you need to let the AI have the SRA with only historical forces resisting the advance.

The AI is much better in AE than WiTP. There are actual multiple pathways that can be followed as far as the AI scripts. So the AI will vary from game to game. That said, any attempt by either player to break the AI will likely be successful. Some one invaded Hokkaido in 1942 a while back and the AI didn't even try to defend it. There are only so many hours in the day and space in the computere as far as writing and storing the AI scripts. Not every possibility can be accounted for by the AI; so if you go off the reservation so to speak you can expect teh AI to collapse

I think I am going to hold Singapore and Burma without providing any reinforcements (including the historic ones). That excludes air--I have sent air units. Especially Singapore was touch and go for a while, but I think I'm secure. I'm guessing I only held Burma because the AI script to attack is triggered by the fall of Singapore (purely a guess on my part).

I know the AI isn't considered as strong for the Allied side, but if I play Japan according to the historical guidelines, how will things go?

(in reply to vettim89)

Post #: 8

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 9:47:05 PM

Chickenboy

Posts: 24520
Joined: 6/29/2002
From: San Antonio, TX
Status: offline

I look at playing against the AI somewhat differently, I guess. The AI is there to break, to abuse and to manipulate as best you can to enable you to learn as much as possible on your pace. It is a useful learning tool, but quickly outlives its usefulness, IMO.

In training (martial arts, boxing, etc.), the addage is to 'train like you want to fight'. If you repeatedly train lackadasically with inferior opponents and 'go easy' on them (and vice versa) you'll never know what you are capable of doing when truly pressed to perform in a competitive environment. The more I convince myself that I *can* hold Singapore against a good IJ opponent or that I *can* hold the Philippines or that I *can* hold Java indefinitely, the more I skew my learning process about the game and short change my own growth as a player. Reverting to the boxing analogy again, if I strove to 'even things out' by literally fighting with one arm behind my back or by tying my feet together (e.g., AI on Ironman, leaving poor fighters in place, etc.), that may even up the fight somewhat and make it more challenging to win that fight. However, it's just not a realistic way to train oneself and maximize one's potential with the engine at hand.

So go ahead and break the AI. Hurt it. No mercy. Let it impale itself on your defenses. Gut it like a fish. Then load up a GC against the Allied AI. Do the same. Then train against a human being, cause there is nothing quite like it.

_____________________________

(in reply to brian800000)

Post #: 9

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 11:21:19 PM

sdhundt

Posts: 300
Joined: 9/6/2002
From: Milwaukee, WI, USA
Status: offline

If you plan on playing a pbem game any time soon my BEST advice would be to at best to play the AI basically as a side show event or consider it a World Wrestling Event where you know nothing really is as it really is. If you continue to play the AI you might get some VERY BAD habits and when you finally play a human player he will most likely crush you. Just some friendly advice :-)

(in reply to Chickenboy)

Post #: 10

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 11:27:49 PM

Bullwinkle58

Posts: 11302
Joined: 2/24/2009
Status: offline

quote:

ORIGINAL: brian800000

I tried playing historically, without the benefit of hindsight. The problem is that this was my first game and especially at first I was figuring out how to set up convoys as much as any grand strategy.

I'm somewhat attached to my game--I've had fun and a part of me wants to play it out. But considering the time commitment to finish the game, I'm not sure I want to finish smashing an opponent that is on the verge of being beaten this early.

Just a suggestion. Load the save PC vs. PC and let it run into mid-1943, then pick it up and play 2-day turns. It's worth seeing the late-war stuff as the Allies. Script playing script ought to probably still leave the Japanese somewhat competitive.

Or, if you like, re-start a Scenario 2 on Hard, let it go PC-to-PC to Jan. 1943, and pick it up on 2-days. This will probably leave the Japanese on much better than historical terms, and you can take it from there.

The second half of the show is very different, and you need that practice moving circa 10,000 units to play a late-war Allied game. (I think Tracker has it at 9000 and change, but who's counting?)

PC vs. PC on no-pause, no reports, no animaiton can get you into mid-war about overnight depending on how fast your box runs.

_____________________________

The Moose

(in reply to brian800000)

Post #: 11

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 11:31:56 PM

sdhundt

Posts: 300
Joined: 9/6/2002
From: Milwaukee, WI, USA
Status: offline

Bullwinkle, good advice, I think I'll try that myself.

(in reply to Bullwinkle58)

Post #: 12

RE: I'm not sure whether to continue much longer - AI Game - 3/1/2011 11:37:14 PM

Bullwinkle58

Posts: 11302
Joined: 2/24/2009
Status: offline

quote:

ORIGINAL: sdhundt

Bullwinkle, good advice, I think I'll try that myself.

I like to think of PC vs. PC as our own little May 1942 Start scenario.

_____________________________

The Moose

(in reply to sdhundt)

Post #: 13

RE: I'm not sure whether to continue much longer - AI Game - 3/2/2011 1:38:28 AM

sanch

Posts: 421
Joined: 10/30/2004
Status: offline

Another option would be to flip to the other side. I did this once in WITP on 1-Jan-1943 after the AI went passive late in '42. Talk about a crash course in Japanese economy - it was totally dead in the water, no wonder the AI went passive. It took about a month to fix the worst of the economy, but then had fun - actually did a couple offensives (bloody tho ...).

(in reply to Bullwinkle58)

Post #: 14

RE: I'm not sure whether to continue much longer - AI Game - 3/2/2011 10:08:33 PM

Alpha77

Posts: 2116
Joined: 9/24/2010
Status: offline

�Well you don�t want revenge for all the pain the Jap inflicts on you early on ?

After running and bleeding the real fun should start in mid 42 or so when you can also do some offensive moves imho. Also I have hurt the AI some more than you but still it is a challenge to mount big offensives. Also the Jap has still enough planes (he seems mostly crippled on the carrier front in my game which is actually historic correct). Also the comp has "stolen" one Marine div. from me (it just dissapeared somehow after was recombined and shipped VERY OLD bug i guess - I am playing older version) so I am already weak on the troop sides (one more challenge, even if the Aussies seem to be capable of throwing the Jap alone in NG)

Oops:

"I think I can hold Singapore, I know I'll hold Java and Sumatra (I sank an invasion force in which the AI took 50,000+ casualties). "

Seems you were unlucky and got a "not so aggressive" script, cause I had no hopes in my game vs AI to hold those....

< Message edited by Alpha77 -- 3/2/2011 10:11:22 PM >

(in reply to sanch)

Post #: 15

RE: I'm not sure whether to continue much longer - AI Game - 3/2/2011 10:15:53 PM

Alpha77

Posts: 2116
Joined: 9/24/2010
Status: offline

Or your just a much better player then me

Well I fared better on the navy I think, but you are better at defending locations which historically have fallen to the Japs without much hope in reality to hold those (Singapore, Java etc.). So you played a more ahistoric game it seems...

Can you post the VP status ? And some details about your game (maybe screensshots?)

You can read about my AI game here:

http://www.matrixgames.com/forums/tm.asp?m=2595850&mpage=1&key=?

I will continue at the weekend....

So will people report a bit more about their games vs. AI ? Seems in the AAR forum isn�t much to find

Except mine few reports....

< Message edited by Alpha77 -- 3/2/2011 11:45:29 PM >

(in reply to Alpha77)

Post #: 16

RE: I'm not sure whether to continue much longer - AI Game - 3/2/2011 11:56:42 PM

John Lansford

Posts: 2662
Joined: 4/29/2002
Status: offline

I'm up to 4/44 in my campaign game vs the AI and it's become a lot like a late war WITP game already. Virtually the entire IJN surface fleet has been sunk; I've not seen anything bigger than a DD on the seas since mid-43. I lost two BB's at Pearl Harbor and those are the biggest ships I've lost the whole game. All my CV's are intact, and the RN still has both Repulse and PoW.

Campaign wise, I've advanced up to the southern Philippine Islands via both New Guinea and the CentPac route, and also retaken half of Timor, Ambon and Kendari by advancing north out of Darwin. The British have finally taken Prome and are on the outskirts of Rangoon, and made an amphibious landing further south on the peninsula. I occupy Guam, Rota and Saipan as well, but for the last 9 months game-wise it's basically turned into a game of 'prepare, build up supplies, load the troops, go take another base". I have thought of invading Formosa next, or another base nearby; right now the only limit to what I can do is how long it takes to build up supplies and fuel for the massive number of ships I'd need.

(in reply to Alpha77)

Post #: 17

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 12:25:49 AM

Alpha77

Posts: 2116
Joined: 9/24/2010
Status: offline

John: "British have finally taken Prome and are on the outskirts of Rangoon"

Ah, my brits did better they are before Rangoon but in Okt 42 :) But I must praise myself planning on this offensive which pretty much rolled back the Jap steamroller

�However the troops did good work.

"but for the last 9 months game-wise it's basically turned into a game of 'prepare, build up supplies, load the troops, go take another base"

And is it still fun, or more "work" now ?
I have to say the level of micromanament starting in mid 42 when the reeinforcements roll in is really a bit too much. All those small battalions etc. some tanks, some engineers per unit. You get "hundreds" of "seabees" and simmilar small units. You cannot select 20 of them at once, even in old HOI you can just send 20 divs at once with 2 clicks. Or in the old "WIR" you can transfer easily severall units to a new corps and move them with 2 mouseclicks or 2 key commands :( In this one you need to select every small unit... and forming TFs could also be easier btw. So this workload maybe be a bit too high on the long run..... nice is also I need to transport single regiments cause of this bug (dissapearing divs). Some INDIAN unit got "teleported" out of nothing to Noumea.. I don�t like this level of "chaos" in a strategic game really. I mostly try to put some order to my assets. Which is not so easy to do in this game....rant off.

Ah I forgot all the Chinese sceleton units which have no use anymore whatsoever

< Message edited by Alpha77 -- 3/3/2011 12:56:18 AM >

(in reply to John Lansford)

Post #: 18

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 2:01:25 AM

brian800000

Posts: 225
Joined: 9/15/2010
Status: offline

I'm trying to upload a picture of the VP screen, but it says the picture is too large (never tried this before).

Essentially, my plan was to try to hold Java and Palembang. No reinforcements would go to Singapore or the PI, but I wouldn't try to evac those bases either (they could be a buffer while I fortified Java and Palembang). I kept two AS's in Manila as long as possible to keep cycle times on subs short, as well as AMs to keep Manila harbor clear (this was very farsighted). Singapore would also be a major sub base, with very short cycle times for subs to attempt to intercept troops to Malaya. These took a terrible toll on unescorted AI convoys.

When I discovered that the AI was using lightly escorted convoys, I started using destroyers off of Borneo (iniitially I was going to hold my fleet back until carriers arrived for the massive battle for Java/Palembang), which ended up costing me some DDs but also picked off a few invasion groups. The Vildebeasts at Singapore as well as 2 PT squadrons I moved there took a horrible toll on the invasion forces (subs did damage too).

For a while it looked like Singapore might fall. But the AI move all its troops into Singapore, and so some of the wrecked remnants I had outside of Singapore were able to march into Jahore Bahru and cut the AI off. Eventually they got pushed out, but the AI forces in Singapore lost supply for a while. I then loaded Palembang with as many Dutch bombers as it can hold without overstacking and have been hammering the ground units in Singapore ever sense. The AI seems to want to set up an airfield at Jahore Bahru, but I'm preventing that with most of my B17s stationed at Batavia keeping that airfield closed. I have total air superiority over Singapore.

In other regions, the AI took Luganville briefly (and very early on--this may have been the AI strategic focus), but got murdered doing so. Through a odd bit of timing, their invasion went off when I had 4 US carriers in the area (some were in transit to Batavia as it hadn't occured to me to use the CT route). My carrier forces slaughtered the invasion forces, followed up by a strong SC TF. I probably sank 50-60 AKs and AKLs in the fight--many loaded with troops. They took the base, but I struck back landing 2 aussie divisions against a very outmanned Japanese base force.

The AI also landed 2 times on Canton Is. with significant carrier support--but I have a regiment at the base and the AI seems incapable of dealing with that, so both attacks failed. They also tried a VERY ill advised attack on Pearl Harbor in April, and I think I sent 2 CVLs and a couple of CAs down without so much as an airstrike against PH launching. Finally, the AI in April tried to attack Java. I have about 2500 AV on the island, and it is now protected by 4 carriers (3 british including the Hermes, 1 American), 2 BBs and a number of cruisers and destroyers. They did launch carrier attacks on Batavia, but my air forces shredded the attacks. Their actual invasion force was completely uncovered, and was a predictable massacre (over 50k casualties).

In Burma, I decided not to try to defend it after the initial reinforcement convoy got hit by LBA. I sent most of my units to Rangoon for a last stand and a minimal garrison force to Mandalay. Enemy troops showed up in January in Pegu, and I'm sure they could have taken Rangoon (and maybe still could), but they just sat there. 1 Japanese division moved North and took much of Burma, and entered Mandalay. They could have wiped me out--I had well under 100 AV there--but instead they just sat there. I put a bunch of bombers in Chittagong to bomb it daily, and eventually it was destroyed. No ground combat ever occurred.

< Message edited by brian800000 -- 3/3/2011 2:08:56 AM >

(in reply to Alpha77)

Post #: 19

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 4:35:09 AM

BigDuke66

Posts: 2013
Joined: 2/1/2001
From: Terra
Status: offline

Maybe one approach can be to play the different nations as allies more separated.
I guess most players ignore nationalities and use all forces as one big army/navy/air force, but reading a bit about the different nations you see that they had different targets, strategies, etc., for example Chinas attitude of conserving their forces for the fight against the Chinese Red Army.
So does anyone know a good comprehensive source that sums up each Nations "way of fighting the war in the Pacific"?

_____________________________

JOIN The Blitz Wargaming Club
JOIN Napoleonic Wargame Club
JOIN American Civil War Game Club

(in reply to brian800000)

Post #: 20

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 4:45:12 AM

sdhundt

Posts: 300
Joined: 9/6/2002
From: Milwaukee, WI, USA
Status: offline

Aside from messing around (I just started Iron Man to play when no human turn ready) or trying out some new tactics, the Old saying "Once you stop playing with yourself you'll never go back", Is very much true.

(in reply to BigDuke66)

Post #: 21

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 11:45:21 AM

koontz

Posts: 274
Joined: 8/27/2009
Status: offline

STOP IT! NOW!

Its mutch more fun to play vs an human.

I think it even would be more even if the AI was better(mayby in the distant future)

then
a human player.

_____________________________

Amateurs study tactics, professionals study logistics.

"All warfare is based on deception. There is no place where espionage is not used. Offer the enemy bait to lure him."

(in reply to sdhundt)

Post #: 22

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 2:14:38 PM

Shellshock

Posts: 533
Joined: 12/31/2010
From: U.S.
Status: offline

quote:

ORIGINAL: BigDuke66

Maybe one approach can be to play the different nations as allies more separated.
I guess most players ignore nationalities and use all forces as one big army/navy/air force, but reading a bit about the different nations you see that they had different targets, strategies, etc., for example Chinas attitude of conserving their forces for the fight against the Chinese Red Army.
So does anyone know a good comprehensive source that sums up each Nations "way of fighting the war in the Pacific"?

You could do something roughly equivalent to this in the original WITP by assigning computer control to certain parts of the map. Giving computer control to China was popular as most Allied players didn't want to mess with that confused and jumbled theater. However, it was frustrating watching the computer stupidly fritter away the British Fleet when I once made the mistake of giving them control of that sector.

China's national priority--avoid provoking the Japanese too much and try to kill Mao.

Britian's national priority--hold on to India and avoid a fleet clash with the Japanese until they were greatly weakened.

USA's national priority--Read Tojo's mail and crush Japan.

(in reply to BigDuke66)

Post #: 23

RE: I'm not sure whether to continue much longer - AI Game - 3/3/2011 2:34:36 PM

beppi

Posts: 382
Joined: 3/11/2004
From: Austria
Status: offline

In general the AI is the AI with a huge amount of weaknesses and things that could been exploited. So restrict yourself to not exploit that stuff. And it is true that a PBEM is much more fun and you can learn much more in a PBEM.

If you want put the AI to very hard, then at least you can train things like. Invading bases against 100 prep forces, learn the overall strategic movements, learn how to prep the proper command line (area hq, corps hq, army hq) as you need the extra AV, nevery to forget you amphib HQ with their AGCs and so on.

As the AI gets almost unlimited planes you can train how to proper rotate air groups when they get attrited, how to defend landings against land based air and so on.

For Naval warfare it is harder against the AI as it tends to trash its carriers, but you can train how to do bombings, how to do interceptions and so on.

Overall you can learn/train against the AI how to build up a supply network and how you best ships you stuff from here to there.

There are always more things to learn as a PBEM is completely different to the AI but some strategic knowledge you can learn and train against the AI.

< Message edited by beppi -- 3/3/2011 2:36:23 PM >

(in reply to Shellshock)

Post #: 24