Pearl Harbor and AI
Moderators: wdolson, MOD_War-in-the-Pacific-Admirals-Edition
RE: Pearl Harbor and AI
Well, guys, we are either talking about playing a "historical simulation" or a "historical game." The simulation should turn things out pretty much as they did historically, IMO. The game could take history and tweak it a bit here and there, not based upon little green men, but based upon concepts that might have been within the grasp of the historical participants, had they some more foresight...since we clearly have the benefit of hindsight. I can live with calling this a well-researched historical game, that takes a liberty or two to better enable me to play it with more enjoyment, and for longer. Yes, a grog can argue that a shortage of such and such type of bomb fuzes could totally negate this and that type of scenario from ever happening, but who is to say what might have happened historically with just a few changed events and conditions? War is all about uncertainty. Writing history is attempting to explain the uncertain. Playing history games is all about enjoying the uncertainty and in some way pretending to be part of it, in your mind and imagination, at least.
Now if this were a War College simulation, I would have to agree, bomb inventories and fuel levels are indeed where it's at. But Matrix is not selling to the US War College primarily, but to all the history and naval buffs out there. And most of them want to enjoy cutting loose their imagination a bit.
That's my two bits, anyway.
Now if this were a War College simulation, I would have to agree, bomb inventories and fuel levels are indeed where it's at. But Matrix is not selling to the US War College primarily, but to all the history and naval buffs out there. And most of them want to enjoy cutting loose their imagination a bit.
That's my two bits, anyway.
"Things are getting better!
...Well, maybe not as good as they were yesterday, but much better than they will be tomorrow!"
-Old Russian saying
...Well, maybe not as good as they were yesterday, but much better than they will be tomorrow!"
-Old Russian saying
RE: Pearl Harbor and AI
ORIGINAL: RevRick
ORIGINAL: AcePylut
Well consider it like this...
If the AI keeps the KB around an extra day or 3, just assume that was what the Japs had planned, and stocked their ships with enough ammo to bomb//torp an extra few days.
It's not too hard to dream up a believable fantasy to account for the extra attacks. Imagine it to be true.
This IS - after all - a fantasy "what if" game.
Okay, if that is what this is supposed to be - I want turbochargers on the P-39s, re-engining of the old BB's so they can steam at 27 kts., torpedoes that work for the USN (that's in!), and a Congress that had the foresight to plan the two ocean Navy in 1939 instead of 1940. I mean, if the IJN can dream up more tankers to refuel the tankers who already fueled the KB, and then more ammo on each ship which the IJN didn't have in the first place, and an Admiral with a pair, what's good for the goose is good for the gander. Oh, by the way, Grumman started work on the F6F a year early. It shows up in January of 1942, along with the TBF.
Say hello to the AE editor, and all your fantasies may come true!
All this is to say that having the KB stick around and obliterate PH just because the game allows it because of the mechanics is only a little bit shy of slightly gamey. I know the USN would not have folded up and quit, but if you want the USN to wait in Pearl and San Fran until Jan of 1944, and then come out hunting, so be it.
I won't stick the USN navy on the West Coast or PH if the KB obliterates PH. That might be your fantasy, but it ain't mine. [:)]
RE: Pearl Harbor and AI
I honestly just dont get it maybe its me but I dont we tried to make the AI play the game as hard as we could to make it interesting.
If thats not your cup of tea there is a scenario that starts 8th Dec which has a 100% historic approach or you can play the stock Ai which will faithfully recreate history and fall over in mid 42 at Rabaul when a player a historically reinforces
I guess it boils down to what do you want
1. Faithfull recreation of history where the Ai loses badly because you know its every move UNLESS you also follow history 100%
2. An AI that will try to do unpredictable things but while following the shape of history will try to do things I would do if I were playing PBEM against you.
If the majority honestly want 1. then I am not sure I haved provided what you want.
Andy
If thats not your cup of tea there is a scenario that starts 8th Dec which has a 100% historic approach or you can play the stock Ai which will faithfully recreate history and fall over in mid 42 at Rabaul when a player a historically reinforces
I guess it boils down to what do you want
1. Faithfull recreation of history where the Ai loses badly because you know its every move UNLESS you also follow history 100%
2. An AI that will try to do unpredictable things but while following the shape of history will try to do things I would do if I were playing PBEM against you.
If the majority honestly want 1. then I am not sure I haved provided what you want.
Andy
RE: Pearl Harbor and AI
That would be great Andy! Those landings are all way too lightly covered, and there are a bunch of English ships that can be there really early. Even the PT boats can make it to Davao to cover.
RE: Pearl Harbor and AI
Okay, ran turn 2 again and this time KB moved off to the west and was near French Frigate Shoals when it launched two attacks in the AM, a small attack against Lihue and single air attack against PH consisting of 57 Kates & 22 Zeros. Damage from this attack was minimal and no other attacks were launched the rest of the turn.ORIGINAL: afspret
I think I found a way to get rid of them, or at least reduce the serverity of follow up attacks. Before running turn 2 I sent all the subs I had in or around PH to the hex KB is supposed to be in and during the combat resolution they showed up several hexes west of their turn 1 location and only launched a small strike on PH (20+ ea Kates & Zeros).
Unfortunately I forgot about the Portland TF, which was apparently closer to KB than PH was. The end result was the TF was wiped out by numerous AM & PM air strikes. Now that I think about it, maybe thats why KB didn't launch a bigger follow up attack on PH, so maybe I'll run turn 2 again and this time move the Portland farther south and see what happens.
John E. McCallum
- Splinterhead
- Posts: 189
- Joined: Fri Aug 30, 2002 11:45 pm
- Location: Lenoir City, TN
RE: Pearl Harbor and AI
ORIGINAL: Andy Mac
I honestly just dont get it maybe its me but I dont we tried to make the AI play the game as hard as we could to make it interesting.
If thats not your cup of tea there is a scenario that starts 8th Dec which has a 100% historic approach or you can play the stock Ai which will faithfully recreate history and fall over in mid 42 at Rabaul when a player a historically reinforces
I guess it boils down to what do you want
1. Faithfull recreation of history where the Ai loses badly because you know its every move UNLESS you also follow history 100%
I don't understand why you think that's what we want. It's not exactly like your AI falls apart the 50% of the time KB doesn't hang around.
2. An AI that will try to do unpredictable things but while following the shape of history will try to do things I would do if I were playing PBEM against you.
One of my non negotiable rules for PBEM is no port attacks by Kates if KB stays a second day.
If the majority honestly want 1. then I am not sure I haved provided what you want.
I think you have a no win situation. A lot of people seem ecstatic with your current version, so unless you plan to do a "no linger" version of scen 1 in addition to the current one, I'd say just ignore us. We do have Dec 8th, after all.
RE: Pearl Harbor and AI
OMG - forget the complaining - I LOVE the surprises! I want as much of a challenge as possible. Andy, you made it sound like there are % chances. Does that mean you should get a significantly different game each time (to some degree at least)?
I had an early raid on PM by a surface fleet that just decimated my resupply ships - awesome! I feel like I have to stay on my toes with everything I do! Subs all over are making me pay dearly for unescorted ships.
Other than the Midway weak invasion we discussed, my only other observation is lack of support for the landings near Rabaul, New Guinea. I sent my carriers down there and they've been eating up transports for lunch every day. Actually a squadron of Dive Bombers at PM sinks about 2 a turn. I just expected them to pull back like I do if I start losing lots of ships. Of course, I'm not sure that is possible.
Of course I'm only up to Jan 30, '42 but I'm loving it so far. Thanks for all your hard work!
Mozo
I had an early raid on PM by a surface fleet that just decimated my resupply ships - awesome! I feel like I have to stay on my toes with everything I do! Subs all over are making me pay dearly for unescorted ships.
Other than the Midway weak invasion we discussed, my only other observation is lack of support for the landings near Rabaul, New Guinea. I sent my carriers down there and they've been eating up transports for lunch every day. Actually a squadron of Dive Bombers at PM sinks about 2 a turn. I just expected them to pull back like I do if I start losing lots of ships. Of course, I'm not sure that is possible.
Of course I'm only up to Jan 30, '42 but I'm loving it so far. Thanks for all your hard work!
Mozo
RE: Pearl Harbor and AI
ORIGINAL: Andy Mac
I honestly just dont get it maybe its me but I dont we tried to make the AI play the game as hard as we could to make it interesting.
If thats not your cup of tea there is a scenario that starts 8th Dec which has a 100% historic approach or you can play the stock Ai which will faithfully recreate history and fall over in mid 42 at Rabaul when a player a historically reinforces
I guess it boils down to what do you want
1. Faithfull recreation of history where the Ai loses badly because you know its every move UNLESS you also follow history 100%
2. An AI that will try to do unpredictable things but while following the shape of history will try to do things I would do if I were playing PBEM against you.
If the majority honestly want 1. then I am not sure I haved provided what you want.
Andy
You got it right as far as I'm concerned.
I bought WITP, and played through the full war a grand total of one time. By the time I captured Tokyo in mid 1944 I was so bored with the game that I've never played another round.
As far as I'm concerned, there is no way I should have been able to achieve such a decisive victory on my first time out. I thought the AI flopped around like a dying whale and would have been happier that s**t if it had done something that surprised me, damaged my forces or upset my plans from time to time. As it was, there was virtually no strategic or tactical challenge.
So, remember...
For everyone who is displeased with the changes to the AI you made, there are probably a half dozen of us who will be very satisfied.
RE: Pearl Harbor and AI
If somebody playing Japan in PBEM game, camped out and hit Pearl Harbor for a week, that person would be banned from PBEM games, no one would play them again! So is it too much to ask that the AI not do things that would be considered gamey in PBEM games, I don't think so.
I have no problems with AI doing things that are non-historic such as invading Midway or Dutch Harbor on Dec 7th.
I do wish the AI would cover invasion forces with more escorts and with carriers, it's too easy for a
couple of DD's to destroy an invasion force. Just put one or two allied DD's to guard the ports in Sarawak and northern Borneo and they will destroy most of the invasion forces.
I wonder if ships are just too fragile in AE?
I have seen it twice where a single DD sank 6 to 8 ships in one battle. No this wasn't FOW, all the ships were reported sunk. Allied subs now sinking a lot of ships with their guns because their torpedoes failed, just doesn't seem right.
I have no problems with AI doing things that are non-historic such as invading Midway or Dutch Harbor on Dec 7th.
I do wish the AI would cover invasion forces with more escorts and with carriers, it's too easy for a
couple of DD's to destroy an invasion force. Just put one or two allied DD's to guard the ports in Sarawak and northern Borneo and they will destroy most of the invasion forces.
I wonder if ships are just too fragile in AE?
I have seen it twice where a single DD sank 6 to 8 ships in one battle. No this wasn't FOW, all the ships were reported sunk. Allied subs now sinking a lot of ships with their guns because their torpedoes failed, just doesn't seem right.
-
- Posts: 6187
- Joined: Wed Jan 01, 2003 1:17 am
- Location: Kansas City, MO
RE: Pearl Harbor and AI
ORIGINAL: pad152
I have seen it twice where a single DD sank 6 to 8 ships in one battle. No this wasn't FOW, all the ships were reported sunk.
Under what circumstances? I'll bet they were quite favorable to the DD..., as in a night action (or limited visability) against a poorly escorted group of transports caught unloading. 30-knot DD with Naval fire control and gun crews at close range facing transports trying to raise steam and anchors. Sounds like a recipe for a one-sided slaughter to me...
RE: Pearl Harbor and AI
ORIGINAL: gunnergoz
Well, guys, we are either talking about playing a "historical simulation" or a "historical game." The simulation should turn things out pretty much as they did historically, IMO. The game could take history and tweak it a bit here and there, not based upon little green men, but based upon concepts that might have been within the grasp of the historical participants, had they some more foresight...since we clearly have the benefit of hindsight. I can live with calling this a well-researched historical game, that takes a liberty or two to better enable me to play it with more enjoyment, and for longer. Yes, a grog can argue that a shortage of such and such type of bomb fuzes could totally negate this and that type of scenario from ever happening, but who is to say what might have happened historically with just a few changed events and conditions? War is all about uncertainty. Writing history is attempting to explain the uncertain. Playing history games is all about enjoying the uncertainty and in some way pretending to be part of it, in your mind and imagination, at least.
Now if this were a War College simulation, I would have to agree, bomb inventories and fuel levels are indeed where it's at. But Matrix is not selling to the US War College primarily, but to all the history and naval buffs out there. And most of them want to enjoy cutting loose their imagination a bit.
That's my two bits, anyway.
Well said!
RE: Pearl Harbor and AI
All the fuel info is public domain.
Distance from the Hitokappu Bay to Pearl - under 4000 miles.
Max cruising range for the shortest legged Japanese CV (Soryu) - over 8000 miles (Obviously, battle speed will reduce this, but the other CVs could go a fair bit further).
But then consider that the any one of the 8 tankers KB had in support could have refuelled Soryu twice over from dry.
Japanese logistic considerations related to stormy North Pacific weather making refuelling hazardous/impossible, not the quantity of fuel the supporting oilers could supply.
Anyone who still disagrees needs to explain why all the First Air Fleet and Combined Fleet staff members who desired / recommended repeat attacks at the time had managed to somehow miss this incredibly obvious point.
Personally, I think repeated hits on Pearl are entirely reasonable and historically feasable, and make for a far better game
Distance from the Hitokappu Bay to Pearl - under 4000 miles.
Max cruising range for the shortest legged Japanese CV (Soryu) - over 8000 miles (Obviously, battle speed will reduce this, but the other CVs could go a fair bit further).
But then consider that the any one of the 8 tankers KB had in support could have refuelled Soryu twice over from dry.
Japanese logistic considerations related to stormy North Pacific weather making refuelling hazardous/impossible, not the quantity of fuel the supporting oilers could supply.
Anyone who still disagrees needs to explain why all the First Air Fleet and Combined Fleet staff members who desired / recommended repeat attacks at the time had managed to somehow miss this incredibly obvious point.
Personally, I think repeated hits on Pearl are entirely reasonable and historically feasable, and make for a far better game

- Splinterhead
- Posts: 189
- Joined: Fri Aug 30, 2002 11:45 pm
- Location: Lenoir City, TN
RE: Pearl Harbor and AI
ORIGINAL: Vladd
Anyone who still disagrees needs to explain why all the First Air Fleet and Combined Fleet staff members who desired / recommended repeat attacks at the time had managed to somehow miss this incredibly obvious point.
Weren't they in favor of a third wave on Dec 7? Do you have a source indicating that anybody urged a multi-day attack?
RE: Pearl Harbor and AI
If you change the AI, I'm going to write a letter saying I'm very very angry with you. Then I'll write another telling you I'm very, very angry with you.
If people want it to be more like the historical PH attack, is there not an editor? Edit to your hearts content. From my lurking and sparsely posting on the original WITP, people were highly disappointed with how hard the AI was - that it was no challenge. Now we got it! Don't mess it up!
If people want it to be more like the historical PH attack, is there not an editor? Edit to your hearts content. From my lurking and sparsely posting on the original WITP, people were highly disappointed with how hard the AI was - that it was no challenge. Now we got it! Don't mess it up!
KurtC in the WITE PBEM module.
RE: Pearl Harbor and AI
We can use a little imagination here. Let's assume Yamamoto planned to have KB stick around and Tomonaga came up with the idea to overload the carriers with extra ammo. Yamamoto told the gutless Nagumo to grow a pair and gave explicit orders to destroy the US carriers or don't bother to come back. On the way to Pearl the flight decks had the first wave bombs and torpedoes stored on the decks and extra ammo was stored in the hangars.
They expected to have the element of surprise and Japan was still at peace with the US so the risk of overloading with ammo considered minimal.
On the way to Pearl they were refueled a few more times than was historically the case.
Just put that into the introduction and I'll be perfectly satisfied....
They expected to have the element of surprise and Japan was still at peace with the US so the risk of overloading with ammo considered minimal.
On the way to Pearl they were refueled a few more times than was historically the case.
Just put that into the introduction and I'll be perfectly satisfied....
RE: Pearl Harbor and AI
I'm with Hgilmer3,
Like ii've mentioned above, I love the new AE AI. From reading most of the posts in this thread, I think most would agree.
Andy, good job on the AI. Keep it mean and aggressive.
Like ii've mentioned above, I love the new AE AI. From reading most of the posts in this thread, I think most would agree.
Andy, good job on the AI. Keep it mean and aggressive.
RE: Pearl Harbor and AI
I haven't seen that much, but i have seen 3 DD's sink 5 to 6 transports. That was a night action. My groups that were patrolling Northern Borneo were in groups of 3 to 5, with force Z being a group of 15. I killed around 100 transports between there and the PI. Not all the groups are destroyers. I also have the Prince of Wales, the CVL Hermes, several corvettes, A few light cruisers, PT boats and HDML's. The British and Dutch forces that can be gathered for the defense of Borneo and the PI are quite formidable.
RE: Pearl Harbor and AI
Well, if it's easy enough, why not just put out one or two scenarios that tone down the agressiveness a tad. The current "historical scenario" can then be considered "tenacious." A new one could be with a less aggressive IJN approach with Pearl and maybe a few other areas.
I think the game is MUCH more exciting, and like the challenge, but perhaps a more "historically-minded" AI would be a welcome option.
Call it the "slightly less agressive AI."
BUT, DO NOT change the scenarios that we have, except to fix errors that have been found. I like them. The surprises are many, and make me wish I had done this or that. But, that said, I would also like to play occasionally against a script that was not quite so ...... "innovative" in its approach.
This is NOT a criticism of the game. I love this dang thing already and I'm just now getting into January '42 (well, after a couple of restarts resulting from gross stupidity). It's an absolutely magnificent game. No doubt the best I have ever seen. It's worth twice the price.
Not bad for a bunch of amatures. [:D]
I think the game is MUCH more exciting, and like the challenge, but perhaps a more "historically-minded" AI would be a welcome option.
Call it the "slightly less agressive AI."
BUT, DO NOT change the scenarios that we have, except to fix errors that have been found. I like them. The surprises are many, and make me wish I had done this or that. But, that said, I would also like to play occasionally against a script that was not quite so ...... "innovative" in its approach.
This is NOT a criticism of the game. I love this dang thing already and I'm just now getting into January '42 (well, after a couple of restarts resulting from gross stupidity). It's an absolutely magnificent game. No doubt the best I have ever seen. It's worth twice the price.
Not bad for a bunch of amatures. [:D]
Occasionally, and randomly, problems and solutions collide. The probability of these collisions is inversely related to the number of committees working on the solutions. -- Me.
RE: Pearl Harbor and AI
Run 6 turns so far and now KB is south of Midway moving west, so I guess chasing away with multiple subs helped keep the number of follow up attacks on PH after turn 1 to a total of two, which were small and did very little damage. Of course now Midway's getting hammered!
John E. McCallum
-
- Posts: 1156
- Joined: Wed Apr 27, 2005 5:10 am
- Location: Los Osos, CA
RE: Pearl Harbor and AI
Perhaps BigJ62 can rewire the editor so that the number of torpedos on a carrier at the start of a scenario can be modified accordingly.
ChadG
ChadG
"If you want peace, prepare for war."