A new WiTP Ae II Version

[DD696]

I reckon it is time that I fess up.

Many years ago I was inclined to put signicicant cash for a super version of War in the Pacific. I deliberated over this for quite a while, but at the time I decided not to do so. The reason was that at this time the AE version was being developed, and I was thoroughly pissed off off that AI players were being ignored by the Powers That Were at this time. I was willing to invest hundreds of thousand of dollars in order to get a GREAT version of War in the Pacific that would be simply glorious to all AI players. But the emphasis was on PBEM players, and I was so griped at PbEM players consistently belittling AI players that I became very aggressive to those that consistently belittled those of us that preferred to play against the AI. I consistently railed that if the game did not have an adequate AI the game would fail in sales.

Well, the game had become some kind of a hit for Matrix and Henderson Field Designs, and they did do a super
job at constructing a viable AI for those of us who have not temperament nor the desire to play against another player who will disappear with the the next sunset. I prefer to play against an opponent who will be here with the next sunrise.

But, as is well known, some dumb Marine Staff Sergent is not going to change what is, or what was. Besides, there were some individuals on the team that were simply better off not being around in the forum. They pissed me off to the extent that they are no longer around. Blame me, or praise me, but the forum is a much better place for it.

[Canoerebel]

[&:]

[DD696]

Agreed. It was a shame.

[Timotheus]

Am an AI player because
1) I am not good enough (yet)
2) This game takes a psychological toll (as Allies when in beginning it is hammer blow after hammer blow, and as Japan 1943+ when it is not Japan that takes hammerblow after hammerblow).
3) This game takes years - up to a decade of real life! - to play.

Extra comment on WITP2:
The UI is atrocious, and needs to die in hellfire and WITPAE needs to have modern, user friendly AI. My fingers bleed while playing this game.

The AI scripts were VERY good for their era, but now they are not. AI CAN and SHOULD be much improved for WITP2, but it CANNOT be done in WITPAE due to game engine limitations.

This is a niche game, and the first thing I will do when I win the lottery is get a team of history nerds and coders and let them loose: WITP2 coming soon! [8D]

[witpqs]

ORIGINAL: Canoerebel

[&:]

I'm with you: [&:]

[PaxMondo]

ORIGINAL: witpqs

ORIGINAL: Canoerebel

[&:]

I'm with you: [&:]

+2

[InfiniteMonkey]

ORIGINAL: Timotheus
The UI is atrocious, and needs to die in hellfire and WITPAE needs to have modern, user friendly AI. My fingers bleed while playing this game.

There are limits to what a "modern UI" can do. While I agree that the interface can be improved, allowing the level of control that exists in WitP:AE means lots of actions by the user to specify them. Things could be done to decrease the number of clicks somewhat, but with control comes clicks.

ORIGINAL: Timotheus
The AI scripts were VERY good for their era, but now they are not. AI CAN and SHOULD be much improved for WITP2, but it CANNOT be done in WITPAE due to game engine limitations.

Not sure I agree... In some ways, the maturity of the game engine makes the AI easier to improve, not harder. The problem is that no one with access to the code is actively working on the game right now.

[spence]

I've played the Allied Player many times because "the blush came off the (production) rose" long ago in the board-games of old. So now I endure a game over and over again for the first 300 moves or so while the IJ Player has fun only to find no opponent at all once I start to get F6Fs, P-47s, etc to play with cause the IJ Player gets discouraged by the endless pounding he is supposedly going to take and disappears.

The biggest problem with the game is there is no mid-1942 to mid 1943 scenario where, with appropriate victory conditions, morale/experience and capabilities of the Allies, the issue is in the balance. Right now the entire system seems to hinge on the idea that the Battle of Midway was a fluke Allied/American victory but that viewpoint entirely seems to hinge on the entirely de-bunked account of the battle of LCDR Fujita (who certainly had a personal ax to grind concerning his (minor) role in the IJN's defeat. The results of The Battle of Midway were by no means guaranteed for either side but the Americans did get 175+/- strike planes on their way to bombing/torpedoing the IJN before the IJN got even a single strike plane airborne in response so it would seem programming superpowers for the IJN is overdoing a bit.

Some kind of "Coral Sea-Guadalcanal Era" game with the entire map providing strategic options for the IJ Player would seem to offer the a great place to start. That era was what the USN called the Defensive-Offensive Stage.

[janh]

Apart from what may be written above, AE has been around for so long and found a large, mature and stable community that this alone may warrant creating a successor.

If there was focus also placed on AI and creating a truly functional, logical scripting language (see e.g. ARMA3 -- allows you to modify pretty much everything) that would be really good news. With such a scripting language, the player community could chime in to systematically improve the AI over time to levels a limited group of developers with limited resources usually cannot. For a complex game like AE with outstanding AI and AI modding capabilities I would readily pay $200. Anyhow, "would be" I guess.

ORIGINAL: Timotheus
The AI scripts were VERY good for their era, but now they are not. AI CAN and SHOULD be much improved for WITP2, but it CANNOT be done in WITPAE due to game engine limitations.

Regarding updrading AE AI, better to ask the developers. I surely does not come down to being possible or not possible at all, but probably to "possible at what cost/effort?". Would be nice if they one day published the source of the game -- I'd really like to take a look and see...

[adarbrauner]

ORIGINAL: Timotheus

This is a niche game, and the first thing I will do when I win the lottery is get a team of history nerds and coders and let them loose: WITP2 coming soon!

+5

[Lowpe]

I read the OP two or three times and I think there must be drink involved or a 2nd language.

I am clueless about what is trying to be conveyed.[&:]

[adarbrauner]

I understand he was at any extent involved in the development, but got frustrated by some design's choices.

[PaxMondo]

ORIGINAL: Lowpe

I read the OP two or three times and I think there must be drink involved or a 2nd language.

I am clueless about what is trying to be conveyed.[&:]

+1

[Alfred]

ORIGINAL: adarbrauner

I understand he was at any extent involved in the development, but got frustrated by some design's choices.

No, the OP was not involved in the creation of AE.

Alfred

[Alfred]

ORIGINAL: InfiniteMonkey

ORIGINAL: Timotheus
The UI is atrocious, and needs to die in hellfire and WITPAE needs to have modern, user friendly AI. My fingers bleed while playing this game.

There are limits to what a "modern UI" can do. While I agree that the interface can be improved, allowing the level of control that exists in WitP:AE means lots of actions by the user to specify them. Things could be done to decrease the number of clicks somewhat, but with control comes clicks.


In an ideal world, an improved interface would exist. That however would require as a minimum a complete graphics rewrite. The database structure would probably need to be rewritten too. In theory not impossible but definitely a major call on any development resources.

As to the number of clicks, the comment that "with control comes clicks" is absolutely spot on. There are many existing labour saving opportunities which complaining players stubbornly refuse to use. To mention just a few, both the Auto Convoy and the CS control, if used correctly, can reduce considerably the amount of clicking. Instead of 1 day turns, multi day turns will cut clicks by at least 50%. The quid pro quo is less player control. Given the choice very few players elect that exchange.

ORIGINAL: Timotheus
The AI scripts were VERY good for their era, but now they are not. AI CAN and SHOULD be much improved for WITP2, but it CANNOT be done in WITPAE due to game engine limitations.

Not sure I agree... In some ways, the maturity of the game engine makes the AI easier to improve, not harder. The problem is that no one with access to the code is actively working on the game right now.


There is no need to have access to the code in order to "improve" the "AI".

When most players refer to the AI they are conflating two different things. The true AI is embedded in the code and yes for that to be altered, access to the code is required. However this is not really a problem for the AI is actually quite good. The AI does a good job of following the "rules". For example, it sends out air units at acceptable altitudes, maintains the overall logistical system, prosecutes the scripts competently (and certainly more competently than I have seen many a human conduct their operations), task forces move smoothly etc.

It is the second element, the scripts, which overwhelmingly is what gets excoriated. The scripts are not AI. Anyone can write new scripts. In fact, upon the release of AE, the devs expressed the hope that new scripts would be written. Andy Mac who wrote the scripts, extensively posted on the technical details involved in writing new scripts. And yet, after all this time and all the forum whinging, not one single script has been produced and presented by a non dev.

Alfred

[Alfred]

ORIGINAL: janh

... Would be nice if they one day published the source of the game -- I'd really like to take a look and see...

It has been stated many times in the past that the code will never be published.

Putting aside the non insubstantial issue of protecting intellectual property on software which is still generating income for both the AE and classical WITP devs, releasing the code would destroy the integrity of PBEM.

Alfred

[Alfred]

ORIGINAL: PaxMondo

ORIGINAL: Lowpe

I read the OP two or three times and I think there must be drink involved or a 2nd language.

I am clueless about what is trying to be conveyed.[&:]

+1

The OP believes that he was a major factor in the departure from the forum of JWE/Symon.

I don't know what religion he holds but a Roman Catholic who felt the need to "confess" might find the standard confessional box somewhat inadequate as there would be no understanding on the other side of the confessed "sin".

Alfred

[geofflambert]

What glory is there to defeating an AI? Any AI. You can ask Kasparov about Deep Blue, but WitP-AE was never going to have Deep Blue. AI is for practice. Glory comes from victory against someone who has a record. I have a record. It isn't superlative. I have always won in short order as the Allied player but I have not yet won as the J player. Working on it.

[janh]

ORIGINAL: Alfred
There is no need to have access to the code in order to "improve" the "AI".

When most players refer to the AI they are conflating two different things. The true AI is embedded in the code and yes for that to be altered, access to the code is required. However this is not really a problem for the AI is actually quite good. The AI does a good job of following the "rules". For example, it sends out air units at acceptable altitudes, maintains the overall logistical system, prosecutes the scripts competently (and certainly more competently than I have seen many a human conduct their operations), task forces move smoothly etc....

It is the second element, the scripts, which overwhelmingly is what gets excoriated. ...

This is were I respectfully disagree, Alfred. You do know the game quite well, so you should also actually be aware that the AI often has suicudal tendencies -- just to pick one example where improvement might be conceivable. For e.g. sending air raids time after time at targets where previous raids have just gotten mauled the day before, or already for weeks day by day. Why? Because the depolyment of air groups is scripted, and the daily orders are generated by the hardcoded parts that we cannot access or modify in this game. But that would need to be extended on to address the suicidal tendencies. Hence, it is quite clear that improving some of the points in AE are issues that can -- in part -- only be addressed systematically at the source level. This at large applies to tactical issues.

In part, because the deployment of AG is scripted, so one could of course also imagine an AI that dynamically beefs up deployments to counter the threat. Yet, the current scripting langauage is static and does simply cannot alter its own assets to react to something that was simply not imagined by the creator of the script at the time of the scripting. This applies at the strategic level mainly. Here an in-depth scripting language with all sorts of triggers/checking functions, logical constructs, game-specific functions etc. could be of much use. However, I am well aware that this would have been and probably would still be well too costly for development of a game with a comparably limited customer base, espcially since this is largely evolved into a PvP game and not primarily or solely developed for a single-player audience.

Anyhow, Alfred, you are of course right - chances of seeing the source code are negligible. None the yet, I would be extremly curious. I have followed what happened when the Falcon 4 code got spread - and what community teams developed that game into. Just awesome what some dedicated people may achieve.

[Aurorus]

ORIGINAL: spence

I've played the Allied Player many times because "the blush came off the (production) rose" long ago in the board-games of old. So now I endure a game over and over again for the first 300 moves or so while the IJ Player has fun only to find no opponent at all once I start to get F6Fs, P-47s, etc to play with cause the IJ Player gets discouraged by the endless pounding he is supposedly going to take and disappears.

I think the problem is not so much the PBEM system as a lack of good Japanese players. The solution to this problem is the forums in which players with more experience can teach new players some best practices for playing Japan. Of course, none of this helps the player who insists on "leaping before looking" and jumping into a grand campaign as Japan before learning some best practices for Japanese play.

At any rate, I learned a lot from reading AARs, so I try to help by putting some helpful advice for Japanese players in my AARs.