New AI oriented RHS scenario 76 uploaded

[el cid again]

This is a variation of RHSEOS intended for use by players who want to have AI control Japan.

It is very similar to RHSEOS except

there are no interior river systems (INTERIOR)

the Russians are passive

there are few Japanese political points (AI won't use em) and units are often assigned to forward commands

some features - notably the passive Russians - will permit it to do AI as the Allies - as well as AI can (which is never good for the Allies)

This is version 7.7594. IF there are no art pointer issues for ship art - which we won't know until ship art reviews are completed - it also will be 7.760

[bradfordkay]

Cid, did you reset the ports to stock base size? Mike Wood indicated that this was a major mistake in the non-stock maps...at least for those ports which are large enough for various ship types to replenish.

[el cid again]

ORIGINAL: bradfordkay

Cid, did you reset the ports to stock base size? Mike Wood indicated that this was a major mistake in the non-stock maps...at least for those ports which are large enough for various ship types to replenish.

I have not been so advised by Mike or Andrew - and I am not inclined to reset them in any case. We have attempted to understand WITP design concepts and to set the ports to the sizes they should be set to. But it is quite true that port size matters - and also that the sum of port and airfield size matters. In this respect, interior river system ports removed for RHSAIO will be less important than they were - as magnets for local supplies. I don't see a way to address that either: the game works better with the river ports as ports - but only humans can manage interior river systems - AI cannot.

If you can point to what Mike said - or if you can describe his concern - I may be able to comment specifically on what impact (or not) this may have on RHS. In general, I seek to get the data right - and Matrix can fix code issues in due course. There is no practical way to reset port sizes to those of stock - we have about 100% more ports - and many of the original settings were incorrect - even given the semi-abstract sense in which port size has meaning. I don't particularly care if some port sizes have problems re replenishment, repair, loading mines, name it: these more or less are "spiritually" true in some abstract sense (there are always limits and special capabilities associated with real places). But some specific issue at some specific point might be something we could address - until Matrix addresses the matter - if it does - in code.


Bottom line: the only port changes were due to errors (I don't remember any) or the removal of interior river system port ratings in RHSAIO.

[bradfordkay]

Here's the comment.

tm.asp?m=1354360&mpage=1&key=&#1354849

"...A number of bases had airfields and or ports reduced, which confuses computer opponent, who expecting to use a base for staging or headquarters suddenly finds the ships it sends there for storage cannot disband or the airfield is too small to use for domination of the area..."


It appears that the AI is expecting to use particular bases for replenishment or staging areas. While it might be nice to see hard code changes in the way the AI handles the new maps, I don't think that we can reasonably expect Matrix to reprogram the game to match our mods. Thus we have to mod the game with certain restrictions in mind, especially if we are trying to make a mod for AI play.

As far as no practical way to reset base sizes to stock, I think that you are overstating the situation here as much as you are understating the likelihood of code changes by Matrix to match the mods...

[el cid again]

Well - I was commenting in general. The problem as presented now is different: all the new ports are not going to be an issue at all: stock hardly expects to use ports that do not exist. Further - many changes were not to decrease the port - see Manila - and increasing it is not going to be an issue here. Further - AI tries to build up ports automatically - so many of these issues will be self correcting. I don't think this is a big problem - and if some place has a problem that matters - it can be addressed. But IRL the ports in the vast ocean area - and the airfields even more so - were not well developed when the war began. Also, in the game construction is too efficient and too fast. Finally - we cannot be responsible for hard code we cannot read. AI should not be sending anything anywhere - and to the extent it does that - we need to know specifics to be able to deal with it. I have figured out how to "program" the AI to send things where it should do that. It requires testing to see how well this is working? And I bet few cases involve ports being reduced too critically. RHS in general increased major port sizes - not decreased them. Mainly it was airfields we reduced.

EDIT: I checked the comment. It is not specific about what locations were a problem, or what sizes were required? Further - it was a bit out of context insofar as it ignored that CHS was done for human vice AI play. But I do not think this is a big problem. After making things hard to take with supply sinks (in RHS), so locations seemed to take forever to fall, continual tweeking has resulted in faster conquest of Malaya than history (if not quite as fast as stock). CHS never had the problem of things took too long to conquer. If we find some specific port problem - we can also change it.

[bradfordkay]

Cid, my comment was based on the idea that this new release is designed for AI play. In that case, if I were the one creating it, I would look at which bases in stock had a port size of 3 or larger and return any changed by Andrew's map to the stock size (unless subsequent revisions made them larger).

I am not as concerned about airbase size as is Mike Wood, as I feel that the level bombers are too omnipresent in the game anyway.

Just my thoughts...

[el cid again]

In a very real sense, bombers are what matters. The WITP was fought over bomber bases - so dominant were they.
Carriers can sustain air power near major land bases only if the carriers are present in numbers (see late war US operations). I agree with Mike Wood it was wrong to reduce bomber bomb loads - and RHS has increased them - but not to max load in most cases: to the load appropriate for the ranges the code allows. We have found even values higher than stock do not hurt too badly - because we changed the effect settings for bombs (and shells) - soft effect is either square root of weight - or square root of 2/3 of weight - which is much more correct than a one for one effect value. Neither battleship shells nor bomber bombs have the "nuclear bombardment" effects of previous versions of WITP.

As for port size, all I care about is relative port size. Is it right for the relative capacity of the port IRL? Both the actual and potential size need to be thought about. But if stock was wrong - I think it should be changed. If the AI was originally set to use the wrong port - I can set it to use the right one - or so it appears. Time to test and find out if that is the case?

[Andrew Brown]

ORIGINAL: bradfordkay

Cid, my comment was based on the idea that this new release is designed for AI play. In that case, if I were the one creating it, I would look at which bases in stock had a port size of 3 or larger and return any changed by Andrew's map to the stock size (unless subsequent revisions made them larger).

I am not as concerned about airbase size as is Mike Wood, as I feel that the level bombers are too omnipresent in the game anyway.

Just my thoughts...

I think that if anyone is planning on having their scenario playable with an Allied AI, then doing this is a very prudent idea, given Mike's comments. When I asked him about it I was informed that it affected the Allies mainly, so for Japanese AI it is presumably not so important.

As mentioned CHS was designed mainly for PBEM play, so I decided not to reverse any of the changes made in that scenario, but for AI play it would be a good idea to ensure no stock ports of size 3 or higher are set to a value lower than 3.

Andrew

[m10bob]

IMHO if their is concern over the number of bombers, the issue should be directed at the number of bombers in-game, not the size of the port nor field.
It is historic fact bomber runways could be (and were) thrown up inside of a week, under the cruelest, most harsh natural conditions,(malarial insects, dysentary, coral, jungle,etc.).
The real fields were thrown up to "rating 4" in less than a week.
How long does it take in game?
As Eric Bergerud said, this was a war of airfields.
The ports supplied them.
I appreciate what CHS and RHS have done for the ports.

[bradfordkay]

Trust me, guys, I appreciate what Andrew has done with the map. I found WITP to be unbearable (what with the Australian RR racetrack) until he released his map.

However, I began to notice some problems with the AI in my games vs the AI. After about 8-9 months, the KB no longer was a factor. It would sail into attacking distance and launch a 4 plane assault on one of my bases. It continued to do this for months. My concern is that it is sending the KB back to the wrong base for replenishment. When Mike Wood posted that comment last January, it struck home with what I was seeing.

I have since switched to PBEM, CHS scen 159, and have been enjoying WITP even more than ever.

Cid, if you can reprogram the AI to return to specific bases for replenishment, then I am all in favour of that - with the bases at whatever size you/Andrew/concensus believes that they should be.

As far as the war being about airbases, well... duh! My comment was mainly concerned with the fact that the Betty/Nell nests will completely close down all approaches to Singapore/Palembang/Port Moresby/etc long before it occurred IRL. My feeling is that, in the absence of tracking torpedoes for all units in the game, there are too many early war level 4 airbases. That is all...

[Historiker]

Is it a known problem, that the "channels" to aden, arount africa etc. don't work? Or have forgotten something?

[witpqs]

They work fine for human players. It was stated in the scenario instructions (on the forum) that the AI cannot be made to work with them. Is that what you mean?

[Historiker]

My ships simply drive through the "borders"!

[drw61]

ORIGINAL: Historiker

My ships simply drive through the "borders"!

This sounds like a pwhex problem to me , make sure you have the latest pwhex version for the mod you are playing.
If you are using RHS the pwhex can be downloaded at

http://rhs.akdreemer.com/RHS%20Scenario ... at%20file/

[el cid again]

ORIGINAL: m10bob

IMHO if their is concern over the number of bombers, the issue should be directed at the number of bombers in-game, not the size of the port nor field.
It is historic fact bomber runways could be (and were) thrown up inside of a week, under the cruelest, most harsh natural conditions,(malarial insects, dysentary, coral, jungle,etc.).
The real fields were thrown up to "rating 4" in less than a week.
How long does it take in game?
As Eric Bergerud said, this was a war of airfields.
The ports supplied them.
I appreciate what CHS and RHS have done for the ports.

Here the game has apparently compromised - but it makes a good approximation. The Japanese needed much longer to do an airfield than the US did - and the US had to learn how to do what it did - the organization was experimental for a while and NOT what had been planned for. It should take longer for the Japanese than it does, and less time for the Allies. Ports build too fast for both IMHO. But we cannot do an improvised harbor - so it may be a reasonable compromise.

[el cid again]

ORIGINAL: bradfordkay

Trust me, guys, I appreciate what Andrew has done with the map. I found WITP to be unbearable (what with the Australian RR racetrack) until he released his map.

However, I began to notice some problems with the AI in my games vs the AI. After about 8-9 months, the KB no longer was a factor. It would sail into attacking distance and launch a 4 plane assault on one of my bases. It continued to do this for months. My concern is that it is sending the KB back to the wrong base for replenishment. When Mike Wood posted that comment last January, it struck home with what I was seeing.

I have since switched to PBEM, CHS scen 159, and have been enjoying WITP even more than ever.

Cid, if you can reprogram the AI to return to specific bases for replenishment, then I am all in favour of that - with the bases at whatever size you/Andrew/concensus believes that they should be.

As far as the war being about airbases, well... duh! My comment was mainly concerned with the fact that the Betty/Nell nests will completely close down all approaches to Singapore/Palembang/Port Moresby/etc long before it occurred IRL. My feeling is that, in the absence of tracking torpedoes for all units in the game, there are too many early war level 4 airbases. That is all...

Well - we did decrease some air bases that were too developed. Also, in RHS many planes do not carry torpedoes in fact - never mind they are torpedo bombers. The only way to know that is the unit name - if it is coded K (ASW) or AP or something other than T - it does NOT load torpedoes - even attacking ships - I think (unless hard code forces it to - which is not 100% clear). But AIO is designed to make Japan stronger to compensate for weak AI - so it is EOS and not CVO. Don't expect everything to be the same as IRL. The things Japan could have done and should have done sometimes are done here. There is NO WAY to approach Singapore safely UNLESS you provide fighter cover - as was really the case IRL IMHO - some writers lament how slow the Brits were to learn the lesson.

[el cid again]

ORIGINAL: Historiker

Is it a known problem, that the "channels" to aden, arount africa etc. don't work? Or have forgotten something?

I do not know of any such problem - and I use them all the time. If you have something specific in mind I could look at it.

[el cid again]

ORIGINAL: Historiker

My ships simply drive through the "borders"!

Sounds like you may have the wrong combination of files. This should not be possible if you have the right ones.

[m10bob]

ORIGINAL: bradfordkay

Trust me, guys, I appreciate what Andrew has done with the map. I found WITP to be unbearable (what with the Australian RR racetrack) until he released his map.

However, I began to notice some problems with the AI in my games vs the AI. After about 8-9 months, the KB no longer was a factor. It would sail into attacking distance and launch a 4 plane assault on one of my bases. It continued to do this for months. My concern is that it is sending the KB back to the wrong base for replenishment. When Mike Wood posted that comment last January, it struck home with what I was seeing.

I have since switched to PBEM, CHS scen 159, and have been enjoying WITP even more than ever.

Cid, if you can reprogram the AI to return to specific bases for replenishment, then I am all in favour of that - with the bases at whatever size you/Andrew/concensus believes that they should be.

As far as the war being about airbases, well... duh! My comment was mainly concerned with the fact that the Betty/Nell nests will completely close down all approaches to Singapore/Palembang/Port Moresby/etc long before it occurred IRL. My feeling is that, in the absence of tracking torpedoes for all units in the game, there are too many early war level 4 airbases. That is all...

I play RHSCVO 6.xxx and my observations have been that the Japanese AI KB uses Truk, Kwajalien for replenishment, mostly, but they have also been over there in the Gulf of Tonkin.

My comment ref the number of bombers was only meant as response to your comment of the number of bombers.
My comments which followed regarding the size of bomber bases was only meant to underline the historic reason those "size 4" bases/runways were built.

Of the Allied 4-engine bombers, I'm glad we are no longer deluged with fleets of LB 87's, (of which all but a few were used as transports IRL).

[el cid again]

Even when they were part of bomber formations - the LBs - about 80 odd total (taken back from the Brits who took over orders for the French - where the designation came from) - were used mainly as recon ships - not bombers. A typical USAAF unit operating them had only one or two on strength at any given time. Even those planes ended up as hack transports in due course. The LB as transport was an ex B-24 - and so was the C-87 - both having the same payload of 8800 pounds (and the unique thing of armor on a transport) - so we give you both as a combined type. While I don't think they were used that way operationally, they ought to be superb for air assault operations. What they did do was significantly transport fuel "over the hump" from India to China.