Home Trending Latest threads New posts Developer diaries

Download CORE 0.62.8

Jump to latest Follow Reply
  • We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Hagar,

I've done a bunch of events re: license production of armoured vehicles. Here is an example of the events:

# 780200 Swiss License Production: 1938 Light Tank dec152000 testing
# 780201 Swiss License Production Agreement: 1938 Light Tank1 dec152000 testing
# 780202 Swiss License Production: 1942 Light Tank dec152000 testing
# 780203 Swiss License Production Agreement: 1942 Light Tank dec152000 testing
# 780204 Swiss License Production: 1945 Light Tank dec152000 testing
# 780205 Swiss License Production Agreement: 1945 Light Tank dec152000 testing
# 780206 Swiss License Production: 1939 Medium Tank dec152000 testing
# 780207 Swiss License Production Agreement: 1939 Medium Tank dec152000 testing
# 780208 Swiss License Production: 1941 Medium Tank dec152000 testing
# 780209 Swiss License Production Agreement: 1941 Medium Tank dec152000 testing
# 780210 Swiss License Production: 1943 Medium Tank dec152000 testing
# 780211 Swiss License Production Agreement: 1943 Medium Tank dec152000 testing
# 780212 Swiss License Production: 1945 Medium Tank dec152000 testing
# 780213 Swiss License Production Agreement: 1945 Medium Tank dec152000 testing
# 780214 Swiss License Production: 1947 Main Battle Tank dec152000 testing
# 780215 Swiss License Production Agreement: 1947 Main Battle Tank dec152000 testing
# 780216 Swiss License Production: 1949 Main Battle Tank dec152000 testing
# 780217 Swiss License Production Agreement: 1949 Main Battle Tank dec152000 testing
# 780218 Swiss License Production: 1938 Infantry Tank dec152000 testing
# 780219 Swiss License Production Agreement: 1938 Infantry Tank dec152000 testing
# 780220 Swiss License Production: 1941 Infantry Tank dec152000 testing
# 780221 Swiss License Production Agreement: 1941 Infantry Tank dec152000 testing
# 780222 Swiss License Production: 1944 Infantry Tank dec152000 testing
# 780223 Swiss License Production Agreement: 1944 Infantry Tank dec152000 testing
# 780224 Swiss License Production: 1947 Infantry Tank dec152000 testing
# 780225 Swiss License Production Agreement: 1947 Infantry Tank dec152000 testing
# 780226 Swiss License Production: 1940 Assault Gun dec152000 testing
# 780227 Swiss License Production Agreement: 1940 Assault Gun dec152000 testing
# 780228 Swiss License Production: 1942 Assault Gun dec152000 testing
# 780229 Swiss License Production Agreement: 1942 Assault Gun dec152000 testing
# 780230 Swiss License Production: 1944 Assault Gun dec152000 testing
# 780231 Swiss License Production Agreement: 1944 Assault Gun dec152000 testing
# 780232 Swiss License Production: 1946 Assault Gun dec152000 testing
# 780233 Swiss License Production Agreement: 1946 Assault Gun dec152000 testing
# 780234 Swiss License Production: 1948 Assault Gun dec152000 testing
# 780235 Swiss License Production Agreement: 1948 Assault Gun dec152000 testing
# 780236 Swiss License Production: 1942 Light Tank Based SPA dec152000 testing
# 780237 Swiss License Production Agreement: 1942 Light Tank Based SPA dec152000 testing
# 780238 Swiss License Production: 1944 Medium Tank Based SPA dec152000 testing
# 780239 Swiss License Production Agreement: 1944 Medium Tank Based SPA dec152000 testing
# 780240 Swiss License Production: 1947 Medium Tank Based SPA dec152000 testing
# 780241 Swiss License Production Agreement: 1947 Medium Tank Based SPA dec152000 testing
# 780242 Swiss License Production: 1950 MBT Based Based SPA dec152000 testing
# 780243 Swiss License Production Agreement: 1950 MBT Based Based SPA dec152000 testing
# 780244 Swiss License Production: 1940 Light Tank Based Conv. TD dec152000 testing
# 780245 Swiss License Production Agreement: 1940 Light Tank Based Conv. TD dec152000 testing
# 780246 Swiss License Production: 1942 Light Tank Based Conv. TD dec152000 testing
# 780247 Swiss License Production Agreement: 1942 Light Tank Based Conv. TD dec152000 testing
# 780248 Swiss License Production: 1943 Medium Tank Based TD dec152000 testing
# 780249 Swiss License Production Agreement: 1943 Medium Tank Based TD dec152000 testing
# 780250 Swiss License Production: 1945 Medium Tank Based TD dec152000 testing
# 780251 Swiss License Production Agreement: 1945 Medium Tank Based TD dec152000 testing
# 780252 Swiss License Production: 1947 MBT Based TD dec152000 testing
# 780253 Swiss License Production Agreement: 1947 Medium Tank Based TD dec152000 testing

Every country with an event could use an appropriate image. Of course some countries wouldn't really have solid choices based on the historical course and could just use some image. The following are the countries that have these events:

AST, AUS, BEL, CAN, HOL, HUN, ROM and SCH.

Also a few other events added that could use images:

AST: Activation of Tech Team Horace Robertson (517)

CAN: Activation of tech team Guy Simonds (919)
CAN: Activation of Tech team Davie Shipbuilding and Repairing (926)

CHC: Volunteers From Zhengzhou (replicated for a number of other provinces, an image of a peasant fighter would do nicely)

CHI: Activation of tech team Chen Cheng (1103)

FRA: Activation of tech Team Alphonse Juin (1425)

GER: Victory in Yugoslavia
GER: The Wartime Plan for Fuel Rationing
GER: Increased Fuel Rationing
GER: Expanded Fuel Rationing

IND: Independent India

ITA: The Wartime Plan for Fuel Rationing
ITA: Increased Fuel Rationing

JAP: The Wartime Plan for Fuel Rationing
JAP: Increased Fuel Rationing

SWE: Activation of tech team Flygtekniska försöksanstalten (3210)

SCH: The War Economy (material shortages)

good luck!

mm
 
Mmh... Thoughts on event 580042:

Replace "JAP" with "-2" in the trigger and remove the "secedeprovince" commands for cheap versatility.
This is one of/the area, which JAP "almost naturally" occupies in a timely manner, so we should probably leave it that.

I would leave out the flag until it is actually used somewhere. Flags getting set without effect is a bit of a problem in CORE (looking at you, promethean army).

--

Great to see proper oil plants in GER's autarky events. Note however, that the AI has only the most basic comprehension of stockpile: It will not convert enough to drain the input resource below IC-reducing limit, but it sure will convert as much as it can up to that point, completely oblivious to net change from other factors.
... We may end up with a slew of complications from that change.
- Still in favour though.

--

I was looking at the japanese AI files regarding the Australia issue and was once again reminded that they are in dire need of refactoring. Probably going to do that post-0.70.0. Apparently Tegetthoff started a few things in 0.62 which ended up unfinished and there is also a lot of old baggage...
Planning for that, I was wondering if the submarine bases historically erected on Wake and Midway would warrant Naval Base 1 buildings; mainly as JAP cannot naturally acquire the reach necessary to invade Hawaii. If that were acceptable, I could look for a good place to hack them into the USA AI files and also (should that AI decide it have a better use for its IC) add the same bases into JAP's rebuilt campaign file.
- The intended effect would be that eventually a strain of bases Marcus-Wake-Midway allowed for a landing on Hawaii, should a pure AI-on-AI campaign allow for such an outcome to be viable; historically, if I am not mistaken, this was the goal of the push stopped at Midway, as Yamamoto had recognised before the war that to win, the USA's ability to project its clearly superior power had to be denied.

--

@dec152000:
Please, code tags! with '[' and ']' instead of '<' and '>':
<code>Code</code>
<spoiler>File Name<code>Code</code></spoiler> for added niftiness.
<list><*>List Element</list>
<list><list><*>Indented List Element</list></list>
 
Last edited:
Hi,

580042: This event also has companion JAP event which is actually the main focus. The main objective is to give JAP better results from capturing the DEI. The design has the files set so that HOL has the DEI as Core and then the provinces were artificially degraded to account for colonial effects. But this completely screwed JAP as they then got non-core penalties applied on top of the existing penalties designed to degrade HOL. While I could set it up with a -2 instead of JAP this would just complicate the coding in the other event for very marginal increased utility. I'd add that this is also designed to speed things up for JAP once the main centers are captured. Right now am inclined to leave as is as I haven't even seen testing reults on this code yet.

Oil: So far the AI seems to run them at 1005 consistently. This is good, as I was worried it would foolishly throttle back. In the old design it was pretty much a given that they would always be 100% which is normally the desireable choice. OTOH, the plants were gaining efficiency from IE which wasn't strictly correct. If this works ok I'll expand the concept to the rares plants for SOV and USA as well.

Midway: AI USA already should build this one as things heat up via event driven production. Base building by the AI is so spotty it's really hard to do any other way with reliable results. Wake might be a good event driven one for AI JAP. Ideally AI JAP should really drive at Hawaii if it is doing well.

mm
 
dec152000 said:
Hi,

580042: This event also has companion JAP event which is actually the main focus. The main objective is to give JAP better results from capturing the DEI. The design has the files set so that HOL has the DEI as Core and then the provinces were artificially degraded to account for colonial effects. But this completely screwed JAP as they then got non-core penalties applied on top of the existing penalties designed to degrade HOL. While I could set it up with a -2 instead of JAP this would just complicate the coding in the other event for very marginal increased utility. I'd add that this is also designed to speed things up for JAP once the main centers are captured. Right now am inclined to leave as is as I haven't even seen testing reults on this code yet.
...
Click to expand...

Would it be possible to code a unique national idea for NL once it has surrendered in Europe to get the desired effects? That way JAP wouldn´t be affected when it conquers DEI.
 
Hi,

We definitely could do some NI coding to get the right effect. But that would cause other issues. At this point I think my "fix" is better than what we had in place for almost all games.

mm
 
Is there any possibility of getting hold of the fixes in this thread? the Czechoslovakia stuff, the French stuff, the Soviet Central planning stuff, the japanese invasion targets, etc to just copy-paste over an existing 0.62.8 install?
 
This, or wait a week or two until we post 0.70.1.
 
Some more changes to ministers_FRA.csv:

Changed availability to 1936, had been Minister of Finance in 1934.
2083;Minister of Armament;Paul Marchandeau;36;SL;Administrative Genius;Medium;M2083;x

Name change and availability to 42 when he joined the cabinet.
2279;Chief of Air Force;Jean-François Jannekeyn;42;PA;Carpet Bombing Doctrine;Medium;M57038;x

Delete this one, dead since 1931.
2106;Minister of Security;Arthur Fontaine;37;SL;Efficient Sociopath;Medium;M2106;x

Delete, no politician at all, retired general since 1933.
2098;Minister of Security;Pierre Auguste Chardigny;36;SD;Prince of Terror;Low;M2098;x
2160;Chief of Army;Pierre Auguste Chardigny;41;SL;Elastic Defence Doctrine;Medium;M2098;x
 
Kennelly said:
Delete this one, dead since 1931.
2106;Minister of Security;Arthur Fontaine;37;SL;Efficient Sociopath;Medium;M2106;x
Click to expand...

I for one would support having a zombie chief of security! :laugh:
 
dec152000 said:
Midway: AI USA already should build this one as things heat up via event driven production. Base building by the AI is so spotty it's really hard to do any other way with reliable results. Wake might be a good event driven one for AI JAP. Ideally AI JAP should really drive at Hawaii if it is doing well.
Click to expand...
There's a JAP AI event for garrisoning Wake after capture, maybe this event could also build a base there?

Jannekeyn is also in the VIC file as 57038, please amend him as well.
Also in the VIC minister file 57037 and 57065 are the same guy, one should be deleted. http://en.wikipedia.org/wiki/Jean-Marie_Charles_Abrial

Amended entries for FRA minister file:
2148;Chief of Staff;Paul Le Gentilhomme;40;SC;School of Psychology;Medium;M2148;x
2151;Chief of Staff;Pierre Armand Billotte;42;SC;School of Fire Support;Medium;M2151;x
2173;Chief of Navy;César Campinchi;37;ML;Indirect Approach Doctrine;Very High;M2173;x
2170;Chief of Navy;Joseph Paul-Boncour;38;SD;Base Control Doctrine;High;M2170;x
2184;Chief of Air Force;Pierre Cot;36;SD;Army Aviation Doctrine;Very High;M2184;x
2188;Chief of Air Force;Guy La Chambre;38;SL;Vertical Envelopment Doctrine;Medium;M2188;x

In FRA event 2103 almost all sleepminister commands are commented out, why?Anyway, a few more FRA ministers who joined the Vichy regime should be slept as well, so add
command = { type = sleepminister which = 2271 } #Abrial
command = { type = sleepminister which = 2176 } #Pietri
command = { type = sleepminister which = 2272 } #Decoux
command = { type = sleepminister which = 2183 } #Deat
command = { type = sleepminister which = 2279 } #Jannekeyn
command = { type = sleepminister which = 2289 } #Romatet
command = { type = sleepminister which = 5294 } #Deat
 
Last edited:
Hi,

I did all the fixes except for the Ministers getting slept for VIC. I don't know why these are commented out. At this point FRA is normally useless after VIC fires d2 game mechanics. As a result, it needs a full overhaul that I don't currently have time for. Maybe then we could fix the Minister issue correctly.

mm
 
Kennelly said:
There's a JAP AI event for garrisoning Wake after capture, maybe this event could also build a base there?
Click to expand...
I was going to try and reduce event-spawned objects, not add to them.
Would like to stick to the policy of not adding things artificially which the AI can build itself, where possible.

... To elaborate, I think that I can get rid of the transport-spawns for JAP, but not of the division-spawns for invasion of Indonesia et al. .
I could also, in theory, eliminate all the Garrison spawns, but the AI would then use arbitrary divisions to guard these places, including such horrible choices as LARM or HQ.
Eliminating the invasion divisions would certainly cause delays between the possibility to invade and actual invasions happening, and I have no cure for that*.

But just building some Naval Base lvl 1 buildings? I can do that, especially when both sides of the conflict want them there.**
Of course, if the islands in question keep switching sides, how could meaningful construction happen there!

* - To elaborate even more, whether JAP AI will provide troops for invasion depends on its max_front_ratios settings and therefore on the number of chinese divisions - even if those end up being GARs; a human CHI could thence block japanese expansion on the ocean fronts by simply spamming divisions with no regard for their combat value (and how CHI AI is set up, it does almost exactly that).
When those divisions, as they do now, spawn on invasion/staging_province entries, JAP AI will use them for that purpose, even though it cannot satisfy its max_front_ratios vs. CHI et al. .
Even if this problem did not exist, however, JAP would still have to move divisions towards their invasion/staging_province entries, for which it needs additional time vs. the current solution.

** - Actually I recall AI reluctance to re-add province improvements which were removed from losing control of the province in question. Never tested sufficiently to determine whether this be rational behaviour and therefore of limited time or a bug and permanent.
- Going to hope for the best first; if it does not work out, well, event-build it is. But at least I would have tried!

--

Of course, when JAP captures Wake the first time, that Naval Base should already exist: USA should start building it in January 1941.

Midway on the other hand should only get its base in 1943 (USA controlled) or as soon as JAP can make it, if they manage to take the province.
(Apparently the trigger for this was the Battle of Midway; having no means to detect that or something like that, I will probably choose the occupation of Wake as trigger, so if that happens earlier than historical, the USA will build sooner... assuming, again, that I can hack this into the USA AI without touching its files too much.)

--

Day-16 Shidan should apparently be named 'Kaki' and placed in Kyoto province on scenario start.

@Tegetthoff: How did you determine the composition of the japanese 14th army? According to Wikipedia it should be composed of 16th, 48th/4th, 56th division, but you assigned to it 16th and 18th division. As the 56th division in 1942 should be in Burma according to two other articles, I somehow doubt that the Wikipedia information is correct.

--

Reading about the area, I found that SIA should gain Kayah State in addition to Shan State from its alliance with JAP.
As the in-game province "Shan States" (1304, typo?) is notably smaller than the real Shan State, I wonder if this would warrant the addition of a core on Toungoo (1301), which is in turn bigger than Kayah State.
... SIA would certainly look more pleasing from it and the minor resource gain should not unbalance things.

On the other hand, I have wondered whether Shan States really should gain a core in the first place. An event to simply switch control of the provinces to SIA once taken by JAP may later provide a useful starting point for a hypothetical puppet Burma (I would refer to my write up on the japanese Sphere plans but that got lost with the old forum).
 
Last edited:
Zsar1 said:
@Tegetthoff: How did you determine the composition of the japanese 14th army? According to Wikipedia it should be composed of 16th, 48th/4th, 56th division, but you assigned to it 16th and 18th division. As the 56th division in 1942 should be in Burma according to two other articles, I somehow doubt that the Wikipedia information is correct.
Click to expand...


I use data from the Nafziger collection, which is usually very good and also quite often the source for Wikipedia articles.

http://www.cgsc.edu/carl/nafziger.asp

Please leave the Japanese garrison AI alone for the moment, it does what it should do.
 
Tegetthoff said:
I use data from the Nafziger collection, which is usually very good and also quite often the source for Wikipedia articles.
Click to expand...
Going to give that priority over Wiki then. Looks more trustworthy for sure.
... But there, too, the 18th division is not mentioned as part of the 14th army for the invasion of the Philippines. Have you boundled the "Other Troops" into that unit to have them represented?
Tegetthoff said:
Please leave the Japanese garrison AI alone for the moment, it does what it should do.
Click to expand...
Mmh, you have a stake in that? When you noted that you would rather take a break from CORE development I thought it fine to claim that part, as I deem it more important than my other focus ITA - but if being back means that you plan to work on JAP AI for 0.70.+, then I will of course reshuffle. You have been the last one to work on it, after all.
 
Last edited:
Hi,

I will add that at times too many chefs can be an issue. That said I've just done some stuff to ITA AI files trying to keep more troops in ITA proper. But once 0.70 starts out I'd love to have someone work the ITA scripting. And the JAP scripting. And the CHI/CHC scripting.

Naval Bases: Part of the issue here is game mechanics. It should not take much of a base to support a few submarines or destroyers. That is what a tender is for. And a full on expedient base can go up very fast if you have the right mobile support. See what the USN did in Ulithi. So some event driven Bases are ok if they are needed to replicate RW things. especially for the AI, which does not stockpile bases like a Human can.

Air Bases: similiar issue. Doesn't take much to fly off single engine aircraft. Especially those designed with grass in mind. Bigger/modern planes not so much. Mechanics just don't support this well. So whatever is needed to get a realistic combat model is fine with me.

mm
 
I do not mind this kind of event, I share your thought. I do mind it being AI-only, though.

Also mind the staggering number of transports given for free to JAP AI and JAP AI being balanced around not needing to produce them. A player is at a distinct disadvantage here and for no clear reason: Having these transports built (not the exact number, but something in the vicinity) is a matter of simple math. ... Sad that I lost my tables on this, but I can always recalculate them on ~a day or so. Then JAP may also naturally build its battleships, if in some cases only two instead of three - all the better.

... Event-spawned GAR and invasion armies appear much more acceptable to me for reasons stated: hardcoded unreasonable choice of divisions for garrisoning and dependency on CHI behaviour for invasions abroad. There is no simple solution here, so they may as well stay.
 
Hi,

Bases: The thing here is a human can build a base and then keep it in the pool for immediate deployment while the AI can not. As long as the pricing is right I think AI only is fine and easier to script in some cases.

BB: I'm currently working to get JAP to actually build the BB normally. Due to the mechanics the semi-event driven production is actually bad for AI JAP as it doesn't generate the proper attachments.

WRT TP this is a direct result of how the Naval AI operates and is pretty much lifted from Vanilla for all AI nations. The AI does a horrible job of utilizing TP and needs more TP to perform normal functions than a player would as it doesn't do things efficiently. Worse yet, it fails to provide proper amounts of escorts when ordering dangerous missions. As a result, the AI loses far more TP than a human ever would. What we are doing here is simply trying to keep the AI somewhat in line capability wise with what a human could do.

I'd add here that the AI still ends up building TP at full cost to replace TP that it lost foolishly. So it's still paying a price for it's behavior to a level that I think the field here slants to the player. Especially since TP are expensive in CORE compared to Vanilla.

mm
 
Actually, in the naval model the game presents, adding escorts to transports (within the same fleet) is a bad idea, as transports are hard-coded to (try to) spawn at max distance, whereas (vanilla bug) combat fleets (including transports) will spawn in firing distance of their shortest-ranged unit (which would be the transports).

I am not sure, whether our positioning values are bad or the system is inherently broken, but a non-carrier fleet which contains any non-transport will spawn far within firing distance, even though its combat action would be to start fleeing immediately.
This, incidentally, leads to unworkable fleet compositions like: BC+CL+DD (GER commerce raider squad), CA+DD+DD (ITA anti-submarine squad, JAP commerce raider squad), as they will spawn <5 km aside a hostile BB-group (I blame the DD, but have not tested) and pretty much disintegrate within the first hour of combat (another vanilla bug: the popup for fleet combat only appears after the first hour of combat; several ships may have been destroyed already; recall the "disintegrating ships" issue I mentioned on the old forum, we actually just always miss the true start of combat).

While the AI certainly needs more transports than a player would need, a player merely holds transports in reserve, as the ones in use may be lost; the AI on the other hand always uses all its transports if necessary.
AI JAP gains twelve transports just like that. Not two, not four, twelve! For free! If I would gain that many I would never build any (unless I lost some).
This truly seems to amount to a bit much.
 
I'm 90% sure this bug wasn't always around - don't know if it's CORE or if it was introduced by a patch, but naval combat was actually one of the major things that AoD improved and I'm pretty sure it functioned well, at least for a while.
 
You must log in or register to reply here.
Top Bottom
Back