Infrastructure building queue for AI in AoD 1.09

[Mr_B0narpte]

There is a general lack of rares on the euroasian continent. This limits factory construction if either some reserves are to be built up by AI or there should be enough resources for the human player to build up effective ic to high levels.

This makes me wonder how realistic AoD's handling of worldwide resources is. AFAIK CORE has given Africa much more resources in comparison to vanilla. Certainly much of vanilla AoD Africa is a barren wasteland, it would be good to see this - and resources worldwide - change.

Another more immediate and basic change is to create the UK puppet of Malaysia (with Singapore remaining under UK control to ensure those rares leave the puppet), which suddenly provides the world another 100 rares daily. It makes much more sense then the UK AI building infra there, effectively a nice gift for Japan.

For the USA i would estimate that TC is a minors concern.

Something sensible could be to make USA build synthetic rares plants. Else its energy and metal cannot by utilized, no matter by whom. Also past 1941 the USA tend to have more than enough oil because it cannot export it to axis because it is at war with them.

So in other words it's 80 infra serials are a complete waste.

 

[Pang Bingxun]

Another more immediate and basic change is to create the UK puppet of Malaysia (with Singapore remaining under UK control to ensure those rares leave the puppet), which suddenly provides the world another 100 rares daily.

That is a viable strategy, but as for a vanilla patch this seems no good solution.

So in other words it's 80 infra serials are a complete waste.

It is only 67(i could of course increase it to 85 if you insist) serials in total at 177 starting effective ic. A great amount of them are good for increasing effective ic so that USA can outproduce the rest of the world at ease, which ultimate also increases tc to high levels. Some other Infra builds increase the output, that a human UK etc. could tap into.

 

[Mr_B0narpte]

That is a viable strategy, but as for a vanilla patch this seems no good solution.

Having the UK AI build infra in Malaysia - and much of the British empire - is also a "no good solution".

It is only 67(i could of course increase it to 85 if you insist) serials in total at 177 starting effective ic. A great amount of them are good for increasing effective ic so that USA can outproducce the rest of the world at ease, which ultimate also increases tc to high levels. Some other Infra builds increase the output, that a human UK etc. could tap into.

Building factories is a much better way to increase effective IC in comparison to building infra in 1-4 IC provinces. I imagine the USA can provide the UK's needs without spamming so much wasted infra - in fact I know it does considering I'm currently in a game where the UK player has 430 effective IC and the USA has 447 base IC and it's only May 1940.

 

[Pang Bingxun]

Having the UK AI build infra in Malaysia - and much of the British empire - is also a "no good solution".

It is not, but it is better than building no Infra there. If i were to play the UK releasing Malaysia would be a priority as it may be the most useful puppet one can have. Rares are rare and having them extracted from national territory is much better.

Building factories is a much better way to increase effective IC in comparison to building infra in 1-4 IC provinces.

True. Outside ouf eurasia this seems like a semi-viable choice. But the USA donnot need factories so much. Nor do they need to spend anything on new military prior to the 1940 elections.

 

[Commander666]

I have made AI create some puppets. So i have developed an understanding of where what kind of difficulties are to be expected. As said in a different thread: Familiarize yourself with the facts first so you gain the ability to formulate useful criticism.

No, I don't need to familiarize myself with your work to have told you that your earlier claim "AI and puppets donnot mix very well." is wrong. So now you state, "I have made AI create some puppets."

How ridiculously funny! Perhaps it really is only your limited English creating the major problems with people accepting your unilaterally decided changes? So, did you make AI create puppets in the last 24 hours on account of my criticism - which is what your statement “I have made AI create some puppets” literally implies when subjected to the running clock of real time as regards this discussion?

Or did you mean to write, "In Set 34 the AIs can create puppets"? If this fits better, how come you didn't point it out sooner to so not get Mr_BOnarpte's criticism in the first place when you and he were discussing? And if the AIs in Set 34 actually do create puppets, I imagine that must have taken a HUGE amount of beta testing by various people to even confirm that really happens. So how come nobody has even mentioned it? How is it that you suddenly admit this when there was never any mention of it before; or confirmation it actually works as regards the many people who beta tested?

Seems to me you only reading files and making assumptions how game might actually work regarding any AI creating a puppet. Let’s not forget your revelation regarding what happened with v1.08 when you admitted that those massive changes had unforeseen consequences. Or do I need to find and post here your admission regarding v1.08 – which stated that the best fix would be to roll back to v1.07, but you had “no intention of doing so.” Remember?

So far this thread has surfaced many legits points of view but no reason why deleting Infra would improve the game.

HA! HA! HA! How absurdly ridiculous your incorrect assessment is!

You know Pang, I suggest you re-read the thread “109 patch beta 1” started over a year ago – with particular attention to the many posts made around October 2014. Next, count every person who back then disagreed with your "massive infra changes". It isn't 2 or 3 like here. Rather it is EVERY person that posted. It is so many people that I would say it is about half of the community if we define community to be those persons who are regular contributors to the “Arsenal of Democracy” sub-forum. You have had massive objections to your massive infra changes for a very long time.

Debating anything with you regarding changes you made to AoD is senseless, since you intend that your mostly one man product will be AoD by whatever authoritarian means necessary. At least we can still enjoy ourselves criticizing your unpopular changes; and load Lord Jarski's mod or join CORE for a much more interesting "AoD experience"; or try TRM for a greater challenge.

Frankly, IMO, you are the reason AoD vanilla has started to suck to the point of little interest playing that peculiar version which is no longer so much like any "vanilla" but seems rather better described as just "Pang Bingxun’s AoD" to so clearly distinguish between the AoD vanillas that existed before you took this game and forced it as to your pre-set ideas for massive and unpopular changes.

The short – and correct – assessment of this whole thread is not what you write, or might wish. Rather, this whole thread – starting with your first well polished post which was a “publicity stunt” designed to support your infra changes – is nothing more than Pang Bingxun thinking “he knows best”. I imagine you would not disagree that you think “you know best.”

Unfortunately for you, your legacy will become “The guy that killed AoD vanilla.”

EDIT: The below is some unknown file which somehow uploaded itself. Strangely, I can't find any code associated to that in the body of my post when I do edit and search for it in order to delete it. Maybe its a sign from God as - at first - it displayed with an all black profile of a human head ... so giving a rather appropriate R.I.P. message. But now it changed itself, but seems to exist nowhere. How weird!

 

[Commander666]

But the USA donnot need factories so much. Nor do they need to spend anything on new military prior to the 1940 elections.

While I agree with the "not so much" view regarding factories, USA certainly does need to spend for new military as regards their navy long before the 1940 elections (and afterwards also). If USA does not spend anything on navy (as you clearly state) until November 1940 (time of that election) then they would have mostly only their 1936 fleets come Pearl Harbour as all capital ships (except CAs and CVLs) take longer than 13 months to construct. Your statement amounts to the equivalent of recommending a program of total obsolescence for the USN until war begins.

What do you intend... that come Pearl Harbour, the USA has only antiquated fleets augmented by only a few new DDs, CLs, SS and TPs and perhaps a CA (that started retooling only after the November 1940 elections) to fight a human player of Japan? They would not even have a single new CV, BB or BC for Pearl Harbour. Not only is that extremely ahistoric, but the result would be greater than the Infamy of Pearl Harbour and more like the Infamy of Pang's Recommendation resulting in the nearly full elimination of the USN within a few months. How do you even come up with such wild statements suggesting the USA should spend nothing on military for first 4.9 years of the game?

So far this thread has surfaced many legits points of view but no reason why deleting Infra would improve the game.

And just to set you straight, there is nothing in this thread that has given any reason why leaving the massive new infra builds improves the game. It is all only your lone point-of-view.

 

[Lord Jarski]

Pangs infra builds work wonderfully.

If you play on Very Hard difficulty as Costa Rica and take no part in the worldly affairs.

Oh that poor newbie playing UK with Very Easy difficulty. How frustrating it must be that the commonwealth assistance is the same as their starting troops untill 1942 or what ever the hell is their supposed "optimal time to build troops".


But anyways, I am preparing a new update for the Improved 1936 Mod, with the good fixes integrated from the 1.09 patch. The infras serials do have a place in the scenario, that is granted, just not in the extent they are at the moment. The mod fixes that, and update will add resources to Africa, CORE bases/installations, new events and much more. Alot of the complaints on this, and many other threads have already been solved in the mods, this discussion has given me the motivation to urge more people to try it out if they are unhappy with the vanilla game.

Give it, CORE, or some other mod a shot and enjoy the game like it should be.

 

[Commander666]

Pangs infra builds work wonderfully.

My God, you gave me a shock!

Yah, I love playing Costa Rica on very hard difficulty; and take no part in the rest of the world. Let's all play AoD 1.09 "HANDS OFF" so we can stop lambasting it.

EDIT: I for one will soon be patching to your highly recommended mod because I sure as heck will not be trying to beat Mr_BOnarpte's UK (which can't construct new DDs or CVLs as to his challenge) on any 1.09 beta 34.

 

[Lord Jarski]

My God, you gave me a shock!

I like to make a shocking entry

And welcome aboard if you indeed patch to my mod, although I suggest you hold out untill the next update is released which brings alot of new additions.

 

[Commander666]

Hi, is the correct checksum for 1.09RCI patched to v1.08 supposed to be LSEV
Thanks!

 

[Pang Bingxun]

The checksum of 1.09 RC1 is supposed to be LSEV. No prior patching to 1.08 is needed, though.

Edit: As 1.09 reaches its final stage we moved to fully integrated patches, so your prior version of AoD does not matter.

Estimated Checksum of Release Candidate 1 is: LSEV

 

[Commander666]

I guess I got it right.

So, the USA AI is constructing from start infra to 200% in 67 provinces... which means that 24 of the provinces have only 2 IC. Just how much concentration bonus do you expect to get from a 2 IC province? Asked differently, what you figure will be the end IC for those 24 provinces? As the extra resources that can be extracted with the greater infra really are not needed from those 24 provinces (considering what massive increase in resource extraction will be gotten because of the other 43 provinces with new infra) , I suppose that increased IC is the only real benefit in those 24 provinces for spending a massive amount of icd to increase the infra there.

Next, fortunately the USA AI does construct one line of CVs from start - as is proper. So why did you make such a looney statement that USA does NOT need to spend anything on new military prior to the 1940 elections?

 

[Commander666]

It helps to check things. I was wrong that all provinces getting infra had at least a factory there. But "No", you also improving provinces to 200 infra that have zero factories. What you think, increasing Los Alamos to 200 infra will speed the future nuclear test site there? Don't think so, unless some new change or AI special benefit. Or you doing that to get a bit more oil... as that's the only thing in Los Alamos.

I wonder if I can find more "zero IC provinces" that Pang has got the AI improving to 200% infra? Does it not occur to you that giving Yellowstone Park, the deserts and some swamps the same infra as New York or Washington seems slightly odd?

 

[Pang Bingxun]

Just how much concentration bonus do you expect to get from a 2 IC province?

2 factories give 2% concentration bonus. 100 factories give 100% concentration bonus. As the amount of factories is not changed in most provinces you donnot refer to concentration bonus, do you?

Asked differently, what you figure will be the end IC for those 24 provinces?

2.652 base ic.

So why did you make such a looney statement that USA does NOT need to spend anything on new military prior to the 1940 elections?

USA should not spend anything into new military prior to the 1940 elections because prior to the change in HoG the USA pay additional 21% icd per unit. So any delay is welcome as it increases efficiency a lot.

More important is however the hawk lobby slider which increases efficiency by 77.8%. All that has already been discussed in a seperate thread about the potential of the USA:

https://forum.paradoxplaza.com/forum/index.php?threads/how-much-can-the-usa-produce.814878/

What you think, increasing Los Alamos to 200 infra will speed the future nuclear test site there?

There is an effect infra has on the repair speed of buildings in the province. That i do know. I donnot know whether Infra has an effect on the output of a research reactor. For now let us assume there is no such effect.

The reason for the Infra there is the oil in the province. At initial ministers, 1938 tech and 200% Infra Los Alamos will produce 39.468 oil a day once the peacetime penalty is eliminated. Sold at 5 oil per $ that is worth 7.8936$.

 

[Commander666]

YUP! Found 16 more provinces which have zero factories but are all being built to 200% infra (also some 1 IC provinces).
Apparently Pang thinks that constructing infra to maximum in Casper (with the only thing there being 13 oil to reap any benefit) or Smoky Hill - with just 9 oil, and nothing else - must be a good investment for the AI.

Lets see.... (where's my calculator?) 24 new infra at Smoky Hill - which is plains and so has no discount for being "terrain blind" takes AI until June 1st, 1940 to construct.

Would that be 4 years x 360 days per year = 1440 days
Plus 5 months x 30 days per month = 150 days
Plus one day for June 1st/40 = 1
.... which totals 1,591 days.

So far so good. Now, at 2 IC per infra line which AI gets 30% discount on = 1.4 IC/ day.

So that would be 2,228 icd to attempt to force some further extraction out of a starting resource of only 9 oil for that province. Of course, there are slider, minister and policy changes that might help reduce the icd somewhat.

BUT THE POINT IS that a country which is "drowning in oil already" given all the other provinces with major oil deposits who also get infra to 200% sure as hell does not need to waste ~2000 icd on improving frigging Smoky Hill - a two-bit hick town on the pony trail between Omaha and Denver provinces with nothing more in it than a fork in the trail down to Dodge City (province).

You should have done your research, Pang, and you might have learned that - historically - Smoky Hill has actually become a ghost town.

You know, I'm just totally at loss of words to describe these ridiculous infra builds. "Over abundant" is much too polite. "Crazy" does not come close to the absurdity of it. Maybe "desperate" fits?

 

[Commander666]

2 factories give 2% concentration bonus. 2.652 base ic.

So, for about 2000 icd USA gets 0.652 more IC in mid-1940? WHOOPEE!

USA should not spend anything into new military prior to the 1940 elections because prior to the change in HoG the USA pay additional 21% icd per unit. So any delay is welcome as it increases efficiency a lot. More important is however the hawk lobby slider which increases efficiency by 77.8%.

Any delay is welcome ... including being at Pearl Harbour with only the three CV-2s gotten at game start??? That is what you will have. Have fun losing your most of your obsolete navy before you can even construct a single new CV if you start military spending only after the 1940 elections.

I donnot know whether Infra has an effect on the output of a research reactor. For now let us assume there is no such effect.

I was not discussing "output of a research reactor". I was discussing "constructing the reactor" and in 1.08 the reactor constructs same speed regardless of terrain or infra as pertains to the human. Unless you changed something, or AI gets some unknown discount, infra don't matter regarding construction speed.

Next, I doubt reactor output increases with increased infra. Rather higher levels of reactor might give greater output... although the nuclear funding probably is more important.

The reason for the Infra there is the oil in the province. At initial ministers, 1938 tech and 200% Infra Los Alamos will produce 39.468 oil a day once the peacetime penalty is eliminated. Sold at 5 oil per $ that is worth 7.8936$.

WHOOPEE! WHOOPEE! WHOOPEE! Instead of USA having to find a market for its more than 500 surplus daily of oil, it now must find a market for 539.468 surplus daily of oil. Have you ever taken an economics course? And do you remember that you will be doing Japan oil embargo (and not supplying Hitler either) to compound your enormous excess oil situation? You get nothing - zero $ - for the extra 30 oil that a built up Los Almos extracts - because you can't even sell the hundreds of oil which Lubbock, Monroe, Houston, Dallas and other far better oil provinces will extract. Your idea is nothing but majorly flawed economics.

Besides, you did not account correctly for the 13 oil Los Almos already starts with - which is also going to benefit from 1938 tech. So - at best - your true increase in oil extraction is only 26/day. Man, you can't even do inventory properly!

WAIT until I load up as Costa Rica and do a hands off to learn the real facts of how astronomically excessive all of USA's daily resource surplus will be with the your unbelievable AI infra builds for the States.

 

[Pang Bingxun]

Any delay is welcome ... including being at Pearl Harbour with only the three CV-2s gotten at game start??? That is what you will have. Have fun losing your most of your obsolete navy before you can even construct a single new CV if you start military spending only after the 1940 elections.

You might choose to recognize that the USA can build CV1938(or CV1941) in 351 days without even using accelerated production. All those details had been discussed in the thread i linked to:

https://forum.paradoxplaza.com/forum/index.php?threads/how-much-can-the-usa-produce.814878/

But as we are discussing AI only this is not even relevant.

 

[Lord Jarski]

I would think the main questions Commander666 asked was:

1. When does an infra serial to a province with 2 IC pay itself back?

2. When does an infra serial to a province with 9 oil pay itself back?

And depending on the answers, whether it is economical for the AI to be forced to build to such serials? Human players can always cancel the serials, but sadly the AI does not have the wisdom to cancel wastefull constructions.

 

[Pang Bingxun]

1. That depends on many circumstances and the alternative. For the USA the return on investment is about 12% p.a. compared to stockpiling money instead and thus getting zero return on investment instead. Ai does not stockpile money.

2. Assumung 5 oil per $ and 2.4 $ per icd the return on investment is about 4.2% p.a..

It might be worth to note that the USA, the UK and France are 3 special cases as their peacetime penalties had been reduced in 1.09. The purpose was to increase resource supply and to allow the human player to fund research properly. As AI has no need to fund its research it has been made to spend the gained icd into some additional Infra, so that the goal, to keep those nations sufficiently unprepared for war at their historic dates of war entry, is not harmed.

 

[Mr_B0narpte]

Why do you keep referencing the $ value of the resources when you know the AI does not trade them away?