Some changes in 1.3

[Evil4Zerggin]

I suspect it's a matter optimization; the quadratic algorithm possibly works faster at the number of nodes we have in vanilla, but gets inefficient (due to the quadratic factor) with more nodes.

I don't see how that would occur in this case. #0 would be computed before the start of the game, #1-3 are directly displayed so there's no way around computing them, and I can't think of a more straightforward way to compute #4. Matter of fact, I don't see how a quadratic algorithm would fit the current system at all, barring AI as dezuman mentioned---but given the less strict definition of AI behavior you can blame any running time on AI, and it would probably involve a lot more than the number of trade nodes. Again, the trade system seems to be designed to allow a linear algorithm; for example, the cycle-less structure and trade good market shares being computed only from local trade value and trade power share.

Or maybe we're just all misinterpreting what Johan meant by "NxN". I and probably others were operating on the assumption that N referred to the number of trade nodes since that was the subject of discussion.

 

[raw]

- You will no longer get AE from inheriting another country.
- Revoking electorate through war now reduces Imperial Authority by 10.
- Reduced AI skill with armies on Low difficulty
- Provincetradepower is now given to controller of province, not owner.
- Trade Steering now also modifies the amount of money that goes forward from a tradenode you have a trader in.

A wonderful dream of love and peace for everyone
Of living our lifes in perfect harmony
A wonderful dream of joy and fun for everyone
To celebrate a life where all are free.

 

[unmerged(80402)]

Declare war on the world. Occupy whole world. Controll whole trade. Win.

EDIT: In order to avoid WE, make sure you have one OPM that is being alone in war with you, is unoccupied.

 

[hauptman]

Declare war on the world. Occupy whole world. Controll whole trade. Win.

EDIT: In order to avoid WE, make sure you have one OPM that is being alone in war with you, is unoccupied.

Rofl you just gave me my first goal...

 

[TheDarkMaster]

This talk of the trade nodes has gotten me worried, in that I may have made my mod much slower then it would otherwise be simply by moding the trade nodes slightly just to reverse one trade path...

 

[Tom013]

Declare war on the world. Occupy whole world. Controll whole trade. Win.

This really sounds like something for an achievement... rather than conquer everything, occupy everything!

 

[aethel27]

This talk of the trade nodes has gotten me worried, in that I may have made my mod much slower then it would otherwise be simply by moding the trade nodes slightly just to reverse one trade path...

Eh.. its also a problem of the engine not operating from the moddir correctly. Since when using vanilla tradenodes in the moddir for meiou makes it slower than custom nodes in the vanilla directory for meiou. Gigua's thread has more info on it.

 

[Xeorm]

I don't see how that would occur in this case. #0 would be computed before the start of the game, #1-3 are directly displayed so there's no way around computing them, and I can't think of a more straightforward way to compute #4. Matter of fact, I don't see how a quadratic algorithm would fit the current system at all, barring AI as dezuman mentioned---but given the less strict definition of AI behavior you can blame any running time on AI, and it would probably involve a lot more than the number of trade nodes. Again, the trade system seems to be designed to allow a linear algorithm; for example, the cycle-less structure and trade good market shares being computed only from local trade value and trade power share.

Or maybe we're just all misinterpreting what Johan meant by "NxN". I and probably others were operating on the assumption that N referred to the number of trade nodes since that was the subject of discussion.

I'd hazard a guess that the NxN would be due to the AI and having to compute where it's best to place traders and ships. It doesn't seem like a very simple calculation in the later stages of the game where there's so many choices for each AI, while also making sure the AI doesn't do silly things like constantly changing where trade ships are assigned.

 

[Evil4Zerggin]

I'd hazard a guess that the NxN would be due to the AI and having to compute where it's best to place traders and ships. It doesn't seem like a very simple calculation in the later stages of the game where there's so many choices for each AI, while also making sure the AI doesn't do silly things like constantly changing where trade ships are assigned.

Again, yes, it could be AI. But Johan didn't make any mention of AI in his post.

Thinking about it some more, I suppose NxN could refer to the size of a Laplacian matrix of the trade node graph which could be used to find the trade behavior similar to the eigenpair method I mentioned earlier. I think this would be slower than the linear-time (edit: in terms of the number of edges---but if the degree of vertices is bounded by a constant, which is reasonable for a trade graph, this is also linear in the number of vertices) topological sort method, though.

 

[lucaluca]

And some more...

- AI: Less likely to switch attitudes and become hostile/rival towards an ally

Mhhh not very happy about this, will aliances become very static? I think countries should be more willing to become rival/hostile AND friend/allies to create more of an ebb and flow depending on missions and obejctives

 

[unmerged(785632)]

And some more...

- AI: Less likely to switch attitudes and become hostile/rival towards an ally

This sounds good, hopefully those non-sense back-stab goes away....., by the way, could you look at the unlawful territory stuffs ? it is pretty non-sense for stacking .....take 300 year to removed ..... that's too much. probably should have a maximum of 50 if each year can improved only by +1...., or just impose a maximum to the other HRE member other than the emperor.

 

[unmerged(785632)]

Mhhh not very happy about this, will aliances become very static? I think countries should be more willing to become rival/hostile AND friend/allies to create more of an ebb and flow depending on missions and obejctives

The AI is already a little bit more likely to be rival than ally towards players, or simply just put all AI on coalitions against player seems fun ? WW lol.

 

[Atr]

- You will no longer get AE from inheriting another country.

Hope it isn't too late to chip in.

To clarify - does that mean countries will still get AE if they integrate a personal union?

Either way, I'm not sure it's a good idea to remove AE for this. For instance, if France were to inherit Spain and become a massively dangerous European power, it makes sense that its neighbours would get AE, which might lead to them forming a coalition to combat this new threat. Inheriting may be legitimate, but surely other powers would be very wary of blobs developing.

What is now the counterbalance to big countries inheriting other big countries?

 

[Al. I. Cuza]

To clarify - does that mean countries will still get AE if they integrate a personal union?

Indeed, he clarifies a couple of pages in.

 

[billclintonmask]

Even though I've never really had a problem with the AI preventing my expansion, I welcome any changes that give a fresh gaming experience. Patch time can't get here soon enough!

 

[tkamin]

Any chances to get American Dream on Linux fixed?

 

[geckoman1011]

To the people who think the changes to diplomacy will create static relations, I think the real change is removing the AIs tendency to be schizophrenic over the short (12mo) term. I hope they left the long term AI diplomacy unchanged, or at least mostly unchanged.

 

[Hans Lemurson]

Declare war on the world. Occupy whole world. Controll whole trade. Win.

EDIT: In order to avoid WE, make sure you have one OPM that is being alone in war with you, is unoccupied.

Occupy All Street(s)!

 

[Saladin Osmanli]

Occupy All Street(s)!

You, sir, deserve a platinum medal for this.

 

[Revshawn]

And some more...

- AI: Less likely to switch attitudes and become hostile/rival towards an ally
- fx files now work properly for steam workshop.
- Great leader for achievement is 5 or 6 fire, and 5 or 6 shock, nothing else now.

Real happy about the first one. No more random "Eff you, you're my enemy now" moments when you have 200+ relations with an ally but happen to border or gain a border with them.