[NOT IN DEVELOPMENT] DOC Community Mod Compilation

[1SDANi]

NEW UPDATE

Xi'an is the Chinese Capital in 600 AD again. World builder refuses to accept it as the Chinese Capital, but Python will happily oblige.

Fixed an oversight in the Moorish UP. May have fixed the Moorish city capture crash. Maybe.

May have added New Zealand to England's Trading Company Conquerors cities. Maybe.

 

[Tigranes]

Wait! How'd they get Colonial Masters? I need to see a screenshot of this! I didn't think the AI could mount a naval invasion!

They are probably unstable because they don't really expand after settling their historical territory and end up stagnating, (I could have them settle Australia, but that's up to you guys. I have no opinion on whether or not they settle more than what they currently do) their economy drops like a rock once a plague hits, and they continue to use Deification even after it becomes bad to use (I'll look at making the same change to them as @Leoreth did the Vikings)

If most of the islands gained Colonial Masters in your tests, why would a manual flip be necessary?

I agree, New Zealand definitely has enough land for a TC invasion. I'll look into it for the next update.

No, no I say -- in real history most islands gained masters without much wars. And then I suggest something similar for the game...

 

[Tigranes]

you once said you are good with implementation -- here is a pressing challenge that never been solved: 3000BC/600AD Berlin. AI usually founds Magdeburg, which gets developed for many centuries until it gets razed in 1700 (together with GPs, maybe) and one gets a Settler 1NE to found Prussian Berlin. Lamentable implementation. Either copy Magdeburg before 1700 raze and paste it as Berlin 1 NE or better yet teach AI Holy Rome to found Berlin instead of Magdeburg and then pass that city to Prussia as capital in 1700, just like Argentina starts with Buenos Aires. Of coure due to Prussian UHV one has to erase previous Great People in case of Human Prussian player.

 

[1SDANi]

No, no I say -- in real history most islands gained masters without much wars. And then I suggest something similar for the game...

Did not know that. Good idea! Though what would stop the human player from just cruising through Polynesia and picking up every island?

you once said you are good with implementation -- here is a pressing challenge that never been solved: 3000BC/600AD Berlin. AI usually founds Magdeburg, which gets developed for many centuries until it gets razed in 1700 (together with GPs, maybe) and one gets a Settler 1NE to found Prussian Berlin. Lamentable implementation. Either copy Magdeburg before 1700 raze and paste it as Berlin 1 NE or better yet teach AI Holy Rome to found Berlin instead of Magdeburg and then pass that city to Prussia as capital in 1700, just like Argentina starts with Buenos Aires. Of coure due to Prussian UHV one has to erase previous Great People in case of Human Prussian player.

I'll get on it once I'm done with my current addition. I'm attempting to add Polynesia to the 600 AD scenario, though currently I'm getting a division by zero error from the starting techs.

 

[Lokki242]

Wait! How'd they get Colonial Masters? I need to see a screenshot of this! I didn't think the AI could mount a naval invasion!

I believe Tigranes meant historically

 

[Tigranes]

Spoiler Tahiti :

Even in shower I cannot escape Polynesians...

Did not know that. Good idea! Though what would stop the human player from just cruising through Polynesia and picking up every island?

Stability. Economy. Lack of interest.

 

[citis]

Now I'm leaning towards not having a western Tibetan city. @Leoreth, it's not that words > facts, it's that you can't explain a fact without using words, and @Tigranes ' words meant more than yours. However, @ozqar 's words were more convincing than Tigranes'.

I have experimented with Kyunglung and to be honest I was happy from the result. The second city in their core is a great destabiliser for Tibet.

 

[1SDANi]

The new update is being delayed by a division by zero error when giving the Polynesians Leverage in the 600 AD Scenario. I am desperate for help right now, the deeper I go into this the less I understand.

EDIT: When I lowered the column to 1, I'm now getting the error from Sailing, I think this is a trade route related error.

 

[1SDANi]

Note to self: If editing a scenario file and a city is not appearing, fix that one BEFORE any Python, XML, or C++ errors.

 

[Tigranes]

Sorry to distract your from you Herculean effort in debugging, but just to be clear - what was the trick that made Polynesians work like this? <DefaultUnitAI>UNITAI_SETTLER_SEA</DefaultUnitAI> for Waka? Or the 3rd settler?

 

[1SDANi]

Sorry to distract your from you Herculean effort in debugging, but just to be clear - what was the trick that made Polynesians work like this? <DefaultUnitAI>UNITAI_SETTLER_SEA</DefaultUnitAI> for Waka? Or the 3rd settler?

A combination of many things. I made them unable to refuse settling their core cities or the cities that lead to Hawaii or Easter Island. The AI refused to settle the useless islands that lead to the two best islands because they had no idea that Hawaii or Easter Island even existed. Likewise, I set the Settler Value of a few select islands to 3 in order to avoid collapses. I made the Polynesian AI like to use unassigned citizens and production specialists more, I set up a series of checks to ensure they always are building something, as in the early game their cities would idle, I made them more willing to build settlers before they grew to size 3, I prevented them from building barracks or walls unless there was absolutely nothing else they could build, AI Polynesian cities start with an Archer when settled, I made it so that once a Waka drops off a Settler it converts to UNITAI_WORKER_SEA, and the AI now builds more UNITAI_EXPLORE_SEA Wakas.

In short, I modified a bunch of the AI in order to allow the Polynesians to accomplish their goal.

If you compare my Harappan AI with DOC's, they build city builders right away, whereas in DOC they usually wait a long off time. A similar change was made so that 3000 BC AI civs build workers right away. Another change is that the AI builds more units, which is something I did so that Pirates would actually be built. I haven't played any games yet, but I've seen a lot of barbarian ships during my simulations so I think it should be working.

I'm currently working on your proposed Berlin fix.

How does everyone feel about the owner of the most populous city in Greece automatically getting Knossos if it's independent, just so that we don't have a barbarian Knossos the entire game, like what I just rolled?

 

[Tigranes]

Wow, so much scripting? I feel defeated, honestly, if one has to babysit AI to this degree -- it takes all the joy away ... On the plus side I am in awe with your dedication, because at the end of the day not every player cares what is hidden under the hood, all that matters is final performance and the effect on the big picture.

And very strange discrepancy: I have the latest git of your modmod and my Waka is SETTLER_SEA, not explore... But I was able to generate your empire with my copy. My unitifo has a green check mark next to it, so I am reasonably sure it is identical to your copy, so how this can be different? Also where is the New Zealand's Sheep?

 

[Leoreth]

Wow, so much scripting? I feel defeated, honestly, if one has to babysit AI to this degree -- it takes all the joy away

 

[1SDANi]

Wow, so much scripting? I feel defeated, honestly, if one has to babysit AI to this degree -- it takes all the joy away ... On the plus side I am in awe with your dedication, because at the end of the day not every player cares what is hidden under the hood, all that matters is final performance and the effect on the big picture.

And very strange discrepancy: I have the latest git of your modmod and my Waka is SETTLER_SEA, not explore... But I was able to generate your empire with my copy. My unitifo has a green check mark next to it, so I am reasonably sure it is identical to your copy, so how this can be different? Also where is the New Zealand's Sheep?

I'm not sure where the Sheep is, maybe it was accidentally deleted? I'm not entirely sure. I'll check DOC and see if I can fix it.

To be fair, all the scripting was changes to AI behavior, except the Archer when settling a city, which I plan to change soon. Other than that, I'm really not sure how this scripting is bad, the code for the Core cities, Hawaii, and Easter Island were taken straight from Leoreth's Korea code, and the rest makes the AI act more like a human.

If something's wrong with my approach I'll be happy to change it but I'm not psychic, instead of telling me you don't like my implementation tell me what about my implementation you don't like.

The explore sea is a DLL script change that affects their AI. It uses the similar unit selection code to Vanilla Civ 4.

 

[Tigranes]

I am not a psychic either, so no way of knowing what could have been done better. You did a great job, and I am only lamenting AI's inflexibility. A general expectation from every civ's AI is the ability to recognize their environment and develop their empire accordingly.

I actually think it is good for them to collapse at some point, because the chief in Tonga was not controlling Statue building guys from Easter Island. But why is it that they don't collapse to the core? After all this effort it would be nice to interact with SOMEONE in Oceania...

 

[FinalLegendZero]

I'm not sure where the Sheep is, maybe it was accidentally deleted? I'm not entirely sure. I'll check DOC and see if I can fix it.

To be fair, all the scripting was changes to AI behavior, except the Archer when settling a city, which I plan to change soon. Other than that, I'm really not sure how this scripting is bad, the code for the Core cities, Hawaii, and Easter Island were taken straight from Leoreth's Korea code, and the rest makes the AI act more like a human.

If something's wrong with my approach I'll be happy to change it but I'm not psychic, instead of telling me you don't like my implementation tell me what about my implementation you don't like.

The explore sea is a DLL script change that affects their AI. It uses the similar unit selection code to Vanilla Civ 4.

The New Zealand sheep isn't on the map initially; it's scripted to appear much later on.

 

[1SDANi]

I am not a psychic either, so no way of knowing what could have been done better. You did a great job, and I am only lamenting AI's inflexibility. A general expectation from every civ's AI is the ability to recognize their environment and develop their empire accordingly.

I actually think it is good for them to collapse at some point, because the chief in Tonga was not controlling Statue building guys from Easter Island. But why is it that they don't collapse to the core? After all this effort it would be nice to interact with SOMEONE in Oceania...

I am planning to revert New Zealand and Hawaii's settler map so that they settle in different parts there, though as for the AI's inflexibility, I'm pretty sure a large contributor to it is that their historical extent is mostly 1 tile islands, barring New Zealand, and I'd rather they not settle Indonesia or Australia. If you want them to I can make them do so, but I'm not entirely sure it's something the AI should be doing.

 

[Tigranes]

I am planning to revert New Zealand and Hawaii's settler map so that they settle in different parts there, though as for the AI's inflexibility, I'm pretty sure a large contributor to it is that their historical extent is mostly 1 tile islands, barring New Zealand, and I'd rather they not settle Indonesia or Australia. If you want them to I can make them do so, but I'm not entirely sure it's something the AI should be doing.

No, no very different civilizations lived in Australia, Melanesia, Micronesia and Polynesia. We can really rest Polynesian case for the moment. My and I think most of the players' only request is working collapse to core mechanic, but without all those free archers being teleported to Tonga...

 

[Tigranes]

The New Zealand sheep isn't on the map initially; it's scripted to appear much later on.

Oh, I see, during colonial time that is... Makes sense, Polynesians did not have Sheep...

 

[1SDANi]

No, no very different civilizations lived in Australia, Melanesia, Micronesia and Polynesia. We can really rest Polynesian case for the moment. My and I think most of the players' only request is working collapse to core mechanic, but without all those free archers being teleported to Tonga...

I didn't make any changes to the existing collapse to core code, so I'm not sure why the AI is getting a complete collapse.

I will say that this discussion has made me look back and rethink how I approached Polynesia, and I've made a few decisions for the next update:

1) Polynesia's AI will build defenders, instead of automatically getting them
2) New Zealand's Settler Map will be changed to allow more variability in settlement locations

As for the progress of the next update: I'm currently having trouble with the suggested Berlin change, but I've made a good amount of progress on completing it.