Feature Requests and Discussion

It is supposed to be significantly faster and less of a memory hog to use xrange rather than range when iterating over a loop (not when defining a list that will be referenced multiple times), but the use of range is far more common than the use of xrange in the game currently. It might be worth changing that, at least for the sake of older computers.
 
Would it be possible to teach the AI how to use Portals?

I just tried to see if I could make a PyRequirement function which let an AI use the spell only if the exit plot is closer to the unit's intended destination, but found that pCaster.lastMissionPlot() was not exposed. (I'm not sure if that is the best function to use anyway, but it was one I came across in the DLL that looked relevant.)
 
Would it be possible to teach the AI how to use Portals?

To teach the AI to use Portals as a means to get somewhere specific would be difficult and beyond my current skills. I think it would require a major change to the pathfinding routine and some sort of store variable system that would keep track of these new "paths".
 
on this subject :
is the AI able to take portals into account for "trading" purpose (especially ressource spreading in an empire ) ?
 
I have a new WorldBuilder for you now, based on Platyping's World Builder 4.16d.

This adds the ability to force units to perform various missions, and so sort units by activity.
 

Attachments

  • World Builder 4.16d for MNAI v 2.63.zip
    147 KB · Views: 249
Platyping just released yet another update, just to make it easier to tell what is blockaded.

It have merged it for you already.

It only changes 2 files from the last version, including a file where the version number was the only change.

This time I decided there was no reason to include files that were not changes from the last MNAI release.

Edit: Platyping reuploaded without relabeling to fix a typo. I am now doing the same, as it does not appear anyone has actually downloaded this yet.
 

Attachments

  • World Builder 4.17 for MNAI v 2.63.zip
    79 KB · Views: 246
Could you expose the new diplomacy functions (like eTeam.setHasNonAggression(iTeam, True), eTeam.isHasPrepareWar(iLoopTeam), and eTeam.setHasPrepareWar(iLoopTeam,True) ) to python?
 
I've missed a decent amount since my last time here but what's considered the version with only minimal changes to the architecture of the .dll (little of the new exposed python and the like). I'd like to build a multiplayer mod off the back of Tholal's and while I know that you've minimized OOS errors, I'd prefer a version that I can work on without having to quash new versions of the same old OOS errors. Or is the latest the best?
 
Sorry if this doesn't belong here but I was wondering if you could tell me which mod you got the Nuclear Diffusion option from.
 
I have a worldbuilder update for you. I don't know that I have it working perfectly yet, but it is an improvement.

I fixed the bug I mentioned earlier plus a couple more. I also fix a couple typos in some text files and got rid of a lot of unnecessary whitespace. If you use these files it should make merging easier for me later on.

I added advanced diplomacy stuff to the Diplomacy screen. I'm not sure why I don't seem to be able to change things like war plans or non aggression pacts yet, but it does at least display them.


Loading a game from a worldbuilder save still seem buggy. I'll look more into that later, but would welcome others to give it a try too.
 

Attachments

  • World Builder 4.17b for MNAI v 2.67Beta4.zip
    1.4 MB · Views: 201
it's going slow now in the old days but hoping we get one or two more updates to tie up loose ends... cheers guys :thanx:
 
I've missed a decent amount since my last time here but what's considered the version with only minimal changes to the architecture of the .dll (little of the new exposed python and the like). I'd like to build a multiplayer mod off the back of Tholal's and while I know that you've minimized OOS errors, I'd prefer a version that I can work on without having to quash new versions of the same old OOS errors. Or is the latest the best?

The 2.7 Beta versions have some new Diplomacy stuff. I cant say if it has been tested in multiplayer or not. The 2.6x versions has had most of the OOS errors fixed. I think there might still be one issue outstanding with creating a Puppet State.


Sorry if this doesn't belong here but I was wondering if you could tell me which mod you got the Nuclear Diffusion option from.

Advanced Diplomacy - http://forums.civfanatics.com/showthread.php?t=447915

I have a worldbuilder update for you. I don't know that I have it working perfectly yet, but it is an improvement.

Thanks!
 
Some things I noticed, most of which Magister told me to ask you. I'm not a coder so I really don't know how many of these can be fixed/implemented or if they would be too hard to do, but I'm just putting it out there :D

  • Open border agreements seem to cancel regularly due to no apparent reason.
  • Early game diplomacy (i.e before currency) is extremely annoying with advanced tactics on. Even for establishing embassies the AI expects to get a tech or resource in return (unless they themselves propose a fair deal, which they do, but not that often). Once you do get currency they usually agree to take 15-20 gold/ 1-2 gpt for such deals, so it's not like the trade was one-sided enough to warrant giving them a tech. Would it be possible to make the AI accept even deals early on (even when they want a bit more)?
  • From one of my previous games:
    Yep.. this time got a stack of about 50 SoKs.. Its stupidity really.
    The AI obsession with certain cities it sets its eyes on is dumbfounding. As Auric, I started out quite close to Kandros Fir and Mahala. I made my second city at a spot roughly equidistant from our capitals. It was a good location for me (riverside ice) but quite pointless for Kandros. However for some reason he was hell bent on taking the city. I killed a total of about a hundred SoKs by turn 200. Due to spamming huge amounts of units in trying to get a city which was absolutely useless for him, he didn't settle a region on his other side which was empty and just grassland and resources. Not to mention how much it would have drained his economy, by the end of the rushes he was last in tech despite having started next to Yggdrasil.
    I could understand a human obsessing over a single city (hurts my ego when my invasion is repelled :lol: ) but why the AI would do this I do not understand.
    When the AI's power is significantly more than yours, do they attack you even when they possibly have other (better) land to settle?
    Also, to counter the early game SoK stupidity I'd suggest changing the power rating of SoKs to that of axemen (I believe they have the same power as champions)
  • The AI doesn't seem to explore lairs (barrows) in/near its own territory.. just kill off the guardian and let the lair be. Due to this they regularly have to suicide their units against Lizardmen when they could just explore the lair. Magister said this was because you had added some code that made them disinclined to explore lairs near them.
  • Just a minor gripe, could you prevent evil leaders from asking good leaders to DoW good leaders (and vice versa)? Happens a lot, civ that completely hate me ask me to wardec my 'closest ally'. Similarly, quite often civs ask for a large tech as "help" even when they're annoyed with me, which inevatiably results in them becoming even more annoyed :lol:
  • Would it be possible to give certain civilizations a high priority for certain techs? It just feels strange when the Hippus are researching animal handling but still haven't done horseback riding :lol:
 
Another thing I had asked in the civ 4 modding forums first, but was told that this would be a better place to ask. Has anybody noticed how inconsistenly AI capitals are spread on most maps? Ocassionally three capitals are within 6-7 tiles of each other, and sometimes one civ is all alone in a large area. At first I blamed this on the Erebuscontinent mapscript, but then I realised that the script was placing the civs right, but they themselves chose to move away from their capital, in some cases by a lot of distance!
So on turn 0 I removed the additional visibility and movement and erased the extra tiles they had seen. A lot of AI's still chose to move. So I don't think it is necessarily because of the starting settler promotion.
Is there a way to make the AI settle on spot, or at least greatly decrease their tendency of moving?
 
Is there a way to make the AI settle on spot, or at least greatly decrease their tendency of moving?

I find replacing the "starting settler" promotion with "sentry" goes a long way towards fixing this ( you can do that in Assets/XML/GlobalDefinesAlt.xml )

there's no good reason why starting settlers need to move so fast, it was just a stopgap measure added early in ffh development when you could start next to the broken sepulcher and get killed by Barbatos' earth elementals
 
Hi all, it's been a while! Every few years there comes a point when I like to reinstall this mod and reminisce about how great it used to be (and still is, even with age). Anyway, I've never been a modder but I used to like using the XML editor to make small tweaks to the game. I thought I'd give it a try once again by adding Shekinah in as a playable Sidar leader (Arcane/Defender).

I can do it easily in the base game, but if I make the same changes in this modmod it causes a bunch of XML errors when trying to load up. Does anyone know of any way that I can get the editor working? Or if that's not possible an easy way of switching Shekinah to a playable leader and changing some of her characteristics?

EDIT: One thing that I tried was making the changes in the vanilla version of FFH (using the editor), and then copying the Civilizationinfos and Leaderheadinfos XML files from the vanilla into my MNAI copy of the mod. This did not work!
 
MNAI adds some new tags to CIV4LeaderHeadInfos.xml, to handle preferences for Advanced Diplomacy. I think there is a change to CIV4CivilizationInfos.xml to allow certain civs to get different yields for riverside tiles on various terrains.

If the file does not match its schema file, then it won't load properly.

Even without your changes the base FfH2 versions of your files would make this mod crash.

I'm not quite sure what you mean by the xml editor, but I'm guessing you mean the excel document with micros that Kael wrote. It has diverged enough from this modmod that I would not suggest trying to use it.

It is probably easier just to use a standard text editor (Notepad is fine, although Notepad++ is better) to modify your xml files directly. It is not hard, expecially for little changes like you suggest.
 
That makes sense, and yes I was trying to use Kael's excel sheet! Okay thanks, I will just give it a try in Notepad.
 
Top Bottom