Good & Bad News - Victory Point Values (To win the capture point) in kv_rules_tables

[Ahiga]

Go to the kv_rules_table, at the very bottom are the two rows:

tickets_multiple_victory_points
tickets_single_victory_point

At least in PFM there seems to be a cap to how high the value can be. I went with 1E+36. That led to me instantly losing the battle after 10 seconds. It's possible the code doesn't like that mathematical a formula for its value, since I had tried to type 999999999999999999999 ect. So it is possible we can set it so high, just not with the PFM right now.



So I switched it to 1000000. Lowest numerical value before it goes 1E+07. This time the capture point did display a numerical value. Last time it didn't.

After Pontus took the point at 57:20, I minimized the game and let it run until the points ran out: http://i.imgur.com/EM8vHn5.jpg

At 46 minutes it was 9993 something something, since the rest of the digits disappear beyond the edge of the window.

At 35:20 it was 9986 something something something.

At 29:49 this was the value: http://i.imgur.com/83EEI89.jpg

Meaning, 30 minutes did not see a full revolution of a hundred million points. It went from 1 billion to approximately 998,000,000.

Simply put: We cannot set the VP to some infinite number, but we can set it high enough that more than two hours should be able to occur and you will not run out of victory points. In fact I am pretty sure you could have a battle going for 24 hours and not run out of VP points.

So that's the worst case scenario - the best being we can remove them for all but sieges. No idea if that is the case.

[Pnutmaster]

Our chief issue is not that the victory points only last a minute, but that the AI suicidally, mindlessly charges towards them, right? Our only solution is no less than eliminating them.

[Yarkis de Bodemloze]

Aren't the ticket values just seconds?

[Ahiga]

Our chief issue is not that the victory points only last a minute, but that the AI suicidally, mindlessly charges towards them, right? Our only solution is no less than eliminating them.

And I am not sure if that is possible yet. Might be, no reason to think it will be or won't be just yet.

Yarkis, they seem to go by faster than seconds.

[.Mitch.]

They are likely game CPU ticks. Every cycle of the game loop or somesuch.

And I agree, no less than removal of these horrid things will be acceptable.

[Ahiga]

Theoretical speculation of mine, but I think the weak point may be in how the game decides what battles have capture points and what do not.

I have fought battles without capture points and battles with them. On the Roman campaign if I attacked the Etruscan stack to the north there was no capture point, and if I let them attack my general and a hastati the next turn there was no capture point. Fighting a Median stack that had forced march by the Tigris brought on a capture point for them, as did attacking the besieging Cyrenians at Alexandria. It is clear the AI can function in a field battle without a capture point. I am not sure we'll be able to remove them for sieges - and honestly I am wondering if the AI didn't prioritize the city center in previous total wars anyway.

As a test I did a Roman campaign and force marched three units to stand by the etruscans and let them attack me. This is the result: http://i.imgur.com/LdK2rqt.jpg

I reloaded the campaign, regular marched three units. This is the result: http://i.imgur.com/zyedRXj.jpg

It seems thus far that if you are part of a siege (attacking or defending) you get a control point. If the besieger is sallied out against, they have a control point to defend.

If you are caught forced marching, you get a control point. Not going to busy myself to see if fortifying or raiding causes a control point. It's a safe assumption that some or all of those marching abilities does so.If you are NOT caught forced marching, and are not besieging or being besieged, you will not (barring a bug) have a control point battle.


We need to find the db file that decides what gets a control point and what doesn't. I will be looking myself, I was hoping "campaign_variables" had it, but so far I didn't see anything about it there - there was a lot of interesting stuff in that db file, though.

[Ahiga]

battle_types_to_victory_conditions.

Mitch and others of code-savvyness, take a look at that db file. I don't know about you but it sure sounds like the big enchilada to me:

-<battle_types_to_victory_conditions record_uuid="{d873d118-6ae0-458c-9bba-21e73e0c649e}">
<attacker_victory_condition>BATTLE_SETUP_VICTORY_CONDITION_CAPTURE_ALL_KEY_LOCATIONS</attacker_victory_condition><battle_type>capture_point</battle_type><defender_victory_condition>BATTLE_SETUP_VICTORY_CONDITION_CAPTURE_ALL_KEY_LOCATIONS</defender_victory_condition></battle_types_to_victory_conditions>-

VS:

<battle_types_to_victory_conditions record_timestamp="1355311905395" record_uuid="{cb7db88a-3e4d-4f4f-844a-c0a121b409cb}"><attacker_victory_condition>BATTLE_SETUP_VICTORY_CONDITION_KILL_OR_ROUT_ENEMY</attacker_victory_condition><battle_type>classic</battle_type><defender_victory_condition>BATTLE_SETUP_VICTORY_CONDITION_KILL_OR_ROUT_ENEMY</defender_victory_condition>


Not sure if it'll be as simple as switching out conditions. There are for instance two river_crossing_Battles - one is kill or rout, the other is capture location and kill or rout.

[.Mitch.]

I'm going to focus all of my energy in this area as my project. It's quite probably the most pressing issue right now.

I'm pretty sure it'd DB related, all of the different battles types are in: battle_types.xml

The table you mention seems to be a possible answer, setting the capture point battles victory conditions to KILL OR ROUT instead of CAPTURE might just do it.

[Ahiga]

I think I was mistaken, battle_db files seem to deal with custom battles. It's campaign_.

I am looking at campaign_bonus_value_battle_xyz, and am noticing some interesting info. It seems like these are the files that decide variables to the campaign battle - there's stuff about if you are raiding or they are raiding, what faction or culture they are, if you own the territory you are waging the battle in, if you are attacking breaking out of a siege or attacking to sally out, ect. However the only reference that could mean the capture point is 'escorting'.

There doesn't seem to be a clear reference to capture points in the campaign_ files, but it's a start.

Best of luck in it, I'll dig around further tomorrow to see if there's anything.

[Radious]

Removing Victory points would be nice, but since Ai is so obsessed with them i would worry what it wil ldo with battle ai during these battles. Since their only goal is to cap that point and even ignore players army, maybe without the point they will do nothnig or just charge on the same place where VP was before and again ignore everything.

[Augustus Lucifer]

Removing Victory points would be nice, but since Ai is so obsessed with them i would worry what it wil ldo with battle ai during these battles. Since their only goal is to cap that point and even ignore players army, maybe without the point they will do nothnig or just charge on the same place where VP was before and again ignore everything.

Well presumably the AI prioritizes what the battle type declares the objective, so they wouldn't wander around if given a new objective. It looks like there are ways to make capture points impossible to take, such as the method posed by Ahiga in the OP, or the values gate_capture_time, keep_capture_time, etc in _kv_key_buildings.xml, but as you say that would just make the AI go there and sit forever since it still thinks it's the way to victory.

I agree with Ahiga's second suggestion about changing the conditions in battle_types_to_victory_conditions.xml. That file and the victory_conditions.xml are the only ones that reference those in the provided DBs however, so we have to assume they're fed somewhere else currently inaccessible where the AI uses them. One thing to keep in mind though is the victory condition for AI defenders in city sieges is already kill/rout, so if one of the problems being experienced is player on the offensive taking the point and AI hurling themselves at the point, that would seem to happen either way since both the point and the troops that need killing are in one location.

[uanime5]

One problem with removing victory points in siege battles is that if the AI has reinforcements then you'll need to defeat the AI army in the city, then leave the city and attack the reinforcements (assuming that the AI reinforcements just stand around doing nothing because they have no reason to go into the city).

Removing victory points from non-siege battles may also cause problems as the player can use forced march or build defensive fortifications with impunity as you're removing one of the key weaknesses from this strategy.

[Nemo84]

One second problem I've just discovered is that removing victory points can make combined land-sea battles unwinnable for the attacker.

I was playing a campaign with the VP timer set to longer than the battle timer, so no VP victory was possible. My Roman army attacked a Spartan army which got reinforced from a nearby Spartan fleet. There was a VP in the Spartan deployment zone. I beat the Spartan army, the Spartan ships land as well and the crews get massacred, so normally I would have won. Except now I didn't. Why? Because the Spartan fleet contained an artillery ship, and these apparently don't disembark on land. My army contained no artillery units nor naval units, so I couldn't touch his lone ship. Timer runs out, I lose the battle.

These victory points serve a distinct purpose in field battles, and disabling them leads to far more problems than they do. Needless to say I currently have them re-enabled in my campaign.

[MathiasOfAthens]

It would take 200 Hours to run through 800,000,000 points.