Extra Traits for C2C

[Sgtslick]

I tweaked populist yesterday or day before in the svn. I deleted the -1 per city coz people were saying it hurts too much at the start of the game and for new cities etc. Which I agree with. So sorry about that hyrdo

Im about to make those fixes to the undefined improvements now, I think i'll have one last quick look at the traits. I was thinking I might lower some of the city maintanence penalties generally, I wanted to nerf anti-clerical a bit, and I want to reduce spiritual happies from state religion to +2 instead of +3. Maybe one or two other small things, ill have a look.

 

[Sgtslick]

Agricultural - Workers build improvements with a -20% penalty.

<iWorkerSpeedModifier>-20</iWorkerSpeedModifier>
This is suppose to translate into your workers buildings stuff 20% faster.

Organised - Workers build improvements with a -20% penalty.
Same thing.

Spiritual - -500% reduction to anarchy time when changing state religions.
Shouldn't this read simply 500% reduction to anarchy time, not -500%.
<iReligiousAnarchyTimeModifier>-500</iReligiousAnarchyTimeModifier>

Comparatively, Anti-Clerical is listed correctly with
"500% increase anarchy time for changes in state religion".
<iReligiousAnarchyTimeModifier>500</iReligiousAnarchyTimeModifier>

Politician - -25% reduction to anarchy time for civic changes.
Shouldn't this read 25% reduction.
<iCivicAnarchyTimeModifier>-25</iCivicAnarchyTimeModifier>
Philosophical for instance has a 20% increase in anarchy time (which is listed correctly)
<iCivicAnarchyTimeModifier>20</iCivicAnarchyTimeModifier>

 

[Flinx]

Spiritual - -500% reduction to anarchy time when changing state religions.
Shouldn't this read simply 500% reduction to anarchy time, not -500%.
<iReligiousAnarchyTimeModifier>-500</iReligiousAnarchyTimeModifier>

Comparatively, Anti-Clerical is listed correctly with
"500% increase anarchy time for changes in state religion".
<iReligiousAnarchyTimeModifier>500</iReligiousAnarchyTimeModifier>

While + 500% is six times a given value, what do you expect from a value of - 500%?
If you would have the opposite of raising the given value about five times then you'd like more to reduce a given value to a fifth, thus a mere -80 % would fit better.

Even better is a -100% modifier, reducing any given value to zero.
A 100% modifier doubles any given value, so again the opposite to doubling any value should be the half of any given value.

 

[Thunderbrd]

Agricultural - Workers build improvements with a -20% penalty.

<iWorkerSpeedModifier>-20</iWorkerSpeedModifier>
This is suppose to translate into your workers buildings stuff 20% faster.

Nope... positive if you want a better work speed.

Organised - Workers build improvements with a -20% penalty.
Same thing.

It's reading correctly. Use a positive in the XML for a better build rate and a negative for a worse rate, as indicated by the text.

Spiritual - -500% reduction to anarchy time when changing state religions.
Shouldn't this read simply 500% reduction to anarchy time, not -500%.
<iReligiousAnarchyTimeModifier>-500</iReligiousAnarchyTimeModifier>

Politician - -25% reduction to anarchy time for civic changes.
Shouldn't this read 25% reduction.
<iCivicAnarchyTimeModifier>-25</iCivicAnarchyTimeModifier>
Philosophical for instance has a 20% increase in anarchy time (which is listed correctly)
<iCivicAnarchyTimeModifier>20</iCivicAnarchyTimeModifier>

Alright. True. These ones should be changed accordingly. Noted and will be corrected on next update. Thanks for the spot. Any more?

 

[Thunderbrd]

While + 500% is six times a given value, what do you expect from a value of - 500%?
If you would have the opposite of raising the given value about five times then you'd like more to reduce a given value to a fifth, thus a mere -80 % would fit better.

Even better is a -100% modifier, reducing any given value to zero.
A 100% modifier doubles any given value, so again the opposite to doubling any value should be the half of any given value.

It will safely reduce to a minimum of 0 rnds. But if other modifiers are in play, going over -100% will potentially eliminate any penalties being received as well.

 

[Sgtslick]

Nope... positive if you want a better work speed.


It's reading correctly. Use a positive in the XML for a better build rate and a negative for a worse rate, as indicated by the text.


Alright. True. These ones should be changed accordingly. Noted and will be corrected on next update. Thanks for the spot. Any more?

I just noticed Foriegn aren't receiving their free point per city.
Nor are deceiver's receiving their free per city for that matter

I think the free trade routes are bogus too.
Currently in my game as Meiji (who should get +2 trade routes)
Cities get 1 trade route free if im not mistaken, then I have trails +1 trade route, and im using open borders +2 trade routes.
All these cities have 4 trade routes though, not the 6 they should have.
So seafaring +1 trade routes and foreign +1 trade routes isn't working.

Thats all the examining I got time for today sry.

 

[Thunderbrd]

I just noticed Foriegn aren't receiving their free point per city.
Nor are deceiver's receiving their free per city for that matter

I think the free trade routes are bogus too.
Currently in my game as Meiji (who should get +2 trade routes)
Cities get 1 trade route free if im not mistaken, then I have trails +1 trade route, and im using open borders +2 trade routes.
All these cities have 4 trade routes though, not the 6 they should have.
So seafaring +1 trade routes and foreign +1 trade routes isn't working.

Thats all the examining I got time for today sry.

Those Espionage are probably showing up as 'Base' rather than being specified as being by the Trait. Laziness I know. But check to make sure that it isn't coming in as a 'Base' value.

On the trade routes, are you certain that there are enough cities in the connection web to trade with? I'm not meaning to doubt, just making sure all the variables that could be throwing off the expected results are being taken into consideration. Though there was, as I recall, some additional work done to get those trade route issues coded out so its very possible something in there is incorrect somehow. If those aren't working then it casts some doubt that the civic and building tags for those are, or if they were in the first place.

Either way, I'd be happy to take some time to ensure they ARE working now that we're into a new cycle of course.

 

[Yudishtira]

While + 500% is six times a given value, what do you expect from a value of - 500%?
If you would have the opposite of raising the given value about five times then you'd like more to reduce a given value to a fifth, thus a mere -80 % would fit better.

Even better is a -100% modifier, reducing any given value to zero.
A 100% modifier doubles any given value, so again the opposite to doubling any value should be the half of any given value.

I agree. -500% anarchy time is patently absurd. It reduces a one-turn anarchy to a -4 turn anarchy - it's over 4 turns before you make the civic change.

 

[Sgtslick]

Those Espionage are probably showing up as 'Base' rather than being specified as being by the Trait. Laziness I know. But check to make sure that it isn't coming in as a 'Base' value.

On the trade routes, are you certain that there are enough cities in the connection web to trade with? I'm not meaning to doubt, just making sure all the variables that could be throwing off the expected results are being taken into consideration. Though there was, as I recall, some additional work done to get those trade route issues coded out so its very possible something in there is incorrect somehow. If those aren't working then it casts some doubt that the civic and building tags for those are, or if they were in the first place.

Either way, I'd be happy to take some time to ensure they ARE working now that we're into a new cycle of course.

No the espionage is not showing up as 'base' I checked it a number of ways, and its simply not showing up. Not showing up in the each base, or in the empires total espionage. Im quite sure about the trade routes too.

 

[Thunderbrd]

No the espionage is not showing up as 'base' I checked it a number of ways, and its simply not showing up. Not showing up in the each base, or in the empires total espionage. Im quite sure about the trade routes too.

Ok, since it seems to only be happening for espionage, something else to check is to make sure the gameoption No Espionage is not on... there IS a filter for that. It may be that the filter is not triggering properly on that option but just make sure the game isn't set to that option if you would.

I'll look into the trade routes certainly.

 

[Sgtslick]

Ok, since it seems to only be happening for espionage, something else to check is to make sure the gameoption No Espionage is not on... there IS a filter for that. It may be that the filter is not triggering properly on that option but just make sure the game isn't set to that option if you would.

I'll look into the trade routes certainly.

I never turn off espionage. Its on 100% coz i've used espionage in the game where I checked
By 'not showing up' I mean the bonus - not espionage in general. The 10% bonus you get for 'foreign' for example is there and working, but just not the free points.

 

[Thunderbrd]

Yeah... ok, makes sense then. Still its probably the filter that's not ONLY being used when the No Espionage option is on that is the likely code culprit. I'll take a look into that too

 

[Sgtslick]

Even better is a -100% modifier, reducing any given value to zero.

Yes i will change it to -100% don't stress guys.

--------
Ok so just made a few quick changes, spiritual change to -100%, humanitarian no longer has the -1 per military unit, very stupid idea that was, don't know what I was thinking. And lastly changed around a few upgrade modifiers for improvements, they aren't all 100%'s now (thx koshling).

 

[RidetheSpiral23]

Don't know where to post this so it's going here.

How are the AI coping with the new traits stuff. Specifically the dynamic traits or whatever its being called. Are the AI players tendencies for trait choices based off of the leaders original traits?

 

[ls612]

Don't know where to post this so it's going here.

How are the AI coping with the new traits stuff. Specifically the dynamic traits or whatever its being called. Are the AI players tendencies for trait choices based off of the leaders original traits?

Not well. THere has been very little AI support for this at all, to my great consternation. I'd recommend against using Leveling Leaderheads in any game for now, as it is still under construction.

 

[ls612]

And here are another three trait proposals PLUS a revised Scientific proposal.

New Scientific:

+ (5/10/15)%
Units can Upgrade anywhere.
+ (2/3/4) per Scientist
- 1


New Scientific: (Normal)

+ 10%
Units can Upgrade anywhere.
+ 3 per Scientist
- 1

Philosophical:

+ (15/30/45)% GPP
Reduced National Revolt Risk
Allows commerce from Non-State Religions (bNonStateReligionCommerce)
+ (20/15/10)% XP needed for Unit Leveling.

Philosophical: (Normal)

+ 20% GPP
Reduced National Revolt Risk
Allows commerce from Non-State Religions (bNonStateReligionCommerce)
+ 20% XP needed for Unit Leveling.

Organized:

- (15/30/45)% Civic Upkeep.
+ (10/20/30) Free Military Units
Max (6/5/4) Turns of Anarchy
- (1/2/3) Flammability per Population Point.

Organized: (Normal)

- 30% Civic Upkeep.
+ 30 Free Military Units
Max 6 Turns of Anarchy
- 1 Flammability per Population Point.

Expansive:

+ (2/3/4)
- (5/10/15)% Maintenance from Number of Cities. NOTE: This is a lot more than it seems because all negative from buildings is treated as maintenance.
Reduced Revolution Distance modifiers
+ (20/40/60)% Worker Speed


Expansive: (Normal)

+ 3
- 8% Maintenance from Number of Cities. NOTE: This is a lot more than it seems because all negative from buildings is treated as maintenance.
Reduced Revolution Distance modifiers
+ 30% Worker Speed

 

[Thunderbrd]

Don't know where to post this so it's going here.

How are the AI coping with the new traits stuff. Specifically the dynamic traits or whatever its being called. Are the AI players tendencies for trait choices based off of the leaders original traits?

I don't believe ls612 gets the spirit of the method being employed. But I do admit that the flavors mechanism as a whole has long been in need of an update to include a great many new C2C factors and more and that will be addressed soon. In so doing, it will aid leaders in getting more of what they should want rather than being pigeonholed into a very limited set of defined personality values that currently make the selections they'd want fit into far too wide brackets. There's very little to play on to get them to hone in on more personalized choices at the moment.

In particular, we have a 'FLAVOR_GROWTH' that has been such a vast catchall for so many things including happiness, health, anything that speeds up anarchy, etc... maintaining such a vastly huge generalizing definition that its next to impossible for leaders with this value to have more reason to pick one type of trait over another as almost all traits have cause to have some - strong growth flavor values. We simply need more definitions for flavors and a rather large overhaul for leaders will be required to achieve this.

But I have a whole method in mind that will hopefully drastically simplify our leaders and cut down a LOT on the amount of necessary defining content on leaders that will at the same time increase the leader personality dimension a great deal as well. I'm not sure if I'll be getting to that on this cycle or not but since it seems to be right on the same track of my current plans then I believe I may begin working on that very soon.

ls612's disagreement with me on this is that he feels that a leader should be considering his current game situation when making trait selections and I feel that those selections should be based far more on the leader's personality with minimal consideration for the current situation because the choices are going to influence the leader's long term plans as much or more than any fleeting short term needs. I also don't feel we need another area that adds another minute onto turn times to make a huge inquiry into the entire game situation the leader is currently in so that he makes a pseudo-wise decision as to what trait to select. AI can be done simply and yet effectively at the same time... we just need a few more definitions to make this more possible for them.

They aren't making bad choices now... just ones that at times may seem contradictory to what you'd expect their natures to be - and that's not at all outside the bounds of what the option should allow given that by the very nature of wanting developing leaders, you'd want leaders you wouldn't expect them to be. If you don't select No Positive Traits on Gamestart, you'll get those leaders starting with the classic picks they usually begin the game with and that enforces them to be a bit more like you'd expect.

 

[MaXimillionZero]

ls612's disagreement with me on this is that he feels that a leader should be considering his current game situation when making trait selections and I feel that those selections should be based far more on the leader's personality with minimal consideration for the current situation because the choices are going to influence the leader's long term plans as much or more than any fleeting short term needs.

Considering how bad the AI is at the game, basing any decisions on flavour rather than what's actually smart is gimping the AI unnecessarily even further. IMO it would be best to have the AI consider any new features based on what would improve its situation rather than what suits its flavour in order to stop it falling even further behind human players.

 

[Thunderbrd]

I can understand what you mean Max, but there are far more critical areas to work on in the AI that will make them vastly more effective. If traits are designed correctly, very few situations make one trait better than another to select, particularly in a long term outlook. There are a few cases where some variance exists based on current situation:

1) Era/trait tier: In the earlier game, leaders should be diversifying their trait portfolio but as the game goes on into middle and late eras, they should prefer to develop more advanced versions of their existing traits rather than branching out into other trait lines. (We don't quite have the additional tier traits in the game yet but they should be there by v29.) The AI is setup to want additional tier traits much stronger than a wider variety and I put a tech prereq tag on traits so that we can hold them off from higher tiers until they are technologically ready for them, enforcing greater diversification when they should be diversifying.

2)Seafaring and perhaps other traits to come that only have purpose for nations with a reliance on coastal cities: This is actually taken into consideration now. The value of Seafaring is modified by an evaluation value of what percentage of cities are coastal.


I'm looking for more specific causes but haven't found any fully valid ones yet. I particularly feel that the warfare vs peaceful natured traits should not be made reactive because we'd end up with leaders following all too defined paths. Under such a structure, a player could figure out how to persuade another leader to select traits in a manner that their overall game strategy would suffer from by declaring war, provoking war etc... And it's very contradictory to the idea that these leaders have personality at all.

And in fact, that's part of the problem of making an overly reactive trait selection method. You get leaders who have traits that don't support their overall pre-programmed game strategy (they do have these as setup in their personality profiles already - those could be better refined, sure but they do have them!) For example, you might get an Aggressive Ghandi due to a reaction to being in a state of irritation with a number of AI opponents, even though Ghandi is programmed to rarely, if ever, attack another nation. As a whole game strategy, Aggressive would be a horrible choice for him even if he does launch an occasional attack. Consequently, a Philosophical Montezuma would be equally bad due to the fact that, even though he may be triggered to select Philosophical due to an unusual extended period of peace he's in when the opportunity to select a trait and the evaluation is run, it won't help him to pull off the frequent warfare he will usually seek to wage whenever he feels strong enough to do so.

If we want to see vast improvements to the way the AI plays, improvements in how they carry out their personality defined game strategies and how they make tactical decisions in the game are far more important than overcomplicating the trait selection mechanism! But certainly some better depth of definition IN their personalities is currently in order...


EDIT: There IS one more consideration in making sure the AI leaders are selecting traits in an effective manner and that amounts to the synergy between the traits selected (this is quite important!) Thankfully, most of those synergies exist in keeping with a similar personality based selection. Thus, for example, one of the more powerful classic combinations is Aggressive/Charismatic. Thanks to both being highly motivating for a military flavored leader, this synergy is actually quite likely for them to end up with. In fact, in the whole of the design as it stands was giving the highest of consideration for this sort of synergistic trait selection method (for as a player, I have personally found this to be the most important thing to keep in mind when choosing your trait set!)

I've had record breaking successful games when selecting my leader and traits based on synergies between those traits and playing to the strengths of those synergies and the most mediocre results when playing leaders with little to no synergy between their traits. So this, really, is the most important factor in getting a strong and difficult to overcome AI player.

 

[Sgtslick]

I really don't think it makes all that much difference either way, all the traits are pretty good and its more about player/AI decisions rather than what traits they have, overall I tend to agree with you though thunderbrd.