1.17 released

I see that in HR1.17 the length of time that a city remains unhappy after capture scales with game speed. Yes? On Odyssey speed, in the Ancient Era, a city of size 4 took 15 turns to be pacified. That is perhaps a fair scaling, but I feel that it reduces playability. However, I haven't any better suggestions to offer! Has anyone else an opinion on this?
 
Small bug:

AI_DIPLO_DEMAND_TRIBUTE_EQUAL_POWER_CLOVIS_1


Reminder:

You were going to see whether the colour of coast tiles could be made different from that of ocean tiles (I'd like the latter to be much darker).


Check:

Skirmishers are causing collateral damage (after Mathematics). Is this intended?
 
Perhaps I was not clear enough. When the conversion is n for 1, I presume that there is a calculation that runs something like:

gold := gold + floor[hammers/n], hammers := 0.

What I intended was:

gold := gold + floor[hammers/n], hammers := hammers - n * floor[hammers/n].

So, I didn't intend it as rounding up, just a mechanism to ensure that there is no wastage due to the side effect of the integer arithmetic. Otherwise, I would be micromanaging to ensure that my city produced a multiple of n hammers.


However, the result is still far in excess of the intention: At 3 for 1, 5 hammers yield 1 gold. I wish that there was a better way to implement this ...

Ah I get what you mean now. That would require changing the calculation itself though which I cannot do. I suppose I could 'refund' the amount determined by the original formula and then apply the new formula but the AI is still going to use the old formula in it's calculations. It might also mess up the financial advisor and city screen.

I see that in HR1.17 the length of time that a city remains unhappy after capture scales with game speed. Yes? On Odyssey speed, in the Ancient Era, a city of size 4 took 15 turns to be pacified. That is perhaps a fair scaling, but I feel that it reduces playability. However, I haven't any better suggestions to offer! Has anyone else an opinion on this?

Yes it now scales with gamespeed, at the same rate as most other factors of game speed (number of turns in particular):

[TABLE="head;width=100px"]Gamespeed|Scaling
Quick|50%
Normal|100%
Epic|150%
Saga|200%
Marathon|250%
Odyssey|300%[/TABLE]

Small bug:

AI_DIPLO_DEMAND_TRIBUTE_EQUAL_POWER_CLOVIS_1

Found a couple more like this too. Fixed for 1.18, thanks.

You were going to see whether the colour of coast tiles could be made different from that of ocean tiles (I'd like the latter to be much darker).

I had a go at it for 1.17 but discovered that the final colour of coasts and oceans is produced by several different textures overlaying and that it's all too easy to screw up the blends between them and the land. I inadvertently created some strange looking beaches. A much trickier task than I expected but I'll give it another shot for 1.18.

Skirmishers are causing collateral damage (after Mathematics). Is this intended?

Yes. I intend to review the Skirmisher at some point, it's role, mechanics, and maybe even make it its own class of units. I'd really like to split it into Slingers (ancient) and Javelineers (classical) and possibly make Longbowmen the medieval equivalent (with Crossbowmen taking over as the era's city defense unit). The concept might extend into later eras too (Grenadiers for example). Biggest hindrance to doing this atm is available artwork.
 
I'd argue for it being the other way around- javelins as an ancient skirmishing weapon, slings as a classical one.

Also, if you're really hard up for artwork... what's lacking? You'd use the same skirmisher artwork for the javelin guys, all you need are slingers. Which ones are missing?
 
Ah I get what you mean now. That would require changing the calculation itself though which I cannot do. I suppose I could 'refund' the amount determined by the original formula and then apply the new formula but the AI is still going to use the old formula in it's calculations. It might also mess up the financial advisor and city screen.

The refund (exactly 1 or 2 hammers) wouldn't ever make a substantial difference to strategic calculations, so it doesn't matter if the AI doesn't know about it. But yes, I hear that you're saying that it's nontrivial to implement ... I'll mull over it and see if I can't come up with another idea. Can you point me to the place where the calculations are made?


Yes it now scales with gamespeed, at the same rate as most other factors of game speed (number of turns in particular):

[TABLE="head;width=100px"]Gamespeed|Scaling
Quick|50%
Normal|100%
Epic|150%
Saga|200%
Marathon|250%
Odyssey|300%[/TABLE]

Thanks for these numbers --- is there a Civilopedia article on gamespeed?


This reminds me of something else that I realised last night. The interest that Financial civilizations get also scales with gamespeed --- it's only paid out every 6 turns on Odyssey. (I presume that this scaling ensures that it it once per turn for Quick.) However, I think that the reasoning leading to scaling with gamespeed is actually not appropriate for the interest, as no other financial transactions scale with gamespeed. A gold-for-resources trade is paid every turn. The commerce generated by working tiles is calculated every turn. Even the effect of `inflation' (whatever that means ... would be nice to eliminate this kludge) is calculated every turn.

To illustrate the result: To maximise the benefit of this bonus, the player should maintain a minimum balance in their treasury. In the Ancient Era, where the cap is 10 gold per `turn', and the interest rate is 1%, that means having a minimum balance of 1000 gold. The result of that 1000 gold is the player's budget is strengthened by 10 gold per `turn'. However, on Odyssey speed, that becomes in practice 10 gold every 6 turns, which is a very poor investment: 0.17% (and this rate doesn't increase with the different caps that arise as the Eras progress). This is just not worth the lost opportunity cost: Hold 1000 gold in the treasury so as to generate less than 2 gold per turn income. The Financial trait becomes very weak, and is possibly actually a hindrance. The reason that it's not fair is that all other income and expenses are calculated per turn, and don't scale with gamespeed.


Whilst on the topic, a side effect of the mechanism of the caps is that they lead to a one-dimensional strategy. There simply isn't any alternative to maintaining that minimum balance (if the returns outweigh the lost opportunity costs!). Perhaps the caps could be better removed altogether? The 1% interest earnt fairly represents a fair return on lost opportunity cost (it's money not spent on technologies, upgrades, etcetera), and doesn't having a large treasury attract the unwanted attention of one's stronger competitors? Yes, it might lead to exploits, but I'd like to be able to demonstrate that. Could you point me to the place where the caps are implemented?
 
I'd argue for it being the other way around- javelins as an ancient skirmishing weapon, slings as a classical one.

Really? Both weapons were used contemporaneously but I always hear of javelins in classical warfare and rarely slings.

Also, if you're really hard up for artwork... what's lacking? You'd use the same skirmisher artwork for the javelin guys, all you need are slingers. Which ones are missing?

Basically there are several different European ones (Balearic, Thracian, Slavic), an Incan one and that's it.

The refund (exactly 1 or 2 hammers) wouldn't ever make a substantial difference to strategic calculations, so it doesn't matter if the AI doesn't know about it. But yes, I hear that you're saying that it's nontrivial to implement ... I'll mull over it and see if I can't come up with another idea. Can you point me to the place where the calculations are made?

The percentages are set in CIVProcessInfo.xml but the calculation itself is hidden away inside either the DLL or BTS itself. Anything we want to attach to it would have to be done via onCityDoTurn in CvEventManager.py. CvMainInterface.py and BugFinanceAdvisor.py would also have to be adjusted to show the changes. Even a simple change would require a ton of testing.

Thanks for these numbers --- is there a Civilopedia article on gamespeed?

No. I should probably add one sometime. It's all very intertwined with map size and difficulty level though.

This reminds me of something else that I realised last night. The interest that Financial civilizations get also scales with gamespeed --- it's only paid out every 6 turns on Odyssey. (I presume that this scaling ensures that it it once per turn for Quick.)

Yes.

However, I think that the reasoning leading to scaling with gamespeed is actually not appropriate for the interest, as no other financial transactions scale with gamespeed. A gold-for-resources trade is paid every turn. The commerce generated by working tiles is calculated every turn.

When I first added Financial's interest bonus it was received every turn regardless of game speed. It was insanely overpowered on the slower game speeds, even with the caps. Drastically stronger than any other trait and Financial leaders were winning the game almost without fail. Never underestimate the power of compound interest.

However, I am going to raise the caps in 1.18 (to 100 per era) and see how that goes. I think that's a much safer and probably more effective solution. If that goes well we can consider removing the caps altogether.

Even the effect of `inflation' (whatever that means ... would be nice to eliminate this kludge) is calculated every turn.

Can't be eliminated without completely overhauling how maintenance is calculated and how everything is priced. A balancing nightmare.
 
When I first added Financial's interest bonus it was received every turn regardless of game speed. It was insanely overpowered on the slower game speeds, even with the caps. Drastically stronger than any other trait and Financial leaders were winning the game almost without fail. Never underestimate the power of compound interest.

However, I am going to raise the caps in 1.18 (to 100 per era) and see how that goes. I think that's a much safer and probably more effective solution. If that goes well we can consider removing the caps altogether.

Please reconsider. The scaling with gamespeed really is a problem. The caps are easier to modify.

Apart from being incompatible with other turn-based financial transactions, scaling with gamespeed means that the Financial trait is actually worthless on Odyssey. The return per 1000 gold invested is currently less than 2 gold per turn. I can get a greater benefit in many ways (perhaps by building wealth!), and that without the lost opportunity cost: The 1000-per-Era gold could be better invested in supporting deficit spending, unit upgrades, technology purchases, etcetera, whilst at the same time not prompting my more avaricious neighbours to make arrogant demands.

If it's not scaled with gamespeed, the current caps lead to the Financial trait being worth only 10 gold per turn in the Ancient Era, and more generally 10 gold per turn per Era. The compounding effect is only visible whilst the treasury is being built up to the `cap-limit' of 1000 gold per Era. If 10 gold per turn per Era really is overpowered (for the AI, for the human player?), then the level of the caps needs to be lowered rather than raised. (Hmm: Perhaps it would be simpler to give the Financial player a treasury bonus of X gold per turn per Era, not call it capped interest, and not require an investment? Not elegant!)

----------------

Aside: I've done some more play-testing and I now have an opinion on the scaling of pacification with gamespeed: The conqueror likes it! The negative effect of having to garrison captured cities longer is nicely offset by the fact that they aren't so quickly smothered by the culture of nearby not-yet captured cities.
 
Workers of the World, unite under the red flag!

HR1.17: See the attached save file: Somewhere around 400BC the Workers started appearing as red splodges. I've not seen this before.
 

Attachments

Please reconsider. The scaling with gamespeed really is a problem. The caps are easier to modify.

I'll have a think about it. Either the scaling or the cap has to stay, removing both would be game breaking.

(Hmm: Perhaps it would be simpler to give the Financial player a treasury bonus of X gold per turn per Era, not call it capped interest, and not require an investment? Not elegant!)

Easier to balance but it removes strategy completely from the equation. Not ideal.

Aside: I've done some more play-testing and I now have an opinion on the scaling of pacification with gamespeed: The conqueror likes it! The negative effect of having to garrison captured cities longer is nicely offset by the fact that they aren't so quickly smothered by the culture of nearby not-yet captured cities.

Good news.

Workers of the World, unite under the red flag!

HR1.17: See the attached save file: Somewhere around 400BC the Workers started appearing as red splodges. I've not seen this before.

I'm not able to load the saved game atm but I'm guessing HC = Huayna Capac? if so, that's due to bad art define for the Mesoamerican worker and is fixed for 1.18.
 
Hi Xyth! I'm a happy fan, enjoying 1.17. Just wanted to leave you a save game/error report. I suspect it's similar to what others have experienced, as it happens in the Medieval Era.

Keep up the excellent work!

Thanks for these. That crash report is related to the display of city buildings on the world map, however I continued your saved game into the 1600s and wasn't able to reproduce it. Does it happen repeatedly (on or near the same turn) for you?

I notice though that you are running BTS 3.17. I don't know if that could be the cause of this particular crash but please update to 3.19 as its bound to be the cause of other problems.



EDIT: I have discovered and fixed one error directly related to Polynesian city art. I didn't cause a crash for me but it may have done on 3.17.
 
Now, we're at it with bad art assets: In one of my games the Thai rushed me with some startlingly pink horse archers (or horsemen or whatever they're called now).

It was largely my own fault misreading the diplomatic situation, but their hues were kind of unsettling.
 
First: the work you've put into this mod is amazing. I've played through a whole game with it and it is phenomenal and complex. Well done.

Is there any chance in the future of the addition of a RFC stability mechanic or something like the revolutions in the Revolution mod? I'm playing on mac, and I wonder if it would require a custom dll? I don't know if you've already addressed this (I'm new to CivFanatics) so please forgive me if you've done so already.

Again, thanks for creating this incredible mod!
 
Yes, something like that would require modifications he can't do without the dev kit, which he doesn't have access to and probably never will.
 
Now, we're at it with bad art assets: In one of my games the Thai rushed me with some startlingly pink horse archers (or horsemen or whatever they're called now).

It was largely my own fault misreading the diplomatic situation, but their hues were kind of unsettling.

Hmm, I'm not able to reproduce that. Pink units means a texture is missing or not referenced properly but both the Thai Horseman and Horse Archer are showing up the expected colours for me. What graphics settings are you using?

First: the work you've put into this mod is amazing. I've played through a whole game with it and it is phenomenal and complex. Well done.

Thank you.

Is there any chance in the future of the addition of a RFC stability mechanic or something like the revolutions in the Revolution mod? I'm playing on mac, and I wonder if it would require a custom dll? I don't know if you've already addressed this (I'm new to CivFanatics) so please forgive me if you've done so already.

Yes, something like that would require modifications he can't do without the dev kit, which he doesn't have access to and probably never will.

The Revolutions mod relies heavily on a custom DLL. As Mac BTS does not support these there is no way I can make it function in HR. That said, it's a concept I want to look into at some point and it might be possible for me to implement something vaguely similar, though much simpler. I'd need to do a lot of testing before confirming either way though.
 
I understand that you made the change so that on cities can be founded on Ice.

However, that leaves a number squares with special resources that can never be exploited.
Annoying.
I wonder whether there is any way to change that.
 
I understand that you made the change so that on cities can be founded on Ice.

However, that leaves a number squares with special resources that can never be exploited.
Annoying.
I wonder whether there is any way to change that.

Please read the following post and subsequent discussions to know why this was done and ideas for alternatives. Input is appreciated.

 
Thanks for these. That crash report is related to the display of city buildings on the world map, however I continued your saved game into the 1600s and wasn't able to reproduce it. Does it happen repeatedly (on or near the same turn) for you?

I notice though that you are running BTS 3.17. I don't know if that could be the cause of this particular crash but please update to 3.19 as its bound to be the cause of other problems.



EDIT: I have discovered and fixed one error directly related to Polynesian city art. I didn't cause a crash for me but it may have done on 3.17.

I continue to crashes in the Middle Ages as well. Still playing Kongo and Anasazi mostly. But they don't generally repeat, so I'm ignoring them.

BTW, speaking of Skirmishers, the Anasazi Atlatl is an awesome early game unit, which can be completely dominant in the very hilly terrains spawned on Terra maps.
 
As with 1.16, I will attempt to increase the default number of civs on Massive World size for my Windows box (just a personal preference). To add some spice I won't (at least initially) unpack the custom assets folder: let's celebrate the work Xyth has done in cleaning them up!

Hopefully you see some decent performance/memory improvements. Loading time with the unit files unpacked is a certainly a lot faster than it used to be, but it's tricky to measure how much difference this optimization it makes in game, only that there should be some. I'd be interested in your findings.

I have an delightful game in progress. Monarch Level; 24 AI Civs on Old World Start; Raging Barbs, Aggressive AI and Fewer Religions. Current turn is 320/600. It's too early, but so far no performance issues save some turn lag.

The one thing I would like to adjust are the number of religions. Is it possible to adjust locally the "Fewer Religions" to 1/2 rather than 2/3?

Having reached the Medieval Era, I have more confidence that my rig's memory allocation will hold to game's end. That is of course if I can make it that far: Civ with these settings has become quite a challenge!
 
Back
Top Bottom