Suggestions and Requests

[merijn_v1]

Adjusting modifiers can be integrated with the new period system.

 

[Kewlipo]

I have decided to try testing some solutions to the runaway science issue. I went ahead and changed the English UP (which was hidden away in CvCity.cpp, I couldn't find it ) as well as bumping the science modifier for the Americans up to 100 (temporary fix).

English UP: instead of capping the distance at 10 for calculating maintenance, the "Distance from Palace" maintenance cost is simply halved. I will report back some results after I have made some test play-throughs, to see if anything has changed. Additionally, something I forgot to mention: I play on marathon, and have my balance expectations calibrated for that. I know a lot of people prefer to play on Normal, so any solution I find might be inconsequential if it only addresses the problem on Marathon speed. I suspect that some of the runaway science lead may be associated with the English having more time to settle cities, but I am unsure.

If anyone else has any balance suggestions for me to test that may address the problem, please share. Coronavirus has given me plenty of time.

 

[Hickman888]

I have decided to try testing some solutions to the runaway science issue. I went ahead and changed the English UP (which was hidden away in CvCity.cpp, I couldn't find it ) as well as bumping the science modifier for the Americans up to 100 (temporary fix).

English UP: instead of capping the distance at 10 for calculating maintenance, the "Distance from Palace" maintenance cost is simply halved. I will report back some results after I have made some test play-throughs, to see if anything has changed. Additionally, something I forgot to mention: I play on marathon, and have my balance expectations calibrated for that. I know a lot of people prefer to play on Normal, so any solution I find might be inconsequential if it only addresses the problem on Marathon speed. I suspect that some of the runaway science lead may be associated with the English having more time to settle cities, but I am unsure.

If anyone else has any balance suggestions for me to test that may address the problem, please share. Coronavirus has given me plenty of time.

Thanks for taking the time to test out some possible solutions. I'd like to hear back on how your fix works out, and your post on why the Americans get so far ahead in tech. I suspect it's just their abundance of , coupled with the best tech modifier in the game, but maybe there's more to it than that. I'm a marathon player as well, and I've also thought that maybe the British/American (B/A) tech lead can be attributed to the fact they are able to settle their cities faster than they are on Normal/Epic speed, but I have yet to play these civilizations, or play on any other gamespeed, so I really can't say.

I will say though, in my own opinion, the problem isn't necessarily that the B/A are too runaway in technology (but that is part of the problem), it's more that the rest of the world struggles to keep up. Maybe your experience is different, but in my Japanese games I played, the B/A were more or less where they should be at historically, researching the end of the Industrial era/beginning of the Global era around 1880/1890 (probably a decade or two too fast), while the rest of Europe is roughly two whole tech columns behind. I'd like to see the Global era be entered by the world around 1900, so I think the problem lies more with every other civilization...

If you are taking ideas on possible balance solutions, I'd suggest letting the "tech spread" research bonus be applied more liberally, at least in the Industrial era and up. I don't 100% know how this function works, but I believe that if a certain number of known civilizations have researched a technology, that technology becomes easier to research, with the bonus becoming greater as more civilizations research it. Maybe you can change this bonus so that it kicks in if only one civilization researches a technology, and increase from there?

 

[Kewlipo]

I will say though, in my own opinion, the problem isn't necessarily that the B/A are too runaway in technology (but that is part of the problem), it's more that the rest of the world struggles to keep up. Maybe your experience is different, but in my Japanese games I played, the B/A were more or less where they should be at historically, researching the end of the Industrial era/beginning of the Global era around 1880/1890 (probably a decade or two too fast), while the rest of Europe is roughly two whole tech columns behind. I'd like to see the Global era be entered by the world around 1900, so I think the problem lies more with every other civilization...

I think this is a good point. However, I have also experienced a Japanese game where America is researching Globalism/Aviation in 6 turns in 1920 (with England not too far behind), which makes the Japanese UHV impossible. It is my hope that any nerf applied to B/A will not only push them back a decade or two to more historical research times but also reign in the extreme cases like the one I experienced.

If you are taking ideas on possible balance solutions, I'd suggest letting the "tech spread" research bonus be applied more liberally, at least in the Industrial era and up. I don't 100% know how this function works,

I have thought about testing changes with tech spread as well. I have not looked at how it is calculated, but I always thought it would be cool if nations that are closer to each other should receive greater benefit from tech spread. That is, the nations nearer to the tech leader (England) should receive a larger research buff from tech spread than faraway nations. Maybe it is already calculated this way, I do not know. However, I don't think that would necessarily help stop the runaway tech lead that B/A develop. I also don't think the solution is to bring everyone up to B/A tech speeds, but rather to slow B/A down, if that makes sense. Either way, I can still look into experimenting with tech spread, it seems like an interesting avenue to investigate.

Edit: Here is the Japan game I was playing. I looked at the Americans and the results confirmed my suspicions:

Spoiler :

A lot of interesting things to glean from this.

• First and foremost is the confirmation that Japan's tech modifier is ridiculously high, especially when the player is expected to out-tech other civs. Just look at the base comparison: 8100 vs 14400 for the same tech.
• America is running at 100% science. When I have played America, running at 100% science is almost given. There are so few costs involved because their modifiers are the lowest in the game. (50 for civic upkeep, holy moly!)
• Do the AI not suffer from tech-spread costs? How does Globalism cost only 8100 for the Americans? I will have to look at how those costs are calculated, that seems suspiciously low.

America has some of the best land in the game and many of the best wonders, however my gut-feeling is that even if those things are nerfed their modifiers will always allow them to excel extraordinarily. I originally did want to avoid touching modifiers, but in America's case I think nerfing them might very well be the most obvious answer.

 

[Leoreth]

There is no tech diffusion of that kind, instead there is an overall rubber banding effect that is based on the overall amount of beakers researched by each civ. The top 25% get a penalty (scaling with rank, so the full effect only applies to the tech leader) and the bottom 25% get a discount (with scaling in inverse). The effect also scales up with game era, but it could certainly be more pronounced. I was fairly conservative, since it applies to the player too.

There also was a time where America was struggling a lot with tech, and I'm pretty sure I buffed them at some point to help. Looks like they don't have as many problems anymore and could be nerfed a little. I'm looking forward to your results. England is a little bit trickier because they need to be balanced in such a way that they are competitive early on with their small territory but do not get out of control once their empire has expanded.

 

[Kewlipo]

There is no tech diffusion of that kind, instead there is an overall rubber banding effect that is based on the overall amount of beakers researched by each civ. The top 25% get a penalty (scaling with rank, so the full effect only applies to the tech leader) and the bottom 25% get a discount (with scaling in inverse). The effect also scales up with game era, but it could certainly be more pronounced. I was fairly conservative, since it applies to the player too.

There also was a time where America was struggling a lot with tech, and I'm pretty sure I buffed them at some point to help. Looks like they don't have as many problems anymore and could be nerfed a little. I'm looking forward to your results. England is a little bit trickier because they need to be balanced in such a way that they are competitive early on with their small territory but do not get out of control once their empire has expanded.

Oops, I just finished editing my post right after you commented. Is there anything abnormal with the 8100 cost for Globalism, or am I missing something? Also, thank you for explaining tech spread, that clarifies a lot. I will write a full summary for the Americans when I finish doing my testing runs with the English (12 600 AD Canada spawns take quite a while to auto-play on marathon )

Edit: Sorry for continually editing my posts, but I just decided to switch to America in my Japan game and the costs actually changed:

Spoiler :

Before I switched, the cost was 8100. It then changed to the correct cost of 12600.

This is very intriguing. By the looks of what you were saying, this does not seem intentional, and may very well elude to a bug that might be causing the crazy tech leads. How strange.

 

[Leoreth]

Could be AI only modifiers.

 

[Kewlipo]

Could be AI only modifiers.

Apologies for so many questions, perhaps I should move this to some other thread... but where can I find those? I could not find them in the Python assets. AI only modifiers would help explain why the Americans and English tech so incredibly quickly, compounding the already low cost of 12600 (~25% reduction due to civ modifier) with an additional ~35% AI only reduction (12600→8100). Unless, of course, AI only modifiers are the same for all civs, which would mean that we would have to exclude from mechanisms to slow AME/ENG tech speeds.

 

[Temüjin VIII]

I've long been wondering about the AI-only research modifier too. Haven't located it yet either in Python or in the DLL. I guess it's the difficulty level? I usually play on Monarch and I notice the difference.

 

[Leoreth]

Apologies for so many questions, perhaps I should move this to some other thread... but where can I find those? I could not find them in the Python assets. AI only modifiers would help explain why the Americans and English tech so incredibly quickly, compounding the already low cost of 12600 (~25% reduction due to civ modifier) with an additional ~35% AI only reduction (12600→8100). Unless, of course, AI only modifiers are the same for all civs, which would mean that we would have to exclude from mechanisms to slow AME/ENG tech speeds.

I've long been wondering about the AI-only research modifier too. Haven't located it yet either in Python or in the DLL. I guess it's the difficulty level? I usually play on Monarch and I notice the difference.

Right, it would mainly be the difficulties defined in CIV4HandicapInfos.xml, but there may also be some code in the DLL that only applies to AIs.

 

[Kewlipo]

Hello, I completed some work on testing the change I made. I did 6 runs with the default UP (distance capped at 10 for "Distance From Palace" calculation) and modifiers and 6 runs with the test UP ("Distance from Palace" cost halved) and modifiers. I know 12 runs is not a lot but that took me about 7 hours to perform, so please understand.

Here are the modifiers I worked with:

Spoiler :

Here is the analysis:

Spoiler :

I wanted to make sure that England was still a tech leader, which meant I wanted their Proportion of Leading Techs to remain high while also reducing the number that they researched overall. In the tests that I have done, that goal seems successful. The number of techs they researched went down by 1.5 on average with that disparity inevitably growing larger as the game progresses. However, I did notice some problems. In the test version, the English did not seem to settle cities in Canada very much and rarely conquered Indian cities with the TC event. I don't know why this is, perhaps I need to run more tests!

As for the English modifiers, I increased DistanceMaintenance and CitiesMaintenance. Leoreth said he wanted England to be competitive early, so hitting these modifiers made the most sense since they only really hinder a larger (and later) empire. I also changed some of the American modifiers with the intent of stopping the AI from running at 100% science... I will be coming back to these 12 runs when I do my American analysis and see if they glean any interesting results.

I have attached all 12 runs so that others can take a look and perform analyses if they wish. I have also included the English UP and modifiers I worked with, in case anyone else wants to test them out.

 

[TJDowling]

Anyone else think the mechanic by which you buy slaves and they pop up in your capital is a bit silly? Not sure how this could be improved but I think it should be.

 

[Temüjin VIII]

I suggest keeping America's research cost modifier to 75 (or bump it up to 80, at the most), but increasing the maintenance or civic upkeep modifiers a bit further, for testing. Bumping up the research cost modifier itself may prevent America from catching up in its early years, unless that is the intended behavior.

 

[Kewlipo]

I suggest keeping America's research cost modifier to 75 (or bump it up to 80, at the most), but increasing the maintenance or civic upkeep modifiers a bit further, for testing. Bumping up the research cost modifier itself may prevent America from catching up in its early years, unless that is the intended behavior.

With the American modifiers I figured it would be better if I aimed too high rather than aimed too low. If the modifier change harmed America to the point where they cannot catch up or be a tech leader then we would at least know that good modifier values should be somewhere inbetween the original values and the test's values. I think you're probably right with the research modifier being ~80, I'll look into it more when I investigate America specifically.

 

[trevor]

This might be a tricky geopolitical question but shouldn't Taiwan count as part of China for Japan's second goal?

 

[Imperator Knoedel]

shouldn't Taiwan count as part of China

Yes, yes it should.

On a more serious note, IIRC most of these "conquer X civ" goals are about controlling the core of that civ, and Taiwan is only Historical for China.

 

[TJDowling]

On a more serious note, IIRC most of these "conquer X civ" goals are about controlling the core of that civ, and Taiwan is only Historical for China.

Taiwan is only historical part of China not core for good reason. Taiwan was not part of China for most of its history. It shouldn't be in anyone's core but in China and Japan's historical, perhaps Dutch/Spanish but that is an even more insignificant period of time than Japan and China. Also, I think it's kind of a moot point about it being considered China for Japan's UHV as AI China NEVER settles Taiwan anyway. If Japan controls the Eastern coast that extends cultural influence over Taiwan (which it must to fulfill the goal) then it controls Taiwan.

 

[Leoreth]

The real question is shouldn't continental China count as part of Taiwan.

 

[trevor]

Yes, yes it should.

On a more serious note, IIRC most of these "conquer X civ" goals are about controlling the core of that civ, and Taiwan is only Historical for China.

Saigon isn't core for either Thailand or Khmer but is part of the Indochina goal.

Taiwan is only historical part of China not core for good reason. Taiwan was not part of China for most of its history. It shouldn't be in anyone's core but in China and Japan's historical, perhaps Dutch/Spanish but that is an even more insignificant period of time than Japan and China. Also, I think it's kind of a moot point about it being considered China for Japan's UHV as AI China NEVER settles Taiwan anyway. If Japan controls the Eastern coast that extends cultural influence over Taiwan (which it must to fulfill the goal) then it controls Taiwan.

AI China 'settled' it in this 600 AD start so it's not impossible for it to exist.

 

[TJDowling]

AI China 'settled' it in this 600 AD start so it's not impossible for it to exist.

Huh, I've never seen it. Extremely rare at least.