Not necessarily, no. It's not just a matter of preference; that kind of test is functionally different from the current Happy*100 / Unhappy test I'm using. Not sure I want to go that way, it pads out pop power quite a bit.
G
New Beta Version - March 14th (3-14)
- Thread starter Gazebo
- Start date
- Status
Not necessarily, no. It's not just a matter of preference; that kind of test is functionally different from the current Happy*100 / Unhappy test I'm using. Not sure I want to go that way, it pads out pop power quite a bit.
G
tu_79
Deity
Well, you could try a run or two. The metric is not much different than the one you are using now. Happy faces / (Happy+Unhappy faces). You don't even need to force a minimum value on any of those, since there's always at least one happy face. On paper it sounds really good. Maybe on practice is not better...
Well, if you're just using adjustable breakpoints, it's sort of just a UI thing, as one is a 1:1 function of the other.
Let Y = Happy/(Happy+Unhappy) and X = Happy/Unhappy
X+1 = (Happy + Unhappy)/Unhappy
1/(X+1) = Unhappy/(Happy + Unhappy)
1 - 1/(X+1) = (Happy + Unhappy)/(Happy + Unhappy) - Unhappy/(Happy + Unhappy)
X/(X+1) = Happy/(Happy + Unhappy) = Y
Essentially, you can convert a "Gazebo" number to a "DeAnno" number by X/(X+1). An example is a Gazebo number of 2 (200%) translates to 2/(2+1) = 2/3 (67%) in my lexicon. The function is 1:1 and monotonically increases, so if you only care about break points for the variable (ie, being higher or lower than set percentages), you can just adjust those break points for your function of choice. The main difference between the numbers is that mine has a Range from 0 to 1 and Gazebo's has a Range from 0 to Infinity, which seems a little more unwieldy.
Of course this isn't quite true if you want your break points to be "pretty", as breakpoints which are pretty in one set of numbers will be ugly in the other and vice versa. Our scales are most different when Happy > Unhappy though, and since that isn't the part of the scale we're most interested in the divergence isn't too extreme.
tu_79
Deity
It's not prettiness what we need. It's more of a sense of how close to the cliff we are. I want to know for how long I can sustain a war if I had to. Or how many cities I can expect to conquer in a row, before needing to care about their development.
For this, we need some testing, cause the variance of the unhappiness in such cases is not linear. War weariness, for example, is quiet for a while, and then it bursts. Any extra city is increasing the requirements for every city, so the bigger the empire, the bigger the requirement jumps. All these together could make unorthodoxal Gazebo's way of calculating happiness more useful... or not. I'm not confident enough to say in advance if any metric will provide a better game experience.
One think looks certain: people are enjoying more the game when happiness is more forgiving. On the other hand, that might be causing faster (read unbalanced) domination games.
Edit. But I have to admit that happiness ranging from 0 to 100% is pretty.
New beta inbound. This one is NOT savegame compatible, unfortunately. Now that I know the 'local' system is staying, I took the opportunity to serialize some values to limit UI hiccups. It means that new memory elements are in play, so no savegames, sorry. Furthermore, underlying changes to happiness system would royally screw players on the existing mechanics.
If there aren't any major bugs this will be the code basis for the next official version.
Link: https://mega.nz/#!3dViUAzZ!_CG0wBVfhUeEbwNJgYG-nI6eyg2X7Adar7xMYOMGSbc
G
Code:
Major changes here:
- Assorted bugfixes from ilteroi and myself
- Some github issue resolution and cleanup
- New luxury system, dropped the pop scaler and local luxury buff and now all luxury 'power' comes through the empire level.
- All unique owned or traded for luxuries are worth 2 happiness each, period. Simple!
- Empire happiness is no longer distributed 'modulo' - is now averaged across all cities, with the division remainder handed out one by one to each city in succession, starting with Capital and continuing until no remainder, well, remains.
- New empire-wide happiness function - is now a happy/unhappy comparison, with parity of happy-unhappy a 100% value for your empire happiness.
- New handicap function - there are now two handicap values, one that is a larger empire-wide value, and another that just affects your capital. Scales based on difficulty.
- New ratios for unhappy/very/super are 60/40/20 (were 75/50/25)
- Adjusted happiness modifiers and metrics to account for other changes
And only balance:
- Shoshone encampment tech bumps moved to gunpowder/riflingIf there aren't any major bugs this will be the code basis for the next official version.
Link: https://mega.nz/#!3dViUAzZ!_CG0wBVfhUeEbwNJgYG-nI6eyg2X7Adar7xMYOMGSbc
G
I'm fine with your happy/unhappy method if you wanna do it that way. I just don’t like the cap at 100%. Forget about it being pottenty misleading. It’s just that the difference between 100:10 and 10:10 is really big, and you should adjust your priorities accordingly. Will we at least be able to see actually uncapped ratio of happy to unhappy in the UI?
Alright I guess that will have to be good enough for me. If enough people complain maybe you will change it but hovering over the smiley face isn’t too tough.
BiteInTheMark
Deity
If too fast domination victories are the concern, there's still the option to stay with a forgiving local happiness system but with increased unhappiness by war weariness and City conquest. (would be probably most enjoyable for players of low difficulty)
Or the most simple solution:
Remove the war weariness reduction in the authority tree. If war weariness is the way to counter warmongers, there shouldn't be a so early way to counter this mechanic. I think, 33% more war weariness should really hurt enough to compensate a forgiving local happiness system.
tu_79
Deity
Yes, that's my concern. But I'm not a warmonger, so maybe the aggressive players should propose in which way they prefer to be delayed in a way that they can still enjoy the game.
civplayer33
King
- Joined
- Sep 11, 2017
- Messages
- 965
This shouldn't be a problem anymore if you have the newest version, which I published here (it should also be included in an updated version of the installer, though I didn't check that, and it should be part of the next release).
The entire tree should be visible if you have the UI size option set to small (activated checkbox), which you can now toggle with the key 'D' as well. The only case where some inconsequential parts of the Promotion Tree UI aren't visible is the scroll bar and the top label bars if using the lowest vertical resolution (768), but even then all the promotion buttons can be viewed and clicked, at least when I tested it at the lowest resolution Civ 5 allows (1024x768).
I'm pretty busy these days so I don't check here often, which is why the late reply...
Cat-du-fromage
Warlord
- Joined
- Nov 17, 2018
- Messages
- 183
as a warmonger player, i don't really like the idea of the removal of the war weariness, maybe a nerf? (25 > 15 for exemple). but authority is all about war, if you remove the war weariness bonus, the tree would lost one of his core.
this is all speculation right now, we need to test if warmongering is really a problem with the new beta
PS: One thing that could be better if you want to hurt warmongers, is maybe to increase the bonus "anti-warmonger".
Aristos
Lightseeker
@Gazebo : just to make sure before making a point... happy + unhappy = total pop, correct?
If yes, then the new method is counter-intuitive, and may even be too soft. Let's take your "very" value, 0.40...
H/U = 0.40 .... (1)
H + U = T ...... (2)
(1) into (2): 0.4 U + U = T = 1.4 U
Therefore, U = T/1.4 ~ 0.71 T --> means "very unhappy".
So, what you call "very unhappy" is 71% of the total population is unhappy... all government buildings are long gone in flames at that level man, even in Latin America.
Point is: the empire wide happiness function should be a function of happy/total, both to make it intuitive (100% becomes the "utopia" level, everyone is happy) but also more in line with realistic results. Me thinks.
If yes, then the new method is counter-intuitive, and may even be too soft. Let's take your "very" value, 0.40...
H/U = 0.40 .... (1)
H + U = T ...... (2)
(1) into (2): 0.4 U + U = T = 1.4 U
Therefore, U = T/1.4 ~ 0.71 T --> means "very unhappy".
So, what you call "very unhappy" is 71% of the total population is unhappy... all government buildings are long gone in flames at that level man, even in Latin America.
Point is: the empire wide happiness function should be a function of happy/total, both to make it intuitive (100% becomes the "utopia" level, everyone is happy) but also more in line with realistic results. Me thinks.
There is a decent amount of neutral citizens as well. They are not included in the empire wide happiness calculation.
Aristos
Lightseeker
U sure? That's what I asked... not sure about that though.
Happy + unhappy does not = total pop. It is happy faces made versus unhappy faces made. Does not discretely account for pop, but the amount each city makes is capped by pop.
G
Is it just me who feels that it is rather hard to guess how global happiness pool will be distributed to each city?
It’s literally in the tooltip at the top.
G
- Status
Similar threads
- Locked
