The Number Crunching Thread

That's not much and somewhat of a disappointment. Certainly will not change the numbers much.

Yeah it has little influence. The +25% increase is applied to excess food only so to fit it into the analysis, just multiply the whole food formula by 1.25. Hospitals work slightly differently but you can get a good approximation by simply halving g (it's not quite perfect because what you will actually keep when going to level n+1 is g(n)/2, not g(n+1)/2)
 
Well the issue is, except for France/the Liberty Policy, each newly founded city produces 0 culture.

There is a lag time from founding of the city (and the policy price boost) to the city reaching cT.

Just like founding a city post colluseums will Always increase your happiness... the city makes an initial hit to happiness, followed by an increase once the buildings are built.

In Civ IV AND Civ V
Additional Undeveloped city=drag on your empire
Additional developed city=bonus for your empire

Yes, you're right - in my example, if you're getting just 1 culture per city and no city states, that's an abysmally slow speed, so I suppose it does make intuitive sense that any additional culture from a new city would help speed it up.
 
Basically yes, if you invest (the correct amount can be found throughout this thread :) )

@ alpaca,

I was thinking the same thing about the WLTKD, but I wasn't sure anymore.
Can we assume the same goes for the Floating Gardens bonus then ?
 
To answer your original question Bibor: Short term you'll have very little impact in SP pace. However if you don't keep up with the increases per SP level, you'll eventually start lagging.

(Pretty hard to translate numbers in my head to something clear for other people to understand, thank god I'm not a teacher :) )

You mean:

Napoleon only takes care of the "number of cities inflation" part of policy prices and you still need buildings to counter the "number of policies inflation".
 
Basically yes, if you invest (the correct amount can be found throughout this thread :) )

@ alpaca,

I was thinking the same thing about the WLTKD, but I wasn't sure anymore.
Can we assume the same goes for the Floating Gardens bonus then ?

Floating gardens are amazing as long as you have a lake(s) around and pretty good as long as you want your city to actually grow. It also costs just 1 not 2 GPT, compared to the building it replaces. It's also ridiculous if you have a 4-5 hex lake that you can put your city near to.

However, I realised painfully that a Lighthouse does not improve lake food as it used to in CIV4.
 
Here's a link that might be of interest for the culture/city analysis. I just posted it in its own thread, but that was before I saw this one.
 
At least they added the extremely handy Seaports now.
As a small point in favor of the Lighthouse: it increases The Great Barrier Reef tiles with 1 food :)

You mean:

Napoleon only takes care of the "number of cities inflation" part of policy prices and you still need buildings to counter the "number of policies inflation".

Absolutely correct.
 
Basically yes, if you invest (the correct amount can be found throughout this thread :) )

@ alpaca,

I was thinking the same thing about the WLTKD, but I wasn't sure anymore.
Can we assume the same goes for the Floating Gardens bonus then ?

I don't know, I never played the Aztecs yet. However, if that is the case the floating gardens are very lame compared to a watermill because 15% of your growth will rarely even match a flat +2 food per turn (would require a food surplus > 13)
 
Border Growth

The cost of a border growth is, with nT the number of tiles already claimed

20 + (10 (nT - 1))^1.1

I noticed today that when you conquer a city, nT gets reset to 0. I have no idea what happens on re-conquest though.
 
Wow, just noticed a huge oversight on my part: representation isn't "+1 culture to all cities connected to the capital"; but instead "+1 culture per city". This makes it slightly better than I thought as it applies both instantly and to the capital as well :)

"Assumptions are the mother of all :):):):)-ups". I assumed it was, and therefore was wrong
 
I noticed today that when you conquer a city, nT gets reset to 0. I have no idea what happens on re-conquest though.

The same. Quite stupid if you ask me. I'm ok with the city keeping the tiles but afterwards the growth speed is fairly high. Of course, all cultural buildings are razed on conquest.
 
Unit Maintenance

Unit maintenance is calculated as follows:

c(t,n) = ((j n)//100 (1 + 2t m/1000))^(1 + 2t/(1000*d))

where n = number of units and t = number of turns. // denotes an integer division. j, m and d are constants from GlobalDefines.xml, j = INITIAL_GOLD_PER_UNIT_TIMES_100, m = UNIT_MAINTENANCE_GAME_MULTIPLIER, d = UNIT_MAINTENANCE_GAME_EXPONENT_DIVISOR

With the vanilla settings, the formula then is

c(t,n) = floor(floor(0.5 n) (1 + 16/1000 t))^(1 + 2/7000 t))


Since there's no easy "each unit in turn t will cost this much" here's a table you can use as a rough reference. The first row is the number of turns, the first column the number of units

Code:
					Turns
Units	1	20	50	100	150	200	250	300	350	400
4	2	2	3	5	7	9	11	14	17	20
8	4	5	7	11	15	19	24	30	36	43
12	6	8	11	16	23	30	38	47	57	68
16	8	10	14	22	31	41	52	64	78	94
20	10	13	18	28	39	52	66	82	100	121
24	12	16	22	34	47	63	80	100	122	148
28	14	18	26	40	56	74	94	118	145	176
32	16	21	30	46	64	85	109	136	168	204
36	18	24	34	52	73	96	124	155	191	232
40	20	26	37	58	81	108	138	174	215	261
44	22	29	41	64	89	119	153	193	238	291
48	24	32	45	70	98	131	168	212	262	321
52	26	35	49	76	107	142	184	231	287	350
56	28	37	53	82	115	154	199	251	311	381
60	30	40	57	88	124	166	214	270	335	411
64	32	43	61	94	133	177	229	290	360	442
68	34	45	64	100	141	189	245	310	385	473
72	36	48	68	106	150	201	260	330	410	504
76	38	51	72	112	159	213	276	349	435	535
80	40	54	76	118	167	225	292	370	461	567
84	42	56	80	124	176	237	307	390	486	598
88	44	59	84	130	185	249	323	410	512	630
92	46	62	88	137	194	260	339	430	537	662
96	48	64	92	143	203	273	354	451	563	694
100	50	67	95	149	211	285	370	471	589	727

Here's an equivalent table detailing the cost per unit

Code:
					Turns
Units	1	20	50	100	150	200	250	300	350	400
4	0.50	0.50	0.75	1.3	1.8	2.3	2.8	3.5	4.3	5.0
8	0.50	0.63	0.88	1.4	1.9	2.4	3.0	3.8	4.5	5.4
12	0.50	0.67	0.92	1.3	1.9	2.5	3.2	3.9	4.8	5.7
16	0.50	0.63	0.88	1.4	1.9	2.6	3.3	4.0	4.9	5.9
20	0.50	0.65	0.90	1.4	2.0	2.6	3.3	4.1	5.0	6.1
24	0.50	0.67	0.92	1.4	2.0	2.6	3.3	4.2	5.1	6.2
28	0.50	0.64	0.93	1.4	2.0	2.6	3.4	4.2	5.2	6.3
32	0.50	0.66	0.94	1.4	2.0	2.7	3.4	4.3	5.3	6.4
36	0.50	0.67	0.94	1.4	2.0	2.7	3.4	4.3	5.3	6.4
40	0.50	0.65	0.93	1.5	2.0	2.7	3.5	4.4	5.4	6.5
44	0.50	0.66	0.93	1.5	2.0	2.7	3.5	4.4	5.4	6.6
48	0.50	0.67	0.94	1.5	2.0	2.7	3.5	4.4	5.5	6.7
52	0.50	0.67	0.94	1.5	2.1	2.7	3.5	4.4	5.5	6.7
56	0.50	0.66	0.95	1.5	2.1	2.8	3.6	4.5	5.6	6.8
60	0.50	0.67	0.95	1.5	2.1	2.8	3.6	4.5	5.6	6.9
64	0.50	0.67	0.95	1.5	2.1	2.8	3.6	4.5	5.6	6.9
68	0.50	0.66	0.94	1.5	2.1	2.8	3.6	4.6	5.7	7.0
72	0.50	0.67	0.94	1.5	2.1	2.8	3.6	4.6	5.7	7.0
76	0.50	0.67	0.95	1.5	2.1	2.8	3.6	4.6	5.7	7.0
80	0.50	0.68	0.95	1.5	2.1	2.8	3.7	4.6	5.8	7.1
84	0.50	0.67	0.95	1.5	2.1	2.8	3.7	4.6	5.8	7.1
88	0.50	0.67	0.95	1.5	2.1	2.8	3.7	4.7	5.8	7.2
92	0.50	0.67	0.96	1.5	2.1	2.8	3.7	4.7	5.8	7.2
96	0.50	0.67	0.96	1.5	2.1	2.8	3.7	4.7	5.9	7.2
100	0.50	0.67	0.95	1.5	2.1	2.9	3.7	4.7	5.9	7.3

Edit: Small caveat, I still have to check if the game floors or rounds the values after calculating the function.

Edit2: Ok, the game floors so I updated the tables
 
IBorder Growth

The cost of a border growth is, with nT the number of tiles already claimed

20 + (10 (nT - 1))^1.1

Does that include 'gold bought tiles'?
and what is the formula for the cost of those tiles?
 
Does that include 'gold bought tiles'?
and what is the formula for the cost of those tiles?

No, gold tiles don't increase culture-cost, but the reverse is true. As far as I know, every tile the city has in excess of it's inner ring adds 5 to the gold buy cost for the next tile, no matter whether it was claimed by culture or bought. I don't know how the base cost is calculated, but it seems like flat terrain and the tiles your city would culturally claim next are cheaper than hills and forests.
 
Disclaimer: I'm horrible at math and have next to zero skill with spreadsheets.

That said, I was wondering if anyone could put some of these calculations together into spreadsheets for relative ease of use to calculate some of the costs. I'm specifically thinking about something where you could input or just cross-reference the number of social policies you have with the number of cities you have founded/annexed to determine when you'll next be getting an additional SP.

I usually hold off for quite awhile to get Renaissance and Industrial age SPs, but I'm never quite sure how many I'll be able to get all at once when I qualify. I'd love to just type in 2 purchased SPs and 5 cities and get a number. Is that possible?
 
Alpaca, are you sure the unit maintenance tables correspond to the current formula listed? Shouldn't the total cost go up linearly with the amount of units, with just a +-1 for rounding as the number of units is only included as a final multiplier? Instead, looking at, say, the turn 300 column at 20 unit intervals, the cost goes up by 82, 92, 96, 100, 101. Quickly checking with a calculator I get ~67.432 per 20 units from the formula listed (0.5n=10).


Ceterum censeo, this is a horribly kludgy way of defining unit maintenance costs. It's the only thing in game affected by inflation, and if it's supposed to model more modern units requiring more maintenance, that would be better done by raising the costs either by the amount of techs or defining it separately for each unit type.
 
Alpaca, are you sure the unit maintenance tables correspond to the current formula listed? Shouldn't the total cost go up linearly with the amount of units, with just a +-1 for rounding as the number of units is only included as a final multiplier? Instead, looking at, say, the turn 300 column at 20 unit intervals, the cost goes up by 82, 92, 96, 100, 101. Quickly checking with a calculator I get ~67.432 per 20 units from the formula listed (0.5n=10).


Ceterum censeo, this is a horribly kludgy way of defining unit maintenance costs. It's the only thing in game affected by inflation, and if it's supposed to model more modern units requiring more maintenance, that would be better done by raising the costs either by the amount of techs or defining it separately for each unit type.

No, it doesn't increase linearly in the amount of units. The formula for a fixed number of turns is c = k*n^k' where k and k'>1 are constants for each turn. This means the formula increases in a higher-than-linear fashion with the number of units, the more strongly the further you are in the timeline. For turn 300, k' = 1.085 approximately so if you go to high unit number you'll definitely notice the non-linearity. The tables were created using the formula, by the way, they are not in-game values. I think they match, but didn't check a huge number of them (instead, I conducted controlled experiments to figure out the parameters).

I agree that this solution for maintenance isn't great.
 
No, it doesn't increase linearly in the amount of units. The formula for a fixed number of turns is c = k*n^k' where k and k'>1 are constants for each turn. This means the formula increases in a higher-than-linear fashion with the number of units, the more strongly the further you are in the timeline. For turn 300, k' = 1.085 approximately so if you go to high unit number you'll definitely notice the non-linearity. The tables were created using the formula, by the way, they are not in-game values. I think they match, but didn't check a huge number of them (instead, I conducted controlled experiments to figure out the parameters).

I agree that this solution for maintenance isn't great.

Ah yes, just noticed that it's (n*k)^(k') , not (n)*(k^k') as I first read.
 
Top Bottom