Solving the mystery: Unit Maintenance

[Horizons]

So basically you have to be a computer programmer/university-level mathematics student in order to understand and play Civ 5 at anything but the lowest difficulty levels?

 

[quill18]

This is why I love civfanatics: The number crunching is really valuable.

 

[lordrune]

I'm not convinced inflation is actually in the game. Why? Because the cost for tile maintenance (roads, railroads), and building maintenance does not increase as the game goes on. It stays static, and if you add the numbers up you will always see this.

Now they might have applied inflation to units and not buildings/tiles, but why would they? I don't think they have. I think unit maintenance simply goes up as you build more of them, but as we all know, it's not linear, but something more complicated.

Edit - I'm sorry, I've read the previous page of posts, and it looks like they may have a sort of inflation in place for units after all.
Why they've found this necessary, I've got no clue.

Why can't they just give each type of unit a fixed maintenance value? E.g., 4 GPT for a rifleman, 2 GPT for a worker, 1 for a warrior, 10 for a fighter, 40 for a GDR - would make things so much simpler and more sensible.

 

[Eomer]

I'm not convinced inflation is actually in the game. Why? Because the cost for tile maintenance (roads, railroads), and building maintenance does not increase as the game goes on. It stays static, and if you add the numbers up you will always see this.

Now they might have applied inflation to units and not buildings/tiles, but why would they? I don't think they have. I think unit maintenance simply goes up as you build more of them, but as we all know, it's not linear, but something more complicated.

Inflation is in the game for unit maintanence.

In the endgame of an epic game unit maintanence goes sth. like this 0->4->4->8->8->13->13

 

[player1 fanatic]

Exactly. Just by not doing anything or not getting any growth you'll eventually start getting less and less money, due to unit maintenance factor alone.

 

[Daniel D]

So basically you have to be a computer programmer/university-level mathematics student in order to understand and play Civ 5 at anything but the lowest difficulty levels?

Not at all, just don't build units right up to the edge of breaking even in GPT without having a plan to either gradually increase your monthly income or to rake in decent coin knocking heads together. V seems to point towards a more aggressive style of play than IV...even the "peaceful" win conditions like culture and diplomacy are made easier by going to war.

 

[ohioastronomy]

Not at all, just don't build units right up to the edge of breaking even in GPT without having a plan to either gradually increase your monthly income or to rake in decent coin knocking heads together. V seems to point towards a more aggressive style of play than IV...even the "peaceful" win conditions like culture and diplomacy are made easier by going to war.

Workers are just as costly as carriers as far as I can tell.

This violates a pretty basic rule for a good game: you should have a way of knowing when you're doing something that gets you into trouble. If you just park a bunch of workers to wake up for the occasional tile growth then you're, in effect, paying a ballooning army cost as if you were sitting on a large pile of tanks and mech infantry.

You really notice this if you capture a bunch of AI workers - your maintenance goes through the roof, and it gets extremely bad in the endgame. Given the steep rising unit cost curve, those workers are *expensive*. And there is no clue in the game that they are.

 

[Daniel D]

Workers are just as costly as carriers as far as I can tell.

Yes, both workers and military units count toward your unit totals in the same way. Someone could change this in a mod easily if they wanted to, there's a "NoMaintenance" flag for each unit in the game data.

The manual states that "Units and buildings both have 'maintenance costs' that must be paid every turn." There's no reason to assume that this wouldn't be the case for workers or great people.

Disband or gift useless units.

 

[Anarchist]

The data suggests that this is exactly what's happening. Given that this is one of the first errors taught in any computer science class, I am very dissatisfied.

Hey, ease up - I've been playing the 'software' game* for dozens of years, and still make that mistake on occasion. It is very easy to make, and to miss.

(*Software Game Goal: make as much money as possible without getting bored. Top ranking: Bill Gates)

 

[Durak]

Why can't they just give each type of unit a fixed maintenance value?

I think this is because of the 1 unit per tile rule. Suppose there was no inflation. As the game goes on, you reach a higher GPT. And, without inflation, more gold means you can support a larger army. Now you propose to make more modern units cost more. But the problem is that civ V allows you to build very old units. So this inflation might be a way to prevent us from spamming the map with obsolete units (which, i'm sure, would dazzle the AI something awful).

 

[Ahriman]

So this inflation might be a way to prevent us from spamming the map with obsolete units

So stop players from building any obsolete units. Mostly this already happens.

The only reason why this would even be a problem is in using cheap obsolete units as blockers (eg to wall off city states from being conquered).

Far simpler to just make units of different eras cost different upkeep.

 

[vendur]

The unit costs are stupid. I'm astounded by the stupidity of this design. how can a tank, a battlesip, a worker and a spearmen all be essentially cost the same? Not just dumb in a logical sense, but dumb in the game design as well.

 

[jwallstone]

Hey guys,

There's actually an explicit Inflation column in the GameSpeeds table. So that might be a factor. Even though some have said gamespeed doesn't matter, it probably doesn't matter at first, but maybe after a large number of turns it starts to change.

 

[harry_flashman]

I can't wait till the bug mod comes out. Delete this worker and you will save 56gpt. Top drawer!

 

[BobbleBobble]

I can't wait till the bug mod comes out. Delete this worker and you will save 56gpt. Top drawer!

THIS. Has it been formally announced?

 

[ohioastronomy]

Having workers be expensive is a pitfall for inexperienced players. Most would probably assume that armies were expensive to build and keep up, but that civilian units were not.

 

[ShaqFu]

I'm more annoyed at Great People costing upkeep. Having an Engineer hanging around to rush the Statue of Liberty should not take upkeep, and draining your treasury via idle Great Artists is extraordinarily obnoxious.

I'm okay with Workers/Settlers taking some upkeep, but Great People should be a reward, not a punishment.

 

[Blitz47]

Since I don't have time to play civ5 for an extended period of time, I did the following.

Start a duel map chieftan level, archepelago map as greek.

Fire up debugger/editor mode.

start plop down units, I dropped worker, warrior, companion calvary, etc. and here is what I observed.

military or civilian does not seem to matter, I can have a mix of 4 worker + 6 warrior give the same maintenance cost as 10 worker or 10 warrior.

The increase is in 2 or 4 unit increments.

maintenance per turn my # of units on the map

-1                              4
-2                              6
-3                              10
-4                              12
-5                              16
-6                              18

As far as the effect of running over your supply limit, each additional unit beyond your supply limit reduce your production by 10%.

I then granted myself technology one era at a time. I'm still at turn 1, up until future tech era, I did not observe any maintenance cost change, it's still at -6 eventhough I edited my greek to have all technology already.

Around turn 15, adopt policy alert box came up, I picked patronage, my maintenance cost went up by 1, to -7. I turned on single player end turn and watched the maintenance cost went from -6 to -4, then -7. I guess I or someone else with more time need to do a slowly turn by turn record keeping, rather than let the game run automatically

turn 25, forgot what stopped my auto turn advance, maintenance up to -8
turn 37, civilization ranking alert box, my unit maintenance is up to -9.
turn 39, adopt policy alert, maintenance up to -10

total units used in test, 18 warrior (although you can have any combination and initial observation confirm it's # of units, regardless it's civilian or military units that matters)

 

[Tyrael]

Poking around in the XML solved one part of the mystery.
The reason people keep reporting that every other unit deleted reduced their maintenance is this (I believe):
After applying some (unknown) formula based on game speed, number of turns elapsed, and number of units, the result is then rounded down to the nearest multiple of 7. This is the "exponent divisor value." Every other place that the word "divisor" is used in the XML, it's in this context (for example, the policy cost divisor is 5, they are based on a formula and then rounded down to the nearest 5).
So it seems that the reason that deleting 2 units tends to make it go down by exactly 7 some of the time but not always depends on the amount of maintenance. If your unrounded maintenence is 48 (rounds to 42) and you maintenance per unit is 5, you have to delete 2 units to see a change in maintenance, but deleting one more unit after that makes it change again! The number is inconsistent because 5 does not divide into 7.
I'm willing to guess that unit costs remain between 0 and 7 for most of the game, if not all of it, which explains why sometimes you can delete a unit and not have any change in maintenance.

By the way, 2 units costing 2.02 times as much is likely just a rounding issue. If you have one unit that costs 7.6, it will round it DOWN to 7. If you have two units costing 7.6, it adds up to 15.2 and then rounds to 15. The ratio will not be exactly two because the fractions are dropped after all of the cost is added up.

To anyone wanting to figure out the formula, the easiest way would be to set the divisor to 1 so you can actually see it changing in amounts smaller than 7, use Tuner to give yourself a huge number of units, then record the amount of maintenance on each turn for a few dozen turns (please keep in mind the game speed makes a difference). If you want to, send me the results you obtain and I'll figure out the formula (I have a math background and love this kind of stuff, I just don't really know how to use Tuner and don't have much interest in pressing end turn 100 times to get a decent sample).

So we should just change the divisor to 1, right? If I'm reading this correctly, that should make for much more accurate and intuitive upkeep costs, where deleting a unit yields -4 cost per turn, another unit -4 per turn, another unit -3 per turn, etc. It seems like the issue where only every other unit costs money isn't [integer]/2, it's the rounding to the nearest 7, so rounding to the nearest 1 should eliminate that behavior, correct? (I'm not concerned with the formula here, just making the upkeep costs more logical).

 

[Pantastic]

The supply limit is so high relative to productive capacity and unit costs that I have never even come close.

I've hit it early on playing as Germany, with only one city (finishing Stonehenge) and a bunch of converted barbarian units.