Understanding and modifying/rebalancing park tile stats

[Walter Raleigh]

I downloaded some user-made park tiles yesterday. I wanted to check that the stats the tiles had been given aligned with the base game's park tile stats, so I had a look at them too.

I am new to the game and don't fully understand the mechanics, but the base game's park tiles' stats look quite strange.

A couple of examples:

- a Dog Park has absolutely identical stats to a Park with Trees in every respect, except their wildly different footprint: 4x5 vs 13x12.

- a Plaza with Picnic Tables costs fifteen times as much to buy and fifteen times in maintenance as a Small Park, and yet its Entertainment Accumulation and Entertainment Radius are the same. The only advantage Plaza with Picnic Tables has over Small Park is it has double tourism, but Park with Trees has the same and costs a third of the price, and has better entertainment stats.

Looking online, other people have commented on these apparent disparities:

https://www.reddit.com/r/CitiesSkylines/comments/2zv6iz/park_comparison_chart_and_gallery/
https://forum.paradoxplaza.com/forum/index.php?threads/rebalancing-parks.906647/#post-22545750

I constructed a quick test city yesterday to try and test the effects of the base game's park tiles. There didn't appear to be a great deal of difference between the strengths of small and large versions of similar tiles, such as Small Playground and Large Playground.

If we compare the small and large versions against each other, then we see that their entertainment radius, the distance projection of their effect, is the same. The entertainment accumulation stat, or power/intensity of the tile's effect, is different, with the small tile versions having 100 and large having 125. Without knowing the precise mechanism, it's difficult to know exactly what's going on, but it looks as though it's just a 25% boost. Placing two small Small Park tiles next to each other produced higher land values than a single tile of its larger version, Park With Trees. Given that the larger tile version is five times the cost and maintenace, this makes no sense. So, purely in terms of entertainment value, the park tiles appear very imbalanced.

The other attribute to take into consideration is the Tourist bonus, and the larger park tile versions have double the tourist bonus than their smaller counterparts. It's difficult to know the power of this. In purely entertainment value terms, the park tiles look really imbalanced, but perhaps the tourist value bonus is powerful enough to compensate for that.

My current thinking is that to better balance the park tiles, the small versions should have their entertainment radiuses halved. This would give them one quarter the coverage area of their larger counterpart. Since the larger versions are five times the cost/maintenance of the smaller versions, and slightly more powerful, this seems a fair trade.

The Tourist bonus is a bit of a complication to this and I'd be interested in hearing other player's comments.

There's a list of park tiles, with an incomplete list of stats here.

Here's the full stats, derived from the assets editor, for four of the park tiles: small/large park and small/large playground.

x = Size, x axis
y = Size, y axis
Ar = Total area of tile's footprint (x multiplied by y)
El = Electricity cost
Cost = Purchase price
FH = Fire Hazard
FT = Fire Tolerance
GA = Garbage Accumulation
MC = Maintenance Cost
T = Tourist bonus
EA = Entertainment Accumulation (power/intensity)
ER = Entertainment Radius (effect range/distance)

Any thoughts?

 

[Walter Raleigh]

Let's compare a couple of similar tiles: Small Park and Small Playground. The difference between the price cost to buy between them is negligible (1000 v 1250) and their entertainment and tourist stats are identical. The Small Playground, however, has three times higher maintenance costs, although lower electricity costs (1 v 2 units) - anyone have a ballpark figure for how much one unit of electricity costs?

The main difference between them is that the Small Playground is almost one third the size of the Small Park, but is that really worth paying three times the maintenance cost for?

Thoughts?

 

[Walter Raleigh]

So I'm starting to formulate some ideas about how I'm going to play around with the stats, starting with the park park tiles.

Firstly, thoughts on range (Entertainment Radius). I looked at some of the Unique Buildings and they have the same range as the other lesser park tiles - 400 units. So I was thinking it would probably be best to leave that as the maximum and not increase it. Which therefore means reducing range would be the way to do any modifications, i.e. small park would have to its range diminished, rather than augmenting the range of large parks.

I first experimenting with halving the range, which would reduce the total coverage area to one quarter. In game-testing, it felt a bit small, so the next thing I tried was to reduce the coverage area to one half of its total area, which produces a range/radius of 283 units (282.843 units). I thought this worked pretty well in game-testing.

Then I tried making the Entertainment Accumulation of the small park half that of the large park, so down from 100 v 125 to 63 v 125. Again, in testing it, while obviously a bit diminished, still seemed to be working okay. These proportions mean that four Small Parks has the same entertainment output as one large Park With Trees.

Now, these are obviously quite big nerfs to the small versions of tiles and will inevitably make the game a bit harder, but that's not necessarily a bad thing.

I did also check that Entertainment Accumulation stacked by testing a 1x1 tile with an E.A. value of 20 (so one fifth of regular). Placing one of them gave only a very faint presence on the Land Value map, but clustering a load of them together made the Land Value map turn blue.

For Tourist values, I thought I'd leave it as is, so smaller versions have half the Tourist values of their larger counterparts. This is because this may also be stackable.

So maintenance value I'm still thinking about. This value would perhaps fit into the equation as a counterbalance to the footprint size of the park tiles, i.e. four Small Parks has a greater total footprint than one Large Park (roughly double).

There's a first draft then. Any thoughts?

[EDIT: Accidentally wrote "economic" instead of "entertainment" a couple of times, and so have corrected it]

 

[Walter Raleigh]

Here are some sample Electricity Consumption values for other buildings.

Elementary School: 30
University, Hosptital: 90

From this, we can probably say that the Electricity Consumption values of the park tiles are so small that they are not a major factor to take into consideration in trying to assess the relative values between them.

 

[Crenickator]

There's a first draft then. Any thoughts?

My only thought is this: Don't be too mechanical with the way small parks and large parks compare with entertainment and tourism values. While it is good on its face, a little more nuance might be in order, such as making smaller parks have a higher relative entertainment rating, and a lower relative tourism rating. In other words, the smaller parks improve the lives of citizens more than they draw tourists, but the bigger parks are the tourist draw, even as the citizens don't necessarily need that much space in their own city. But this should not be an ironclad rule either, but something to add to the mix so it's not pure 1 + 1 = 2.

Come up with a base value for each square of park, which you seem to be on the path to doing, and then modify the finished product for size and just what kind of park it's meant to be. Playgrounds, dogparks, basketball courts are more fun for citizens. Larger parks and monuments are more attractive to tourists.

 

[Walter Raleigh]

Thanks for the suggestions. There is a difficulty that the leisure/happiness system and tourism systems are two different things and with two things going on, it's difficult to weigh the importance each of the two components has to a tile's overall value. So I'm trying to work on a basic framework, and am still contemplating how tourism fits into the picture.

 

[Walter Raleigh]

The smaller versions of park and playground cost five times less than the larger versions. Maintenance is between five and seven times less. If cost has a direct relationship with performance, then these are the results if they are both using the same Entertainment Radius (400, as they do in the game currently). Total Power is Entertainment Radius multiplied by Entertainment Accumulation. (Since power reduces with distance, all actual total power values would be halved).

A pretty feeble 25. So there'd be a massive difference in living next to a small park than to living next to a large park. You'd need to build five small parks immediately surrounding the house to get the same bonus as a large park.

So the solution is to use smaller radiuses for smaller tiles, so that their total economic power value is concentrated on the nearer tiles.

Here is a set of data using a system with tiers, with each tier having half the total Economic Coverage Area and half the Entertainment Accumulation as the tier above it. I adjusted the E.A. value from 125 to 128 to make things more easily divisible.

This system would make a large park worth around 2.83 small parks. I opened up the game and placed a large park and some small parks next to each other. In my opinion, 2.83 small parks looked of better equivalence to one large park. At least for the game's basic parks. There are a lot of nicer user-made parks available to download, and with this system, they would have the potential to be strengthened at increased cost to buy/maintain.

The parameters could be adjusted according to what one was going for. For example, if one wanted to make a small park exactly three times or five times less in cost, then other parameters could be adjusted.

The next stage, is to factor in footprint size, which I'm still working on, and which will be another layer of purchase cost modifiers.

 

[Walter Raleigh]

[I accidentally wrote "economic" instead of "entertainment" a few times in this thread and so have corrected this - I blame staying up late ]

 

[Walter Raleigh]

My brain's not great for mathematical stuff, but I've come up with a basic system that I'm going to try out.

Here are some points about my aims and the method:

a) A system that can work for all sizes of park tiles, right down to 1x1 square tiles.

b) The smaller the park tile, the less entertainment power it has (Economic Accumulation).

c) A diminishing radius, so the smaller the tile, the smaller distance it can project entertainment (Entertainment Radius).

d) The game currently appears to have most tiles set as Entertainment Radius of 400, including the special Unique buildings, so this system sets 400 as the radius cap, and so making tiles more powerful than this would only involve increasing Entertainment Accumulation, and not Economic Radius.

e) The entertainment power per square of a park tile increases as its tile area diminishes. It is set as being that a 1x1 independent square park tile has twice the entertainment power per square than the large park tile. It is set as this for gameplay balance reasons, although this can alternatively be thought of as that a 1x1 park will try and condense entertainment into smaller space.

f) To compensate, the smallest park tile size, 1x1, costs four times per square more than a single square contained within a large park tile.

This effectively means that if one had $100,000 to spend on park tiles, and one spent it all on the large park tiles, one would get 100% of the entertainment value. If one spent the $100,000 all on 1x1 tiles, one would only be getting half that amount. So the smallest tiles are half as efficient as the largest tiles.

Because of the need to square root down to 1, This system is based on having 128 as the theoretical maximum tile size. The "Large Park With Trees" is 9x8 with an area of 156. The square root of 156 is 12.489, not an easy number to work with. So 12x12 = 144 is also a possibility as the largest size.

So to calibrate this system with the game it will involve a scaled modifier, with the 128 value being stretched to either 144 or 156, depending on which system is decided upon, while the lowest tile size, 1x1 remains unmodified. This will result in a bonus for the largest tile size of either +12.5% (for 144) or +21.875% (for 156), with a sliding scale down to +0% for the smallest 1x1 tile.

My mathematics is not strong and one set of values was done by graph, as I don't have the maths know-how to calculate it, and so I did it by compressing a graph in an image-editing program, and so I'm not sure if it is mathematically correct. If there's a maths whizz passing by, I'd be interested in knowing how to calculate this: how to proportionally squash this curve from eight points to five. So compressing I to VIII into I to V, so that 128 and 1 at either extreme are unaltered. I am not sure if the values in the right column are mathematically correct.

So the end result will be something like this graph, which has been plotted from the figures in the table above. Green is the Entertainment Accumulation, Blue is the Entertainment Radius and Red is the Cost. One takes a reading by looking up a park tile's tile size at the bottom and then getting readings for the three values. The red cost line has come out a bit funny, but that may be related to the issue mentioned in the previous paragraph.

As I say, maths isn't my strong point so there's a possibility that there are fatal flaws in this system which I haven't foreseen.

 

[Walter Raleigh]

In the game, playgrounds are three or four times the (maintenance) cost of park tiles. So a rule for calculating playgrounds could be that to halve a park tile's footprint area means doubling its cost, and to reduce it to one-third, like playgrounds, means tripling the cost.

 

[Walter Raleigh]

I feel that playgrounds do not represent good value for money if going by the rule that three times smaller is three times more expensive.

What I might try instead is a rule that playgrounds halve the footprint size but double the cost. So that would they'd be slightly less powerful than currently, but slightly cheaper.

Also, mods, feel free to move this thread into the Modding sub-forum.

 

[joewagner501]

An interesting thought experiment with the parks. I've always thought parks were a bit unbalanced and it's interesting seeing how you are working out a possible framework for a solution.

 

[Walter Raleigh]

Thanks. It seems to be working as hoped in game-testing.

I have been wondering how to fit tourism into this.

My current feeling is to apply exactly the process to it as has been done with entertainment output. At least for a first play-through with these settings, I think it would be best to have everything balanced to better judge how it is working.

So the Tourism output would diminish with the tile's area in the same way that Economic Accumulation does.

I can see the logic behind having certain types of leisure tiles produce more tourism than others, such as a park having more tourism value than a playground, it's just a matter of knowing an exchange rate between entertainment and tourism - if one tile gets a boost for tourism, how does one calculate its relative value compared to, say, a decrease in entertainment value? I'm not sure if that's something that can be mathematically calculated, and it seems to me more like a subjective decision and so something that would need a better level of understanding of how it's currently working.

The same goes for adjusting tourism types - low-wealth, medium-wealth and high wealth. Would doubling the weight/power to low-wealth be equivalent to doubling it the high-wealth?

So I'm open to suggestions, but otherwise my thoughts are, for the time being, to keep tourism in a neutral setting.

 

[Walter Raleigh]

I believe I've mathematically solved the issue regarding the theoretical tile area in the chart I posted previously.

This has meant recalculating a lot of the numbers so I've posted a new table below. Remember that this is calibrated for a 128 tile area maximum, whereas it fits the game better calibrated to 144. So these are not the final numbers, which will need stretching, so to speak.

I have also renamed Tiers as Levels and reversed the order, as it makes sense that a 1x1 tile is Level 1, so park tile sizes can be created beyond Level 5 (Level 6 etc).

I am learning how to make line graphs in LibreOffice, so I'll post one when I've created one.

 

[Crenickator]

Good work.
I assume tile area in the table is meant to be rounded off to 1, 4, 12, 40, and 128?
I'm not quite following you if that's not the case.

As for tourism, what is the base tourism going to be for each tile? Extrapolate that out to the size of park you need, and then halve it for "local" entertainment like playgrounds and dog parks, leave it as default for "neutral" entertainment like parks and plazas, and double it for "tourist" entertainment like monuments/beach volleyball.

 

[Walter Raleigh]

Good work.
I assume tile area in the table is meant to be rounded off to 1, 4, 12, 40, and 128?
I'm not quite following you if that's not the case.

The tile area readings would be taken from a line graph. There's probably a mathematical formula to do it, but that's beyond my capability. So a 2x2 tile would have slightly more powerful stats than a theoretical Level 2 tile, which cannot be created in reality, since its area is 3.3636.

As for tourism, what is the base tourism going to be for each tile? Extrapolate that out to the size of park you need, and then halve it for "local" entertainment like playgrounds and dog parks, leave it as default for "neutral" entertainment like parks and plazas, and double it for "tourist" entertainment like monuments/beach volleyball.

As I say, the tourism output would scale in a similar way to entertainment output. Regarding halving and doubling tourist value, it's just a matter of knowing how such changes would affect cost price of the tile and so how much entertainment output and tourist output are worth in relation to each other.

A small park or playground in the game has, compared with the larger versions, 100 versus 125 entertainment and 25 vs 50 tourist output. So it's not clear exactly where a point of equivalence lies. Does 125 entertainment equal 50 tourist output? Would halving tourist output therefore reduce tile cost by one quarter?

 

[Walter Raleigh]

I'm considering making a new modified chart. I feel that a radius of 80 feels better than 100 for the smallest 1x1 tile, which would make it one fifth, rather than one quarter, of the maximum radius of 400.

I may also apply a different curve for range and accumulation values.

 

[Walter Raleigh]

Behold! What I believe to be the final data set, now calibrated to the game.

Changes:

1) Entertainment Radius for smallest 1x1 tile reduced from 100 to 80.
2) Maximum tile size (beyond which Ent. Radius does not increase) increased to 156, so exactly the same as the base game's Park With Trees tile.
3) Entertainment Accumulation for the maximum tile size reduced from 128 to 125, so returned to the same number as the base game.
4) Smallest 1x1 tile Ent. Acc. increased from 8 to 10.
5) Power modifier per square difference between smallest and largest tile increased from x2 to x 2.5.
6) Cost modifier between smallest and largest tile per square increased from x2 to x2.5.

So a 1x1 square has 2.5 times more power per square than the largest size but 6.25 times the cost (2.5 x 2.5), although this is better viewed as being that a larger park tile has better economy of scale than a small tile. So it's not that small tiles are penalised, but rather that large tiles are subsidized.

I think I will also add a reduced class of tile. A 1x1 park tile represents a small condensed park with trees and benches and suchlike or plaza with benches and street light or other decoration. There will also be a basic/reduced version of these tiles that will have half the Ent. Acc. and half the cost, and will be represented by plain versions of these tiles. So a basic/reduced 1x1 plaza square would just be plainly paved and a 1x1 basic/reduced park tile would be plain grass/lawn. This will help fill in blank spaces on the map without costing too much.

 

[Walter Raleigh]

I have created my first chart in LibreOffice! The program automatically plots the chart, including the curve, when given the data. All lines read from the vertical y axis on the left, except for Tile Cost, which reads from the y axis on the right.

I need to figure out how to add more grid lines to make it easier to take readings. I will probably take readings for each tile size and make it into a table, so it's easier to get the stats needed for any given tile size.

 

[Walter Raleigh]

This extended graph shows what happens to park tiles greater than 156 squares in area:

- Entertainment radius is capped at 400.
- No further modifiers to cost per square or power square.
- So increase in tile area only changes Entertainment Accumulation (and subsequent Cost).

The curves on the graph can come out a bit funny with too much going on and in certain graphical dimension settings, so this graph is just to illustrate what happens beyond 156 square tile areas (and would also be absolutely straight lines).

I've also worked out how to add lines to the graph to make it look like graph paper, which will facilitate taking readings.