Official Everybody Edits Forums

Slabdrill

Formerly 12345678908642

From: canada

Joined: 2015-08-15

Posts: 3,402

Website

Custom sized worlds

max area 90k

energy cost - area^0.735?
625 (small) - 113.5 (100)
2500 (medium) - 314.4 (250)
10000 (large) - 871 (500)
20000 (wide) - 1450 (1000)
22500 (big) - 1581 (1000)
31800 (ultra wide) - 2038 (1500)
40000 (massive/tall) - 2412 (2000)
80000 (great/vgreat) - 4015 (4000)
90000 (huge) - 4379 (4000)

if width exceeds 636 or height exceeds 476, minimap cannot be enabled.

---

suddenly random sig change

Gosha

Member

From: Russia

Joined: 2015-03-15

Posts: 6,215

Re: Custom sized worlds

if width exceeds 636 or height exceeds 476, minimap cannot be enabled.

That is just stupid

there are bigger than 636x476 worlds already, they are fine with minimap

for example PWJesse - 2000x2000 world, cool minimap

drunkbnu

Formerly HG

Joined: 2017-08-16

Posts: 2,309

Re: Custom sized worlds

Just why?

Tomahawk

Forum Mod

From: UK

Joined: 2015-02-18

Posts: 2,863

Re: Custom sized worlds

To all intents and purposes, a world that's been downsized with fully black blocks on the top and left side is a custom sized world. That limits the possible dimensions though, as you need to own a world larger than the one you want.

Using indices for a pricing formula isn't great. Here's the area-to-price ratio for some worlds in the shop:

Size --> blocks per 1 energy

25x25     --> 6.25
50x50     --> 10
100x100 --> 20
150x150 --> 22.5
200x200 --> 20
300x300 --> 22.5
------------------
400x50   --> 20
636x50   --> 21.2
400x100 --> 20
400x200 --> 20

From that, I think something like Area / 15 (and then rounded up to the nearest 10 if not a whole number) would be a decent formula, but a non-linear formula would fit better with the ratio found in the shop.

---

One bot to rule them all, one bot to find them. One bot to bring them all... and with this cliché blind them.

LukeM

Member

From: England

Joined: 2016-06-03

Posts: 3,009

Website

Re: Custom sized worlds

Following on from this discussion in a slightly unrelated topic:

▼peace wrote:

just a customable world size setting (to a limit ofc) and the game calculate steh energy neededfor it

▼AnatolyEE wrote:

If the world is a * b, we need a formula with f(a, b) = c Energy. Let's take ideas what we need to know:

200x200 should have another payment then 100x400. Reason: Different amount of border blocks, more space.

a*b - (a-2)(b-2) = Blocks not border:
ab-(ab-2a-2b-4) = 2(a+b+2)

Scaling 2*18 is not really better then 6x6.

We can get this square world with the same blocks amount by √(ab) * √(ab) = √a√b * √a√b

|a - √(ab)| = Difference in x
|b - √(ab)| = Difference in y

So:    ab / (|a - √(ab)|)(|b - √ab|) = 1

Maybe anyone know how to edit this for a good formula?

▼LukeM wrote:

If there was a formula, it would have to cost more than any of the preset world sizes (as it would be better than them), and tiny worlds can't cost almost nothing, otherwise you could buy hundreds of them, so there would have to be a constant first term (or some other means of having a fairly large intercept).

Also I don't think border blocks should cost less, because 1. You can now remove them, and 2. The more rectangular the world is, the more 'specialised' it is, so perhaps it should cost more instead possibly?

▼AnatolyEE wrote:

If there was a formula, it would have to cost more than any of the preset world sizes (as it would be better than them), and tiny worlds can't cost almost nothing, otherwise you could buy hundreds of them, so there would have to be a constant first term (or some other means of having a fairly large intercept).

Also I don't think border blocks should cost less, because 1. You can now remove them, and 2. The more rectangular the world is, the more 'specialised' it is, so perhaps it should cost more instead possibly?

Yes point 2 is what I tried to explain:

Scaling 2*18 is not really better then 6x6.

Maybe there was a bit confusing to understand me.

I've joined EE to check the prices:

• Wide World (400x50, 1000 Energy)

• Big World (150², 1000 Energy)

• Ultra Wide World (636x50, 1500 Energy)

• Large World (100², 500 Energy)

• Massive World (200², 2000 Energy)

• Medium World (50², 250 Energy)

• Small World (25², 100 Energy)

• Great/ Vertical Great World (400x200, 4000 Energy)

• Huge World (300², 4000 Energy)

• Tall (100x400, 2000 Energy)

Firstly let's sort them per the product of a and b and specify whether they are a square a : b = 1 : 1 or a rectangle e.g. Wide World a : b = 400 : 50 = 8 : 1

Sorted by Product

• Huge World (90000, 4000 Energy, 1:1)

• Great/ Vertical Great World (80000, 4000 Energy, 2 : 1 or 1 : 2)

• Massive World (40000, 2000 Energy, 1 : 1)

• Tall (40000, 2000 Energy, 1 : 4)

• Ultra Wide World (31800, 1500 Energy, 318 : 25 ~ 127 : 10 ~ 25 : 2 ~ 12 : 1)

• Big World (22500, 1000 Energy, 1 : 1)

• Wide World (20000, 1000 Energy, 8 : 1)

• Large World (10000, 500 Energy, 1 : 1)

• Medium World (2500, 250 Energy, 1 : 1)

• Small World (625, 100 Energy, 1 : 1)

Now let's, for being sure, sort out the 1 : 1 worlds and check them single:

Sorted by Product, Only square worlds

• Huge World (90000, 4000 Energy)

• Massive World (40000, 2000 Energy)

• Big World (22500, 1000 Energy)

• Large World (10000, 500 Energy)

• Medium World (2500, 250 Energy)

• Small World (625, 100 Energy)

Let's divide the product by 625, so every number is as small as possible and stays natural. Divide energy by 50.

Sorted by Product, Only square worlds

• Huge World (144, 80 E)

• Massive World (64, 40 E)

• Big World (36, 20 E)

• Large World (16, 10 E)

• Medium World (4, 5 E)

• Small World (1, 2 E)

Can we do anything with this information? The problem is that the current square worlds are also unregularly build.

I am suggesting for a and b: Haven't view many values currently but seems to work.

E = 5 * |(√(ab) - ( | a - b | + | b - a | )²  )|

▼LukeM wrote:

I decided to imput all the world sizes into a spreadsheet and see what it looked like, and see how well it could be modelled, heres my results:
https://image.prntscr.com/image/Kb6XmRA … EnhMtQ.png
The original google sheet

The red data series is just the auto-calculated trendline, the green is that with a small decrease in cost the closer it gets to square, and the yellow is somewhere in-between

▼Slabdrill wrote:

I actually just said area^0.735 was a "good enough" formula for this. It's actually rly bad but works for what it's made to do

also get back on topic? I have https://forums.everybodyedits.com/viewt … p?id=39231 if needed i guess

(Ive added your area^0.735 formula to the sheet)

I dont think this would work well, because currently it isnt an exponential (at least that isnt the overriding factor), so its matched better using a linear equation (shown by red)

Slabdrill

Formerly 12345678908642

From: canada

Joined: 2015-08-15

Posts: 3,402

Website

Re: Custom sized worlds

Following on from this discussion in a slightly unrelated topic:

▼peace wrote:

just a customable world size setting (to a limit ofc) and the game calculate steh energy neededfor it

▼AnatolyEE wrote:

If the world is a * b, we need a formula with f(a, b) = c Energy. Let's take ideas what we need to know:

200x200 should have another payment then 100x400. Reason: Different amount of border blocks, more space.

a*b - (a-2)(b-2) = Blocks not border:
ab-(ab-2a-2b-4) = 2(a+b+2)

Scaling 2*18 is not really better then 6x6.

We can get this square world with the same blocks amount by √(ab) * √(ab) = √a√b * √a√b

|a - √(ab)| = Difference in x
|b - √(ab)| = Difference in y

So:    ab / (|a - √(ab)|)(|b - √ab|) = 1

Maybe anyone know how to edit this for a good formula?

▼LukeM wrote:

If there was a formula, it would have to cost more than any of the preset world sizes (as it would be better than them), and tiny worlds can't cost almost nothing, otherwise you could buy hundreds of them, so there would have to be a constant first term (or some other means of having a fairly large intercept).

Also I don't think border blocks should cost less, because 1. You can now remove them, and 2. The more rectangular the world is, the more 'specialised' it is, so perhaps it should cost more instead possibly?

▼AnatolyEE wrote:

If there was a formula, it would have to cost more than any of the preset world sizes (as it would be better than them), and tiny worlds can't cost almost nothing, otherwise you could buy hundreds of them, so there would have to be a constant first term (or some other means of having a fairly large intercept).

Also I don't think border blocks should cost less, because 1. You can now remove them, and 2. The more rectangular the world is, the more 'specialised' it is, so perhaps it should cost more instead possibly?

Yes point 2 is what I tried to explain:

Scaling 2*18 is not really better then 6x6.

Maybe there was a bit confusing to understand me.

I've joined EE to check the prices:

• Wide World (400x50, 1000 Energy)

• Big World (150², 1000 Energy)

• Ultra Wide World (636x50, 1500 Energy)

• Large World (100², 500 Energy)

• Massive World (200², 2000 Energy)

• Medium World (50², 250 Energy)

• Small World (25², 100 Energy)

• Great/ Vertical Great World (400x200, 4000 Energy)

• Huge World (300², 4000 Energy)

• Tall (100x400, 2000 Energy)

Firstly let's sort them per the product of a and b and specify whether they are a square a : b = 1 : 1 or a rectangle e.g. Wide World a : b = 400 : 50 = 8 : 1

Sorted by Product

• Huge World (90000, 4000 Energy, 1:1)

• Great/ Vertical Great World (80000, 4000 Energy, 2 : 1 or 1 : 2)

• Massive World (40000, 2000 Energy, 1 : 1)

• Tall (40000, 2000 Energy, 1 : 4)

• Ultra Wide World (31800, 1500 Energy, 318 : 25 ~ 127 : 10 ~ 25 : 2 ~ 12 : 1)

• Big World (22500, 1000 Energy, 1 : 1)

• Wide World (20000, 1000 Energy, 8 : 1)

• Large World (10000, 500 Energy, 1 : 1)

• Medium World (2500, 250 Energy, 1 : 1)

• Small World (625, 100 Energy, 1 : 1)

Now let's, for being sure, sort out the 1 : 1 worlds and check them single:

Sorted by Product, Only square worlds

• Huge World (90000, 4000 Energy)

• Massive World (40000, 2000 Energy)

• Big World (22500, 1000 Energy)

• Large World (10000, 500 Energy)

• Medium World (2500, 250 Energy)

• Small World (625, 100 Energy)

Let's divide the product by 625, so every number is as small as possible and stays natural. Divide energy by 50.

Sorted by Product, Only square worlds

• Huge World (144, 80 E)

• Massive World (64, 40 E)

• Big World (36, 20 E)

• Large World (16, 10 E)

• Medium World (4, 5 E)

• Small World (1, 2 E)

Can we do anything with this information? The problem is that the current square worlds are also unregularly build.

I am suggesting for a and b: Haven't view many values currently but seems to work.

E = 5 * |(√(ab) - ( | a - b | + | b - a | )²  )|

▼LukeM wrote:

I decided to imput all the world sizes into a spreadsheet and see what it looked like, and see how well it could be modelled, heres my results:
https://image.prntscr.com/image/Kb6XmRA … EnhMtQ.png
The original google sheet

The red data series is just the auto-calculated trendline, the green is that with a small decrease in cost the closer it gets to square, and the yellow is somewhere in-between

▼Slabdrill wrote:

I actually just said area^0.735 was a "good enough" formula for this. It's actually rly bad but works for what it's made to do

also get back on topic? I have https://forums.everybodyedits.com/viewt … p?id=39231 if needed i guess

(Ive added your area^0.735 formula to the sheet)

I dont think this would work well, because currently it isnt an exponential (at least that isnt the overriding factor), so its matched better using a linear equation (shown by red)

https://image.prntscr.com/image/cfcTcY9 … 0ufsrA.png

My goal with mine was "make sure it remains above the cost of purchasing the world now", though i think minor adjustments to mine would work.
Yours also does work - I was looking at the default area - cost ratios and noticed they tended to lower as you got bigger worlds.
I think the 0.055 with a reduction in the non-square price reduction, and something to make mediums cost more (increase constant? tiny worlds are never rly used anyway) could work.

---

suddenly random sig change

LukeM

Member

From: England

Joined: 2016-06-03

Posts: 3,009

Website

Re: Custom sized worlds

My idea would be to just make custom worlds cost a fair bit more (as they are 'better' than normal worlds), then you can just use the best model you have so they all match the original prices fairly well (just a bit more expensive)

Anatoly

Guest

Re: Custom sized worlds

Based on Slabdrill suggestion: Maybe you could try making a graph?

2 * (XY ^ 3/4 - |X-Y| ^ 3) ^ 3/4

Basically

Area ^ A - XYDifference ^ B)

Joeyc

Guest

Re: Custom sized worlds

if we had a 100,000x100,000 we wouldn't even have to change worlds, yet 10 years to reach the side of the map