Map balancing and editing tool

[plako]

To keep Darrell, happy, thanks for being a crap map maker

Just trying to beat you there .

[Sareln]

This is an awesome tool. Great work!

[Ellimist]

Just trying to beat you there .

Burn!

[scooter]

I have a feature suggestion based on my balancing attempts for PBEM20. For those who are NOT playing in that game, head over to the lurker thread to see an example of what I'm about to suggest.

The short of it is - I'd suggest developing some sort of statistic that calculates a "quality of land" number. At it's most simple, this could be a simple ratio of food/land tiles... It could be done at a few different levels:



1. Simple calculation of tiles per food bonus. So take tiles / food resources and you have a rough ratio that gives a VERY basic idea of land quality.

2. Expand "food resources" based on value - meaning: sheep != corn != crabs. This could be done a few different ways - but the simplest would be the amount of food this is worth. If it could calculate in irrigation that would be wonderful... So wet corn is worth 6 while plains cow is worth 3, and use the sum of those values in the ratio. This would be more accurate than #1, but also more complex.

3. Expand #2 to add up the food of all tiles. This SHOULD be feasible right? This might actually be the way to go - adding up all food/hammers available in the land and calculating a "Land Quality" ratio for each person.




So the idea is that, at a glance, you could see "land tiles" for each player and "land quality" for each player, and figure who has the best start. I'm suggesting this as I THINK it should be pretty realistic to implement, as opposed to calculating the effects of jungle/peaks which could be more difficult.

Novice, if you'd like any help coding, I might be able to help some... Depends some on my free time (or lack of it) as well as complexity (I'm certainly no expert, just a college student), but I do think I could probably do some of it.

[Kylearan]

No one uses assembly any more .

Not true. Besides me using it almost daily, there's also an xml parser written in pure assembly, in case anyone wants to parse the Civ XML files. :neenernee

[mostly_harmless]

Here is PBEM16

Excellent. Almost exactly like my internal estimations of my sphere of influence. Shame Krill did not agree to a fair split of lands.

mh

[novice]

I have a feature suggestion based on my balancing attempts for PBEM20. For those who are NOT playing in that game, head over to the lurker thread to see an example of what I'm about to suggest.

The short of it is - I'd suggest developing some sort of statistic that calculates a "quality of land" number. At it's most simple, this could be a simple ratio of food/land tiles... It could be done at a few different levels:



1. Simple calculation of tiles per food bonus. So take tiles / food resources and you have a rough ratio that gives a VERY basic idea of land quality.

2. Expand "food resources" based on value - meaning: sheep != corn != crabs. This could be done a few different ways - but the simplest would be the amount of food this is worth. If it could calculate in irrigation that would be wonderful... So wet corn is worth 6 while plains cow is worth 3, and use the sum of those values in the ratio. This would be more accurate than #1, but also more complex.

3. Expand #2 to add up the food of all tiles. This SHOULD be feasible right? This might actually be the way to go - adding up all food/hammers available in the land and calculating a "Land Quality" ratio for each person.




So the idea is that, at a glance, you could see "land tiles" for each player and "land quality" for each player, and figure who has the best start. I'm suggesting this as I THINK it should be pretty realistic to implement, as opposed to calculating the effects of jungle/peaks which could be more difficult.

Novice, if you'd like any help coding, I might be able to help some... Depends some on my free time (or lack of it) as well as complexity (I'm certainly no expert, just a college student), but I do think I could probably do some of it.

New version is up here with basically what Scooter suggests above. The program does not take freshwater lakes, rivers or irrigation into account. (Assumes irrigation for all grain resources).

[Kylearan]

Hi,

1. Randomly place a set of N starting positions, with a reasonable minimum spacing.

2. Compute the Voronoi diagram for the map based on those N points. If some of the cells end up with what you deem to be too many or too few neighbors, try again from step 1.

3. For each of the N starting positions, generate an equal-sized random land shape contained within its Voronoi cell, constituting its sphere of influence.

How do you account for mountains and water tiles if you create the Voronoi diagram first and fill in the terrain later? I think adding mountains and water tiles would (should) change the shape and size of the spheres of influence, so I don't think it will be that easy to do.

Still possible I guess, and a very nice idea. (Damn, I wish I had more time!)

[zakalwe]

Basically, my idea is to start with all water (or perhaps some other background terrain, such as desert). Within each Voronoi cell, you would then have to generate a contiguous, traversable land mass connected to the capital site. So you would have to avoid placing peaks in places where they might potentially break connectivity, which shouldn't be too hard to ensure unless you want massive mountain ranges.

That said, my proposed algorithm probably needs to be refined if it's going to work with very "snaky" land shapes. You might have to take a long detour to reach a land tile within your Voronoi cell, which means that it might actually be closer to another starting position, in practice. In other words, if you were to generate a map using my algorithm, the spheres of influence as calculated by Novice's tool might still be unbalanced.

An alternative approach is to devise an algorithm for positioning all starts in a balanced way, given a complete generated map. You could then tack that algorithm onto any number of pangea-style map scripts.

[novice]

New version, with island support.

http://dl.dropbox.com/u/15215428/MapEdit..._v1.02.zip