Hexagon Grids

[Tyrael]

Just get rid of grids. That would be more realistic than squares and hexagons

And while we're at it, let's make units able to be half in forest and half on grassland! If it's a stack, every other time it gets attacked will get the 50%, the other time 0%. Because there's only half of them in the forest.

 

[Perfection]

Just get rid of grids. That would be more realistic than squares and hexagons

Realism blows. Gameplay is key.

 

[Larklight]

Realism blows. Gameplay is key.

hehe, if it was totally realistic, it would be totally balanced too!

 

[hgthechinese]

hehe, if it was totally realistic, it would be totally balanced too!

Definately. Only crazed people think the Dutch can't beat the US in real life.

 

[The Flame]

Definately. Only crazed people think the Dutch can't beat the US in real life.

*cough* sure....

 

[Traitorfish]

"Great spirits have always encountered violent opposition from mediocre minds." - Albert Einstein

So anyone who has proposed an idea that met general opposition is automatically correct? Hm, I guess those crop-circle guys were right after all. Guess I'll get round to making one of those aliminium hats they were telling me about. Don't want those UFOs reading my mind!

Realism blows. Gameplay is key.

Then play Chess- if you want abstract games, stick to abstract games. Civ is an historical simulation, realism is part of the fun.
It's a balance, you know- we don't want obsessive realism at the expense of gameplay, but we don't want our gameplay to be flavourless. It's about achieving a happy medium, which, by and large, Firaxis are pretty good at. (Even if a lot of us, myself included, have a tendency to criticise them more than we really should.)

 

[Perfection]

So anyone who has proposed an idea that met general opposition is automatically correct? Hm, I guess those crop-circle guys were right after all. Guess I'll get round to making one of those aliminium hats they were telling me about. Don't want those UFOs reading my mind!

Nah, it's just somethin' snappy I say to people who who don't have arguments and just browbeat.

Then play Chess- if you want abstract games, stick to abstract games. Civ is an historical simulation, realism is part of the fun. It's a balance, you know- we don't want obsessive realism at the expense of gameplay, but we don't want our gameplay to be flavourless. It's about achieving a happy medium, which, by and large, Firaxis are pretty good at. (Even if a lot of us, myself included, have a tendency to criticise them more than we really should.)

Of course you're right, I'm just saying that overarching realism over gameplay sucks, and while great rhombitrihexagonal tiling may not make much sense from a purely physical standpoint, it adds an element of strategy that will more then compensate for the problems.

 

[The Flame]

Nah, it's just somethin' snappy I say to people who who don't have arguments and just browbeat.

Of course you're right, I'm just saying that overarching realism over gameplay sucks, and while great rhombitrihexagonal tiling may not make much sense from a purely physical standpoint, it adds an element of strategy that will more then compensate for the problems.

Hmm... interesting... I probably only understood half of what you said (not really)

So what does "browbeat" and "rhombitrihexagonal" mean?

I also agree with the people who say to trash the grid system entirely. Or at least make a smaller grid with cities/improvements/units taking up several tiles. It allows more flexibility.

 

[Perfection]

jus' wiki it.

 

[Grotius]

Hexes are the traditional grid for wargames, for reasons stated many times in this thread. I'd prefer a hex-based Civ map too.

I wonder, though, how it would affect the AI. Hex-based computer wargames generally don't have great AI. By contrast, the AI in GalCiv2 and Civ 4 can give you a good, challenging game, especially on higher difficulty levels. Maybe the difference is that wargames can't so easily be weighted to favor the AI. But I wonder whether the grid plays a role too.

 

[Mr. Blonde]

At the moment I am used and fine with the sqare based system.

THE big difference with hexagons would be te possibility to implement zones of control more comprehensively so that units can restrict enemy´s movements which would deepen startegy. I also liked the system of artillery support in the Pazer General series where when unity were attacked artillery adjacent to the unit under attack.

So zones of control would mean that in case presented below B can not move
to the lower right tile although they have a direct border.

. B A
A x x
. x x
The big question is... will the AI be able to handle this feature?

 

[Larklight]

I also agree with the people who say to trash the grid system entirely. Or at least make a smaller grid with cities/improvements/units taking up several tiles. It allows more flexibility.

Would allow for cool organic growth of cities. Sprawling metropolis, growing past their walls. Cities merging! Maybe even working on several projects at once in different sectors.

 

[hgthechinese]

Hexes are the traditional grid for wargames, for reasons stated many times in this thread. I'd prefer a hex-based Civ map too.

I wonder, though, how it would affect the AI. Hex-based computer wargames generally don't have great AI. By contrast, the AI in GalCiv2 and Civ 4 can give you a good, challenging game, especially on higher difficulty levels. Maybe the difference is that wargames can't so easily be weighted to favor the AI. But I wonder whether the grid plays a role too.

Good point there. However, i think many AI suck on hex not because it is hex, but simply because the AI sucks. I mean, I know so many other sucky AIs on square games too.

 

[Perfection]

At the moment I am used and fine with the sqare based system.

THE big difference with hexagons would be te possibility to implement zones of control more comprehensively so that units can restrict enemy´s movements which would deepen startegy. I also liked the system of artillery support in the Pazer General series where when unity were attacked artillery adjacent to the unit under attack.

So zones of control would mean that in case presented below B can not move
to the lower right tile although they have a direct border.

. B A
A x x
. x x
The big question is... will the AI be able to handle this feature?

But imagine the zone of control awesomeness on my tiling!

 

[DisruptiveIdiot]

You guys are looking at it all wrong, specifically, you're using the Euclidean metric (which only works in a Euclidean n-space) whereas Civ uses Tchebychev distance as its metric. Under the latter metric, moving diagonally and orthogonally (with respect to the grid) are equivalent in distance, there is no geometric incongruity.

The apparent gain in distance while travelling diagonally is an illusion because you're seeing a Euclidean space but the units don't move according to the mtric of Euclidean spaces.

Thus, the only reasons to adopt a hexagonal grid system (which would operate under a Euclidean metric like would seem intuitive) are gameplay considerations. Since I don't have the patience to reply to every user who has voiced their (legitimate, in my eyes) concerns, I'll only reply to one post that particularly caught me eye, Lord Olleus's.

1) This can be mitigated rather easily by having the city expand into a larger square rather than a fat cross, perhaps leaving the corners outside of the city's production area.

2) This is a rather minor consideration and can be mitigated similarly to the solution for the previous point. In Tchebychev distance, squares are the equivalent of Euclidean circles and the equivalent of rings of hexagonal cells in that sort of grid.

3) This is a matter of taste. For me, granting that much power to blockades would be too much of a mechanic change. Fortifications have their uses, as strongholds away from cities where you can assail enemy forces from in the cases where such a position would be optimal. Anyway, it only takes two extra unit stacks to defend the flanks of a blockade and truly seal it.

4) You may be seeing a pattern here. Though the game is based in Tchebychev distance, many mechanics are designed with Euclidean considerations. They should have squares for ranges, but they don't. Anyway, this isn't a serious flaw in game mechanics.

5) Perhaps the most serious issue of them all, but in the end rectifying it is *not* worth a radical change in game mechanics.

Squares are the answer. =)

 

[occam]

You guys are looking at it all wrong, specifically, you're using the Euclidean metric (which only works in a Euclidean n-space) whereas Civ uses Tchebychev distance as its metric. Under the latter metric, moving diagonally and orthogonally (with respect to the grid) are equivalent in distance, there is no geometric incongruity

First of all, it is not the Tchebychev distance at all, but its close cousin, the Occam-Tchebychev distance that includes a 'case' statement for world-wrapping. (Yep, I just invented it.)

More importantly, your assertion that this is what Civ uses is wrong. Take a look at, say, the distance corruption formula. Etc. Civ tries to be Euclidean, it is not MY blind spot that is making me argue in that direction.

I would offer kudos for the clever(?) idea of referring to non-Euclidean ideas, except that has already been done to death whenever hexagons come up, including on this very thread. You can't swing a cat in a hexagons thread without hitting a couple Lobachevskys, apparently.

- O

 

[DisruptiveIdiot]

First of all, it is not the Tchebychev distance at all, but its close cousin, the Occam-Tchebychev distance that includes a 'case' statement for world-wrapping. (Yep, I just invented it.)

Ah, quit with the pedantry. Applying the Tchebychev metric to a torus isn't difficult at all, and you forget to account for the map scripts that really are finite planes.

More importantly, your assertion that this is what Civ uses is wrong. Take a look at, say, the distance corruption formula. Etc. Civ tries to be Euclidean, it is not MY blind spot that is making me argue in that direction.

Rather, this brings up the point that the developers were inconsistent when they first designed the basic principles of civ. They should have stuck to a strictly Tchebychev metric for all mechanics because it makes the most sense for civ. Doesn't dimunitize my argument.

I would offer kudos for the clever(?) idea of referring to non-Euclidean ideas, except that has already been done to death whenever hexagons come up, including on this very thread. You can't swing a cat in a hexagons thread without hitting a couple Lobachevskys, apparently.

I don't understand what hyperbolic geometry has to do with my arguments

Anyway, the points still stand, civs movement system is perfectly valid and consistent, except the developers made a few oversights in its implementation (sight and culture mainly, everything else calculated with a Euclidean metric is negligible) that create a few minor issues. Certainly nothing worth revamping the entire system over. Civ has always used its system, it shouldn't change. It isn't perfect, but making it perfect is much more practical than uprooting the whole thing and replacing it with hexagons.

Besides, eight directions is more than five

 

[occam]

Change the grid or change the metric, both work. There isn't a thread for changing the metric advocates, but it does sound like you are with us in principle at least.

Applying the Tchebychev metric to a torus isn't difficult at all, and you forget to account for the map scripts that really are finite planes.

You underestimate the 'case' statement.

Rather, this brings up the point that the developers were inconsistent when they first designed the basic principles of civ.

To use the phrase "brings up" is horrifically wrong. This entire time advocates of hexagons have claimed that the square moves are not true to Civ's own distance metric. We're not just discovering this on page 5 of the thread. Please, give your fellow posters some credit. That has always been the objection, it didn't start as some misguided Euclidean campaign ("stamp out funny metrics"?) that has unwittingly unearthed an inconsistency.

- O

 

[Perfection]

Great rhombitrihexagonal tiling has the benefit of having a more realistic deistance calculation but still one that differs from standard euclidean so it's a safe intermediate.

 

[Mr. Blonde]

Great rhombitrihexagonal tiling has the benefit of having a more realistic deistance calculation but still one that differs from standard euclidean so it's a safe intermediate.

I guess that having tiles with different size makes handling of resources, ect. a bit complicated and counterintuitive.