PDA

View Full Version : Hex Grid vs. Square Grid



Mr. Mask
2016-04-29, 05:56 PM
The other thread is not about hex grids. I shall remedy this.

What is your preference of grids? I like that with hexes all directions are equal distances unlike squares... but, it feels weird to me, having only six directions of movement. Certain movements can get awkward and zig-zaggy. Lines of battle also seem a bit... off angle in relation to their movement.

But those are my nitpicks and personal feelings. What do you think? What's your experience with them?

halfeye
2016-04-29, 06:04 PM
You can more or less map hexes onto a sphere, the nearest you can get with squares is a cube, or a unended cylinder.

Thrudd
2016-04-29, 06:35 PM
Both have their drawbacks in mapping realistic movement, as you have noted. That is a sacrifice that comes with the desire to simplify something into an abstract form.
I prefer the hex grid, because it allows more granularity in facing and does not require special consideration of diagonal movement that is required by using a square grid. Straight lines do appear wonky in certain directions, but I think that is better than needing to do calculations to figure out diagonals, or allowing a gain in movement or range by measuring on the diagonal.

My preference, really, is to not use a grid at all but measure distance with tape or ruler, when we're talking about D&D or similar tactical combat games. Not only does it allow more natural positioning and a built-in means of determining line-of-sight but it removes the need to purchase special mats and maps for your table. Area effects can be easily measured by creating a paper template of appropriate sizes and natural shapes.

DigoDragon
2016-04-29, 07:14 PM
I prefer square grids because they work better with scenes constructed from LEGO. :3

Delwugor
2016-04-29, 10:10 PM
Both are fairly equivalent to me, though I prefer hex slightly over square because of the smoother looking area affects.

Lvl 2 Expert
2016-04-30, 02:26 AM
It depends on what I'm mapping. Most buildings are much easier to map in square grids, and with 1.5xcost for moving diagonally they're pretty good at mapping semi-realistic movement. For outdoor environments hex grids can help by looking less like a rigid unnatural framework, it's easier to do things like meandering rivers on hexes. For large overview maps it's probably best to draw the world first and then just mark the provinces/whatevers, Risk-style.

If you really want to confuse your players use a triangular grid some time. The movement is pretty good if a single step allows you to move to any of the 12 triangles that touch one of the points of your current one. It just looks really weird, and it gets even weirder for large characters. Might be good for some sort of magic crystalline cave?

valadil
2016-04-30, 05:40 AM
Hexes are fine in the wilderness. I never got used to them in cities or dungeons.

Squares are fine in cities or dungeons. They're also fine in the wilderness.

OldTrees1
2016-04-30, 08:30 AM
In my dungeons I like 60 and 90 degree angles
I consider a hallway acceptable if: No drunken zigzagging while walking down a hallway & the hallway always has the same number of tiles in width
Square grids work for 90 degree angles
Hex grids work for 60 degree angles

To give a concrete example:
I have 7 rooms shaped like a hexagon. Each outer room has hallways leading to the inner room and both adjacent rooms. I addition the outer rooms have halls leading away from and halls running perpendicular to the inner room. (So the halls form a hexagon inscribed inside another hexagon). I have yet to figure out a grid that works for all the hallways much less allows me to neatly make rooms with both 90 and 120 degree corners.


Sidenote: If you want to make the ground seem iiregular or alien, you could always use a grid with various polygons. I made one using hexagons, squares & irregular pentagons(square + hex hybrid) that I used for caverns or limbo.

Chronos
2016-04-30, 08:37 AM
As far as approximating the real world where pi = 3.14159... and circles are round and so on:

Hexes are better than poorly-implemented squares.
Well-implemented squares are better than hexes.
And it's not hard to implement squares well.

Basically, it comes down to how you approximate circles. With poorly-implemented squares (without the one-and-a-half rule on diagonals), a circle is a square. With hexes, a circle is a hexagon. And with properly-implemented squares (with the one-and-a-half rule on diagonals), a circle is an octagon.

Plus, as others have noted, squares work a lot better for buildings and the like.


Sidenote: If you want to make the ground seem iiregular or alien, you could always use a grid with various polygons. I made one using hexagons, squares & irregular pentagons(square + hex hybrid) that I used for caverns or limbo.
And if you want to make the ground feel really natural and unplanned, use a Penrose tiling.

LibraryOgre
2016-04-30, 09:22 AM
Sidenote: If you want to make the ground seem iiregular or alien, you could always use a grid with various polygons. I made one using hexagons, squares & irregular pentagons(square + hex hybrid) that I used for caverns or limbo.

That's a very interesting thought. I could see that also being used in dreamscapes and the like.

Templarkommando
2016-04-30, 09:27 AM
From the players point of view, a hex grid means that you can only be surrounded by 6 medium characters. On the one hand this means that it's easier to surround your character, but on the other hand being surrounded by six characters isn't quite as bad as being surrounded by eight. So there's that trade-off.

As a DM, I've found that squares are a lot easier for mapping dungeons - especially those that tend to have square rooms. I'm sure that you can get hex paper, but I can't remember the last time that I saw a blank hex grid. The only time that I can remember seeing a hex grid in recent years was as part of an already assembled setting. I imagine I could order some off of the net or print some off by doing a google search for hex grid, but I'm just too lazy to do that.

Mr. Mask
2016-04-30, 09:33 AM
All this talk of squares makes me think we should just attach little distance trackers to miniature bases. They might be three inches long, wide enough to fit a miniature into them. The miniature can slide from one side of it to the other.

Like this:



_____
O
-----


You could then slide the miniature forward to move to the right:



_____
-> O
-----


You could then, without moving the miniature, move the slider underneath it forward:



-> _____
O
-> -----


Rinse and repeat in whatever direction you want to move your miniature. I bet someone has made something like this already.

Thrudd
2016-04-30, 09:58 AM
All this talk of squares makes me think we should just attach little distance trackers to miniature bases. They might be three inches long, wide enough to fit a miniature into them. The miniature can slide from one side of it to the other.

Like this:



_____
O
-----


You could then slide the miniature forward to move to the right:



_____
-> O
-----


You could then, without moving the miniature, move the slider underneath it forward:



-> _____
O
-> -----


Rinse and repeat in whatever direction you want to move your miniature. I bet someone has made something like this already.

Or just use tape measure

Mr. Mask
2016-04-30, 10:16 AM
I don't know, it seems more clunky in a way. If you want to be exact, you have to work out a standard of where you measure from, and at what point the miniature stops on the tape measure. That can take some fiddling.

Example A



1 - 2 - 3 - 4
-------------
OO --> OO
OO --> OO



Example B



1 - 2 - 3 - 4
-------------
OO --> OO
OO --> OO



There's about a 45% difference in distance in those examples, using the same tape measure. It should be possible to just decide on a method and run with it, but I see a lot of fiddlyness when put into practice.

ExLibrisMortis
2016-04-30, 10:22 AM
I like hexes on principle. Mostly because I prefer the CivV grid to the CivIV grid, and because there are no diagonal squares.

Thrudd
2016-04-30, 11:26 AM
I don't know, it seems more clunky in a way. If you want to be exact, you have to work out a standard of where you measure from, and at what point the miniature stops on the tape measure. That can take some fiddling.

Example A



1 - 2 - 3 - 4
-------------
OO --> OO
OO --> OO



Example B



1 - 2 - 3 - 4
-------------
OO --> OO
OO --> OO



There's about a 45% difference in distance in those examples, using the same tape measure. It should be possible to just decide on a method and run with it, but I see a lot of fiddlyness when put into practice.

Traditionally, that is the center of the base. That is why it is important to use miniatures of consistent scale and base size. If you're using commonly available miniatures from GW or Reaper, or the D&D plastic miniatures game, that's taken care of. You work all that out for your game in the beginning, so there's no question and it isn't an issue. I think the fiddlyness of estimating the center of the base is made up for with the convenience of not needing special mats, and the ease and naturalness with which the miniatures can be positioned relative to each other and the terrain. The system works perfectly well in miniature war games, none of which use grids.

Psykenthrope
2016-04-30, 05:39 PM
I prefer square grids because they work better with scenes constructed from LEGO. :3

That might be the best reason for using a square grid.

Now I'm wondering if there's a hexagonal product that fills a similar niche to LEGO's squares.


That's a very interesting thought. I could see that also being used in dreamscapes and the like.

Yeah, I might have to use it.

Player: "Why can't we move properly?"
Me: "Because this realm runs partially on dream physics."

GreatWyrmGold
2016-04-30, 07:24 PM
Basically, it comes down to how you approximate circles. With poorly-implemented squares (without the one-and-a-half rule on diagonals), a circle is a square. With hexes, a circle is a hexagon. And with properly-implemented squares (with the one-and-a-half rule on diagonals), a circle is an octagon.
Octagons look (slightly) more like circles, but I'd argue that hexagons act more like them.
http://www.informatik.com/mathfun/images/circlearray.png


Funny grids for funny places
I like the sound of that.


I don't use grids often, but when I do it depends on how I'm mapping them. I have a personal preference for hex grids—they feel more right to me for some reason—but it's nothing serious.
It doesn't matter much in practice, though, since when I GM I usually don't have any sort of table-map at all, or just a reference sketch. I've sometimes meant to, but it's usually more work than it's really worth. Especially when you're playing in a setting with sniper rifles...

NamelessNPC
2016-04-30, 08:12 PM
We've recently started to use hexagons in our group, and one of the obstacles we've come across is how to deal with large characters. How do you do it? We settled on a triangular shape, comprising 3 hexagons, but I don't like it very much

Beneath
2016-04-30, 10:44 PM
Sidenote: If you want to make the ground seem iiregular or alien, you could always use a grid with various polygons. I made one using hexagons, squares & irregular pentagons(square + hex hybrid) that I used for caverns or limbo.

If you want to make the world seem downright Lovecraftian you can add heptagons (http://www.roguetemple.com/z/hyper/rlyeh.png) into your tiling (source (http://www.roguetemple.com/z/hyper/gallery.php)).

Templarkommando
2016-05-01, 03:44 AM
There is actually a system where you just do away with a grid altogether and use rulers or some other kind of measuring stick.

This is a fairly short and decent explanation of that sort of system along with a good way to make measuring sticks: https://www.youtube.com/watch?v=XTT1ZU5pDBg

Knaight
2016-05-01, 03:55 AM
I have a certain fondness for hex grids, but square grids work much better for buildings and that tends to win out. I mostly play gridless games though.

Chronos
2016-05-01, 06:54 AM
If you want to make the world seem downright Lovecraftian you can add heptagons (http://www.roguetemple.com/z/hyper/rlyeh.png) into your tiling (source (http://www.roguetemple.com/z/hyper/gallery.php)).

Or for the same effect, use square rooms, but put five of them around each corner rather than four. Though either way does make it harder to draw the DM's copy of the maps, too.

ApocalypseSquid
2016-05-01, 09:23 AM
Neither/both! I use non-gridded maps for battles (using a 1-inch in game = 5 feet out of game, and using tape measures and such to do movement system), Square gridded maps for urban/dungon exploration/architecture, and either non-gridded or hex-gridded maps for wilderness exploration.

GreatWyrmGold
2016-05-02, 07:33 AM
We've recently started to use hexagons in our group, and one of the obstacles we've come across is how to deal with large characters. How do you do it? We settled on a triangular shape, comprising 3 hexagons, but I don't like it very much
Why not? I mean, it's not a perfect approximation of the square-D&D Large creature space, but that's as impossible as perfectly approximating a GURPS magician in D&D.

OldTrees1
2016-05-02, 10:17 AM
Why not? I mean, it's not a perfect approximation of the square-D&D Large creature space, but that's as impossible as perfectly approximating a GURPS magician in D&D.

The triangle has 120 degree rotational symmetry which differs from the hexagon's 60 degree rotational symmetry. This leads to the question of how to count 60 degree rotations of the triangle(inverting the triangle). Does it cost 5ft? If so then what if they need to rotate twice on a turn, is that overly harsh to charge 10ft? Personally I think counting it as 2.5ft is fair(since a 5ft slide is 2 rotations) and thus would count 0,5,5,10,10,15 ... for multiple rotations.

Spojaz
2016-05-02, 10:47 AM
I greatly prefer a square grid with no extra cost for diagonal movement.
A square grid ends up looking cleaner than hexes, and lends itself to understandable architecture a lot better.
Not using diagonal equals 1.5 or √2 or whatever is easy, doesn't add to the mental load of taking a turn or make you use complex shapes for AOE, and if it's good enough for kings in chess, it's good enough for me.

If anyone at my table is a stickler for realism, I can use the "space works a little differently in the magical world D&D" argument. In a world where pi is four, you can have an equilateral right triangle, geometry doesn't need so many complex numbers and you can fit a bit more beer in a glass. Everything is simpler and better in such a world.

Typewriter
2016-05-02, 10:52 AM
I like Hexes for movement and the mechanics of combat, but I prefer squares for building dungeons and buildings because of the ease of use regarding those shapes. I recently built a massive 'castle' out of a mix of Heroscape tiles (Hex) and Dwarven Forge (Squares).

halfeye
2016-05-02, 02:54 PM
The triangle has 120 degree rotational symmetry which differs from the hexagon's 60 degree rotational symmetry. This leads to the question of how to count 60 degree rotations of the triangle(inverting the triangle). Does it cost 5ft? If so then what if they need to rotate twice on a turn, is that overly harsh to charge 10ft? Personally I think counting it as 2.5ft is fair(since a 5ft slide is 2 rotations) and thus would count 0,5,5,10,10,15 ... for multiple rotations.

Think of a hexagon as 6 triangles.

Making a triangle of hexes might thus involve treating the hexes as divisible, but that should work out.

OldTrees1
2016-05-02, 03:22 PM
Think of a hexagon as 6 triangles.

Making a triangle of hexes might thus involve treating the hexes as divisible, but that should work out.

I don't understand what you mean relative to what I was talking about. I am not talking about rotating in place. Form a triangle with 3 coins. Moving the triangle to the right 1 step is equivalent to 5ft of movement. Now by moving only 1 coin, leaving the other two where they are, invert the triangle and move it slightly to the right with that 1 move. How much movement does that cost? (the coin moves right 1 hex + up&right 1 hex or right 1 hex + downright 1 hex depending on the facing of the original triangle) I called it rotation because it is equivalent to a 60 degree pivot centered on/holding 1 hex constant.

This is quite relevant since when a hex grid hallway makes a sharp turn it might take a different amount of movement to negotiate the curve depending on the facing of the triangle towards or away from the turn.

halfeye
2016-05-02, 03:41 PM
Huh? How is that relevant to what I was talking about? I am not talking about rotating in place if that is what you thought. Form a triangle with 3 coins. Moving the triangle to the right 1 step is equivalent to 5ft of movement. Now by moving only 1 coin, leaving the other two where they are, invert the triangle and move it slightly to the right. How much movement does that cost? (the coin moves right 1 hex + up&right 1 hex or right 1 hex + downright 1 hex depending on the facing of the original triangle) I called it rotation because it is equivalent to a 60 degree pivot centered on/holding 1 hex constant.
I don't know whether it's relevant to your issues, however it is geometically true. From a hex of six triangles it's trivial to make a triangle of nine lesser triangles, or one from a partial hex of four triangles. If you can see your circles as hexes divisible into triangles, then you can do this, if not then you have to use the circles as the bases for triangles and build hexes, if you want to, from there.

OldTrees1
2016-05-02, 03:45 PM
I don't know whether it's relevant to your issues, however it is geometically true. From a hex of six triangles it's trivial to make a triangle of nine lesser triangles, or one from a partial hex of four triangles. If you can see your circles as hexes divisible into triangles, then you can do this, if not then you have to use the circles as the bases for triangles and build hexes, if you want to, from there.

Ah, so you were talking about "a 3 hex triangle is defined as the 3 hexes that join at a shared vertex" which was assumed background information for anyone reading my post. I was talking beyond that about the consequence of a gray area in the movement rules when introducing a 3 fold symmetry space into a movement system that uses 6 fold symmetry. That gray area being the movement cost of a pivot to invert the triangle when maneuvering.

halfeye
2016-05-02, 06:03 PM
Ah, so you were talking about "a 3 hex triangle is defined as the 3 hexes that join at a shared vertex" which was assumed background information for anyone reading my post. I was talking beyond that about the consequence of a gray area in the movement rules when introducing a 3 fold symmetry space into a movement system that uses 6 fold symmetry. That gray area being the movement cost of a pivot to invert the triangle when maneuvering.
I don't know what you're talking about here really, but it sounds as if what I was thinking of would translate into suggesting you pick the three or six fold symetry on the fly, depending on which makes movement easier to calculate.

OldTrees1
2016-05-02, 07:39 PM
I don't know what you're talking about here really, but it sounds as if what I was thinking of would translate into suggesting you pick the three or six fold symetry on the fly, depending on which makes movement easier to calculate.


http://i.imgur.com/tnDNf6f.png

Look at the brown hexs. The 3 hexs represent a large creature and has 3 fold rotational symmetry (I will presume you will look up "rotational symmetry" if you are unfamiliar). The 7 hexes represent a huge creature and has 6 fold rotational symmetry. Everything I talk about from here on only applies to the 3 hex large creatures and is a direct result of their 3 fold rotational symmetry while living on a hex grid.

Now look at the red creatures. If the red creature moved from the left location to the right location that would be 5 hexes of movement (all 3 hexes moved 5 spaces )and thus 25ft.

But if we look at the green creatures we notice the 3 green hexes don't all move the same distance. One of those green hexes only moved 4 spaces unlike the other 2 green hexes that moved 5 spaces. At some point in the path the green creature pivoted 60 degrees around one of its hexes. How much movement would it cost total (20? 25? 3050?)?

Finally look at the blue and purple creatures. Consider the various step & pivot combinations they could use to negotiate the turn in the corridor.

This is a uniquely Large creature on hex grid result.

Now, where should I be clearer?

Aotrs Commander
2016-05-03, 04:43 AM
Dungeon floor plans and combat on a square grid (because I have to draw it (and not just the actual map, the grid itself - I have only once draw a hex grid and it's not something I have time to care to repeat), plus the size issues and the fact that they map to square dungeons better).

Large scale mapping - no grid at all. Any distance the PCs are going to be travelling will be pre-calculated (if they decide to wander of in a random directionn, there will be larger problems with the game than mapping) and if it's my own games, the map is a digital CAD image where I can measure the distances as accurately as I need to, and local details are generally unecessary to map anyway outside of actual combat areas.

Chronos
2016-05-03, 08:44 AM
You use digital CAD images, and yet you still draw all of your graph paper by hand?

Aotrs Commander
2016-05-03, 01:31 PM
You use digital CAD images, and yet you still draw all of your graph paper by hand?

Who said anything about graph paper?

Nah, the DM's maps are CAD, but I draw the maps for the actual combats on what is essentially a A1 plastic wallet containing (various colours of) hand-drawn square grid.

(Strangely enough, I don't have room for an A1 printer in the house, or it would be even easier...)

Best idea I ever had, far better than any other solution were ever had.

Lvl 2 Expert
2016-05-04, 12:21 PM
If you really want to mess with your players you could always pull an Escher-y disguise over the grid.

http://images4.ravelrycache.com/uploads/janukke/306241370/Fish_6_medium2.jpg
http://dorotheo.pbworks.com/f/1367240118/Butterflies-by-Escher-43392.jpg
http://www.josleys.com/htmlgalleries/Escher/tn_escher027g.jpg

Instant confusion.

OldTrees1
2016-05-04, 12:53 PM
If you really want to mess with your players you could always pull an Escher-y disguise over the grid.

Instant confusion.

That last one is a tessellation of irregular pentagons. Specifically is a square grid with some diagonal steps disabled.

Lvl 2 Expert
2016-05-04, 11:47 PM
That last one is a tessellation of irregular pentagons. Specifically is a square grid with some diagonal steps disabled.

Jup. To be fair, it does still look pretty weird without the humans.

http://www.wisfaq.nl/bestanden/cairo.gif

OldTrees1
2016-05-05, 06:47 AM
Jup. To be fair, it does still look pretty weird without the humans.

It does indeed.

Now with the clear image, I can see how the conversion from 45 degree rotated square grid to pentagons resulted in 2 hex grids with 90 degree offset.

This is fun. The first one is a triangle grid with the triangles trisected into kites. This forms both the 4 corner intersections where 2 triangles meet side-to-side and the 3 corner intersections in the middle of each triangle. (Alternatively you could view this one as a hex grid with each hex divided into 6 kites.)

The second one has pentagons with 4 3corners and 1 6corner. Is that a hex grid with the inner hex of every 7 removed? I think so.

Lvl 2 Expert
2016-05-06, 02:11 PM
The second one has pentagons with 4 3corners and 1 6corner. Is that a hex grid with the inner hex of every 7 removed? I think so.

I'm not sure. It might be. I can see it's (also) a triangle grid with each triangle split into three butterflies. The edges of the full triangles are ragged but rotationally symmetrical. Butterfly/color wise the "pointing left" and "pointing right" triangles are rotated versions of each other, not mirror images, and if you follow the pointing left ones up or the pointing right ones down the butterflies move one step clockwise with every step.

It's probably one of the more confusing examples of this kind of art I've seen. The pattern is pretty well hidden.

EDIT: O right, I haz the Gipmz.

http://s32.postimg.org/swbfnqdtx/butterflygrid.png

This is it I think. Dividing lines starting from the middle of the triangles, shifted away from being perpendicular to the line they end up on, creating a windmill pattern.

Not the best grid for gaming perhaps. :smallbiggrin: Although I do suddenly feel an urge to get myself one of those fancy tiled hallways.

EDIT again:

This is pretty much just the same thing as the shark, isn't it? Only those lines joined in the middle of the triangles. But same hexagon-triangle stuff going on.

Ah well, at least I can pretty.

http://s32.postimg.org/9z6gz5sp1/butterflygridcolor.png

Asmodean_
2016-05-06, 02:16 PM
Well if you end up adventuring in a four-dimensional hypercube for some reason then square grids clearly have their advantages.

halfeye
2016-05-06, 06:35 PM
Look at the brown hexs. The 3 hexs represent a large creature and has 3 fold rotational symmetry (I will presume you will look up "rotational symmetry" if you are unfamiliar). The 7 hexes represent a huge creature and has 6 fold rotational symmetry. Everything I talk about from here on only applies to the 3 hex large creatures and is a direct result of their 3 fold rotational symmetry while living on a hex grid.

What I was saying was that the grid could be thought of as like the hex in the middle of the seven hex "creature". As such, you'd probably give everything twice the moves, and have them move on the triangles, which are half the length (because 1/2 * 2 = 1)


Now look at the red creatures. If the red creature moved from the left location to the right location that would be 5 hexes of movement (all 3 hexes moved 5 spaces )and thus 25ft.

But if we look at the green creatures we notice the 3 green hexes don't all move the same distance. One of those green hexes only moved 4 spaces unlike the other 2 green hexes that moved 5 spaces. At some point in the path the green creature pivoted 60 degrees around one of its hexes. How much movement would it cost total (20? 25? 3050?)?

Finally look at the blue and purple creatures. Consider the various step & pivot combinations they could use to negotiate the turn in the corridor.

This is a uniquely Large creature on hex grid result.

Now, where should I be clearer?

I suspect we are entirely missing each other's points all the way along, no harm done and fun while it lasted, but probably no common point of reference either.


Well if you end up adventuring in a four-dimensional hypercube for some reason then square grids clearly have their advantages.

Sure, squares are also very good on a torus, but for a sphere you want hexes (and some pentagons that turn into irregular triangles on which you can draw irregular hexes).