Matrix Games Forums

ORIGINAL: Neilster

Say there's a spherical global map. Might not a roughly circular, blobby main screen be able to have hexes about the same shape? Dunno. Bigger in the middle I guess. Maths might be easier? Doesn't fit so well on a rectangular monitor screen perhaps. Maybe less game area.

Possible subtle elegance to solution in that displaying the main display map in MWiF is like it is captured by a satellite in a low orbit around planet WiF. That's essentially what a spherical global map implementation would do, putting a virtual camera above a virtual 3D map and displaying what it sees. Would the satellite return a circular view of the planet?

I'm sort of lost here which map do you want to replace [&:]

Currently CWiF displays 3 maps at the same time. Which can be modified (or eliminated) at the players discretion.

(1) #1 Map Window (Linked) - detailed flat map (covers an area about the size of Poland in hexes)

(2) Globe - Only active air and naval units are displayed. Example of displayed area:
(North - South) Norway to Kenya
(East - West) Urals to Greenland.

(3) #2 Map Window (Linked) - Example of displayed area:
(East - West) Benghazi to Alexandria and about 6 hexes to to bottom.

I was arguing the merits of replacing the current flat global map with a spherical one. Imagine a globe of the Earth tiled with WiF hexes.

The Earth is a sphere; so why not model it that way if we can, using a computer to keep track of what's happening on the surface of a hexy, virtual Earth, 1939?

I proposed that do so may not be terribly hard to program and has some advantages. It would also "unify" the game onto a continous, true scale map, where the movement/combat engine can have free reign with no "edge of map" effects. This may make possibe some simplication of the rule set and provide opportunities for gameplay efficiencies.

Shouldn't there be a reduction in the hex count, as many useless distorted polar hexes disappear? I think so. As I understand it, at the moment, that alone would speed gameplay later in the war when counter densities rise.

The flat global map in the CWiF demo doesn't really exist. It's just a bunch of zeros and ones in your PC's memory. The computer doesn't find the nasty 3d maths required to book-keep a spherical map any harder really. Programming the representation is the tricky bit.

To display the "detailed flat map" or "#1 Map Window" as you say Mzlin, we just display the relevant bit of the 3d map that's in memory. That's what CWiF does now. It's just that currently the global map in its memory is flat. Gameplay-wise, things could stay the same. I think we might be able to get away with 2 maps but we could have three too. It's just a another, zoomed out, view of the hex tiled globe.

Can anyone tell me the dimensions of the latest global map? Is it about 300 hexes by 200 hexes? That's 60 000 hexes square. I'm trying to do some maths.

Cheers
Neilster

I'm just going to throw this out there cause my brain is so simple...yes I'm a simpleton! One world=2 hexes. Each hex=1 hemisphere. Look at a hex and imagine abstractly that it is a circle, then imagine concentric circles. Take a hex...now add another group of hexes on its circumference....another set of hexes on the added hexes' circumference and on and on. Set the scale for the basic subhex and fill the hemishere hex with subhexes. Allow the players to choose any parental size hex representing the area of their chosen observation and then allow max zoom to as low as the basic hex. All will be in perspective as the polar regions will have fewer basic subhexes compared to the greatest amount of basic subhexes at the equator.

The Fuller projection is modular in such a way that any two adjacent triangular panels can be displayed as adjacent in a flat way... no need to show any large black regions, just bring the two faces of the icosohedron together when scrolling over that connection. The only change would be the compass, which would shift by 30 degrees, but with the mechanics of weather zones, direction never comes into play anyway. Also, the vertices where the corners of 5 faces meet could simply include a "partial" hex with one slice missing, and as you scroll away from the vertex the map could automatically "refold" itself to stay flat from that perspective.

Looking at that animation of the earth being transformed into the icosahedron and then flattened out, it occurs to me that perhaps it would make sense to use triangles instead of hexagons for the map grid if you wanted to use the Fuller projection. After all, a hexagon is really just 6 adjacent triangles.

ORIGINAL: rtamesis

Looking at that animation of the earth being transformed into the icosahedron and then flattened out, it occurs to me that perhaps it would make sense to use triangles instead of hexagons for the map grid if you wanted to use the Fuller projection. After all, a hexagon is really just 6 adjacent triangles.

The code for this game largely exists already in the form of Chris' CWiF beta, and Matrix have said that they aren't going to make major changes. That's why I suggested changing the map in such a way that everything else stays the same. I don't think radical changes to the hex-based nature of WiF are going to get up.

In the current demo, we can circumnavigate the world by continuously scrolling sideways, so any change must still enable this to occur. If, as stated by coregames above, a Fuller projection implementation would involve a sudden 30 degree change in direction of the map when moving from one surface to the next, it will look crap and be very annoying.

I think the angles of the surface of the map must vary evenly and continuously in every direction. The Earth is a sphere, so why, if the computer can look after all the annoying details, don't we use a model that closely resembles one? That's why I suggested a spheroid constructed of hexes. Nothing else needs to be changed except the way the map is stored in memory and displayed. We would be able to do a North-South circumnavigation with a spherical map too.

I have to do some study on theoretical mathematical tilings.

Cheers
Neilster

Actually, now that I've taken a look at some online resources, Fuller has a flat icosohedral projection that unfolds and keeps all the land masses contiguous, so that the change moving over a corner of the icosohedron would only happen over water, thus preserving the hex grid without refolding. Higher and lower lattitudes would be much more accurately depicted than in the current cylindrical CWiF map.

Here is a link to a Fuller map site: http://www.bfi.org/map.htm

and here is a clue to how accurate it is:



Also, I see Patrice's point that the unified scale could at least affect the game in a balanced way, as long as you were playing with divisions and artillery to help fill in that sparse Chinese line.

[:)] Coregames you might want to look at Sedecula's post 8/5/2004 3:50:30 PM on this thread. [:)]

I'm been doing some reading on different map projections, and it seems that Fuller's icosahedral projection of the world is still the best from the standpoint of minimal distortion of the land masses and adaptability to wargaming. While it is already unlikely that the existing mapping system in CWIF will be changed in MWIF, this discussion might be of use in future games that involve the world (eg a WWIII game).

I think that this projection can be easily mapped onto hexes. The most difficult and time consuming part will be determining the scale of the projection onto the hex grid. From an object oriented programming standpoint, if each hex is then numbered and considered a cell or an object that stores information about the area it represents (movement point cost to enter a side, combat modifiers, graphic display, etc), then the different hexes/objects can be stored in a database. A mapping display class/object could then be in charge of assembling these hexes to dynamically display a particular area of the world and rotating the dynamic map in either 30 or 60 degree increments.

ORIGINAL: rtamesis

I'm been doing some reading on different map projections, and it seems that Fuller's icosahedral projection of the world is still the best from the standpoint of minimal distortion of the land masses and adaptability to wargaming. While it is already unlikely that the existing mapping system in CWIF will be changed in MWIF, this discussion might be of use in future games that involve the world (eg a WWIII game).

I think that this projection can be easily mapped onto hexes. The most difficult and time consuming part will be determining the scale of the projection onto the hex grid. From an object oriented programming standpoint, if each hex is then numbered and considered a cell or an object that stores information about the area it represents (movement point cost to enter a side, combat modifiers, graphic display, etc), then the different hexes/objects can be stored in a database. A mapping display class/object could then be in charge of assembling these hexes to dynamically display a particular area of the world and rotating the dynamic map in either 30 or 60 degree increments.

Now that you said it this way, I can't help thinking that WiF FE would be the best candidate to have such a map. It would add a lot !!!!
After all, part of the pleasure we have in playing WiF (at least for me) is :
- Having the whole world mapped and ready to receive troops.
- Having very beautiful counters.
- Being such a good wargame !!!!!

So if the map of the Computer version could be both beautiful and exact to reality, and best of all, movable so that I see it from where I want, thsi would be a jewel !!!!!!

Cheers !!!

Patrice

In case people want to read more about it, here are some informative sites that discuss mapping hexagons onto an icosahedron representation of the world:

http://c2.com/cgi/wiki?HexGridIcosahedron
http://www.galaxy.gmu.edu/~dcarr/lib/v8n2.pdf
http://www.epa.gov/wed/pages/staff/white/envmon.00.pdf
http://hcl.harvard.edu/mercatorglobes/a ... gator.html
http://www.fs.fed.us/pnw/wenlab/pdf/gdggs03.pdf

If you didn't want to use fuller's projection (since that is patented), then an equal area icosahedron will still work as far as mapping to hexagons:

http://www.ilstu.edu/microcam/map_proje ... on_E_a.pdf
http://www.progonos.com/furuti/MapProj/ ... jPoly.html

Here's sites for software that generate hexagonal map grids from icosahedrons:
http://www.epa.gov/wed/pages/staff/white/getgrid.htm

For an example of a hexagonal grid mapped unto a flattened icosahedron:

http://www.galaxy.gmu.edu/~dcarr/lib/v8n2.pdf

http://www.sjgames.com/gurps/books/Spac ... netMap.gif

There are many more sites out there that you can find using Google.

Here's a site where you can download a nice icosahedral representation of the Earth:

http://www.solarviews.com/cap/ico/icoearth.htm

And here's another one:

http://www.progonos.com/furuti/MapProj/ ... edron.html

does anyone care that most wargame maps are "fudged" to make game systems work? all this attention to detail is thrown out the window when it comes time to recreate historical movements and capabilities.

no map remains un-retouched when it comes in contact with a game system.
no initial unit movement rate survives when it is time to recreate historical ranges of operation.
no logistics calculations remained untouched when it comes time to play-balance a game.

IMO spherical flat-map projections are ugly and not very much fun to play wargames on.

remember the operative word here: "Game"

Finally, this geodesic map grid divides the spherical surface of any globe into twenty triangles, each of which is further divided into hexagons. This can be used as a starting point for mapping the world onto hexagon cells/objects for storage in a database.

http://www.downport.com/bctc/Forms/IS_F ... rldMap.pdf

The point of all this discussion is not to create a spherical flat map projection but to find a way to map the entire world into hexagons for database storage. These hexes can then be dynamically assembled to create a local map if you want to focus on one part of the globe such as Europe or the Middle East or Southeast Asia. You will then be able to continuously scroll to other parts of the globe and have the software assemble the necessary hexes to display the new area and be assured that there is minimal distortion in terms of area, which is important for accurate movement of units.

minimal distortion in terms of area, which is important for accurate movement of units.

my point is though, that the above statement is not true of the game design process. every (board) game i have designed and co-designed has tweaked (ie. fudged) the maps to preserve historical accuracy of unit capabilities.

designer's may want you to believe design is a scientific process; but it's not, it's an artistic process.

Creating a database of a hexagonal grid of the world does not preclude some artistic license when it comes down to drawing the details of the map itself. By taking each hex and then subdividing it further into smaller hexes, you can create as much detail as you want in an area and then modify the exact placement of geographic features depending on the circumstance and the game's requirements. Having such a database of the world as a reference should actually make the creation of maps a little easier and back up the art of wargame map making with some science.

I've seen the Marinacci spherical map. Apart from the graphics which were a step back to the early version of WiF, I thought it was totally awesome. I think it'll work fine. You should be able to toggle through your units better. Perhaps by allowing them to be ordered according to type and supply source HQ would help to avoid 'forgetting' a front .

ORIGINAL: macgregor

I've seen the Marinacci spherical map. Apart from the graphics which were a step back to the early version of WiF, I thought it was totally awesome. I think it'll work fine. You should be able to toggle through your units better. Perhaps by allowing them to be ordered according to type and supply source HQ would help to avoid 'forgetting' a front .

Whoa. I didn't know there was one. Do you have a link or can you send me an image by PM? Does it use flat hexes (and is hence about a 50 thousand-sided spheroid thingy) or curved ones? I was thinking of keeping the hexes flat and changing the angle from hex to hex by about 1 degree (if the map is about 360 hexes wide, wrapped around a 360 degree equator, that should work).

Cheers, Neilster