Spherical global map

[macgregor]

Yes , they are flat and yes it is a spheroid 'thingy'. I like to think of it more as cylindrical. I believe I saw a hyperlink to it on one of these posts.

[Montbrun]

...It won't work. Look at WitP. Matrix FUBARed the map with some sort of pseudo-mercator projection, except they didn't apply any scale factor. Distances between bases are incorrect. Hex sizes would have to vary across the map in order to maintain the correct distances. Land masses would have to be distorted beyond recognition, which creates new problems. I'm a Licensed Professional Land Surveyor, and create maps every day. Let's keep it simple, and just use a "flat map," with consistent hex sizes, reflecting actual distances.

Brad

[Montbrun]

...actually, the earth is an ellipsoid, not a spheroid. Therefore, you have different scale factors depending on where you are on the surface of the planet.

Brad

[Montbrun]

I've whipped up a quick drawing to illustrate my point. This drawing is obviously an exaggeration, but it will work. The distance between two points at the earth's surface is based on the "ellipsoid distance." The "surface distance," in most cases, will be longer than the "ellipsoid distance." The coordinates of the surface points are calculated on a point-by-point basis, depending on the elevation above the ellipsoid. This is how your hand-held GPS units work. The surface elevation above the ellipsoid in NC, where I live, is between 20-40 meters. Also, the ellipsoid surface is based on local mean sea level.

[Neilster]

ORIGINAL: Brad Hunter

...actually, the earth is an ellipsoid, not a spheroid. Therefore, you have different scale factors depending on where you are on the surface of the planet.

Brad

I got this from the web...
"The circumference of the earth at the equator is 40,076 kilometers, and the equatorial diameter is 12,757 kilometers. The circumference of the earth through the poles is 40,008 kilometers, and the polar diameter is 12,714 kilometers. Therefore, the earth is slightly bulged at the equator and slightly flat at the poles. Thus, the true shape of the earth is best defined as an oblate spheroid."

Going by this, the circumference difference is only 68km. Isn't that close enough for a single scale? There are already plenty of simplifications with the current map.

Why can't we construct a roughly spherical object out of flat sided hexes to simulate the Earth?

Edited because our posts crossed:

I understand what you are saying about the difference in the distances but with hundreds of hexes, won't the difference be negligible?

Cheers, Neilster

[Montbrun]

The distances will be perfect at the "base-line." Assuming that the equator is our "base-line," the further we travel from that "base-line," the greater the distortion (toward the poles). This is the "pie-wedge" effect. An example is photogrammetric photos. When a plane flies along a given flight-line, taking photos, the surface detail at that flight-line is correct. The further away, towards the edge of the photo, that we travel, the worse the distortion is. We can also see this from readily-available on-line satellite photos.

If you've ever seen pictures of a Mercator Projection (which is intended for sea-travel and navigation), you'll see that, for example, Greenland is the size of, or larger than, the US. This is unacceptable for a wargame map. In order to maintain the integrity and "size" of the map, using some sort of projection, the hexes would have to vary in their size, or scale. You might have a hex that represents 120 miles across at the equator, but is only 60 miles across in, say, Canada.

I'm not saying that the map can't be done, I'm just saying that, seeing the map for WitP, I'm afraid that Matrix is not capable of "doing it right." Having said this, I'm not bashing Matrix or their games. As far as I'm concerned, Matrix (and 2by3) is the premier computer wargame producer, and deserves major cudos, producing games in a limited niche market, as they do.

Brad

[Neilster]

Are you familiar with the CWiF demo and the way the maps currently work Brad? I think I understand what you mean and I may have posted about it earlier when describing how a section of a spherical global map would be displayed in the main playing screen...

"I think a little distortion is necessary, probably toward the edges of the screen. As the map scrolls, some of the distorted hexes will move toward the centre of the screen and look normal again. The normal counters can still be placed on the edge hexes because they would only be a slightly different shape and a little bigger. Also, because the map is so big and the part of it we are displaying relatively small, the distortion required wouldn't be significant."

I've one of the lecturers from my university maths dept. investigating the feasibility of the hexagon surfaced spheroid I'm envisioning but he's an absent-minded professor and a busy man.

It's not the end of the world (no pun intended) if MWiF has a flat global map. I just thought there might be some advantages to a representation that, while still constructed of hexes, more closely modelled the (basically) spherical Earth.

Cheers, Neilster.

[Montbrun]

Interesting concept. I know that for WitP, they had initially considered having a "variable" movement system. Each TF would do a movement check based on where the TF was on the map, and this would determine distance travelled for the turn. Sadly, this concept was not implemented.

Anything can be accomplished with enough time, money and computer horsepower...

Brad

[coregames]

It is definitely true that game design is an art form, albeit facilitated through scientific and technical means. Hex-bases games, however, do not have the advantage of area movement games, where the shape of the spaces can be tailored to the needs of the situation. Too much fudging in a hex-based historical game can be noticably different than known geography and interfere with the credibility of the game. I think the icosohedral projections are great for games (i.e., GDW's Invasion: Earth). That being said, I wouldn't want WiF to be converted to an icosohedral projection, as cool a mapping convention as it is. Even as Patrice warms to the icos idea, I feel myself chilling towards it. If Rowland and Pinder had wanted to use such a projection it would have been part of the design from beginning. One small concession that might aid in the realism of the game might be to slice out some of the hex dots where possible in ocean areas at extreme lattitudes, but even that might be too much of a departure from what we recognize as WiF.

[coregames]

ORIGINAL: Neilster

Why can't we construct a roughly spherical object out of flat sided hexes to simulate the Earth?

Even with use of a Fuller geodesic, which completes a sphere using triangles, to be translated to hexagons will require that the vertices be depicted as pentagons, putting it right back to the icosahedral projection. I like the idea of using a more accurate projection for a global game, but not for WiF, since then the "faithful adaptation" will be out the window.

[SamuraiProgrmmr]

Already, I expect we have a cylindrical map that wraps east to west. I don't think any more is really needed for two reasons.

1. The polar areas which are most deformed are not playable anyway. No one is going to mount an invasion of Antartica and the Artic ocean is impassible for this time frame.

2. Any deformation can be 'absorbed' by the sea area. That is to say, the land hexes can always be the same size and the sea areas will be shaped funny in order to make the maps work.

The only place where this will be a big problem, IMO, will be North Central Russa. Little or no action really takes place there, so why bother. Wouldn't it be better to have a map that is easy to look at than one that is geographically precise? It is not as if the size of the hexes is challenged by the volume of units that are occupying them.

[coregames]

ORIGINAL: SamuraiProgrammer

The only place where this will be a big problem, IMO, will be North Central Russa.

In the game we're playing right now (just finished Jul/Aug 41), despite a strong CW that can't be dislodged from Denmark or the Netherlands (defensive shore bombardment), Germany has Moscow and Stalingrad, and Japan has Vladivostok and three Russian resouces. The Wermact is preparing to drive on those northern Asian Map factories, while Japan has cut rail at the bottleneck in Chita (offensive chit) for a lot of Russian HQs and other units to get out of Siberia. For some reason, the Russian player thought the strong CW presence meant he could avoid going to peace with Japan (big mistake); Turkey will probably be aligned before the end of '41.

I am noticing as Japan that each hex around Chita represents a subtantially smaller area than those closer to the equator, based on to-scale maps of the Chita region. It's frustrating having so many hexes to trace supply through and still be secure against partisans. Russia's defense in the north is made much easier by the cylindrical map.

Still, operations that far north are exceptional, and the maps are as they are. At least the Scandinavian map is its own creature, and probably is a much more accurate reflection of the geography there. I would like to see a better projection implemented in a different global game, hopefully by Matrix.

[coregames]

Since I'm shotgun posting in this forum tonight (not randomly though), I remembered this thread, and thought I'd point out to you WiFers out there that I took this idea to a thread in the General > General Discussion forum: tm.asp?m=806123 and got some fascinating feedback.