Other World Mapper Forum

Big Mac wrote:Yay! That is exactly how I wanted it. Thanks Alejandro!

What did you think of the last update?

- Alejandro

Kanchou wrote:

Big Mac wrote:Yay! That is exactly how I wanted it. Thanks Alejandro!

What did you think of the last update?

I just saw it! It is amazing.

From what I can see in August Update and Dev Log I think you have actually solved the polar problem.

I'm considering your comment on the addition of other mapping projections. I don't know enough about the different projections to know what ones might be useful. I do know that there are lots of different projections and that each type would take time for you to include more than the six you have already incorporated. If I can work out any specific projections that would have some sort of advantage to world building, I will let you know. But I suspect that the six projections you have chosen yourself cover everything I would ever need to do with my small amount of talent. I think that, by the time I've got good enough to get the full potential from OWM 1.0, you will already have had the time to add more, if necessary.

One thing you might need to consider, at some stage, are some projection types that work for fantasy worlds. I know that some mathematicians have written about mathematical models for different types of worlds, but I'm not sure there are full-blown cartographic projection methods that do the different things that the various spherical projections do.

A "flatworld projection" would be pretty simple, but a "cubeworld", "torusworld" or "hollow world" projection method might be something you need to invent. (I think you should get the people on your programming team to draw straws to see who gets a fantasy projection method named after them first. You might be able to have Wikipedia articles written about any fantasy projection methods you can create, especially if you can turn them into industry standards and get other people using them. )

I've created a topic called: [cartography] OWM team solves "the polar problem" at The Squishy Bits forum at The Piazza, that you might want to keep an eye on.

EDIT: That Piazza topic is now the featured story on the Facebook page for The Piazza and the Facebook group for The Piazza.

Big Mac wrote:I'm considering your comment on the addition of other mapping projections. I don't know enough about the different projections to know what ones might be useful. I do know that there are lots of different projections and that each type would take time for you to include more than the six you have already incorporated. If I can work out any specific projections that would have some sort of advantage to world building, I will let you know. But I suspect that the six projections you have chosen yourself cover everything I would ever need to do with my small amount of talent. I think that, by the time I've got good enough to get the full potential from OWM 1.0, you will already have had the time to add more, if necessary.

I've just found my old reply to the "Mapping the planets of Spelljammer" topic at the SJ forum at The Piazza, where (after three years of not being able to work out why the maps were shaped that way) I finally discovered that they were in Eckert IV projection (or possibly Robinson projection or Kavryskiy VII projection).

These broken maps (with "Pac Man poles") are the actual reason I've been wanting to learn cartography for several years. I might be able to get a close approximation with Winkel Tripel projection you have already made, but I don't think it would be close enough to make me happy that they are "fixed", because I want to fix the broken poles, but not change the "feel" that the TSR cartographer was trying to create. If you are able to add a Eckert IV Transformation, at some point, I can import the maps in Eckert IV projection fix the maps, in Transverse Mercator and send them back to Eckert IV projection, to make sure I've not messed up the rest of the world.

I can also then change my fixed maps from Eckert IV to something that looks less "weird". (My friend Silverblade, who makes 3D space art says I need to make 1:2 ratio rectangles to wrap around spheres, and I would prefer to do most of my world maps in that format.)

Big Mac wrote:
These broken maps (with "Pac Man poles") are the actual reason I've been wanting to learn cartography for several years. I might be able to get a close approximation with Winkel Tripel projection you have already made, but I don't think it would be close enough to make me happy that they are "fixed", because I want to fix the broken poles, but not change the "feel" that the TSR cartographer was trying to create. If you are able to add a Eckert IV Transformation, at some point, I can import the maps in Eckert IV projection fix the maps, in Transverse Mercator and send them back to Eckert IV projection, to make sure I've not messed up the rest of the world.

I'll add Eckert IV to our wish list We'll see if we can have it in with the first release.

Big Mac wrote:
I can also then change my fixed maps from Eckert IV to something that looks less "weird". (My friend Silverblade, who makes 3D space art says I need to make 1:2 ratio rectangles to wrap around spheres, and I would prefer to do most of my world maps in that format.)

That's what the equirectangular projection is. The equirectangular projection has no use in real cartography because of the types of distortion it creates. (i.e. no useful for navigation). However, because it is perfect for texture mapping (mapping to a sphere) it's used a lot in computer graphics. It's also very simple because the grids are equidistant.
Ideally, you would create a map in a projection that has less distortion to keep you doing things right (i.e. Winkel Tripel), then use a polar projection to fix the poles (i.e. Transverse Mercator) and then transform the finished map to an equirectangular projection and export it for easy texture mapping.

Also, you can use create an equirectangular map to begin with and use the other projections as guidelines to fix the maps.

I saw you talked about that in the Piazza groups, but I didn't want to just interject into the middle of the conversation . Hope this makes more sense!

- Alejandro

Some other projections I would be happy to see (but they're a “nice to have” at this point - the existing projections cover the basics):

- Mollweide: it has two advantages - we have no equal-area projections so far, I think, and it is one of the most popular projections used for world maps
- Eckert VI is another nice equal-area projection if you don’t like Mollweide for some reason
- something like Peirce quincuncial projection would be extremely useful to me, personally, because I have a world centered around a huge polar continent (alternatively something like azimuthal equidistant but Peirce is especially nice because it is on a square)
- a gnomonic or stereographic projection of both hemispheres with selectable central point - or some other projection that allows to show two hemispheres as separate circles

Two more that might be worth considering but are more situationally useful:

- interrupted sinusoidal - equal-area with little distortion
- General Perspective projection for people who want to make maps “as seen from space”

Stefan,

We'll try to add at least some of them for the first V1.x release . We setup the code to be pretty modular, so Mollweide and Eckert VI shouldn't be too bad. Interrupted sinusoidal would be a bit harder to add, but we'll have it as part of one of the V1.X releases.

Thanks!

- Alejandro

Kanchou wrote:Stefan,

We'll try to add at least some of them for the first V1.x release . We setup the code to be pretty modular, so Mollweide and Eckert VI shouldn't be too bad. Interrupted sinusoidal would be a bit harder to add, but we'll have it as part of one of the V1.X releases.

Thanks Alejandro! I didn’t expect you to prioritize any of this. As I said - the basics are already covered.

Any chance of adding Peirce quincuncial at some point? It’s a pretty rare one but it would be immensely useful (at least to me).

stefanstr wrote: Thanks Alejandro! I didn’t expect you to prioritize any of this. As I said - the basics are already covered.

That's one of the perks of being in the alpha/beta

stefanstr wrote: Any chance of adding Peirce quincuncial at some point? It’s a pretty rare one but it would be immensely useful (at least to me).

That one I had to look up. I'll give it a try if the math isn't too crazy. Else, it will come in one of the post release patches/small feature additions.

- Alejandro

Kanchou wrote:

Big Mac wrote:These broken maps (with "Pac Man poles") are the actual reason I've been wanting to learn cartography for several years. I might be able to get a close approximation with Winkel Tripel projection you have already made, but I don't think it would be close enough to make me happy that they are "fixed", because I want to fix the broken poles, but not change the "feel" that the TSR cartographer was trying to create. If you are able to add a Eckert IV Transformation, at some point, I can import the maps in Eckert IV projection fix the maps, in Transverse Mercator and send them back to Eckert IV projection, to make sure I've not messed up the rest of the world.

I'll add Eckert IV to our wish list We'll see if we can have it in with the first release.

That would be really awesome.

I'm not quite ready (I think I might need to buy a new computer and then see if I can upgrade my laptop after that is online) but if I could start fiddling with Eckert IV maps, I can hopefully impress some of the Spelljammer fans that have been looking at those broken maps with me for several years.

Kanchou wrote:

Big Mac wrote:I can also then change my fixed maps from Eckert IV to something that looks less "weird". (My friend Silverblade, who makes 3D space art says I need to make 1:2 ratio rectangles to wrap around spheres, and I would prefer to do most of my world maps in that format.)

That's what the equirectangular projection is. The equirectangular projection has no use in real cartography because of the types of distortion it creates. (i.e. no useful for navigation). However, because it is perfect for texture mapping (mapping to a sphere) it's used a lot in computer graphics. It's also very simple because the grids are equidistant.

Ideally, you would create a map in a projection that has less distortion to keep you doing things right (i.e. Winkel Tripel), then use a polar projection to fix the poles (i.e. Transverse Mercator) and then transform the finished map to an equirectangular projection and export it for easy texture mapping.

Also, you can use create an equirectangular map to begin with and use the other projections as guidelines to fix the maps.

Thanks. That sounds fairly logical.

I think I might eventually pick up the flow of this and be able to do this without scratching my head too much.

Kanchou wrote:I saw you talked about that in the Piazza groups, but I didn't want to just interject into the middle of the conversation . Hope this makes more sense!

If you are willing to join in the conversation (and talk to cartography fans who might not necessarily have backed OWM) please feel free to butt in any time you want.

I've gone from someone who knew nothing about cartography to someone who has learned enough to "give it a shot" by meeting people who are better at maps than me who have been willing to talk to people who are not as good as them. In fact, I wouldn't have seen this Kickstarter, if it wasn't for Anna Meyer PMing me on Facebook and telling me to jump onboard.

And if it wasn't for Thorfin Tait talking to me about the polar problem, I would not have seen this project as something that could solve a mapping problem that I would like to solve, but which I did not have the ability to solve.

So I'm not saying that letting Piazza members talk to you will get you new customers, but it might inspire some more people to try out maps. And if more people make maps, that's good for fans of maps, in general.