My attempt at a watery Mars...

[galileo]

Just a quick glimpse of a texture I'm working on. It's my first texture, so please bear with me. Just 2048x1024 for now. I can create and release a 16k version if there's enough demand.

Anyway, comments and critiques are most welcome.


Colour Map


Bump


Specular

The above 800x400 jpegs link to 2048x1024 pngs. The bump map was stitched and process by myself from Mars MOLA data. The colour map started life as the redmarsv3 beta texture from space-graphics.com.

Just stick the PNGs in yer hires dir, and set up an AltSurface or however you want to do it. If I release the 16k version, I'll do it properly with the ssc files, etc.

Here's a file I called waterymars.ssc, which might work o.k.

Code: Select all

AltSurface "Mars Con Agua" "Sol/Mars"
{
    Texture "bluemars-color.*"
    BumpMap "bluemars-bump.*"
    BumpHeight 4
    NormalMap None
    Color [ 0.27 0.3 0.27 ]
    SpecularTexture "bluemars-spec.*"
    SpecularColor [ 0.35 0.3 0.2 ]
    SpecularPower 50.0
}

[Don. Edwards]

Hey Galileo,
Nice to see someone else giving this a try. I was planning on revisting this project again but I am trying to figure a way of implementing weathering of the bumpmap to show the natural erosion that would take place. With that much water being introduced into the atmosphere and on the ground there would be a great deal of water based erosion. Just something to think about for your next version. Also you might consider taking the saturation values down several clicks. That is probably the first thing a new texture artist has to learn. Less color is, in reality better.
Good Luck.

Don. Edwards

[selden]

It looks like a great start!

A quibble, though: the maps are offset by 180 degrees from Celestia's standard orientation. They have 0 degrees of longitude at the edges but Celestia expects 0 degrees of longitude to be in the center. This means that Celestia will draw your Mars with the wrong side toward the sun, for example.

You can fix this by cutting the maps in half and exchanging the left and right halves.

[rthorvald]

It looks like a great start!

A quibble, though: the maps are offset by 180 degrees from Celestia's standard orientation. They have 0 degrees of longitude at the edges but Celestia expects 0 degrees of longitude to be in the center. This means that Celestia will draw your Mars with the wrong side toward the sun, for example.

You can fix this by cutting the maps in half and exchanging the left and right halves.

Oops, this is something i haven?t thought of, and am completely ignorant about. Forgive me for stupid questions... But are you talking about where to position the meridian on the sphere?

If so, is there any way to describe this in the ssc instead of doing something with the graphics? Easier if you have a few hundred VT puzzle pieces...

Another thing, too: if one can define this with numbers, what is a good resource for _laymen and idiots_ to (learn and) find such data for the planets?

-rthorvald

[selden]

Right: Celestia expects the meridian to be down the center.

To put it another way, Grant painstakingly went through solarsys.ssc, setting the RotationOffset of every body for which we have surface maps. Their meridians all are in the correct relationships to Celestia's coordinate system. All of those planets, moons and comets will be drawn with the correct phase if their surface texture maps have 0 degrees of longitude in their centers. For example, if you observe them telescopically from the earth, you'll see the correct surface features (allowing for light time delays, of course). Also, the correct areas will be sunlit at any given time -- you'll see the sun rising over the Spirit rover when it should.

You can change the definition of a body's RotationOffset to compensate for a different offset on any particular map (add or subtract 180 degrees in this case), but that map's surface features still won't align with the same features on other maps of the body. Having the right alignment is particularly important, I think, when you have several different maps of a planet and want to flip back and forth among them to compare the differences.

Added slightly later:
Don't forget that you don't have to use a GUI interface and strain your eyes and wonder if you managed to cut the image in the right place. Most paint programs have a script interface to simplify these kinds of operations.

Also, if you have freeware graphic tools like NetPBM or ImageMagick packages installed on your system, making corrections like this is easily specified numerically and in a reusable script. For example:

Code: Select all

pnmcut -left 0 -width 1024 map1.ppm >tmp1.ppm
pnmcut -left 1024 -width 1024 map1.ppm >tmp2.ppm
pnmpaste -lr tmp2.ppm tmp1.ppm >map180.ppm

swaps the halves of a 2K map image.

[maxim]

With a VT you should get along by simply shifting the column numbers. You don`t have to reassemble the whole thing (at least when you`re 180° off). Try '1-4a rename' for that task. It does almost every (im)possible batch renaming task you can think of.

maxim

[galileo]

0 long down the centre, eh? will do thanks for your comments everyone. i definitely want to add things like erosion, weathering, clouds, etc. i'll fix the longitude problem and update the files. thanks!

edit: okay, uploaded. refresh thanks again.

[JimBim]

An idea: you could also add cloud maps (like earthclouds.*), maybe not very realistic, but since there is water on Mars, there must be some clouds around.

[eburacum45]

Lovely texture; wish mine looked that good...
I would have expected the desert areas to retain some hint of red colour, though; if you look at the Sahara from space it does look reddish...

although advanced terraforming techiques might make the whole of Mars a green paradise, and thereby make it more comfortable for humans.

[John Van Vliet]

this remindes me ,the offset is something i keep forgetting there is a new mars normal up but it is 180 deg. offset
hera have a look at mine , it's old but looks fairly good