Slashdot Mirror
Blue, Not Red: Did Ancient Mars Look Like This?
astroengine writes "Using elevation data from NASA's Mars Reconnaissance Orbiter, software engineer Kevin Gill was inspired to create a virtual version of the red planet with a difference. 'I had been doing similar models of Earth and have seen attempts by others of showing life on Mars, so I figured I'd give it a go,' Gill told Discovery News. 'It was a good way to learn about the planet, be creative and improve the software I was rendering it in.' He included oceans, lakes, clouds and a biosphere — a view of a hypothetical ancient Mars that looks wonderfully like home."
So the typical Martian was one ugly motherfucker, then? "Ain't got time to bleed!"
Props for realizing that a Mars covered with water would be blue, too. Such insight!
This is where using 2-3 minutes to read the fine article would have helped you out.
It is a software guy who just wondered what it would look like with earthly features. This is not based on any kind of facts other than the elevations.
It is an interesting exercise.
But I notice the renderings show a lot of nicely circular lakes, suggesting meteor impact craters. If Mars at any time had this amount of water and a thicker atmosphere there would likely be less craters and those that did remain would probably have different shapes due to erosion. It would suggest the meteorite impacts happened after the water evaporated and the atmosphere thinned.
Call me cynical, but this is pretty much a case of "Look, ma! We got some fancy 3D graphics now!". But it's not particularly interesting or novel from a technical viewpoint - even bad Hollywood movies have more professional graphics than this.
I mean, all he did was slap some Blue Marble textures onto a Martian height field globe. Wo-hoo, score one for physical simulation, and all that. As someone else has said, score one for the realisation that the planet would have been blue, if there had been large amounts of surface water. Wo-hoo! :-)
Now if he had done some actual simulation on where large bodies of surface water have likely existed: seas are sort of obvious, but what about rivers and lakes - these are extremely important for life, due to being sources of fresh water, as opposed to the inevitable salt water in the oceans. That, coupled with a simulation how life could have spread. Parameterised by how advanced the lifeforms are - move a slider from "basically just slime in the ocean" to "higher plants", and watch the green spread into those regions that could sustain it... that would be news. But this? They cover texture mapping and in-painting in computer graphics 101 these days.
That having said, the images *are* pretty, so it's not all bad. :-) Just not that much of news for nerds.
People seem to forget that after its formation the sun was somewhat LESS bright than it is now so Mars would have been even colder in its current orbit. If there ever was large amounts of water on Mars I suspect that it would have spent most of its time locked up as ice sheets with the occasional melting due to impacts. Pretty much the way it is today.
All this warm wet life on mars stuff strikes me as nothing more than wish fulfillment - the same way people used to imagine Venus was a tropical paradise. Until the probes went there and proved those predictions to be some of the worst ever made in astronomical science.
Is it just me, or is that planet getting closer?
I've been (very slowly) doing something a bit similar with the moon --- see here --- although differently; I've been trying to render everything and producing ground-level views rather than producing a painted sphere like TFA. (His looks better from a distance. Mine looks better close up.) I've been trying to use procedural texturing and atmospheric effects. The pictures above are rather out of date; rendering your own from SVN will look better.
Unfortunately rendering things the size of planets from very close up runs into big problems with floating point precision. The only renderer I've found which will do it at all is Povray, and even then there are loads of bugs --- volumetric effects for things like clouds is well buggered at this sort of scale. See this picture for an example. Plus Povray's is really slow at procedural surfaces.
Right now I really need to start again from scratch using higher-resolution terrain and gravity data from some of the recent lunar probes, and I also probably want to switch to a different renderer which works at higher precision. Any suggestions of a fast raytracer that does procedural isosurfaces, volumetric effects and works at double precision will be gratefully appreciated...
I will also share this test render with you, which I think is delightfully surreal...
It implies that somebody (perhaps the submitter?) thought that the simulation is intended to be accurate.
As parent says - read TFA, it's meant to be a creative exercise.
Also read Kevin Gill's own explaination.
If Mars ever had chlorophyl containing life, it would have left and oxygen atmosphere which Mars hasn't got.
Yes, because Oxygen is non-reactive and couldn't "disappear" into complex molecules.
Upward mobility is a slippery slope - the higher you climb the more you show your ass.
Show the nerds a beautiful picture and they'll totally miss the point and dissect it to death. Good thing this is not a beautiful naked woman. They'd be complaining that the angle of the elbow isn't quite right and prove it with a mathematical formula.
How about a moderation of -1 pedantic.