Page 1 of 17
Experiments..
Posted: Tue Jun 12, 2007 1:34 am
by OnoSendai
Some procedural skies rendered in Indigo.
Top one uses an atmospheric scattering simulation, bottom one is just a cloud scattering model in front of the usual physical sky background approximation.
Clouds are constructed using procedural Perlin noise of several frequencies.
Will post more info in a blog post soon.
On the topic of non-uniform scattering, is there anything in particular people want to see? (Volume emitters should be doable as well.)
Posted: Tue Jun 12, 2007 1:46 am
by Phoenix
Wow looks very interesting!
I'm greatly anxious to see more about it!
And I hope it'll find its way in one of the next releases soon.
Posted: Tue Jun 12, 2007 1:46 am
by SmartDen
Wow, it looks beautiful, gorgeous, wonderful, awesome!!!
Posted: Tue Jun 12, 2007 1:57 am
by psor
Posted: Tue Jun 12, 2007 2:29 am
by StompinTom
looks awesome! i definitely like the top one. very surreal. be nice to see it completely smooth.
Posted: Tue Jun 12, 2007 2:34 am
by Zom-B
thats totally awesome dude!!!!!
The implementation of both modes would be nice too
How about render times / Muts per pixel compared to an cloud free sky?
Would love to see the atmospheric clouds rendering longer
now I have to stay drunk till first 0.9t1 release, to avoid getting lost in total hysteric impatience
Posted: Tue Jun 12, 2007 2:40 am
by OnoSendai
Top image is ~ 6 hours, 2 threads, Core ~1.6 Ghz
Bottom image is a few hours on same.
There's a lot of work to do on making the sampling more efficient tho.
Posted: Tue Jun 12, 2007 2:47 am
by SimonLarsen
now I have to stay drunk till first 0.9t1 release, to avoid getting lost in total hysteric impatience
Great idea!
But almost out of beer.
Posted: Tue Jun 12, 2007 2:56 am
by Kram1032
Posted: Tue Jun 12, 2007 3:00 am
by OnoSendai
a few hours = i can't remember how many exactly, i think 2 or 3
finally... heh
I have more free time now, so I finally got round to doing the sky thing
Another thing... This is still in the very early stages. I have no idea if my sky scattering coefficient constant is correct; it may be off by several orders of magnitude.
And another cool thing.. in the top image, the camera is two meters above a 6300km radius sphere - the whole earth is modeled, and shots from space should be possible
Posted: Tue Jun 12, 2007 3:03 am
by SimonLarsen
And another cool thing.. in the top image, the camera is two meters above a 6300km radius sphere - the whole earth is modeled, and shots from space should be possible
Damn, that would be so cool!
Posted: Tue Jun 12, 2007 3:04 am
by Kram1032
O.o!!
With day and NIGHT?
And:
There automatically is an "infinite plane", which is a huge (earth)-sphere[oid]?
OMG!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! INCREDIBLE!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
(Now, a huge texture is missing xD)
[ok, and different settings, like "Venus" "Mars".... and CUSTOM!!!!!]
You really do incredibly great work!
Posted: Tue Jun 12, 2007 3:06 am
by OnoSendai
Well.. i guess if you position the camera on the other side of the sphere from the sun, it will be night
I guess i need to add some stars....
Also.. .to get a little crazy, and if the precision of the floating point numbers holds up, I could model the moon as a big sphere
Posted: Tue Jun 12, 2007 3:09 am
by Kram1032
O.o Moon!
*turning into a werewolf*
*hoooooooooooooooooooooowwwwwwwwwwwwwwwl*
That's crazy! Don't forget the craters!
Posted: Tue Jun 12, 2007 3:13 am
by SmartDen
OMG! If you realize all this stuff, Terragen will be not longer needed
*very exited*