[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[glob2-devel] New clouds
From: |
Erik Søe Sørensen |
Subject: |
[glob2-devel] New clouds |
Date: |
Wed, 12 Sep 2007 21:31:52 +0200 |
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; da; rv:1.7.13) Gecko/20060717 Debian/1.7.12-1.1ubuntu2 StumbleUpon/1.9995 |
Leo Wandersleb skrev:
to get nice structured clouds i guess the only option is to use
prerendered ones like we had before.
i thought perlin was the answer to get really dynamic clouds and was
frustrated when i saw how expensive it was. perlin clouds like in your
screen shot would need a far finer scale of structures.
now we render blocks of 32x32 pixels i guess but still you can see
steps some times (what is due to some bug maybe).
i did maths so i would have plenty of ideas to make the clouds
prettier and more structured but it gets too expensive even now.
Well, maybe we'll have a bit more room now for quality/luxury -
I've just checked into the 'optimized' branch a faster cloud generator,
which (a) uses Perlin's later Simplex Noise algorithm, and (b) uses
floats only (except outside the main loops).
The noise generator is 4x faster, and the generator in total is 6x
faster now.
Kieran: Care to test & measure? :-)
Leo et al.: For nice could lighting, maybe we could let the noise
generator calculate the differential too (it's not more work), and save
the info in another array - so that the two (cloud intensity + slope)
could be combined at draw time, for that fluffy feeling?
Everybody: Assess quality, please :-)
(It was a bit jump/jagged at one point, but I think I've fixed that. The
exact cloud size etc. probably changed somewhat.)
/Erik
- Re: [glob2-devel] To Erik: My Profiling Information, (continued)
- Re: [glob2-devel] To Erik: My Profiling Information, Erik Søe Sørensen, 2007/09/10
- Re: [glob2-devel] To Erik: My Profiling Information, Leo Wandersleb, 2007/09/10
- Re: [glob2-devel] To Erik: My Profiling Information, Kai Antweiler, 2007/09/10
- [glob2-devel] Optimizing calculations, Erik Søe Sørensen, 2007/09/10
- Re: [glob2-devel] To Erik: My Profiling Information, Kai Antweiler, 2007/09/11
- Re: [glob2-devel] To Erik: My Profiling Information, Leo Wandersleb, 2007/09/11
- Re: [glob2-devel] To Erik: My Profiling Information, Kai Antweiler, 2007/09/11
- Re: [glob2-devel] To Erik: My Profiling Information, Leo Wandersleb, 2007/09/11
- Re: [glob2-devel] To Erik: My Profiling Information, Joe Wells, 2007/09/11
- Re: [glob2-devel] To Erik: My Profiling Information, Leo Wandersleb, 2007/09/11
- [glob2-devel] New clouds,
Erik Søe Sørensen <=
- Re: [glob2-devel] New clouds, Kieran P, 2007/09/12
- Re: [glob2-devel] New clouds, Erik Søe Sørensen, 2007/09/12
- [glob2-devel] trouble using hg again :/, Leo Wandersleb, 2007/09/12
- Re: [glob2-devel] trouble using hg again :/, Kieran P, 2007/09/12
- Re: [glob2-devel] New clouds, Kieran P, 2007/09/12
- Re: [glob2-devel] New clouds, Erik Søe Sørensen, 2007/09/12
- Re: [glob2-devel] New clouds, Erik Søe Sørensen, 2007/09/12
- Re: [glob2-devel] New clouds, Kieran P, 2007/09/12
- Re: [glob2-devel] New clouds, Erik Søe Sørensen, 2007/09/12
- Re: [glob2-devel] New clouds, Kieran P, 2007/09/12