Brainstorming: Noise optimization
#3
I believe it will make a quite substantial performance improvement.
Current noise has to do all these for each point queried:
1, Floor all coords to integral values
2, Calculate underlying noise value at 4x4x4 integral neighbors
3, Cubic-interpolate each layer (4x4x4 -> 4x4), then cubic-interpolate each column (4x4 -> 4), then finally cubic-interpolate the final value (4 -> 1)

With the new system, point 1 and 2 will be done only occasionally (rough expectation - about 5 % of all times) and I think even the interpolation could be tweaked somehow to save a few operations.
Reply
Thanks given by:


Messages In This Thread
Brainstorming: Noise optimization - by xoft - 03-28-2013, 05:54 AM
RE: Brainstorming: Noise optimization - by xoft - 03-28-2013, 07:29 AM
RE: Brainstorming: Noise optimization - by ThuGie - 01-15-2014, 10:29 AM
RE: Brainstorming: Noise optimization - by xoft - 01-15-2014, 10:56 PM
RE: Brainstorming: Noise optimization - by xoft - 01-16-2014, 06:18 AM
RE: Brainstorming: Noise optimization - by xoft - 01-17-2014, 05:06 AM
RE: Brainstorming: Noise optimization - by ThuGie - 01-17-2014, 07:17 AM
RE: Brainstorming: Noise optimization - by xoft - 01-18-2014, 08:13 AM
RE: Brainstorming: Noise optimization - by xoft - 01-19-2014, 01:02 AM
RE: Brainstorming: Noise optimization - by xoft - 01-19-2014, 01:20 AM
RE: Brainstorming: Noise optimization - by xoft - 01-19-2014, 01:45 AM
RE: Brainstorming: Noise optimization - by xoft - 01-19-2014, 02:05 AM
RE: Brainstorming: Noise optimization - by xoft - 01-19-2014, 07:29 AM
RE: Brainstorming: Noise optimization - by xoft - 01-19-2014, 05:14 PM
RE: Brainstorming: Noise optimization - by xoft - 01-24-2014, 06:37 PM



Users browsing this thread: 2 Guest(s)