Timeline for Infinite detail inside Perlin noise procedural mapping
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 12, 2012 at 16:18 | comment | added | House | Midpoint displacement (via the diamond-square algorithm) is an excellent way to create detailed landscape. Especially when you want the edges to match up without much work. | |
| Nov 12, 2012 at 16:15 | comment | added | Dave Jellison | Byte56, is there any reason I couldn't apply the diamond-square algorithm after I have my 4 points of Perlin noise data for each corner? I would have to alter my initial approach as currently I have (x,y) = tile so I only hold 1 value of perlin noise per tile vs. 4. Does that sound like a reasonable first-timer approach? | |
| Nov 12, 2012 at 16:06 | comment | added | Dave Jellison | Aha! knew I was searching the wrong thing. OK cool. Thanks for the detail as it may help others as well. I don't think I would need any blending/lerping for my scenario (yet) as I'm using very static 50x50 tiles. So long as the initial view makes sense, mountains bordered by hills, water by coastline or plains, etc. I'm cool (for my uber-basic implementation). But the fractal approach appears to be what I'm after thanks! | |
| Nov 12, 2012 at 16:03 | vote | accept | Dave Jellison | ||
| Nov 12, 2012 at 15:53 | history | answered | House | CC BY-SA 3.0 |