Chaos Corona Forum
Chaos Corona for 3ds Max => [Max] Feature Requests => [Max] Resolved Feature Requests => Topic started by: JViz on 2019-11-28, 10:10:40
-
1. Fstorm introduced a new feature. a texture tiling randomiser. it takes a texture and randomises its locations and rotations over geometry while blending the edges in a way that removes seams, it effectively makes it possible to use, for example, a 2x2 meters textures over a surface that is 100x100 meters or more with no visible tiling which is extremely useful in Archviz especially for glossiness textures.
2. Fstorm has a noise map that is so natural in its structure it might be the best out there, if you compare the structure of that noise to real life objects you realise it's really smart and emulates a wide range of medium and small scale noise that happens naturally in paper plastics leather stone, you name it. maybe it's time to have a dedicated corona noise map.
thanks team
-
Duplicate. Seamless tiling for stochastic textures already requested here: https://forum.corona-renderer.com/index.php?topic=24174.0
Regarding the noise. Did you try Bercon noise and Siger noise maps yet? They offers several improved noise types, might be something that you're looking for.
P.S. please post one request per topic.
-
I do know Siger and Bercon, they're just not as good.
-
It would be interesting to see some direct comparison.
-
If anything to do random texture tiling, then how about:
"On Histogram-Preserving Blending for Randomized Texture Tiling" paper fromBrent Burley ( (http://jcgt.org/published/0008/04/02/)Walt Disney's Hyperion)
(http://jcgt.org/published/0008/04/02/)Abstract
To support interactive authoring of high-resolution randomly tiled textures, we modify the histogram-preserving tiling algorithm of Heitz and Neyret to avoid any lengthy preprocessing. Instead of calculating a 3D histogram transformation by optimal transport, which can take minutes even at low resolution, the input texture is transformed using per-channel 1D lookup tables, constructed trivially from the input histogram on texture load. Three sources of clipping are described. Modifying the algorithm to use a truncated Gaussian distribution and a novel soft-clipping contrast operator avoids clipping artifacts while retaining very high rendering performance. Per-channel histogram preservation is sufficient for most textures, but some will produce unwanted colorations; these can often be avoided by performing histogram preservation only on luminance. Exponentiating the blending weights can reduce ghosting artifacts and better preserve structured texture details.
(https://i.imgur.com/TB4ZMSJ.png)
Edit:
You can try it here: https://benedikt-bitterli.me/histogram-tiling/
;)
-
looks interesting. the theory is out there I'm sure, it's just that the implementation is still waiting in the pipeline for corona, the question is how long and how well are they going to do it.
-
done