Planetoid Crater Shader testing

Started by WAS, August 09, 2019, 05:19:21 PM

Previous topic - Next topic

WAS

Well, somehow, my Terragen folder got deleted without a trace (literally no event log for it being deleted or recycling bin being emptyies). I can only imagine someone manually used IOBit Unlocker I have installed to delete it without a event. I sure didn't. I found out by trying to save a project that was open windows complaining about no such location.

Because of this I had to restart my crater project as the backup I had was 2 days old. This is the THIRD time rebuilding this. First, first time rebuilding based on colour adjusts, which I didn't like, than rebuilding from merge shaders, and now rebuilding from scratch cause I lost everything but Hetzen's original WIP.

All the displacement in this scene is by craters from large to micro impact. There are complex craters, which are high speed impacts which create the center peaks in the crater floor, and than simple craters, from low speed impacts with no peaks in the crater. This version is even faster than my previous version (which was way faster than tons of merge shaders), so not too mad about rebuilding it again. Lol

Forgive the PNG, my screenshot tool was set to PNG from trying to troubleshoot a bitmap problem with someone. I forgot.

Dune

Looks good. Maybe some more crispness and grey variation.

archonforest

Dell T5500 with Dual Hexa Xeon CPU 3Ghz, 32Gb ram, GTX 1080
Amiga 1200 8Mb ram, 8Gb ssd

Stormlord

#3
Very good start and a cool project!
Looking forward to see the moon more crisp and also mapped in true colours.
(Grey is boring... a subtle bluish gray looks much better, but this is just my opinion.)

STORMLORD

WAS

Hahaha thanks guys. This is just a solid sphere, no shading at that point I hadn't recreated the crater dist map or the ejecta map. Should have another test soon. Also divided smaller craters up so they weren't based on the huge craters heights and look better. Also masked out craters this time for better distribution and less overlays and crater walls just mixing together everywhere.

WAS

At MPD 0.6 AA 2 it took almot 2 and a half hours. This is likely though due to the addition of the crater distribution maps and ejecta maps which are a bunch more adds and mixes. Additionally the base colour texture is warped by the craters at high amplitude.

bobbystahr

I like where you're going with this Jordan, and condolences on yer data loss...that sucks...
something borrowed,
something Blue.
Ring out the Old.
Bring in the New
Bobby Stahr, Paracosmologist

Dune

That looks much better. But what are the kinda straight lines right in the center?

WAS

#8
Quote from: Dune on August 11, 2019, 01:37:48 AMThat looks much better. But what are the kinda straight lines right in the center?
Absolutely no clue. The craters are masked on large scale to some degree so maybe edges of some of them masked out? But it's so straight. I actually vaguely remember this in other scenes with really high scale and soft Perlin noise so it could even be the base fractal terrain. Some sort of stretching on Y near the equator.

My new setup I am at 2e+006 radius (from default) for the planet for a better lunar/dead planetoid look and I don't think I see them. Though a new render is starting now so can't check.

WAS

Some more testing. One is just testing the planetoid faults shader (which I like that version better than the edits in the crater test).

Not happy with surface colour textures.

Oshyan

The faults look pretty great! I like the one in the first image better myself, reminds me of Vales Marineres on Mars.

- Oshyan

Hannes


archonforest

Dell T5500 with Dual Hexa Xeon CPU 3Ghz, 32Gb ram, GTX 1080
Amiga 1200 8Mb ram, 8Gb ssd

cyphyr

Good and interesting.
Are you going to try to work a way of making overlapping craters remove the preceding craters wall?
Also are you using a really long lens on the last two images, something odd about the perspective, the spheres look "flat" (could be just me).
www.richardfraservfx.com
https://www.facebook.com/RichardFraserVFX/
/|\

Ryzen 9 5950X OC@4Ghz, 64Gb (TG4 benchmark 4:13)

WAS

#14
Quote from: cyphyr on August 14, 2019, 03:19:27 PMGood and interesting.
Are you going to try to work a way of making overlapping craters remove the preceding craters wall?
Also are you using a really long lens on the last two images, something odd about the perspective, the spheres look "flat" (could be just me).

That is always incorporated to a degree. The issue I found there is once your distribution starts piling up, it's hard to actually control masking between preceeding layers.

This last test incorporates a Colour Adjust on each iteration of the mask to more control it so it doesn't mask out the entire layer due to piling up of scalar. Maybe I overdid it there.

Here is an example of the setup exaggerated (multiplied x10)

I'll admit, my original, slower setup was better, I spent a lot more time on, and I even increase scales and clamping, which increased the distance between craters, giving a better overall distribution. I should do that again. But I do like how it is now as well.