Latest rocktests

Started by Dune, August 09, 2019, 01:12:18 AM

Previous topic - Next topic

Dune

I can't stop myself  ::)

Dune

There seems to be a restriction to images not wider than 2500px. Just tried upload a 2500px image, but had to resize to 2499px.

DocCharly65

Quote from: Dune on August 09, 2019, 01:12:18 AMI can't stop myself  ::)
Good news ;D

Great adventure feelings in those new renders :)
Nice, Ulco!

WAS

I love that first shot, and the latest image you just uploaded (and had to resize). I like the displacement. Looks like some PF for the vertical cutting?

It would be cool to have larger resolution images so long as they fall within file size limits.

Dune

Well, without even one compute terrain the one with the horseman rendered in half an hour in RT. PT took 4h though. Still have to check differences.
I'll tell you guys (or what's left of the 'community') a little something I just discovered yesterday, and which you can add to your documentation  ;)  Fake stones have an inbuilt way of computing lateral displacement, which I kind of knew but never really explored.
So I took some huge fake stones and just added some lateral displacement into the surface imput from a mix of PF and image map greytones. As simple as that.
I'm working on another cliff where I raise ground to a steep side, clad it with stones, which can have the lateral displacements. Hence no compute terrain needed.

N-drju

These really are just fake stones? ;)

Wow, I knew they were actually quite powerful tool but have indeed never used them as mountainsides. Good thinking.
"This year - a factory of semiconductors. Next year - a factory of whole conductors!"

Dune

Well, this is what experimentation leads to  ;)  Just 30m fake stones indeed with a mix of lateral and normal displacement, merged with some smaller ones (highest raise) of course.

WAS

Quote from: Dune on August 09, 2019, 01:34:30 AMWell, without even one compute terrain the one with the horseman rendered in half an hour in RT. PT took 4h though. Still have to check differences.
I'll tell you guys (or what's left of the 'community') a little something I just discovered yesterday, and which you can add to your documentation  ;)  Fake stones have an inbuilt way of computing lateral displacement, which I kind of knew but never really explored.
So I took some huge fake stones and just added some lateral displacement into the surface imput from a mix of PF and image map greytones. As simple as that.
I'm working on another cliff where I raise ground to a steep side, clad it with stones, which can have the lateral displacements. Hence no compute terrain needed.


It's ironic you just played with that more too. When I was doing those river stones awhile back (I think you helped me with that too), I came back to them to add to a scene but was having trouble isolating my colour adjustments with using the stones as a mask, so I had to work through the surface shader of the fake stones. I than decided to try and offset the stones displacement to try and remedy some explosions in the displacement, and found it was rounding the stones just like my compute normal and offset lower down the chain. So I was essentially wasting resources before.

Dune

Yes, offset raises them and displacement displaces them. You can also use a masked surface shader with an offset to have a variable offset by vertical or normal (lateral doesn't really do much), and an extra lateral or normal displacement in that same surface shader to further bulge up or sideways.

N-drju

Quote from: Dune on August 09, 2019, 01:59:01 AMWell, this is what experimentation leads to  ;)  Just 30m fake stones indeed with a mix of lateral and normal displacement, merged with some smaller ones (highest raise) of course.

Very good, really. These pictures look like blue-noded to perfection. ;)

I remember I tried it once on a "War Free Zone" mesa, but was to lazy to try and exaggerate the effect.

Quote from: WASasquatch on August 09, 2019, 02:15:39 AMI than decided to try and offset the stones displacement to try and remedy some explosions in the displacement, and found it was rounding the stones just like my compute normal and offset lower down the chain.

I had a simmilar problem back in the days. I noticed that my fake stones lean accordingly to the terrain angle... So I put them before compute terrain. They stopped leaning, but I was no longer able to apply any distribution masks to them... Because they all require compute terrain... I'm afraid I have never solved this problem really...
"This year - a factory of semiconductors. Next year - a factory of whole conductors!"

mhaze

Love the mood of the last two.

j meyer

Nice mood indeed, a bit more intense in the path traced one,
at least in my eyes.

Dune

So here's another version of a fakestoneclad mountain. No compute terrain needed. Not perfect of course (too little blocks for instance, due to a mix with normal perlin), and I'll experiment with an added compute terrain soon....

René

These look really good, especially 'SquareRock_v15_05-08-19 test'.
But if you can't stop yourself, I have to warn you; you've taken a dangerous path. I know all about that. ;)

Dune

 ;D  I know. I guess it's my drive to find solutions to self-imposed problems. I don't even have a (professional) need for rocks.