Laci's wip

Post your tests, experiments and unfinished renderings here.
Post Reply
tokiop
Posts: 42
Joined: Fri Apr 13, 2018 8:11 pm

Re: first steps

Post by tokiop »

lacilaci wrote: Mon Oct 22, 2018 4:44 pm It is in the water, But I think that I will need to create a water tension geometry to get some highlights on contact.... This is where bevel shader with option to blend multiple objects would come in handy... also During this project I realized how much I miss triplanar mapping with blending from cycles :D
Indeed, surface tension seems tricky to make dynamically. It was also the case for volume intersections and we now have a nice and clean solution, so there is hope ! Could this be related to a shader-based bevel, sometime mentioned for CAD projects and others ?

Meanwhile, maybe baking a occlusion-type shader with a small radius, or dynamic paint addon, and using it as displacement map on the liquid, could help ?

Thanks for sharing your wip, the process and attention to the shaders, lighting and composition is inspiring !
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: first steps

Post by lacilaci »

tokiop wrote: Mon Oct 22, 2018 6:45 pm
lacilaci wrote: Mon Oct 22, 2018 4:44 pm It is in the water, But I think that I will need to create a water tension geometry to get some highlights on contact.... This is where bevel shader with option to blend multiple objects would come in handy... also During this project I realized how much I miss triplanar mapping with blending from cycles :D
Indeed, surface tension seems tricky to make dynamically. It was also the case for volume intersections and we now have a nice and clean solution, so there is hope ! Could this be related to a shader-based bevel, sometime mentioned for CAD projects and others ?

Meanwhile, maybe baking a occlusion-type shader with a small radius, or dynamic paint addon, and using it as displacement map on the liquid, could help ?

Thanks for sharing your wip, the process and attention to the shaders, lighting and composition is inspiring !
Thanks, I will see if and how I will go about it.
Yes shader based bevel with the ability of including/excluding other objects would help here I think.
I don't like the idea of baking textures...

Anyway, I will deal with this once everything else is done, and maybe I forget and be happy with the result anyways... Or do it with actual geometry.
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: first steps

Post by lacilaci »

Ok, so another problem.
If I want to render high res (4K+) I have to use tilepathocl cause denoiser won't let me use anything else.
However it seems some bright spots do appear as renderer progresses that won't go away, denoiser doesn't handle them as fireflies and I think those are unresolved caustics. So I hope this one actually is solved with the recent clamping+tilepath fix or rendering this on gpu is just not doable.

in attachment sobol vs tilepath, approximately same sampling level, note how the pool area shows "caustics?" in tilepath rendering
Attachments
sobol.JPG
tilepath.JPG
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: first steps

Post by lacilaci »

Time to figure out the house design I guess...
Attachments
oB.jpg
User avatar
B.Y.O.B.
Developer
Developer
Posts: 4146
Joined: Mon Dec 04, 2017 10:08 pm
Location: Germany
Contact:

Re: first steps

Post by B.Y.O.B. »

Nice progress.
lacilaci wrote: Tue Oct 23, 2018 5:44 am note how the pool area shows "caustics?" in tilepath rendering
Try to disable specular paths in the sun visibility settings.
Attachments
2018-10-23_10-33-18.png
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: first steps

Post by lacilaci »

B.Y.O.B. wrote: Tue Oct 23, 2018 8:35 am Nice progress.
lacilaci wrote: Tue Oct 23, 2018 5:44 am note how the pool area shows "caustics?" in tilepath rendering
Try to disable specular paths in the sun visibility settings.
Thanks a milion times... That helped, so I guess it was caustics then..? Interresting how they don't appear in normal pathtracing.
User avatar
B.Y.O.B.
Developer
Developer
Posts: 4146
Joined: Mon Dec 04, 2017 10:08 pm
Location: Germany
Contact:

Re: first steps

Post by B.Y.O.B. »

lacilaci wrote: Tue Oct 23, 2018 8:55 am so I guess it was caustics then..?
Yes, the sun is an intersectible light source, so a unidirectional pathtracer can compute its caustics (but very inefficient when the sun is small).
zeealpal
Posts: 59
Joined: Wed Jan 10, 2018 12:25 pm

Re: first steps

Post by zeealpal »

lacilaci wrote: Tue Oct 23, 2018 7:19 am Time to figure out the house design I guess...
Wow! Really liking how this is shaping up!
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: first steps

Post by lacilaci »

A basic iteration on the design...
btw I wanted to try thick glass with 2 panes per windows(no archglass) The performance hit was insane :D gonna use archglass, but I guess having the ability to hide glass from gi would help and make glass pretty :)
Attachments
4ktest_int.jpg
4ktest_int2.jpg
User avatar
lacilaci
Donor
Donor
Posts: 1969
Joined: Fri May 04, 2018 5:16 am

Re: first steps

Post by lacilaci »

A small tweak to the house and a quick and dirty late hour test..
Attachments
oB.jpg
Post Reply