Different Works in Progress

Post your tests, experiments and unfinished renderings here.
User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Comparison Path depth 42 to 6

Post by FarbigeWelt » Wed Apr 10, 2019 7:12 pm

Four Colors Fluid Simulation in a Tube_GPU path_500 samples, 6 depth_dnpg  with caustics_64'000 bubbles_8h58m28s.jpg
Four Colors Fluid Simulation in a Tube_GPU path_500 samples, 6 depth_dnpg with caustics_64'000 bubbles_8h58m28s
Four Colors Fluid Simulation in a Tube_difference 500 to 461 samples, pgi_with_caustics path_depth 6 to pgi_no_caustics path_depth 42; gamma 2.0, brightness 4_.jpg
Four Colors Fluid Simulation in a Tube_difference 500 to 461 samples, pgi_with_caustics path_depth 6 to pgi_no_caustics path_depth 42; gamma 2.0, brightness 4
This time total path depth showed an impact of render speed (approx. 25%) but also very obvious on image quality. For me it is a little mistery how good the bubbles are visible and how wrong the glass tube looks like. One may know the tube is multi layered, it cconsists actually of two similar glass tube of which one is slightly smaller. This way collision detection quality is increased but does not make much sense for the final render. The next shot is with total path depth 6 and the other three depths set to 5. The same setting as for the picture in this post but with second tube set invisble.

Code: Select all

time		Samples	mSamples/s
11:32:53	461	25.9
06:20:13	238	24.2
10:52:59	445	25.3
08:58:28	500	29.7
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

Fox
Posts: 274
Joined: Sat Mar 31, 2018 11:17 am

Re: Different Works in Progress

Post by Fox » Wed Apr 10, 2019 7:55 pm

The bubble line is not straight with glass wall. Can you boolean modifier cut the bubbles with glass mesh and ctrl-L delete all bubbles that got unmanifold?

User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Re: Different Works in Progress

Post by FarbigeWelt » Thu Apr 11, 2019 9:25 am

Fox wrote:
Wed Apr 10, 2019 7:55 pm
The bubble line is not straight with glass wall. Can you boolean modifier cut the bubbles with glass mesh and ctrl-L delete all bubbles that got unmanifold?
With particles from simulation it is not so easy to follow your suggestion. It is easier to increase the size of the tube to avoid overlapping with the bubbles. I will set up this as next.
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Strange look with Bubbles Not Touching the Tube

Post by FarbigeWelt » Thu Apr 11, 2019 9:23 pm

Total path depth 22, the other dephts 20.
Four Colors Fluid Simulation in a Tube_GPU path_166 samples, 22 depth_dnpg with caustics_larger tube_64'000 bubbles_3h40m19s.jpg
Four Colors Fluid Simulation in a Tube_GPU path_166 samples, 22 depth_dnpg with caustics_larger tube_64'000 bubbles_3h40m19s
Not really what the image is supposed to look like. :(

Further ivestigation is required, to understand what is going here with the bubbles in the water inside the tube. Might be a priority issue of volumes or so. Thought up to date this should follow air p=0, tube p=1, liquid p=2, bubbles made from water with air outside and inside. Because it is actually an at least three phase problem I don't know exactly what to set here.

The tube contains a liquid (water). Around tube and above the liquid is air (clear volume).

The liquid has a top surface, the first outer volume (air), the liquid is in the tube, the second outer volume (glass) and the liquid has its own inner volume.

The bubble is made from water and air. Its inner outer volume is air, its inner volume is water, is outer outer volume is the liquid (water) or air (if the bubble is on the surface of the liquid).

I am bit lost due to missing logical solutions of these two trilemas.
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

Fox
Posts: 274
Joined: Sat Mar 31, 2018 11:17 am

Re: Different Works in Progress

Post by Fox » Thu Apr 11, 2019 10:39 pm

p2 glass
p1 air
beer 0 :ugeek:

User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Re: Different Works in Progress

Post by FarbigeWelt » Fri Apr 12, 2019 12:30 am

Fox wrote:
Thu Apr 11, 2019 10:39 pm
p2 glass
p1 air
beer 0 :ugeek:
Well, thank you, at least it was a try worth. I permuted the few possibilities but none of them showed just any close to real image.

It was not me if I did not try a new approach by following my intuition after the first not appropriate looking pictures. I should not have rejected this solution. The first 175 samples look promising. I will post it as soon I am confident with the result.

Thanks for the little punch. :)
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Four colors liquid

Post by FarbigeWelt » Sat Apr 13, 2019 8:14 am

All path depths are 16. The gif is animated.
EDIT: first version of gif replaced, due to wrong sized frames.
Four-Colors-Bubbles.gif
Four-Colors-Bubbles
Last edited by FarbigeWelt on Sat Apr 13, 2019 10:04 pm, edited 2 times in total.
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

Fox
Posts: 274
Joined: Sat Mar 31, 2018 11:17 am

Re: Different Works in Progress

Post by Fox » Sat Apr 13, 2019 11:32 am

Collisions still look dark, :roll:
Can you share the scene?
------
EDIT

I guess it's the material color, when red green blue get in same area, light can't get through. So it's correct.
With null material and small absorption for bubble volume, the light path depth would continue forever and the speed should be much slower with high path depth.

Example:
Bubbles_Fox.jpg

User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Re: Different Works in Progress

Post by FarbigeWelt » Sat Apr 13, 2019 5:24 pm

Fox wrote:
Sat Apr 13, 2019 11:32 am
Collisions still look dark, :roll:
Can you share the scene?
------
EDIT

I guess it's the material color, when red green blue get in same area, light can't get through. So it's correct.
Yes, absorption of the colored liquid-air bubbles is a bit too strong, what leads to black if r,g,b is substractive mixed.

My actual idea was always to share the scene as soon as I am confident with it. The scenes is close to be ready to be shared.
I will reduce the absorption and setup two or three particle systems ready to be simulated if has one enabled plugin ‚Molecular Script‘ in Blender user preferences.
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

User avatar
FarbigeWelt
Donor
Donor
Posts: 778
Joined: Sun Jul 01, 2018 12:07 pm
Location: Switzerland
Contact:

Four colors liquid: The scene

Post by FarbigeWelt » Mon Apr 15, 2019 5:44 am

Few Bubbles in Reaction Tube.zip
Few Bubbles in Reaction Tube
(547.75 KiB) Downloaded 22 times
Change the amount of bubbles in each emitter to the value you wish.
Run the plugin 'Molecular Script' simulation and then render the image or the animation.
160.8 | 42.8 (10.7) Gfp / Windows 10 Pro, intel i7 4770K@3.5, 32 GB | AMD R9 290x+RX 5700 XT, 4/8 GB
17.3 | 19.0 ( 4.7) Gfp / macOS X 13.6, iMac 27'', 2010, intel i7 870@2.93, 24 GB | ATI Radeon HD 5750, 1 GB
#luxcorerender | Gfp = SFFT Gflops

Post Reply