Page 3 of 6

Re: Mediabook, Welcome to Your Age of Quantum Consciousness

Posted: Fri Sep 27, 2019 12:42 pm
by FarbigeWelt
B.Y.O.B. wrote: Thu Sep 26, 2019 12:23 pmCool!
Thank you much for your short feedback!
:D

Brilliant Revealed Luxcorey

Posted: Fri Sep 27, 2019 1:31 pm
by FarbigeWelt
- The great and amazing .. - \ Bingg Elliv' Aand /- ..

Hidden, unrevealed, private and known capital punishment to see it -- Until now!
Today, all press is overwhelmed by the opening of the best guarded secret of this world.

All grace for the yesterday crowned, very noble and exceeding broad minded successors,
the twin siblings, the well protected children of our most deeply admired former majesty.
May the royal siblings rule for ever equitable and merciful,
with all the healthiness precious human beings deserve!

- - Here it finally is! - -
The first allowed to take picture since photography's invention.
The unbelievable fifteen centimeters radiant,
the biggest, the most brilliant, the one and only pink diamond on earth.
Brilliant Revealed Luxcorey
Brilliant Revealed Luxcorey
;) :D 8-)


openCL, path hybrid, Sobol
250324 samples, 02h16m
2.42M pixels


keywords: diamond; refraction; inner reflections; dispersion; gold; reflections; blue matte; indirect light; direct light

Re: Brilliant Revealed Luxcorey

Posted: Sat Sep 28, 2019 6:06 pm
by arescet
FarbigeWelt wrote: Fri Sep 27, 2019 1:31 pm - - Here it finally is! - -
The first allowed to take picture since photography's invention.
The unbelievable fifteen centimeters radiant,
the biggest, the most brilliant, the one and only pink diamond on earth.

Brilliant Revealed Luxcorey.png
;) :D 8-)

openCL, path hybrid, Sobol
250324 samples, 02h16m
2.42M pixels


keywords: diamond; refraction; inner reflections; dispersion; gold; reflections; blue matte; indirect light; direct light
Awesome!
i hope you don't mind, but can you tell us more about the settings you used for this beauty here?
what i mean is how did you get 250324 samples in only 2 hours? that to me is enviable sorcery! :lol:
also, i am very interested in how you set up that pink diamond material :D :D

Re: Brilliant Revealed Luxcorey

Posted: Mon Sep 30, 2019 1:38 am
by FarbigeWelt
arescet wrote: Sat Sep 28, 2019 6:06 pm Awesome!
i hope you don't mind, but can you tell us more about the settings you used for this beauty here?
what i mean is how did you get 250324 samples in only 2 hours? that to me is enviable sorcery! :lol:
also, i am very interested in how you set up that pink diamond material :D :D
Thank you for your feedback. :D

First, I have to graphic cards running, amd rx 5700 xt.
Second, the scene and the materials are simple.
The room cuboid is based on solidified cube and has real room size.
There are four point lights. Below each there is a flat cuboid reflector. This way there is only direct light from a spot light with a narrow angle.
The table is a cuboid. The brilliant is based on standard Blender add-on, one for additional meshes.
The golden shell is made with cloth physics. Its material is a golden yellow mirror.
The brilliant's material is glass with the IOR of diamonds, with a faint pink. All other materials are matte.
There is no cache used but light tracking for the few caustics.
Due to indirect lightening and the caustics the render needs at least 15'000 to look okay and 25'000 samples to look great.
Optimized clamping avoided bright fireflies, soft fireflies left are removed by denoising with OIDN.

In contrast to more complex scenes this jewel renders with 8.5 MSamples/s.
Revealed Luxcorey, openCL hybrid, 25000 S, screensho
Revealed Luxcorey, openCL hybrid, 25000 S, screensho
:ugeek:
Bing Ellivaand Stats
Bing Ellivaand Stats
Revealed Luxcorey, openCL hybrid, 22977 S
Revealed Luxcorey, openCL hybrid, 22977 S
The attachment Revealed Luxcorey, openCL hybrid, 02733 S.png is no longer available
Revealed Luxcorey, openCL hybrid, 02733 S
Revealed Luxcorey, openCL hybrid, 02733 S
Revealed Luxcorey, openCL hybrid, 16989 S
Revealed Luxcorey, openCL hybrid, 16989 S

Re: Brilliant Revealed Luxcorey

Posted: Mon Sep 30, 2019 7:16 pm
by arescet
FarbigeWelt wrote: Mon Sep 30, 2019 1:38 am
Thank you for your feedback. :D

First, I have to graphic cards running, amd rx 5700 xt.
Second, the scene and the materials are simple.
The room cuboid is based on solidified cube and has real room size.
There are four point lights. Below each there is a flat cuboid reflector. This way there is only direct light from a spot light with a narrow angle.
The table is a cuboid. The brilliant is based on standard Blender add-on, one for additional meshes.
The golden shell is made with cloth physics. Its material is a golden yellow mirror.
The brilliant's material is glass with the IOR of diamonds, with a faint pink. All other materials are matte.
There is no cache used but light tracking for the few caustics.
Due to indirect lightening and the caustics the render needs at least 15'000 to look okay and 25'000 samples to look great.
Optimized clamping avoided bright fireflies, soft fireflies left are removed by denoising with OIDN.
Thank you for well thought out and detailed response, i appreciate it very much!

Your hardware is making me drool man, nice! :D
maybe i will sell both my kidneys someday for two GPUs too :lol: :lol:

sorry for the new-guy question but, what do you mean by "light tracking"? did you mean light tracing?
also in the last few images, what do you mean by "factor"?
i would completely understand if you just give me links so i can do some reading on my own about light tracking and the factor you mentioned :D
thanks again for your response!

Re: Brilliant Revealed Luxcorey

Posted: Tue Oct 01, 2019 9:34 am
by FarbigeWelt
arescet wrote: Mon Sep 30, 2019 7:16 pm
sorry for the new-guy question but, what do you mean by "light tracking"? did you mean light tracing?
also in the last few images, what do you mean by "factor"?
Ooops! Yes, I meant light tracing with light tracking. I think tracing and tracking is synonymous. The BlenderLuxCore uses the term light tracing.

The material of the brilliant has an estimated value for dispersion, dispersion equals coefficient B see Cauchy's equation. You should be able to find other sites with coefficient B values for the equation, including one for diamond. But for some unknown reason there are different notations of Cauchy's equation with different units for the coefficients and additional coefficients. I tried to rearrange some formulae in an algebraic matter to proof their equality and to transform the coefficient into one unit. Unfortunately I was not successful in the time I wanted to spend. In addition, I found no regularity in comparison of index of refraction and Cauchy's coefficient B. Finally, if there is no value in Wikipedia, the best way is to guess and trial error until result looks appropriate.

With factor nothing else than zoom factor is meant. I zoomed into parts of the image to show progress / converging depending on Sample count.

Cheers!

Re: Brilliant Revealed Luxcorey

Posted: Wed Oct 02, 2019 7:45 pm
by arescet
FarbigeWelt wrote: Tue Oct 01, 2019 9:34 am
Ooops! Yes, I meant light tracing with light tracking. I think tracing and tracking is synonymous. The BlenderLuxCore uses the term light tracing.

The material of the brilliant has an estimated value for dispersion, dispersion equals coefficient B see Cauchy's equation. You should be able to find other sites with coefficient B values for the equation, including one for diamond. But for some unknown reason there are different notations of Cauchy's equation with different units for the coefficients and additional coefficients. I tried to rearrange some formulae in an algebraic matter to proof their equality and to transform the coefficient into one unit. Unfortunately I was not successful in the time I wanted to spend. In addition, I found no regularity in comparison of index of refraction and Cauchy's coefficient B. Finally, if there is no value in Wikipedia, the best way is to guess and trial error until result looks appropriate.

With factor nothing else than zoom factor is meant. I zoomed into parts of the image to show progress / converging depending on Sample count.

Cheers!
oooh cauchy!
i see, so about dispersion then.
i get what you mean with factor now, thanks for the reading reference! :D :D

Re: Brilliant Revealed Luxcorey

Posted: Thu Oct 03, 2019 12:25 pm
by Fox
FarbigeWelt wrote: Tue Oct 01, 2019 9:34 am Unfortunately I was not successful in the time I wanted to spend. In addition, I found no regularity in comparison of index of refraction and Cauchy's coefficient B. Finally, if there is no value in Wikipedia, the best way is to guess and trial error until result looks appropriate.
You probable already know the refractiveindex.info database
https://refractiveindex.info/?shelf=3d& ... ge=diamond

I get very high B value, maybe my formula is wrong :roll:
cauchy A: 2.381942809424
cauchy B: 0.012035304932
cauchy equation diamond.7z

Re: Brilliant Revealed Luxcorey

Posted: Fri Oct 04, 2019 1:43 pm
by FarbigeWelt
Fox wrote: Thu Oct 03, 2019 12:25 pm You probable already know the refractiveindex.info database
https://refractiveindex.info/?shelf=3d& ... ge=diamond

I get very high B value, maybe my formula is wrong :roll:
cauchy A: 2.381942809424
cauchy B: 0.012035304932
cauchy equation diamond.7z
Thank you for sharing your calculations. Your idea is good to list several wavelengths and refracting indices because one requires at least two wavelengths and two refraction indices for the calculation of two coefficients, A and B, of Cauchy's equation for dispersion.

I got a slightly different result (with sodium D-line and solar G-line):
A 2.4061 (A means B in Cauchy's equation, en.wikipedia, 04. Oct 2019)
B 3.65E-03 [µm^2] (B means C in wikipedia)
Cauchy's Equoation - Diamond
Cauchy's Equoation - Diamond
I completed your spread sheet with references and formulae including my algebraic isolation of coefficient B.
The attachment cauchy equation diamond_fw.7z is no longer available

Déjà vu: Prisms

Posted: Fri Oct 18, 2019 6:15 pm
by FarbigeWelt
Last time -CPU BiDir ..
A Cascade of Prisms
Image
.. this time path openCL with light tracing, rather strong clamping and OIDN treated with 8% raw part to keep some soft residual noise.
Prism Park.png
1.5 HD, 26'890 samples, 04h19m. Raw image is quite noise free but now in comparison with picture from 29. April 2019 clamping has been to strong this time. (I guess the current picture needs an update.)