CoolColJ's test pics thread

Get feedback from others on your works in progress
User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Sat Apr 28, 2012 11:40 pm

Oscar J wrote:Looking really realistic! Forgive me for asking a rookie question, but are you using a specular material on the water? Could you share any of your values?
The water should be in the Material database. I remeber someone put it up for me. It uses tabulated spectrum data I took from some PDF research papers rather than just simple specular colour absorption values. Just do a search for "coolcolj" and a few entries for clear water comes up when doing a search from within Indigo itself in the online material database :)
There is also a dirty water varient, with values from water with particles and organism from another research paper

As for the surface, I used Indigo's shader language FBM noise to subdivide and displace the surface.
It's a bit coarse if you get close, but now that Indigo is so much faster at subdividing I might bump it up :lol:

User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Sun Apr 29, 2012 12:10 am

Just an old test scene I rendered a few days ago. I've tested a lot of renderers with this one from a decade ago -
Surprisingly slow this one :shock:
Attachments
im1335487126.jpg

User avatar
OnoSendai
Developer
Posts: 6241
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Re: CoolColJ's test pics thread

Post by OnoSendai » Sun Apr 29, 2012 12:16 am

CoolColJ:
That's a neat test scene, can you post it or send it to me? (nick@indigorenderer.com)

User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Wed May 02, 2012 11:50 am

I did this render with default BiMLT max change of 0.01, but had Large Mutation probability at 0.01, and Max Number of Consec. Rejections at 10000. Default MNCR of 1000 is clearly not enough, but 10,000 worked fairly well

I let this one render out for a long time and reached almost 90,000 Samples/pixel :lol:
But still the submerged displaced objects are not fully converged :shock:
The water caustics converged way earlier though, around the 6000-8000 samples area
im1335839463.jpg
Once this is sorted, Indigo should be able to handle anything just about :)

User avatar
OnoSendai
Developer
Posts: 6241
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Re: CoolColJ's test pics thread

Post by OnoSendai » Wed May 02, 2012 8:06 pm

CoolColJ wrote:I did this render with default BiMLT max change of 0.01, but had Large Mutation probability at 0.01, and Max Number of Consec. Rejections at 10000. Default MNCR of 1000 is clearly not enough, but 10,000 worked fairly well

I let this one render out for a long time and reached almost 90,000 Samples/pixel :lol:
But still the submerged displaced objects are not fully converged :shock:
The water caustics converged way earlier though, around the 6000-8000 samples area
im1335839463.jpg
Once this is sorted, Indigo should be able to handle anything just about :)
While I'm asking for scenes, can you send me that one as well? :)

User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Thu May 03, 2012 12:30 am

OnoSendai wrote:
CoolColJ wrote:I did this render with default BiMLT max change of 0.01, but had Large Mutation probability at 0.01, and Max Number of Consec. Rejections at 10000. Default MNCR of 1000 is clearly not enough, but 10,000 worked fairly well

I let this one render out for a long time and reached almost 90,000 Samples/pixel :lol:
But still the submerged displaced objects are not fully converged :shock:
The water caustics converged way earlier though, around the 6000-8000 samples area
im1335839463.jpg
Once this is sorted, Indigo should be able to handle anything just about :)
While I'm asking for scenes, can you send me that one as well? :)
Lycium should have it, I remeber sending it to him. If not, let me know

User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Sat May 12, 2012 9:06 pm

Old sky system vs new simulated one

I like the new one, it actually looks sunny! That's was my peeve with the original sky system, I could never get that real sun burnt look!
CCJ_Athena_NEWsun2.jpg

Original sky system, flat and bland looking.... Interestingly there is a bit of caustic at the front corner of the base developing - which the render above has not shown yet, even after 40+ mins....
CCJ_Athena_NEW_ORGsun.jpg

User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Sat May 19, 2012 8:28 pm

added a bump map to the tiles and boy did that slow down the pool caustics!
im1337193623.jpg

StompinTom
Indigo 100
Posts: 1828
Joined: Mon Sep 04, 2006 3:33 pm

Re: CoolColJ's test pics thread

Post by StompinTom » Sat May 19, 2012 9:48 pm

I love this consistent series of tests. Great way to compare and analyze a difficult rendering situation between versions of Indigo!

User avatar
Polinalkrimizei
Posts: 647
Joined: Sat May 02, 2009 6:59 am

Re: CoolColJ's test pics thread

Post by Polinalkrimizei » Mon Jun 11, 2012 8:55 pm

This thread is awesome.
And I really don't want to hijack it, but I'm having trouble with a scene full of caustics and felt this is the place somehow...

The water is CoolColJ's Clear Water from MatDB, with sss disabled (gives funny colors with all newer versions of indigo, btw how did you fix that CoolColJ?!).
This render is made with bidirectional MLT, and rendered til 7800 ssp. What bothers me are the white dots... they appear already pretty early, and also with plain pathtracing.
Scene is lit by the updated sun/sky system, the water is an actual mesh, no bump or displacement here.

How do I get rid of them?
Attachments
Pool 2 7800ssp.jpg

User avatar
Zom-B
1st Place 100
Posts: 4700
Joined: Tue Jul 04, 2006 4:18 pm
Location: ´'`\_(ò_Ó)_/´'`
Contact:

Re: CoolColJ's test pics thread

Post by Zom-B » Mon Jun 11, 2012 9:26 pm

Polinalkrimizei wrote:How do I get rid of them?
For me it looks like your water mesh is not subdived enough, you can see the polygons even in the caustics.
The "White Dots" could be some direct Sun reflections based on a so edgy water surface that can look... strange!
For a smooth look of possible fireflies, try to max out The Super Sampling for your render, at least 3 is the way to go...
Also use radial splatting and sharp downsize filter for a crispy result... yours look somehow "unsharp" :)

Sun + Pool Caustics is possible with MLT only atm, I just recently tried a pool scene too.
polygonmanufaktur.de

User avatar
CoolColJ
Posts: 1738
Joined: Mon Jun 25, 2007 1:47 pm

Re: CoolColJ's test pics thread

Post by CoolColJ » Mon Jun 11, 2012 10:11 pm

Polinalkrimizei wrote:This thread is awesome.
And I really don't want to hijack it, but I'm having trouble with a scene full of caustics and felt this is the place somehow...

The water is CoolColJ's Clear Water from MatDB, with sss disabled (gives funny colors with all newer versions of indigo, btw how did you fix that CoolColJ?!).
This render is made with bidirectional MLT, and rendered til 7800 ssp. What bothers me are the white dots... they appear already pretty early, and also with plain pathtracing.
Scene is lit by the updated sun/sky system, the water is an actual mesh, no bump or displacement here.

How do I get rid of them?
What do you mean by funny colours?
If dispersion is present you will get multi-colours until converged

The dots look like sun specular reflections that have been distorted because of the filtering.
Certain modes and settings of the filter can alter these bright spots. Turn off the filter to see what I mean.
Guassian, Sharp or MN with blur of 0 and ring of 0.5, radius 2 are some to try

Subdividing the mesh and displacing it in Indigo at render time is a way to get a smoother look without a massive polycount model

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Re: CoolColJ's test pics thread

Post by CTZn » Tue Jun 12, 2012 1:14 am

polinalkrimizei, could I have the pigs of the water mesh alone please ?

I don't think that the faceted look would be a problem in itself but I'm curious about one detail in the mesh declaration.

Or, if that's not what you have done already, you can set all normals smooth, activate subdivisions to one level but let surface smoothing off. Just in case.

Clearly those are reflections of the sun. In some ways when one looks at a ripped water plane under the sun millions of such spots can be seen. It's even more obvious from successive video frames.
obsolete asset

User avatar
lycium
Posts: 1216
Joined: Wed Sep 12, 2007 7:46 am
Location: Leipzig, Germany
Contact:

Re: CoolColJ's test pics thread

Post by lycium » Tue Jun 12, 2012 1:50 am

Some aperture diffraction would make them bloom more realistically, instead of staying sharp circles.

User avatar
Polinalkrimizei
Posts: 647
Joined: Sat May 02, 2009 6:59 am

Re: CoolColJ's test pics thread

Post by Polinalkrimizei » Tue Jun 12, 2012 1:59 am

Thanks for the fast replies guys!!

Aperture diffraction was turned on already, and the scene was rendered with a supersampling factor of 6.

It is definitely the reflections of the sun, and with a different sun position and altered filter settings they almost disappear completely 8) thanks Zom-B and CoolColJ!

CTZn, I attached a .pigs of the pool if you want to check that one out. I did increase the subdivision of that mesh by now, until I really liked the caustics (3x subsurfed subdivisions in blender). The attached file contains the original mesh.

The attached image is at about 10k spp and still cooking, I think I will get a decent render out of this scene with your help.
Attachments
Pool_CTZn.pigs
(225.58 KiB) Downloaded 331 times
Pool progress.jpg

Post Reply
695 posts

Who is online

Users browsing this forum: Google [Bot] and 8 guests