Page 1 of 1

Caustic preview & physical accurency

Posted: Thu Apr 11, 2013 5:52 am
by kiku70
Hi,


1- I would like to know if there is a quick way to see the caustic's patterns ?

2- What is the physical accuracy of the Octane system with caustics ?

3- Compare to lux render, mitsuga, indigo. Octane is using the same calculation system ?

Thanks in advance.



I checked those but it was 2010.

http://render.otoy.com/forum/viewtopic. ... nterfacing
http://render.otoy.com/forum/viewtopic. ... 2&start=30


ps. Octane needs - undo and OBJ export :) thank you

Re: Caustic preview & physical accurency

Posted: Thu Apr 11, 2013 9:30 am
by mib2berlin
There is a setting in the new 1.12 RC called "Paralemism".
It renders slower but prefers caustics.
2. ?
3. Octane use Metropolis Light Transport with Monte Carlo calculation (PMC kernel).
I think Octane is the only GPU rendering engine with MLT support.


Cheers, mib.
EDIT: Use Daylight system if possible, it is much faster with caustics.

Re: Caustic preview & physical accurency

Posted: Thu Apr 11, 2013 3:11 pm
by kiku70
Thanks,

What I try to replicate is the work of Thomas Anagnostou- so i will need artificial lights:

http://www.maxwellrender.com/forum/view ... ight=prism

Any suggestions .


By the way the link's file of the dxf filem simplfied prism dxf, is broken.
Any idea where I can get it. thanks in advance.

Re: Caustic preview & physical accurency

Posted: Thu Apr 11, 2013 3:36 pm
by kiku70

Re: Caustic preview & physical accurency

Posted: Thu Apr 11, 2013 8:05 pm
by mib2berlin
I can get anything realistic without the orginal file, material setting and laser design.
It need long render times, thought.
I post an image later.

Cheers, mib.
EDIT: 8000 Sample PMC parallelism 2
Specular default material seams realy wrong here.

Re: Caustic preview & physical accurency

Posted: Fri Apr 12, 2013 11:39 am
by kiku70
Hi,

thank you for your test.
I guess the power and position of light

I could get the Thomas Anagnostou's dxf file, old. I contacted several forums. For now nothing, I let you know.
I just find this image

I'm gonna be away till wed.16 ap.

Take care

Re: Caustic preview & physical accurency

Posted: Fri Apr 12, 2013 12:35 pm
by mib2berlin
Yep, saw your post in lux forum.
Would be nice to see what Octane can do today, this test was from 2005!

Cheers, mib.

Re: Caustic preview & physical accurency

Posted: Fri Apr 12, 2013 10:06 pm
by mib2berlin
Hi, another test with ies light and dispersion coefficient 0,0035.
I think the angles and a very small light source is important.
But looks also nice. :)

Cheers, mib.

Re: Caustic preview & physical accurency

Posted: Fri Apr 12, 2013 11:10 pm
by LudovicRouy
not an expert but if you make a more longer cylinder an reduce the opening (with light more deaper inside) light would be more "linear" no ? I suggest that as i've made some test with shadows that produce more sharp one with small, powerfull and distanced light source. I realise that it will raise up the render time too, not simple to tweak...