|
|||||
|
|
#1 |
|
Registered User
Join Date: Mar 2004
Posts: 256
Thanks: 0
Thanked 0 Times in 0 Posts
|
mental ray caustics
First of all just a quick question...are there any renderers out there which can accurately model all of the various properties of light (at once)?
- destructive and constructive wave interference - color related to wavelength...and index of refraction - diffraction, evidenced by single/double/n slit interference - refraction, evidenced by Newton's rings in thin-film interference, and light separation in prisms, and telescopes finalRender stage-1 looks like it can do all of this but, I don't know for sure Anyway...I'm trying to use the caustics in mental ray (3dsm7) and having complete failure! I know mental ray can't do 90% of the thigns I listed but the sample pictures show it doing reflection and some pale attempt at refraction. 1) mental ray renderer selected 2) caustics enabled in render dialog 3) double sided sphere encompassing entire surfacr to absorb waylaid photons 4) light aimed through glass lense at plane 5) plane and sphere double sided, both set to recieve caustics 6) lense set to generate and recieve caustics 7) light source set to generate caustics 8) mental ray glass (physics phen) material applied to lense, mental ray DGS applied to other things...also tried using a default material instead of DGS...tried using all default materials... no matter what I do, it always says "no photons stored" in anything and won't even render
__________________
"In a world of compromise, some don't." Last edited by Stuh505; 24-11-2004 at 04:40 AM.. |
|
|
|
|
|
#2 |
|
Guest
Posts: n/a
|
|
|
|
|
#3 |
|
Registered User
Join Date: Mar 2004
Posts: 256
Thanks: 0
Thanked 0 Times in 0 Posts
|
I would like to understand why the tutorials which come with max do not seem to be working...I'm doing all the steps listed in the tutorials as I said above and its not happening..those photons just arent getting caught even though the objects are set to recieve caustics. do they need a special kind of material to recieve photons or something? GI works though..
__________________
"In a world of compromise, some don't." |
|
|
|
|
|
#4 |
|
Not a zombie.
Join Date: Jan 2004
Location: P-town
Posts: 1,547
Thanks: 7
Thanked 32 Times in 20 Posts
|
Have you checked your mental ray message window? You might be having other issues. I can't really imagine what though.
|
|
|
|
|
|
#5 |
|
Guest
Posts: n/a
|
hello stuh505
check the message window for error.... if it won`t even render then the photon map needs more memory than can be stored and it won`t render. I had similar problem with Brazil render... above 60 million photons...the memory to be stored was too much and my render would abort.This was with 1 gig ddr available. How many photons are being emmitted from your light source?How large is your photon map?How much memory your system have? ciao |
|
|
|
#6 |
|
Registered User
Join Date: Mar 2004
Posts: 256
Thanks: 0
Thanked 0 Times in 0 Posts
|
I do not think that is the problem.
the average photons per light is set to 10,000 (defasult) The only thing that the mental ray message window says is, "error: 100,000 photons emmitted and none stored, warning: 10,000 photons emmitted and none stored" I have 1 light souce, 1 reflective object, 1 non reflective object, and one non reflective object surrounding the scene to catch any missed photons... I have 512 MB ram and it is set to use 1536 MB. i cant imagine that this process would use anywhere near the full extent of my ram, I have set up the simplest possible scene and none of the settings are high
__________________
"In a world of compromise, some don't." |
|
|
|
|
|
#7 |
|
Registered User
Join Date: Jun 2004
Location: Bavaria
Posts: 84
Thanks: 0
Thanked 0 Times in 0 Posts
|
10000 caustic photons are way enough to test, stuh.
Dont know how this works in 3dsmax, but generally a material in mental ray needs a corresponding photonic material to get the photons computed. If you have a dgs (both surface mat and photonic mat) with specular set to white to get a reflective material your photons should be well. To get nifty glass/water/liquid effects, a dielectric mat should be used, since it follows fresnel's formulas. But bear in mind to apply it to surfaces that actually have two sides with normals that point in different directions because mental ray needs to distinguish either side of the surface/interface. So dont use planes for dielectric mats, use a box or something alike. Some tips: Dont use point lights to emit photons because they emit 360 degree which is not necessary in most cases, use spot lights. Get rid of the surrounding object, it doesnt make any sense to 'catch missed photons'. If your photons arent stored where they should be, why do you want to store them where they are completely useless? Youre just tampering your photon statistics and get inaccurate results.
__________________
May You Live in Interesting Times |
|
|
|
|
|
#8 | |
|
Guest
Posts: n/a
|
Quote:
|
|
|
|
|
#9 |
|
Registered User
Join Date: Jun 2004
Location: Bavaria
Posts: 84
Thanks: 0
Thanked 0 Times in 0 Posts
|
I guess he meant the total amount of virtual memory he owns.. not the one being used by mr.
![]()
__________________
May You Live in Interesting Times |
|
|
|
|
|
#10 |
|
Guest
Posts: n/a
|
hello Stuh505
I wish I had comments for you..... I just tried the tutorials myself...and they all work fine for me.... Mabey there is just one little setting you have forgotten? It`s hard to say why it won`t work.... best luck ciao |
|
![]() |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | |
| Display Modes | |
|
|