Page 1 of 1

Maxwell mask material & mental ray camera matching?

Posted: Tue Sep 13, 2005 10:41 am
by GumboYaYa
Hi there,

I am missing a shader for Maxwell which can be used like the surface shader in Maya/Mental Ray. I am using this shader for example to create RGB masks for compositing.

For the case a shader like that does not exist (and maybe never will), I wouldn't mind rendering the masks with Mental Ray, but I never was able to match a mental ray camera to the maxwell camera. There is always an offset which is increasing towards the left and right sides of the image. Therefor Mental Ray does not seem to be an option.

So how do you guys create masks for final compositing? I find Maxwell to be quite useless for professional work, as long as mask creation is not working.

Posted: Tue Sep 13, 2005 12:01 pm
by Neil Evans
Maxwell saves a separate alpha/mask file... Have you tried this?

Posted: Tue Sep 13, 2005 12:22 pm
by GumboYaYa
Yes ... but this is the mask for the complete scene. I need masks for secific objects in the scene, so I can composite and adjust them separately i.e. in Photoshop later on.

Usually I create an RGB mask file with Mental Ray. That means I have 4 surface shaders (black, red, green, blue) and applying them to the objects I want to coloradjust later. That way I have 3 masks in one file.

With Maxwell this is not possible yet - I didn't find a way at least. Maybe I could switch off Primary Visibility for all objects except the objects I need an alpha for. Then I would get an alpha, but first I don't know if Primary visibility is recognized by Maxwell, second I still would not have the correct alpha (since overlapping objects aren't calculated with Primary Visibility) and third I would be quite tedious to create one file and on rendering per alpha. I would end up with 10-20 seperate mask files. And I don'T even want to think of the summarized rendertime of all those mask files :?

Well ... I am still eagerly waiting for solutions :)
Thanks in advance ...

Posted: Tue Sep 13, 2005 5:33 pm
by 3dtrialpractice
pabl0 wrote:yes, mayall only supports 24 fps :?

and another trick :
in the expression of maya camera's horiz and vertic aperture replace :
"cameraShape1.horizontalFilmAperture = MXCamera1.filmWidth * 0.039"
with
"cameraShape1.horizontalFilmAperture = MXCamera1.filmWidth * 0.03937"

so your open GL view match exactly with your Maxwell render view !
(if you want to blend two sorts of render in 2D compositing !)
8)

because 1/25.4 = 0.0393700 (to convert mm in inch)
hey Pabl0 posted this exression correction for the mxCamera, this will help you match your camera renders, from maya to mental ray, the problem you describe is one I face myself, for now, Ive been sorting my geometry into layers, and render 2-3 alpha passes for thos foreground mid and background elements. -good luck maybe well have new mayall soon

from this thread:
http://www.maxwellrender.com/forum/view ... riz+vertic

Posted: Thu Sep 15, 2005 11:31 am
by GumboYaYa
Ok ... this sounds interesting! Thank you!

Changing this expression works fine, but I am not able to deactivate Depth of Field in the cameraShape. As long as this is activated, I get wrong results again, when rendering with MR.

Argh ... :shock:

Help greatly appreciated :)

Posted: Thu Sep 15, 2005 4:43 pm
by 3dtrialpractice
GumboYaYa wrote:Ok ... this sounds interesting! Thank you!

Changing this expression works fine, but I am not able to deactivate Depth of Field in the cameraShape. As long as this is activated, I get wrong results again, when rendering with MR.

Argh ... :shock:

Help greatly appreciated :)
--Ok great,, I came accross this too!
so I render with no depth of feild on my alphas if i can get away with it,
heres how to render without it in software or mental ray:

1) select your camera (view>select camera)
2)in your CHANNEL box on the right under camera shape, RIGHT CLICK> Channel Control
3) in channel control window, select the LOCKED tab, find DEPTH OF FIELD> select it and hit the move button to Unlock depth of field
4)now in your camerashapes's attribut editor, roll down depth of field, and now you can uncheck the box for depth of field.

ALTERNATLY,
If you DONT want to DISABLE Depth of field using the aboave method, but still want the render to not be overblured,

Simply, go back to the Depth of Field roll down in your Camerashapes' atribute editor, and change the value for FOCUS DISTANCE to a large number ie 1000, ( i think the default is 5), youll have to play with this, sometimes i can get ti close, but sometime i have to enter up to 10000, wich essentially removes all blur,


Hope this helps,
happy rendering

-Luke