Indigo 1.1.2

Post by **OnoSendai** » Fri Jun 06, 2008 2:55 am

Wow, cool modelling aleksandera.

One thing: I would suggest you use a photo/scan of the retinal inner surface of the eyeball, instead of a diffuse black.

Kram1032 · Post by **Kram1032** » Fri Jun 06, 2008 3:33 am

uhm, just as an example, what did you do before? xD
So that's a full complex eye, very close to a real one? cool

You shouldn't use entirely black diffuse for the background, though...

aleksandera · Post by **aleksandera** » Fri Jun 06, 2008 4:55 am

ThX!!

Nope, next time I'l just texture the back of the eyeball black,
ore maby I'l add some retinal inner bla bla surface displacement + albedo
on it.
Hey here is one: http://www.nervenet.org/papers/images/h ... nCones.gif

Look at Ono's avatar, there is nothing but black inside O.o
The mistake I made was not to set the precedence low on
the specular material - thats why the render did go crazy.

The whole thing looks realistic enaugh right now, so I will try to make some serious art work, and not only test stuff.

Ok, there is still a problem with the displaced head wich I hawe posted
on the Blendigo thread.

ANYBODY !!??

(sorry for my b.a.)

CTZn · Post by **CTZn** » Fri Jun 06, 2008 5:46 am

Ah thanks for the report !

Very nice head ! Have you tried the skin-specific material ?

Kram1032 · Post by **Kram1032** » Fri Jun 06, 2008 6:01 am

I kinda am pretty sure that you tried that.... But anyway: Did you try to remove doubles? (And obviously the two things, you should always check

)

BbB · Post by **BbB** » Fri Jun 06, 2008 9:11 am

Nice to know Alex. I've been fighting with that one. I remember trying to set the precedence lower. Didn't know it had to be set MUCH lower.
Good to know (even though weird).

zsouthboy · Post by **zsouthboy** » Fri Jun 06, 2008 11:53 am

I have something interesting re: the mesh splitting problem that's being reported.

I finally came across it myself, and it's reproduceable (and on a C1 surface, to boot!)

Blender -> create an IcoSphere (i picked level 2, it doesn't matter)
Unwrap the UVs using the auto unwrap function (edit mode -> U key -> Unwrap (smart projection)

Pick a procedural texture (i've got wood here), stick in the COL layer of your object you just unwrapped.

bake that procedural into a new texture (UV/Image view -> Image -> New... -> pick something -> select all faces of mesh -> CTRL ALT B -> Textures Only)

Save that texture you just made somewhere.

Apply that texture as the displacement for your mesh.

The mesh will be non continuous at the edges of the UV mapping you did.

Is this an export problem or something else?

zsouthboy · Post by **zsouthboy** » Fri Jun 06, 2008 12:14 pm

and here's a sphere-unwrapped render of the same mesh:

CTZn · Post by **CTZn** » Fri Jun 06, 2008 1:22 pm

Seems like an exporter issue. I reproduced your workflow in maya and fisrt insight shows things ok here. I will return as soon as I get something decent to show.

Ah, it reached leaf 18/32... Maybe I'm going to lower subdivision settings.

- - -

Nope there is something odd here too, same apart that it looks like the shells are connected while they are not in you test, zsouthboy.

CTZn · Post by **CTZn** » Fri Jun 06, 2008 2:03 pm

Observing closer, ther should be no reason to this mismatch at UV shell border.

I believe Indigo is stretching the UVs near borders. This seems to be inefficient withe low density meshes, wich defeats the point of subdivisions a bit.

That's how I understand the issue right now, I'm not pretending to be correct in my interpretation.

zsouthboy, are you sure you didn't render with 1.1.1 because we have so different results...

zsouthboy · Post by **zsouthboy** » Fri Jun 06, 2008 2:42 pm

Yep I'm sure it's 1.1.2 - check the stamp (damn that's useful Ono, thanks for that)

CTZn · Post by **CTZn** » Sat Jun 07, 2008 12:38 am

Yup you're right. The surface stops in 1.1.1 where the UVs start to strectch in 1.1.2.

However I can't explain why your uvs are not stretching, and the shells separated. Oh yes: are you using <c> ? I think I've set it to

Code: Select all

-0.25*<b>

IIRC. Althought in most recent tests

c = -b

looks fine (edit: but still stretched).

leope · Post by **leope** » Sat Jun 07, 2008 10:00 pm

BbB wrote:There you go.
This is a smoothed, low-poly sphere with a diffuse mat and bump map. Rendered with 1.0.9_3
I'm also attaching the scene.

Not sure but it seems like yafray artifacts, managed by the parameter Bi
(Shadow ray bias to avoid self shadowing).
If i remeber good i posted (long time ago) a similar post at the kerkythea forum.

StompinTom · Post by **StompinTom** » Sun Jun 08, 2008 2:17 am

leope wrote:
BbB wrote:There you go.
This is a smoothed, low-poly sphere with a diffuse mat and bump map. Rendered with 1.0.9_3
I'm also attaching the scene.
Not sure but it seems like yafray artifacts, managed by the parameter Bi
(Shadow ray bias to avoid self shadowing).
If i remeber good i posted (long time ago) a similar post at the kerkythea forum.

...
if only this was yafray.

leope · Post by **leope** » Sun Jun 08, 2008 10:40 am

StompinTom wrote: ...
if only this was yafray.

I don't mean a similar parameter is needed for indigo... i can only see analogy on rendered images and i don't know if self shadowing is a yafray-only issue to manage... so sorry for my previuos post and for my bad english...

Indigo 1.1.2

Who is online