feature requests

[patro]

hey,

nice test... but what about the black linear area to see under the shelf in first image....
or the black area to see in the teapot in the center?

it's a glossy material... but as I know the glossy one is now called glossy lossy and there is a new glossy material... that works better with 0.7 version.

anyway nice tests and incredible progress in image quality

[hedphelym]

black inside teapot is because it's not a "closed" mesh. just thin "paper".
there is no mesh on the inside.

the black under the shelf is happening because the mesh is not dense enough,
this is before:

result:

I then apply a "turbosmooth" to the shelf:

and the result is flawless, so don't worry about black lines, just make sure mesh is dense enough.

this is also the same thing with chamfer boxes etc.. (just so you know), the mesh is not dense enough as "default" when you create the object.

[patro]

thank you for the answer

so this mean that I most probably will not have problem with my imported obj from rhino
they are just dense enough

[Lord Crc]

the black under the shelf is happening because the mesh is not dense enough

Looks awfully lot like smoothed mesh without "edgesplit" as we call it in Blender, causing normals to point into the wall or something.

[hedphelym]

this is a question I've been wondering about for a while:

When you split the mesh like that, does that not cause renders to take longer? or that light leaks into model?
(If I understand the splitting correctly, then you "detach" the faces based on smoothing angle \ texture), correct?

[patro]

the black under the shelf is happening because the mesh is not dense enough

Looks awfully lot like smoothed mesh without "edgesplit" as we call it in Blender, causing normals to point into the wall or something.

the edgesplit for 3dsmax is the smoothing group option..... while the turbosmooth a modifier is like the blender smooth but an heavier version.
the black patch will avoid by tessellating the surface through the turbosmooth.... and not by subdividing the surface in groups surfaces.

as I know... for max & LuxRender works better increasing face subdivision instead to use smoothing groups

[hedphelym]

all shaders is now working.
I just have some minor tweaks to do in the camera GUI, and then I can release this first scripted version of luxmax.

I hope to finish tomorrow, but might be a couple of days.
I'm rendering out my test scene now over night, I'll post it tomorrow.

the luxmax exporter (with all scripted materials, cameras etc) is now about 4000 lines of script
Growing fast.

[hedphelym]

Im proud to announce the first version of the new exporter...

http://www.luxrender.net/forum/viewtopic.php?f=33&t=4926&p=51367#p51363

[patro]

this is a question I've been wondering about for a while:

When you split the mesh like that, does that not cause renders to take longer? or that light leaks into model?
(If I understand the splitting correctly, then you "detach" the faces based on smoothing angle \ texture), correct?

could this topic viewtopic.php?f=8&t=4923 be part of the answer?

[Lord Crc]

When you split the mesh like that, does that not cause renders to take longer? or that light leaks into model?
(If I understand the splitting correctly, then you "detach" the faces based on smoothing angle \ texture), correct?

Yes thats pretty much what happens, if an edge is shared by two faces with different smoothing groups (iirc), or the faces make a sharp angle along that edge in EdgeSplit, you duplicate the vertexes of the edge and assign one pair to each face. Since the vertex positions remain unchanged this will not affect rendering as such, other than normals.

However it did affect things like subdiv until last week