Cloudy plastic artifacts

Started by Xidor, March 28, 2012, 09:45:05 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Xidor

Hey guys,

I have a strange artifact I can't seem to figure out how to resolve.

I am trying to render a plastic part in white that has a LED behind it that glows through. I used the White Shiny Cloudy Plastic material for this. It worked okay as the LED behind is showing through. The problem I am having is in a ball shaped socket in the part that has a grid like artifact.

The only thing I did to the material is change the Diffuse color to solid white. I tried upping the shadow quality setting to 5, I tried upping the ray bounces to 8. I tried increasing the brightness in the environment. Nothing I did based on past experience is working. I even tried indirect illumination.

The image below is the section I am having trouble with. I included the area that has the LED glowing through so you see what I'm trying to do. The LED showing through is okay, not great.



I'd appreciate any advise. Perhaps another material would work better? Anything that would allow the emissive material I used on the sphere behind to glow through. But why is that grid artifact showing? And how to eliminate it?

Any advice would be greatly appreciated! Thanks!

Nelson

caden2010

Hi there,

I also have a similar problem. I am using the "Clear rough plastic - white" material for a plastic facia with a label on the rear surface. There is a gap between the plastic facia and the main body component as I know that mating faces can cause problems.

For my renders I have had to use the shadow quality of 1. This is OK but there needs to be a solution as this is currently compromising the image quality. The higher the shadow quality the worse the problem appears to be.

With shadow quality set at 1 there are still problems (shown within the red ellipse) but the general problem is not so noticeable on the final render.

Xidor

Thanks Caden. I tried shadow quality setting adjustments as you suggested. It didn't work for me.

I'll keep at and hope to resolve it.

Nelson

m2tts

I would try upping my ray bounces first to above 12. The indirect illumination tab only affects your realtime graphics view and if your using the realtime raytracer. The regular raytracer will always use indirect illumination.

Xidor

Thanks for the suggestion to up the ray bounces. I'll give it a shot and see what happens.

guest84672

@m2tts - not anymore. This has changed in 3.1.

To the rest - are you using KeyShot 2 or 3?

Xidor

Hi Thomas,

I'm using version 3. I think I have the most current build. And this sample render I show in the first post was done on the Mac side of my computer. I'm running Bootcamp now, so I can't switch over to check just now.

I hope you have recommendation on how to eliminate those artifacts.  :)

Thanks!

guest84672

Try realtime render mode - set it to max time. Set the time limit high. And then check back periodically. You can stop the rendering as soon as you have the desired result.

Xidor

Hi Thomas,

I've never done anything like that before in Keyshot. I'll see if I can see how that works.

Thanks, Nelson

guest84672

Yes - realtime mode is quite good and often times preferred.

Xidor

I tried realtime rendering. Assuming I did it right, I let it go for about 30 minutes. the image kept getting better and better but the gridded artifacts in the sphere remained.

So I created another model I could actually show as that other model is for client work. This model is actually bigger, so the effects of the artifacts are lessened. But still there.

Here's the new render. Same Glossy Cloudly material. I set shadow quality to 5. That was all I did.



You can still see the grid or whatever that is on the upper area of the sphere. Very strange.

By the way, this part has a 1mm thick wall, so it's not solid.

KeyShot

Can you share the model with support then we can take a look at it?

Xidor

Hi!  Do you want the actual CAD model or bip file?

Thanks, Nelson.

caden2010

Hi, I'm using KeyShot version 3.1.48.

If you wish I can send the BIP, well at least the relevant parts.

Thanks, Stuart

guest84672