Hybrid MLT+Bidir glass material is darker than CPU MLT+Bidir

Please use this forum for general user support and related questions.

Moderator: coordinators

Forum rules
Please include your operating system type/version, LuxRender version and Exporter version used when submitting a support post.

Make sure you have read the Release forum thread for Release and RC (Release Candidates) builds as these threads contain information on known problems and workarounds: Test Builds Forum

Hybrid MLT+Bidir glass material is darker than CPU MLT+Bidir

Postby B.Y.O.B. » Fri Feb 17, 2012 12:49 pm

Hi, out of boredom I made a test today to compare the speed of Hybrid vs. CPU.
The scene has 1,007,905 tris and each image was rendered for 20 minutes.

Yes, Hybrid is slower (GPU was a GTX 460 1GB), but what I also observe is that the Hybrid Bidir glass2-material is darker than in the other two renders.
I know, Hybrid Bidir isn't finished yet, but I just wanted to ask if this is already a known issue.

I also attached the scene.
Attachments
20min-Hybrid-MLT-Path.jpg
Hybrid MLT+Path
20min-Hybrid-MLT-bidir.jpg
Hybrid MLT+Bidir
20min-CPU-MLT-bidir.jpg
CPU MLT+Bidir
untitled.blend.zip
Blender file
(80.41 KiB) Downloaded 3 times
User avatar
B.Y.O.B.
 
Posts: 1880
Joined: Wed Nov 10, 2010 4:10 pm
Location: Germany

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby rendermagic » Fri Feb 17, 2012 5:10 pm

I noticed a similar thing also a couple of weeks ago. I'm at my other job right now, so I don't have blender on this machine to look at your attached *.blend file. I think I remember having a problem with the metal materials too and it wasn't just the Glass material. Are you using a hemi light in your scene or even an area light? For whatever it's worth, I noticed that when I changed the output render resolution, I had varying results both good and bad. I wasn't able to pin point exactly what was causing it other than one common denominator....hybrid bi-dir.
Render Magic
----------------
i7 950 - Not OC'd
24G DDR3 RAM
2 GTX 580s
rendermagic
Developer
 
Posts: 141
Joined: Wed Mar 23, 2011 2:32 pm
Location: Leading edge of a photon (California USA)

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby paco » Fri Feb 17, 2012 5:29 pm

Just out of curiosity I don't suppose you did a CPU MLT + path?
paco
Developer
 
Posts: 456
Joined: Sun Feb 07, 2010 1:37 am

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby B.Y.O.B. » Sat Feb 18, 2012 1:18 am

rendermagic wrote:Are you using a hemi light in your scene or even an area light?

The only light in the scene is a meshlight on the right hand side of the camera, the background is completely black.
paco wrote:Just out of curiosity I don't suppose you did a CPU MLT + path?

No I didn't, but I can make one if you want.
User avatar
B.Y.O.B.
 
Posts: 1880
Joined: Wed Nov 10, 2010 4:10 pm
Location: Germany

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby paco » Sat Feb 18, 2012 3:55 am

B.Y.O.B. wrote:
paco wrote:Just out of curiosity I don't suppose you did a CPU MLT + path?

No I didn't, but I can make one if you want.


Yes please! I'd be very interested to see the result.
paco
Developer
 
Posts: 456
Joined: Sun Feb 07, 2010 1:37 am

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby B.Y.O.B. » Sat Feb 18, 2012 9:57 am

There you are :)

20 minutes MLT + Path (32 bounces, like the others).
Attachments
20min-CPU-MLT-Path.jpg
User avatar
B.Y.O.B.
 
Posts: 1880
Joined: Wed Nov 10, 2010 4:10 pm
Location: Germany

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby paco » Sat Feb 18, 2012 4:03 pm

Thanks BYOB!

Looks the same as bidir to me, so the CPU side seems to working fine. Interesting stuff :)
paco
Developer
 
Posts: 456
Joined: Sun Feb 07, 2010 1:37 am

Re: Hybrid MLT+Bidir glass material is darker than CPU MLT+B

Postby rendermagic » Sat Feb 18, 2012 4:08 pm

paco wrote:Thanks BYOB!

Looks the same as bidir to me, so the CPU side seems to working fine. Interesting stuff :)


Exactly what I noticed. The CPU side seems to work exactly as expected when compared between bi-dir and path. The bug is constrained to bi-dir, specifically between CPU and Hybrid.I expect, however, that bi-dir is probably going to be a little buggy since it's still really new.
Render Magic
----------------
i7 950 - Not OC'd
24G DDR3 RAM
2 GTX 580s
rendermagic
Developer
 
Posts: 141
Joined: Wed Mar 23, 2011 2:32 pm
Location: Leading edge of a photon (California USA)


Return to LuxRender User Support

Who is online

Users browsing this forum: No registered users and 0 guests