OctaneRender™ Standalone 3.08 RC 3

Forums: OctaneRender™ Standalone 3.08 RC 3
A forum where development builds are posted for testing by the community.
Forum rules
NOTE: The software in this forum is not %100 reliable, they are development builds and are meant for testing by experienced octane users. If you are a new octane user, we recommend to use the current stable release from the 'Commercial Product News & Releases' forum.

OctaneRender™ Standalone 3.08 RC 3

Postby abstrax » Wed Mar 28, 2018 11:11 pm

abstrax Wed Mar 28, 2018 11:11 pm
Hi all,

This is the third release candidate of 3.08, fixing more bugs and various platform specific issues. We will release a stable 3.08 next week if no major issues pop up until then.

For an overview of all new features of 3.08, please read the new features thread.

As noted before, we are now using CUDA toolkit 9.1 (to allow for Volta support) on Windows and Linux, which requires NVIDIA drivers of version 387 or higher.

Due to issues with the current CUDA drivers for Mac we are using only CUDA toolkit 8.0 on OSX, which is why any CUDA driver should work that supports CUDA 8.0 or higher.



Changes since version 3.08 RC 2:

  • Fixed rendering hanging on Linux.
  • Fixed GPU failure if a scene has displacement triangles with degenerated UV coordinates.
  • Displacement vectors/normals are now "sanitized" to not lie exactly in the triangle plane, which caused GPUs to fail.
  • Fixed triplanar and box mapping to not use the bump/normal mapped shading normal (see viewtopic.php?p=333691#p333691).
  • Fixed unnecessary resending of all image textures to net render slaves if only some of the image textures changed (see viewtopic.php?p=333478#p333478).
  • Fixed artifacts while rendering scattering media with the direct lighting kernel.
  • Fixed reported error when a shader contains an unsupported type of metadata.
  • Fixed various compilation errors when trying to parse OSL shaders with weird syntax.
  • Fixed the default values of various texture inputs to match the values we have in 3.07 when those inputs are not connected.
  • Since 3.07, the default hair thickness and particle radius are given in the import unit specified in the import preferences. Added a conversion for older files, where these parameters were always given in meters.
  • Fixed undo behaviour of the film region picker.
  • The fallback texture now behaves the same as in 3.07 when used in a medium: It always returns the minimum value if the mode "normal vs eye ray" is used.


Downloads for users with a regular license

regular version for Windows (installer)
regular version for Windows (ZIP archive)
regular version for Mac OS X (DMG image)
regular version for Linux (ZIP archive)


Downloads for users with a subscription license

subscription version for Windows (installer)
subscription version for Windows (ZIP archive)
subscription version for Mac OS X (DMG image)
subscription version for Linux (ZIP archive)


Happy rendering,
Your OTOY NZ Team
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
abstrax
OctaneRender Team
OctaneRender Team
 
Posts: 5112
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby rashadcarter » Thu Mar 29, 2018 2:42 am

rashadcarter Thu Mar 29, 2018 2:42 am
Can someone, anyone help me get to grips with whats happening with Specular materials and refraction settings? Any setting above 1.0 results in horrible results. Feedback is greatly appreciated.
rashadcarter
Licensed Customer
Licensed Customer
 
Posts: 70
Joined: Thu Sep 11, 2014 1:16 am

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby abstrax » Thu Mar 29, 2018 2:47 am

abstrax Thu Mar 29, 2018 2:47 am
rashadcarter wrote:Can someone, anyone help me get to grips with whats happening with Specular materials and refraction settings? Any setting above 1.0 results in horrible results. Feedback is greatly appreciated.

Sound like you have got your normals wrong. You can check the normals using the info channel kernel, mode "Wireframe" and with "Wireframe backface highlighting" enabled. If the mesh is rendered red, the normals are flipped.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
abstrax
OctaneRender Team
OctaneRender Team
 
Posts: 5112
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby rashadcarter » Thu Mar 29, 2018 3:02 am

rashadcarter Thu Mar 29, 2018 3:02 am
abstrax wrote:
rashadcarter wrote:Can someone, anyone help me get to grips with whats happening with Specular materials and refraction settings? Any setting above 1.0 results in horrible results. Feedback is greatly appreciated.

Sound like you have got your normals wrong. You can check the normals using the info channel kernel, mode "Wireframe" and with "Wireframe backface highlighting" enabled. If the mesh is rendered red, the normals are flipped.


There are some facets with inverted normals in the scene, but none of them are assigned specular materials. Normals inversion can't be the issue. This has been a problem for me in all releases that have included Anisotropy. Not that anisotropy is the issue, but I suspect that a change that was made to specular materials during that time has caused this problem.
rashadcarter
Licensed Customer
Licensed Customer
 
Posts: 70
Joined: Thu Sep 11, 2014 1:16 am

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby abstrax » Thu Mar 29, 2018 3:06 am

abstrax Thu Mar 29, 2018 3:06 am
rashadcarter wrote:
abstrax wrote:
rashadcarter wrote:Can someone, anyone help me get to grips with whats happening with Specular materials and refraction settings? Any setting above 1.0 results in horrible results. Feedback is greatly appreciated.

Sound like you have got your normals wrong. You can check the normals using the info channel kernel, mode "Wireframe" and with "Wireframe backface highlighting" enabled. If the mesh is rendered red, the normals are flipped.


There are some facets with inverted normals in the scene, but none of them are assigned specular materials. Normals inversion can't be the issue. This has been a problem for me in all releases that have included Anisotropy. Not that anisotropy is the issue, but I suspect that a change that was made to specular materials during that time has caused this problem.

Could you post an example or send it to me via PM? Thanks.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
abstrax
OctaneRender Team
OctaneRender Team
 
Posts: 5112
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby cfrank78 » Thu Mar 29, 2018 9:29 am

cfrank78 Thu Mar 29, 2018 9:29 am
Hi Marcus!

Back on my machine and tested now the newer 391.35 driver. still gettong no cuda. The log shows:
CUDA error 999 on device -1: unknown error
-> failed to initialise CUDA driver API
Missing or invalid 'seconds_to_expiration' field from ORC server API response
ORC subscription status cannot be found

Install 388.71 and everything works fine. Here the clou: Tryed it now on a win 7 pro machine with a 780 and a 670 and it works with 391 driver too. tryed it with rc2 and rc3.
so it seems its a pure win 8.1 problem. hope that helps!

Have a nice day - Chris!
cfrank78
Licensed Customer
Licensed Customer
 
Posts: 328
Joined: Sat Nov 17, 2012 11:44 pm

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby aufwind » Thu Mar 29, 2018 10:08 am

aufwind Thu Mar 29, 2018 10:08 am
Hi,

thanks for fixing the bump map issue with box projection.

I think I have found another issue with bump mapping.
The appearance of shiny materials with a bump map applied look quite different in 3.08 compared to 3.07.
Is this intentionally? When bump strength is set to 0 both look the same.

There's a lot of noise on the chairs on the right side, which didn't occur on 3.08 RC2

Render 3.07
3.07.jpg


Render 3.08 RC3
3.08RC3.jpg
Win7 | Gigabyte G1.Sniper 3 | I7 3770K | 2x Titan (Kepler) | 1x Titan X (Pascal)
aufwind
Licensed Customer
Licensed Customer
 
Posts: 120
Joined: Sun Mar 31, 2013 4:47 pm

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby mac3d » Thu Mar 29, 2018 10:40 am

mac3d Thu Mar 29, 2018 10:40 am
Thank for the linux version.
My Computer:i7 2,6ghz ram 8gb schede video nvidia 9600tso 512mb gtx560ti 1gb mainboard asus p6t se sistema operativo:Ubuntu 14.10 64 bit
mac3d
Licensed Customer
Licensed Customer
 
Posts: 103
Joined: Sun Jan 15, 2012 10:35 pm

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby sbstnc » Fri Mar 30, 2018 11:33 am

sbstnc Fri Mar 30, 2018 11:33 am
Sorry if this is a silly Q, but I thought I read on the RC1 or RC2 pages that we need 388.XX or above to run?
Mac Nvidia web driver is 378.xx only no? Can anyone confirm 3.08RC2 or RC3 running on Mac? Internal or eGPU?
Cheers
S
sbstnc
Licensed Customer
Licensed Customer
 
Posts: 116
Joined: Mon Jul 25, 2016 1:21 am

Re: OctaneRender™ Standalone 3.08 RC 3 [latest 3.xx]

Postby bepeg4d » Fri Mar 30, 2018 2:18 pm

bepeg4d Fri Mar 30, 2018 2:18 pm
Hi,
I can confirm that 3.08 RC 2 and 3 works correctly on Mac with internal Nvidia or via eGPU, and you need to update both CUDA and GPU drivers to 387.xx, in both Sierra or High Sierra:
CB51AD5C-A9D0-4ABD-A62D-B5D560F8AA4F.jpeg

ciao Beppe
User avatar
bepeg4d
Octane Guru
Octane Guru
 
Posts: 6748
Joined: Wed Jun 02, 2010 6:02 am
Location: Italy
Next

Return to Development Build Releases


Who is online

Users browsing this forum: No registered users and 1 guest

Mon Dec 09, 2019 9:28 pm [ UTC ]