OctaneRender™ 4 RC 7

Forums: OctaneRender™ 4 RC 7
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™ 4 RC 7

Postby abstrax » Wed Oct 24, 2018 3:43 am

abstrax Wed Oct 24, 2018 3:43 am
Hi all,

This is another release candidate fixing various issues in the materials as well as the material preview rendering. We really hope that this is the last one and we can get the stable version out next week.

As mentioned in the previous release, we are still looking at various stability issues, especially one issue that was reported by various users where Octane rendering fails with spurious CUDA error 2 (out of memory). It seems to happen at specific points during the scene initialization phase but during CUDA calls that don't allocate any memory. For the one scenario where we could reproduce the issue, we have found a workaround to "fix" it, but since it might be just some obscure side-effect of some other problem, it might pop up again in some other scenario. I.e. we need more data from users and more failure cases and any help by you is very much appreciated.

Please note that this build will expire in 120 days, this is now shown on the bottom right of the Standalone release.

As before, some introductory material on V4 can be found here: first release post of V4.

Please make sure you have NVIDIA driver 387 or newer on Windows and Linux, and any CUDA driver for Mac that supports CUDA 9.1. Please also note that architecture improvements mean we can no longer support Fermi GPUs.


Changes since OctaneRender 4 RC 6:

  • Added option to the batch render script and job to allow saving of both the main and the denoised main passes (see viewtopic.php?p=348116#p348116).
  • Fixed broken preview rendering if the main render module hasn't been initialized yet (see for example here: viewtopic.php?p=348636#p348636).
  • Fixed darkening of the planar material preview rendering when a glossy, meetal or universal material is being rendered.
  • Fixed artifacts in the sheen / coating layers of the universal material (see viewtopic.php?p=347984#p347984).
  • Fixed fireflies when sheen and coating are both enabled in the universal material.
  • Fixed clear coat of universal material to have no effect if its IOR is 1.
  • Fixed energy gain at glancing angles when the clear coat roughness is low.
  • Fixed excessive usage of system memory during the compilation of heavily overlapping geometry.
  • OSL: Added support for specifying a step size for integer inputs via the sensitivity metadata.
  • Standalone: Fixed incorrect progress bar updates on Windows 10.
  • Standalone: Reimplemented rotation handle which now allows you to rotate objects multiple times around the picked axis.
  • Lua: Added octane.project.setSelection() which clears the current selection and selects the given node items and pins.


Downloads for users with a regular license
OctaneRender for Windows (installer)
OctaneRender for Windows (zip)
OctaneRender for Mac OS
OctaneRender for Linux

Downloads for users with a Studio subscription license
OctaneRender for Windows (installer)
OctaneRender for Windows (zip)
OctaneRender for Mac OS
OctaneRender for Linux


Happy rendering,
Your OTOY 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: 5483
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby KeeWe » Wed Oct 24, 2018 5:54 am

KeeWe Wed Oct 24, 2018 5:54 am
Thx for the update: I've a question about the new licensing system: As far as I know we will be able to use GPUs in our network as soon as V4 is stable without getting another license (as holder of a perpetual license). Is this correct? And if so: will this new licensing system go online the day you release a stable v4?
6850k // 32 GB // 1080, 1080 Ti, 2080 Ti // Win 10 // C4D 19.068
User avatar
KeeWe
Licensed Customer
Licensed Customer
 
Posts: 278
Joined: Wed Nov 27, 2013 11:32 am

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby grimm » Wed Oct 24, 2018 6:13 am

grimm Wed Oct 24, 2018 6:13 am
I have noticed that error when I'm also running another program that makes very heavy use of the GPU. In my case it was when I was running 3D-Coat along with Octane, they do not work well with each other. :)
Linux Mint 20 x64 | Nvidia GTX 980 4GB (displays) RTX 2070 8GB| Intel I7 5820K 3.8 Ghz | 32Gb Memory | Nvidia Driver 460.56
User avatar
grimm
Licensed Customer
Licensed Customer
 
Posts: 1321
Joined: Wed Jan 27, 2010 8:11 pm
Location: Spokane, Washington, USA

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby funk » Wed Oct 24, 2018 6:47 am

funk Wed Oct 24, 2018 6:47 am
Thanks OTOY
Win10 Pro/ Ryzen 5950X / 128GB / RTX 4090 / MODO
"I am the resurrection, and the life: he that believeth in me, though he were dead, yet shall he live" - Jesus Christ
User avatar
funk
Licensed Customer
Licensed Customer
 
Posts: 1204
Joined: Mon Feb 07, 2011 1:24 pm
Location: Australia

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby glimpse » Wed Oct 24, 2018 1:55 pm

glimpse Wed Oct 24, 2018 1:55 pm
"I've found reducing parallel samples and max tile samples seems to help reduce instances of Octane crashing. I get the CUDA 700 error with heavy scenes when editing shaders. Happening a lot less often with 8/16 for those settings rather than the default. (C4D btw). " source: from comment in https://www.facebook.com/groups/OctaneR ... 940645516/
User avatar
glimpse
Licensed Customer
Licensed Customer
 
Posts: 3714
Joined: Wed Jan 26, 2011 2:17 pm

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby coilbook » Wed Oct 24, 2018 4:08 pm

coilbook Wed Oct 24, 2018 4:08 pm
one of my slaves (3gpus + 4 gpus in netstor box, 1080s and titans xps) constantly has an error "out of memory". Even when I enable out of core. All other slaves work fine. Motherboard is ASUS ROG Rampage VI Extreme LGA2066 DDR4 M.2 U.2 X299 EATX Motherboard with onboard 802.11AD WiGig Wi-Fi USB 3.1, 32 gb of ram, i9-7900x abd nvme for 250GB only. Windows 10 PRO
It happens a lot after octane viewport is used and then rendering begins. I really think it has something to do with ASUS boards too, PLX chips, etc. This is the same PC that would just freeze without any errors. Complete freeze. This is the only slave that has 7 GPUs others have 5, 3 and 3


Four slaves are connected to the netgear 10 Gbit switch. And this is the only slave out of 4 messing up. Other slaves have the same video cards 1080s, titans and one more netstor box. This is the only slave with built in 10 Gigabit port. Others have intel 10 Gigabit network cards installed
Attachments
mem.jpg
coilbook
Licensed Customer
Licensed Customer
 
Posts: 2984
Joined: Mon Mar 24, 2014 2:27 pm

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby linvanchene » Wed Oct 24, 2018 8:08 pm

linvanchene Wed Oct 24, 2018 8:08 pm
Nishita Daylight Model

I do not see any effect when adjusting the Sky Turbidity parameter.


2018-10-24 (18).png



Update / Edit:

Not seeing an effect when adjusting Sky Color and Sunset Color as well in the Nishita Daylight Model.


2018-10-24 (19).png


- - -


Any other parameter that needs to be adjusted in combination?
Or are those parameters simply not used in the Nishita Daylight Model?

In the New Daylight environment adjusting Sky Turbidity alone shows a huge difference in shadow softness as expected. Sky and Sunset Colors also change drastically as expected.
Win 10 Pro 64bit | Rendering: 2 x ASUS GeForce RTX 2080 Ti TURBO | Asus RTX NVLink Bridge 4-Slot | Intel Core i7 5820K | ASUS X99-E WS| 64 GB RAM
FAQ: OctaneRender for DAZ Studio - FAQ link collection
User avatar
linvanchene
Licensed Customer
Licensed Customer
 
Posts: 783
Joined: Mon Mar 25, 2013 10:58 pm
Location: Switzerland

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby wallace » Wed Oct 24, 2018 9:22 pm

wallace Wed Oct 24, 2018 9:22 pm
linvanchene wrote:Nishita Daylight Model

I do not see any effect when adjusting the Sky Turbidity parameter.


2018-10-24 (18).png



Update / Edit:

Not seeing an effect when adjusting Sky Color and Sunset Color as well in the Nishita Daylight Model.


2018-10-24 (19).png


- - -


Any other parameter that needs to be adjusted in combination?
Or are those parameters simply not used in the Nishita Daylight Model?

In the New Daylight environment adjusting Sky Turbidity alone shows a huge difference in shadow softness as expected. Sky and Sunset Colors also change drastically as expected.


Hi,

Nishita Sky does not work with sky color and sunset color given that it is a physically based model. Turbidity however should work, I'll look into it.
User avatar
wallace
OctaneRender Team
OctaneRender Team
 
Posts: 188
Joined: Sun Dec 18, 2016 10:38 pm

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby profbetis » Wed Oct 24, 2018 10:46 pm

profbetis Wed Oct 24, 2018 10:46 pm
The Universal Material has a strange behavior at low specular roughness values. It has these harmonic ring looking artifacts. These artifacts fade away as roughness approaches 0.002 and up.
Image

I tested the same on the metallic material and no such artifacts exist.
Ray Epsilon has no affect.
The rings shift around when the metallic parameter is adjusted, in my case around the values 0.9 to 1.0
GTX 1080Ti 11GB (3x), Water-cooled
Intel i7-5820K 6-core @ 3.3GHz
Windows 10 Pro 64-bit, 32GB RAM
User avatar
profbetis
Licensed Customer
Licensed Customer
 
Posts: 279
Joined: Tue Dec 21, 2010 11:00 am

Re: OctaneRender™ 4 RC 7 [latest version 4]

Postby coilbook » Thu Oct 25, 2018 12:44 am

coilbook Thu Oct 25, 2018 12:44 am
Can anyone at otoy explain to me when to use AI lights and Ai light update? When I have a lot of lights. I dont see any difference using it or not. Denoiser works the same and noise levels are the same
Thanks

UPDATE:
I tried a scene with a lot of lights and yes there is a big difference when using AI light but there is a possible bug I posted in max forum when ai lights work in octane viewport and not during rendering viewtopic.php?f=80&t=69152&start=70
coilbook
Licensed Customer
Licensed Customer
 
Posts: 2984
Joined: Mon Mar 24, 2014 2:27 pm
Next

Return to Development Build Releases


Who is online

Users browsing this forum: No registered users and 5 guests

Tue Mar 19, 2024 7:21 am [ UTC ]