Version 4.0-RC6-R2 (Obsolete release candidate) 19.10.2018

Forums: Version 4.0-RC6-R2 (Obsolete release candidate) 19.10.2018
Sub forum for plugin releases

Moderator: aoktar

Re: Version 4.0-RC6 (Latest release candidate) 18.10.2018

Postby divasoft » Tue Oct 23, 2018 2:20 am

divasoft Tue Oct 23, 2018 2:20 am
aoktar wrote:1-What do you see in console output about Octane when C4D loaded?
2-Did you try without any other plugins?
3-What's your nvidia driver and OS?
4-What's your hardware (gpus/cpu/mobo/ram)?


1.
Cinema R20 with no other plug (R20 NOT produces bug with mat prev increasing exposure)
Started logging on 23.10.18 09:08:18
OctaneRender 4.00 RC6 (4000016)
CUDA error 400 on device 0: invalid resource handle
-> failed to launch kernel(kernel63)

Cinema R19
Started logging on 23.10.18 09:10:42
OctaneRender 4.00 RC6 (4000016)
invalid offline data block
Could not retrieve data from OctaneLive.
CUDA error 400 on device 0: invalid resource handle
-> failed to launch kernel(kernel63)

2. R19 produces this bug with no other plug
3. win10 64, NV driver 416.34
4. 2x 1080ti nvidia, i7 930, mobo?, 24Gb DDR3 HyperX Fury
User avatar
divasoft
Licensed Customer
Licensed Customer
 
Posts: 348
Joined: Thu Jul 01, 2010 3:59 am
Location: Russia

Re: Version 4.0-RC6 (Latest release candidate) 18.10.2018

Postby abstrax » Tue Oct 23, 2018 3:02 am

abstrax Tue Oct 23, 2018 3:02 am
divasoft wrote:
aoktar wrote:1-What do you see in console output about Octane when C4D loaded?
2-Did you try without any other plugins?
3-What's your nvidia driver and OS?
4-What's your hardware (gpus/cpu/mobo/ram)?


1.
Cinema R20 with no other plug (R20 NOT produces bug with mat prev increasing exposure)
Started logging on 23.10.18 09:08:18
OctaneRender 4.00 RC6 (4000016)
CUDA error 400 on device 0: invalid resource handle
-> failed to launch kernel(kernel63)

Cinema R19
Started logging on 23.10.18 09:10:42
OctaneRender 4.00 RC6 (4000016)
invalid offline data block
Could not retrieve data from OctaneLive.
CUDA error 400 on device 0: invalid resource handle
-> failed to launch kernel(kernel63)

2. R19 produces this bug with no other plug
3. win10 64, NV driver 416.34
4. 2x 1080ti nvidia, i7 930, mobo?, 24Gb DDR3 HyperX Fury

Thank you for your reply. It's the same issue as this: viewtopic.php?p=348689#p348689
It affects all material/texture preview rendering if no normal rendering has been started yet.
We are planning to get another build out tomorrow, but still working on some issue in the universal material. Apologies for the hassle.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
abstrax
OctaneRender Team
OctaneRender Team
 
Posts: 5484
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby divasoft » Wed Oct 24, 2018 5:07 am

divasoft Wed Oct 24, 2018 5:07 am
Ahmet, please, make possible to save passes in LV like saving beauty pass. PLEEESE. With sRGB profiling and tonemapping.
User avatar
divasoft
Licensed Customer
Licensed Customer
 
Posts: 348
Joined: Thu Jul 01, 2010 3:59 am
Location: Russia

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby chrisbrearley » Wed Oct 24, 2018 9:11 am

chrisbrearley Wed Oct 24, 2018 9:11 am
When using Render Layers the Black Shadow and Reflection passes do not get denoised. This negates the whole denoising process on the beauty pass as we still need to render with the same amount of samples as we did with v3 to get clean shadows and reflections.

Please add denoising to all Render Layer passes.


On another note, having the beauty intersect and cutout the shadow and reflection passes with Render Layers makes it awkward when applying depth blur in post. The mode 'only side effects' keeps the full shadow and reflection passes without anything cutout but then it doesn't render the beauty. Is there anyway to get all of the shadow and reflection like you get with 'only side effects' whilst keeping the beauty?
chrisbrearley
Licensed Customer
Licensed Customer
 
Posts: 23
Joined: Mon Jul 11, 2016 11:09 am

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby chrisbrearley » Wed Oct 24, 2018 9:19 am

chrisbrearley Wed Oct 24, 2018 9:19 am
Also getting a huge amount of crashes as others have reported. It seems to happen a lot when changing material properties. Changing out-of-core to off also has no effect. I would have HUGE objections if this current version was to be released as stable anytime soon.

Loaded Octane DLL:OctaneRender 4.00 RC6 version:4000016
Octane Render for Cinema 4D V4.0-RC6-R2 built for:200 C4D vers:20028

Octane Render SDK version: 4.00 RC6
HDRI Link V(1.054), 20028 Studio, Windows
HDRI Browser V(2.146), 20028 Studio, Windows
init Octane Api...
Activation time=5.316 sec.
Daemon listener -> slave0: THE_BEAST state:4 IP:192.168.0.31
Daemon listener -> slave1: BIRKSY state:4 IP:192.168.0.40
Daemon listener -> slave2: FREELANCE state:4 IP:192.168.0.42
Daemon listener -> slave0: THE_BEAST state:4 IP:192.168.0.31
Daemon listener -> slave1: BIRKSY state:5 IP:192.168.0.40
Daemon listener -> slave2: FREELANCE state:4 IP:192.168.0.42
Daemon listener -> slave0: THE_BEAST state:4 IP:192.168.0.31
Daemon listener -> slave1: BIRKSY state:5 IP:192.168.0.40
Daemon listener -> slave2: FREELANCE state:5 IP:192.168.0.42
Scene created in plugin version 3080400
Scene plugin version is older than version:V4.0-RC6-R2
Plugin will convert the parameters of displacement texture.
Animated attributes should be changed by manually!
Daemon listener -> slave0: THE_BEAST state:5 IP:192.168.0.31
Daemon listener -> slave1: BIRKSY state:5 IP:192.168.0.40
Daemon listener -> slave2: FREELANCE state:5 IP:192.168.0.42
Daemon listener -> slave0: THE_BEAST state:5 IP:192.168.0.31
Daemon listener -> slave1: BIRKSY state:5 IP:192.168.0.40
Daemon listener -> slave2: FREELANCE state:5 IP:192.168.0.42
--------------------------- EXPORT LOG ---------------------------
Export materials time= 16754.636 ms
Collect objects time= 24.025 ms
Mesh creation time = 88.053 ms.
Total export Time = 17084.16 ms
VRAM used/free/max:3.845Gb/638Mb/8Gb Out-of-core used:640Mb total used RAM:27.757Gb
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> failed to download symbol(stats data)
OCT:device 0: path tracing kernel failed
----------------------------------------------------------------------------------------
<<< Render failure detected!!! >>>
Please check render statistics to solve the problem.
MB:0/0 ST/MOV:0/13 Nodes:87 Tris:432k DispTris:118k Hairs:0 Meshes:16
Device:0 TotMem:8Gb rtData:650Mb film:28Mb geo:1.764Gb node:14Kb tex:1.059Gb unavailable:3.163Gb temperature:40
Device:1 TotMem:8Gb rtData:650Mb film:28Mb geo:1.764Gb node:14Kb tex:1.434Gb unavailable:1.393Gb temperature:40
OCT:CUDA error 700 on device 1: an illegal memory access was encountered
OCT: -> failed to wait for event
OCT:device 1: path tracing kernel failed
--------------------------- EXPORT LOG ---------------------------
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> failed to unregister pinned memory
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> could not get memory info
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> failed to unregister pinned memory
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> could not get memory info
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> failed to unregister pinned memory
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> could not get memory info
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT:CUDA error 700 on device 1: an illegal memory access was encountered
OCT: -> failed to unload module
OCT: -> failed to unload module
OCT:CUDA error 700 on device 0: an illegal memory access was encountered
OCT: -> failed to create link
OCT:CUDA error 700 on device 1: an illegal memory access was encountered
OCT: -> failed to create link
OCT:device 0: failed to create OSL compile context:
OCT:device 1: failed to create OSL compile context:
OCT:device 0: failed to compile module 0
OCT:device 1: failed to compile module 0
----------------------------------------------------------------------------------------
<<< Render failure detected!!! >>>
Please check render statistics to solve the problem.
MB:0/0 ST/MOV:0/13 Nodes:87 Tris:0 DispTris:0 Hairs:0 Meshes:0
Device:0 TotMem:8Gb rtData:650Mb film:28Mb geo:1.764Gb node:14Kb tex:1.059Gb unavailable:3.163Gb temperature:38
Device:1 TotMem:8Gb rtData:650Mb film:28Mb geo:1.764Gb node:14Kb tex:1.434Gb unavailable:1.393Gb temperature:36
Export materials time= 172.149 ms
Collect objects time= 16.741 ms
Mesh creation time = 29.968 ms.
Total export Time = 557.993 ms
chrisbrearley
Licensed Customer
Licensed Customer
 
Posts: 23
Joined: Mon Jul 11, 2016 11:09 am

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby chrisbrearley » Wed Oct 24, 2018 9:24 am

chrisbrearley Wed Oct 24, 2018 9:24 am
One more thing. Regarding the denoiser and animations. It's not very good. Instead of sharp noise you just get soft blurry noise. It looks great on stills but for anything moving we have to render enough samples to render out the noise. It looks no better and I find that post denoisers such as Neat video do a better job.
chrisbrearley
Licensed Customer
Licensed Customer
 
Posts: 23
Joined: Mon Jul 11, 2016 11:09 am

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby tomsvfx » Wed Oct 24, 2018 9:33 am

tomsvfx Wed Oct 24, 2018 9:33 am
Hey chrisbrearley, did you tried enabling Static Noise?
For me it helps very much for interior scene rendering.
But samples cant be too low, I use around 300-1000 samles to get it stable.
User avatar
tomsvfx
Licensed Customer
Licensed Customer
 
Posts: 92
Joined: Mon Oct 30, 2017 10:19 pm
Location: Riga

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby lucioing » Wed Oct 24, 2018 9:45 am

lucioing Wed Oct 24, 2018 9:45 am
abstrax wrote:
lucioing wrote:Hi, Ahmet!
Just testing your great job, but I've some problems with materials...
An image is better that 1000 words.
Thank you again for your GREAT job!

L.
Schermata 2018-10-22 alle 22.05.19.png

That's a bug that got introduced with the CUDA error work around of RC6. We are looking to push out an update as soon as possible (either today or tomorrow). Apologies for the hassle.


Ok, thank you for your reply... ;)

L.
Mac Pro 5.1 -16 GB- 2 x Titan Black 6GB - Titan Z - MacOS Yosemite 10.10.4
User avatar
lucioing
Licensed Customer
Licensed Customer
 
Posts: 229
Joined: Wed Feb 19, 2014 4:01 pm
Location: Rome - Italy

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby chrisbrearley » Wed Oct 24, 2018 9:56 am

chrisbrearley Wed Oct 24, 2018 9:56 am
tomsvfx wrote:Hey chrisbrearley, did you tried enabling Static Noise?
For me it helps very much for interior scene rendering.
But samples cant be too low, I use around 300-1000 samles to get it stable.


Yes, we've used that in the passed but it looks like someone has just overlayed a static piece of noise over the whole footage. It works on some occasions but more often than not it can be very off putting.
chrisbrearley
Licensed Customer
Licensed Customer
 
Posts: 23
Joined: Mon Jul 11, 2016 11:09 am

Re: Version 4.0-RC6-R2 (Latest release candidate) 19.10.2018

Postby divasoft » Wed Oct 24, 2018 12:42 pm

divasoft Wed Oct 24, 2018 12:42 pm
chrisbrearley wrote:One more thing. Regarding the denoiser and animations. It's not very good. Instead of sharp noise you just get soft blurry noise. It looks great on stills but for anything moving we have to render enough samples to render out the noise. It looks no better and I find that post denoisers such as Neat video do a better job.

+1 I also had similar troubles with blurry noise in animation.
User avatar
divasoft
Licensed Customer
Licensed Customer
 
Posts: 348
Joined: Thu Jul 01, 2010 3:59 am
Location: Russia
PreviousNext

Return to Releases


Who is online

Users browsing this forum: No registered users and 9 guests

Sat Apr 27, 2024 1:01 am [ UTC ]