OctaneRender 2019.1 LightWave Production build 2019.1.4.0

[Carlos3D]

One of the things I miss so much and would be nice to have is something that sticks the walls vertically.

Something like the clip map but in vertical line.

Maybe it could be in the Oject properties.

So we could get a null to pierce the walls.

With shape modes such as Cubic and Spheric.

I think it would be very valuable.

Thankful.

[WoutTgh]

Hi,

I'm getting this error when F9 rendering in LW 2015.3:


Tried to access pin via an invalid ID P_AI_LIGHT_STRENGTH (385)
Tried to access null node pin
Can't write float to static pin 'unknown' (P_UNKNOWN, type PT_UNKNOWN) of node '' (NT_UNKNOWN, 00007FFFDFD71F90)
Tried to access pin via an invalid ID P_RENDER_PASS_LAYER_COLOR_SHADOWS (164)
Tried to access null node pin
Can't write bool to static pin 'unknown' (P_UNKNOWN, type PT_UNKNOWN) of node '' (NT_UNKNOWN, 00007FFFDFD71F90)
CUDA error 702 on device 0: the launch timed out and was terminated
-> failed to create link
device 0: failed to create OSL compile context:

device 0: failed to compile module 0
Tried to access pin via an invalid ID P_RENDER_PASS_LAYER_COLOR_SHADOWS (164)
Tried to access null node pin
Can't write bool to static pin 'unknown' (P_UNKNOWN, type PT_UNKNOWN) of node '' (NT_UNKNOWN, 00007FFFDFD71F90)
CUDA error 702 on device 0: the launch timed out and was terminated
-> failed to create link
device 0: failed to create OSL compile context:

device 0: failed to compile module 0

This also happens with 2019.1.1 & 2019.1.2 btw.

[juanjgon]

What NVIDIA drivers are you using? If it happens with any scene, this problem sounds like a drivers issue.

Thanks,
-Juanjo

[WoutTgh]

Hi Juanjo,

I'm using the Studio Drivers 431.86, I'm downloading the 441.12 drivers now to see if that makes a difference.

[WoutTgh]

Installed new drivers, It freezes on render with F9, in the IPR it doesn't show anything and then jumps to render complete and shows the render. Before that, nothing.

[juanjgon]

Do you know if Octane 2019.1.4 Standalone works fine on this system? And the older Octane versions, like Octane 2018.x?

Thanks,
-Juanjo

[WoutTgh]

I don't know, I never use the Stand Alone versions ..

[WoutTgh]

So I downloaded the Standalone 2019.1.4, I'm still looking for the fastest way to render particles with Octane. The instancer method in Lightwave works, with limitations( no motion blur) but gets very slow for loading the scene.
A 'render points as spheres' option in Lightwave would be very nice !

In the standalone, I can't seem to get the particles to change their size, no matter what setting I enter in the particle radius under the alembic import.

Also how would you go about creating colour variation on imported alembic files with particles ? It seems to apply the material to all of the particles/spheres..

I'm also getting this error :

OctaneRender Studio 2019.1.4 (6000400)

CUDA error 702 on device 1: the launch timed out and was terminated
-> kernel execution failed(kernel3)
device 1: picking failed
CUDA error 702 on device 1: the launch timed out and was terminated
-> failed to load symbol data to the device(kernel data)
device 1: direct light kernel failed
CUDA error 702 on device 0: the launch timed out and was terminated
-> failed to download symbol(stats data)
device 0: direct light kernel failed
CUDA error 702 on device 1: the launch timed out and was terminated
CUDA error 702 on device 0: the launch timed out and was terminated
-> failed to load symbol data to the device(deep_data)
-> failed to load symbol data to the device(deep_data)
device 0: failed to upload the deep params
device 1: failed to upload the deep params

[juanjgon]

Well, if the Standalone doesn't work you have a system or drivers problem for sure. Here I'm working with the 436.15 drivers without problems so far. Can you please install this drivers version to check if it helps?

Thanks,
-Juanjo

[WoutTgh]

Okay , both my gpu's are now on 441.42 and I've been playing around with the Octane demo for Houdini and it's so smooth and fast that I've decided to switch my subscription from Lightwave to Houdini.
If anyone is interested in the reasons:

Lightwave' SDK doesn't send motion data for instances on deformed geometry to 3rd party renders, that means no motion blur on particles.
The workaround is using the motion blur passes, with full scene reload. Scene loading with a lot of instances becomes very slow and remember that happens every pass so 6 times minimum for fairly decent looking MB.

Lightwave has very poor Alembic support, so no changing point count, no access to attributes etc...


Now in Houdini the workflow is super smooth with Alembic and Octane as it reads Alembic particle sequences super speedy. Then Octane loads the scene almost instantly with the "render as sphere particles" and of course with Motion Blur without the need for passes.

So it's a huuuge difference for me as I work with particles a lot

Thanks Juanjo