SItoA/Arnold bug with motion blur and transmissive lights
Sven Constable
Hi,
anybody still here? :)
Recently I was working on an old test scene again to compare arnold and redshift (and even mr for a good amount of fun). Well, in that particular scene arnold crashes when using motion blur. I nailed it down to settings in the light shader (arnold quad light).
The values in the light shader for camera visibilty as well as transmission visibility are both zero by default. Maybe for a reason. Then it renders but changing them to any other value will crash arnold when using motion blur. Unfortunatly you would need transmission visibility for correct reflections in the glass.
Not sure if its an arnold or SItoA bug. I’m using SItoA 7.2.2.0.
I worked around it using grids for the transmissive highlights and switched off anything but the specular transmission for the grids. A bit awkward... that’s what I would expect from mental ray, not arnold. :(
Does anyone know if it’s a known bug, limitation or maybe something else?
You can see the issue here (the glass model has a 360 rotation over 50 frames)
www.imagefront.de/tmp/testscene_armadillo_comparison.jpg
Sven
Jens Lindgren
--
You received this message because you are subscribed to the Google Groups "Softimage Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to xsi_list+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/xsi_list/000801d9cbbb%242138be30%2463aa3a90%24%40imagefront.de.
Stephen Blair
--
Sven Constable
Hey Jens,
I needed to clean up the scene a bit:
www.imagefront.de/tmp/2307_imagefront_benchmark_arnold.zip
It’s the XSI project folder with relative file pathes so it should work straight away.
The scene has a readme annotation which explains it a bit.
Basically just draw a region (pass settings are used), arnold will render the hilights. Then turn on MB...
It also contains the photoshop file I used for direct comparison.
BTW the bug is with CPU and GPU rendering in Arnold.
Sven
To view this discussion on the web visit https://groups.google.com/d/msgid/xsi_list/CAJabFfO4dU_WoT%2BpjpxXVbr7sOe0-7Eg9Omg6smFdsK_43CybA%40mail.gmail.com.
Sven Constable
Hi Stephen,
thanks for chiming in. I will try an ass export.
Sven
From: xsi_...@googlegroups.com <xsi_...@googlegroups.com> On Behalf Of Stephen Blair
Sent: Thursday, August 10, 2023 10:54 PM
To: xsi_...@googlegroups.com
Subject: Re: SItoA/Arnold bug with motion blur and transmissive lights
Hi Sven
To view this discussion on the web visit https://groups.google.com/d/msgid/xsi_list/CANi5zx%3DHrxuB%3D6C5rmQ3PP46qBJWG7ji8HY5EYy4P1-yfRC%2BVw%40mail.gmail.com.
Sven Constable
Arnold has some interesting terminology...not sure if its foul language but I kicked an ass file. However, it does not crash but it doesn’t render the highlights either.
Similar happened from within XSI and the render region...sometimes it will render when scrubbing in the timeline back to the original frame. Even when it renders, it doesn’t render the hilights.
With kick, it renders straight away but no hilights are rendered when MB is turned on.
Both XSI and kick give these error:
' ERROR : [arnold] [polymesh] : different number of motion keys in vertex/normal arrays: #normal keys: 1, #vertex keys: 2
' ERROR : [arnold] [polymesh] : different number of motion keys in vertex/normal arrays: #normal keys: 1, #vertex keys: 2
' ERROR : [arnold] [polymesh] : different number of motion keys in vertex/normal arrays: #normal keys: 1, #vertex keys: 2
What seemes like an issue with scene geometry isn’t actually related to a polymesh in the scene. It happens even if all geometry is deleted. Three errors with three lights in the scene....
I suspect it to be the arnold area lights itself. Maybe when arnold is building a geometry for the visible area lights to render them.
I might dig into that further and post it on the more appropriate channels regarding arnold. But so far it seems like an arnold issue.
Sven
To view this discussion on the web visit https://groups.google.com/d/msgid/xsi_list/001801d9cbd6%243cb07f20%24b6117d60%24%40imagefront.de.
Isis Hathor
- -
Isis Hathor <marian....@gmail.com> hat am 10.09.2023 14:34 CEST geschrieben:
--
You received this message because you are subscribed to the Google Groups "Softimage Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to xsi_list+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/xsi_list/CAC%2BEufODYCZHcAOuay9epdWjyAsQ7jCiTYdOakjojLO9JEeAuw%40mail.gmail.com.
Walter Volbers
+49 (0) 69.48 000 55.0
Save a tree! Please don't print this e-mail unless you really need to.