Variation of brightness during animation
Forum rules
Please upload a testscene that allows developers to reproduce the problem, and attach some images.
Please upload a testscene that allows developers to reproduce the problem, and attach some images.
-
- Posts: 234
- Joined: Thu Jan 18, 2018 6:02 pm
Variation of brightness during animation
Hi,
I've rendered this object from different perspectives.
The results looks like the light becomes more bright in some frames randomly.
The estimated result ist a constant fall off for the side perspectives, but not such flash light.
This is the belonging blend file: Thanks for your help.
BR
I've rendered this object from different perspectives.
The results looks like the light becomes more bright in some frames randomly.
The estimated result ist a constant fall off for the side perspectives, but not such flash light.
This is the belonging blend file: Thanks for your help.
BR
OS - Windows 7 X64
CPU - Intel CORE i7
GPU1 - Variants of notebook card from nVidia
GPU2 - Variants of notebook onboard card from Intel
Lux - Latest possible relaease
CPU - Intel CORE i7
GPU1 - Variants of notebook card from nVidia
GPU2 - Variants of notebook onboard card from Intel
Lux - Latest possible relaease
Re: Variation of brightness during animation
Can you upload some of the frames where it occurs (+ one before and one after, for comparison) and does it always happen at the same frames (if yes, which ones exactly)? Or is it random each time you render?
Re: Variation of brightness during animation
Try to disable camera volume.lighting_freak wrote: ↑Thu Jul 04, 2019 6:42 am The results looks like the light becomes more bright in some frames randomly.
CPU Bidir + Metropolis | Core i5-4570
Re: Variation of brightness during animation
It's precision issue.lighting_freak wrote: ↑Thu Jul 04, 2019 6:42 am Hi,
I've rendered this object from different perspectives.
The results looks like the light becomes more bright in some frames randomly.
The estimated result ist a constant fall off for the side perspectives, but not such flash light.
Animation.gif
This is the belonging blend file:
Brightness_Variation_During_Animation.blend
Thanks for your help.
BR
Preview
Original gif You have Camera Clipping End set too high (1km) causing precision issues. Simply lower it (1m is enough). It'll also render faster
& simply keep in mind - single float precision - safe range is ~ 7 digits (ie. 1 - 1 000 000)
specifically:
"All integers with 6 or fewer significant decimal digits, and any number that can be written as 2n such that n is a whole number from -126 to 127, can be converted into an IEEE 754 floating-point value without loss of precision." - Single-precision floating-point format (wiki)
Re: Variation of brightness during animation
I don't think this is the correct solution. because his camera is 2 meters from the object, so 1m clipping is obviously too short (although I wonder why it still renders then...)
Re: Variation of brightness during animation
I am beginning to think the whole problem you have with this scene might come from the Volumetrics of the AirTo Silicone object. It is only a plane in your case, which could lead to some issues, even though it seems otheriwse correct with the emmiter having its exterior defined as silicone. However, I can't turn it into a consistent result either by changing things...
Re: Variation of brightness during animation
Well, visually it works.
I don't know better or why. Didn't took time to optimize & set the scene ideally to 1 BU (it's 0.001 now).
Knowing Blender, this ain't good - units on display are just there as psychological, mental helpers.
Which is why I also have a long lasting doubt about dimensions transformation between Blender & Lux, causing quite a lot of precision issues.
I have no doubts about Lux, but transformational consistency is not something Blender is known for. IIRC there is/was a bug when scaling an object to extremes corrupted the whole scene.
PS
All of the mentioned "solutions" have been tested.
I don't know better or why. Didn't took time to optimize & set the scene ideally to 1 BU (it's 0.001 now).
Knowing Blender, this ain't good - units on display are just there as psychological, mental helpers.
Which is why I also have a long lasting doubt about dimensions transformation between Blender & Lux, causing quite a lot of precision issues.
I have no doubts about Lux, but transformational consistency is not something Blender is known for. IIRC there is/was a bug when scaling an object to extremes corrupted the whole scene.
PS
All of the mentioned "solutions" have been tested.
-
- Posts: 234
- Joined: Thu Jan 18, 2018 6:02 pm
Re: Variation of brightness during animation
Hello all,
Thanks to all your reply.
Thanks a lot, I'll send my results tomorrow in the morning (CET)
BR
Thanks to all your reply.
I'm about to rerun the complete animation with a different random seed. It will take until tomorrow to see the result.
What does happen internally while deactivating the auto detection and simply don't enter anything in that appearing menu? Do all my rays start in silicone then, the light- and eye path?
I thought the camera clipping is for viewport only?
Well, using exact volume descriptions is one of USPs of luxcorerender, isn't it? I know that this enlarges calculation time, but my approach is to create most realistic results which means as detailed as possible even in light source.
Since especially the volume properties (absorption and diffusion) are dependent on the real dimensions this transformation must be done.
Thanks a lot, I'll send my results tomorrow in the morning (CET)
BR
OS - Windows 7 X64
CPU - Intel CORE i7
GPU1 - Variants of notebook card from nVidia
GPU2 - Variants of notebook onboard card from Intel
Lux - Latest possible relaease
CPU - Intel CORE i7
GPU1 - Variants of notebook card from nVidia
GPU2 - Variants of notebook onboard card from Intel
Lux - Latest possible relaease
Re: Variation of brightness during animation
Camera clipping settings should have no impact on precision in LuxCore.
These problems only arise in rasterizers that use a depth buffer with limited precision, with range from near to far clipping plane, for example the OpenGL renderer drawing the Blender viewport. But LuxCore traces rays in world space, not camera space.
Are you sure the effect you observed was caused solely by the clipping settings and not something else?
I am currently fixing this during the 2.8 port.kintuX wrote: ↑Thu Jul 04, 2019 3:20 pm I don't know better or why. Didn't took time to optimize & set the scene ideally to 1 BU (it's 0.001 now).
Knowing Blender, this ain't good - units on display are just there as psychological, mental helpers.
Which is why I also have a long lasting doubt about dimensions transformation between Blender & Lux, causing quite a lot of precision issues.
I have no doubts about Lux, but transformational consistency is not something Blender is known for. IIRC there is/was a bug when scaling an object to extremes corrupted the whole scene.
For details of my plan see https://github.com/LuxCoreRender/BlendLuxCore/issues/97
The rays will check if the material they hit has an exterior volume set and use that as their assumed starting volume.lighting_freak wrote: ↑Thu Jul 04, 2019 3:53 pm What does happen internally while deactivating the auto detection and simply don't enter anything in that appearing menu? Do all my rays start in silicone then, the light- and eye path?
Note: auto-detection of the camera volume is only robust if your volume setup is immaculate (which is sometimes hard to achieve).
It might be better to assign the camera volume manually in an animation.
If the checkbox above it is enabled, it is used by Lux. Otherwise, it is used by Blender's OpenGL viewport only, but not by Lux.
-
- Posts: 234
- Joined: Thu Jan 18, 2018 6:02 pm
Re: Variation of brightness during animation
Hi,
I mean in my setup are rays that see through the silicone resin and will be set to air or silicone (the exterior is air in that case, but the beam has been emitted in silicone) Further more if an emitted ray hits the LED body first it will be in silicone since this surface is being defined as having exterior volume silicone. In that case I were would have two classes of rays from the the same emitter.
BR
What will happen to rays that emit from the source while using BiDir?B.Y.O.B. wrote: ↑Thu Jul 04, 2019 4:08 pm lighting_freak wrote: ↑ What does happen internally while deactivating the auto detection and simply don't enter anything in that appearing menu? Do all my rays start in silicone then, the light- and eye path?
The rays will check if the material they hit has an exterior volume set and use that as their assumed starting volume.
I mean in my setup are rays that see through the silicone resin and will be set to air or silicone (the exterior is air in that case, but the beam has been emitted in silicone) Further more if an emitted ray hits the LED body first it will be in silicone since this surface is being defined as having exterior volume silicone. In that case I were would have two classes of rays from the the same emitter.
BR
OS - Windows 7 X64
CPU - Intel CORE i7
GPU1 - Variants of notebook card from nVidia
GPU2 - Variants of notebook onboard card from Intel
Lux - Latest possible relaease
CPU - Intel CORE i7
GPU1 - Variants of notebook card from nVidia
GPU2 - Variants of notebook onboard card from Intel
Lux - Latest possible relaease