Its been a while since I have done any animation in vray. I mainly do stills, so I was wondering if I could run through what I think needs to be done and people could advise if this is correct. I have read the spot3d vray tutorials on walkthrough animations.
The scene is an interior scene with no moving objects. Only the camera moves. Final frame size is 1920x1080.
My process:
Resolution to 1920x1080:
1. Set both first and secondary bounces to lightcache.
2. Lightcache to flythrough mode.
3. Render lightcache and save it.
Resolution to 960 x 540 (half res):
4. Set first bounce to Irradiance map
5. Add to incremental mode and save irr map.
6. Max set to every nth frame.
7. Set lightcache to load from file saved out.
8. Render irrmap
9. When finished, render final animation using saved irr map and lightcache file.
If I have a lot of glossies am I right in thinking I should have “use lightcache for glossy rays” on, and then when rendering the final frames to turn off the secondary bounce (lightcache) ?
I am trying to render out the irrmap using DR…is this possible as I can do the first frame, then I seem to have problems with 2 of my render nodes not joining in the rest of the calculation ??
Your workflow is basically ok but there are a couple of problems.
The main one is that when you render the final frames you should set the lightcache to render from file and use the one you saved in point 3.
The other is that it would be better to combine points 1 to 8 into one process. To calculate the IR map at half res just lower the min and max values under the irradiance map settings.
The only thing I’m not sure about is whether your can use DR for the IR map calculation. I always send it to one node using backburner.
Cheers Dan. I read on another thread that if you have the lightcache for glossies button checked when calculating the irr map, the lightcache merges with the irrmap and there is no need to have lightcache (as secondary bounce) on when rendering the final image…is this incorrect ?
If this is wrong, what is the lightcache for glossies button for, and should I have it on or off ?
Hi Dan. Wondering if you or someone can help me out. I am doing another type of animation for this project. It has moving objects in it. Am I right in thinking that I can do a lightcache pre-calc and then use brute force for the first bounce and the saved lightcache for the secondary ? Would this give good results…there is way too much movement to think about using the spherical gizmo and the atmospheric evaluation..
A saved lightcache would never work on an animated scene - seeing as the samples are being saved as light information in 3d space you’d get a lot of inaccuracies which may drive render times up or cause visual artifacts. Set it to single frame and use brute force for primaries. If it’s something like a character walking through an environment then maybe look at a split render with the environment in one pass with cached light and a brute force / lc combo for the moving character?
Ok, so what would I use to render out an animation on the fly. The scene is fairly complex and so compositing parts together is not really an option. Im guessing I would use brute force for primary, but what for secondary ? Brute force again ? or would I just set the lightcache to single frame, fairly high and calculate on the fly ?
Yep - If it’s an exterior then brute force is going to be fairly quick and the nice thing is once you get the grain down to an acceptable level it won’t give you any surprises in terms of GI errors. Light cache I’ve had the odd problem with a little bit of flickering but as you say you only need to turn the samples up high enough in single frame mode to get it working.
Interior wise light cache will probably fill out the space quicker than brute force and should give you faster renders.
