Tweet
Hi,
I have been evaluating Vray demo for the past few days, and reading through a lot of the posts in this forum to get a better idea how VRay works in production environment. So far, so good. Visual quality of images is top notch and the interface is fairly easy to use.
From what I can tell, rendering GI for still images is fairly straight forward. My only concern is animation. Please correct me if any of the following assumptions are incorrect.
1) Rendering animations where the only movable object is camera, can be done with the use of irradiance maps. General workflow is to compute an irradiance map for every 10th frame of camera movement, with "incremental add to current map" option turned on. This generates a big map that contains "samples" necessary for correctly rendering GI solution from different camera angles.
2) Rendering animations where some or all objects are moving, requires one of the following:
a) creating irradiance map for every single frame, using "Single Frame" option.
b) using "Bucket mode" with distributed rendering.
Now, my only concern are visual differences between frames when new irradiance map is computed for each frame. This seems to be the Achilles Heel of all GI engines, and I'm curious how VRay handles the issue. I am planning to render out a simple sequence soon to test this, but I would love to hear other's experiences in real production with complex full blown animations. Is GI viable for character animation? I imagine if the settings are high enough that the GI solution would be similar for sequence of frames, but at very high setting, irradiance map computation seems to slow down considerbably which may not be production friendly if you need to crank out 5000 frames.
Does bucket rendering help this issue? I don't know much about this feature, since it seems to be tied to Distributed Rendering, which is disabled in Demo.
Eldar
I have been evaluating Vray demo for the past few days, and reading through a lot of the posts in this forum to get a better idea how VRay works in production environment. So far, so good. Visual quality of images is top notch and the interface is fairly easy to use.
From what I can tell, rendering GI for still images is fairly straight forward. My only concern is animation. Please correct me if any of the following assumptions are incorrect.
1) Rendering animations where the only movable object is camera, can be done with the use of irradiance maps. General workflow is to compute an irradiance map for every 10th frame of camera movement, with "incremental add to current map" option turned on. This generates a big map that contains "samples" necessary for correctly rendering GI solution from different camera angles.
2) Rendering animations where some or all objects are moving, requires one of the following:
a) creating irradiance map for every single frame, using "Single Frame" option.
b) using "Bucket mode" with distributed rendering.
Now, my only concern are visual differences between frames when new irradiance map is computed for each frame. This seems to be the Achilles Heel of all GI engines, and I'm curious how VRay handles the issue. I am planning to render out a simple sequence soon to test this, but I would love to hear other's experiences in real production with complex full blown animations. Is GI viable for character animation? I imagine if the settings are high enough that the GI solution would be similar for sequence of frames, but at very high setting, irradiance map computation seems to slow down considerbably which may not be production friendly if you need to crank out 5000 frames.
Does bucket rendering help this issue? I don't know much about this feature, since it seems to be tied to Distributed Rendering, which is disabled in Demo.
Eldar
Comment