Cineversity

Hi,
I thought I’d throw this out to see if someone can notice something in this scene file.
I’m using CV-VR cam which is working great, and I render my text separate from the scene so I can make changes without full long renders.  In this scene, if you render take 1 which is the first piece of text to go with the scene, it’s taking quite a long time per frame (on my system about 14 secs).  I notice that the buckets are rendering in areas where nothing exists.  I would think the renderer would only have to evaluate areas where the text is?
Does this have something to to with the take system, CV VR cam render, or a strange setting.  IE, if you look at the render, the area where text exists is only about 2 secs, but all the black around it is being evaluated for another 12 secs.  Thought I’d ask.

PS. - I could not post the scene file here for some reason, I kept getting an error message, so I posted the file to my Microsoft one drive account here:

https://1drv.ms/u/s!AuVYWNZS3S-EjjL-wlJ7wr13x28J

Mike

Hi Mike,

I have send you a up-load link. (Cineversity Private message)

All the best

Hi Mike,

To see what the rendering takes without anything in the scene, place each of the two groups completely in a Layer in the Layer Browser. (Create two layers and each time select children while the top parent is selected and drag it on the Layer). Then switch anything off in the layer, visibility, render object manager, etc. Everything.
Render.
No switch the Anti Aliasing to None.
Render
Compare results.
Here is the major part, as each pixel, empty space or not will calculated with the given AA and checked for the threshold, etc.

My suggestion, set the AA for the text render pass to 1x1 globally (render settings) and a very high threshold. Make sure the “Use Object Properties” is checked on in the Render Settings> Anti Aliasing.
To get a proper AA for the Text and other objects in question, set up a Compositing Tag and overwrite the AA settings there. Tab> Tag> Force AA.

Note that some objects in the “Scene 6 Base” were enabled to be rendered.
The Soft Shadow for an omni light calculates six maps each frame for each light. If there is no need to recalculate this, (e.g., no animation) then the Render Settings> Options allows for a cache of these. A spotlight calculates only one map, BTW. These are not real shadows and perhaps one should consider to use other options. Just a thought.

Since I had no images, my render times might have varied heavily.

All the best