So me and Adam have been sending stuff off to the farm to see how its coping with frames and why certain issues are occuring.
After speaking with Simon from IT, its apparent there is no point using the farm for rendering off small jobs, rendering locally means it'll split the job better. Makes sense really. Anyway, the render farm copes better with jobs of 100 frames above, it seems to distribute the work better.
Adam is in the process of splitting the scene file in render layers and were testing the longest frame by rendering an occlusion pass of the initial pan shot. From this we can determine how much we need to split up the scene file into layers of props etc. Removing the shrurbs already cut 5 minutes off the render time but it might not be worth it.
After speaking to Alex Hulse he confirmed we'd have to get ultra smart about what needed rendering etc and sort out the layers now. Think we may need further discussions with him about what exact layers we want and will actually benefit the film...no point rendering what isn't necessary. Will speak to Dalli about potentially setting up a meeting for this, as it would be a good opportunity to ask some ques and sort ourselves out now for post.
No comments:
Post a Comment