1. Here is the poster for the final animation.
    I’m not happy at all with the final outcome of the project and had to struggle with some (really time-consuming) problems in the end…after copying all the files via FTP to my computer and composing them in After Effects I realized that some frames were missing…they got lost during the process of copying the files from the render server to another server from which I could get the files then. So I had to rerender as many frames as I could in tha last minute and use my machine as well for rendering. In the end I had like 30 minutes left to add the audio, so it’s pretty crappy as you can imagine. Due to this I didn’t have time to add the motion blur or the depth pass either, which would have improved the looks a lot. Well I simply should have checked each of the 24 folders with render data right away after copying.
    It would have also been A LOT nice if I could have added grass dynamics, so that it doesn’t stand there stiff and still but moves with the wind and especially when the character walks through it. But I turned off the dynamics for decreasing the render time and especially because they can cause serious problems when being rendered via the netrenderer (different processors calculate the dynamics differently which can cause the hair to ‘flicker’). Since I didn’t have time to render everything twice in order to calculate the render times probably and to check if the dynamics work I simply turned them off. I could have also worked with a lot more hair (although the landscape has a total of 200k), but my workflow was already incredibly slow…when I wanted to check how the current frame looks I had to wait for more than 10 minutes until my machine had rendered that stuff. Really annoying.
    However, I think the character animations and camera angles/movements are quite nice in some parts and I definitely learned a lot about that stuff during this projects wherefor I’m very grateful.
    I’ll definitely keep this character, improve or extend the rig (including the eyes for example) and will do some more animation stuff with it, probably simple but nice and focussing mainly on it’s movement, which I want to be more ‘pixar’ like, quick and easing.

     

  2. Funny render fail ONLY in this one frame. Wtf haha, no idea what happened there.

     

  3. Render management. Currently copying all the renders via FTP. It’s funny because the copying process actually takes way longer than the rendering itself. I definitely underestimated the amount of frames I would need for the second sequence as you can see. Well, it simply takes some time to tell a story..

     

  4. That’s the final look I guess. Almost no saturation and a little bit of diffusion and haze. Contrast is pretty high as well and I love it, since the characters body – for whom I’m currently thinking about a name by the way – gets way darker and the glowing eyes really pop out.
    My uni’s netrenderer is running with 38 machines at the moment and is really killing the render jobs. My own machine needs more than 10 minutes per frame, the netrender blasts 100 frames in less than half an hour…bam!

     

  5. Some issues..a couple of frames come out like this one (and can’t get rendered in After Effects). I figured out that it was a problem due to the copying process via FTP for the first scene and could fix it by reloading the images from the server. Hope it’s really only the copying-problem, since I wouldn’t have time to re-render all of these frames.

     

  6. While the other scene from the second sequence are still rendering I’m already starting with the post-production of the sky sequence. Additionally to the beauty pass I also rendered a depth pass and the motion vectors (for creating some motion blur).
    It’s really weird that the camera apparantely doesn’t see the pyroclusters and due to that renders a weird depth pass. However, for this sequence the motion vectors are a lot more important since there is quite some movement in the scene (especially that fast spinning propeller). Let’s see how it works out in After Effects.

     

  7. Since the (color of the) grass is the main issue for the not so nice looking images I just tested how it may look in black and white, and I really like that look somehow. Well let’s see how it looks with the moving footage.
    As you also can see, I removed the hair from the characters body. I did this for several reasons: first of all, I wanted to keep that cell shader look (actually even I don’t know why myself haha), and second and third I had some issues with it. It simply didn’t look good on close-ups, especially the roots. And somehow it didn’t render in front of pyroclusters which is really really weird and looked totally stupid. When importing the model to C4D it was still low-poly in order to play the animations in full speed in the viewport. So I added a hypernurbs for the subdivision in the renders. But when activating the nurbs, the hair didn’t render (although I ‘told’ the hair to use the hypernurbs as a reference). However, it was simply not working that well. And somehow I started to like the simpler look more and more. 

     

  8. Introduction of the glassy spheres. The first 400 frames of the whole sequence are currently rendering…and waaaay faster than expected. So I added Ambient Occlusion to make it at least lot a bit nicer. Still looks kinda crappy, so it will need a lot of tweaking, as expected. The depth pass should come out kinda alright I guess.

     

  9. Overview of the little grass world. The big pit in the back is the place where the glassy spheres will come out.
    23 minutes for this really crappy render…and 20 of them only for the hair calculation. Damn grass. I’m rendering without any effects, so no GI and not eben Ambient Occlusion. Will have to do a lot of tweaking in After Effects when all renders are done. And I hope that the depth pass renders properly because I couldn’t give the grass the detail I wanted in order to keep the render time as short as possible, so depth of field would increase the image quality quite a lot.

     

  10. Modelling some rocks for the environment. Gonna be Stonehenge-like haha.