This past few months I’ve been beavering away at Lola Post on 2 series of shows, creating VFX of a weathery, Earth-scale nature for Britains’ Most Extreme Weather, and shots of all scales for series 3 of How The Universe Works.
Ordinarily I’d put together blog posts before a show goes to air, but in the case of Britain’s Most Extreme Weather it slipped from my mind as soon as I rocked back onto How The Universe Works. Much of my weathery input was particle systems and strands, either using existing setups from previous shows or creating new ones as appropriate. A particular favourite of mine was a system showing the movement of air around cyclones and anticyclones; A strand system that rotates particles around many points, allowing them to move fluidly from one direction to another as if air, all wrapped around a lovely spherical Earth.
How The Universe Works is a series I’ve been on for many many months now. I first started on it in November I think. The first episode, all about our Sun, is to be shown on 10th July on Science in the USA.
For that show I took Lola’s existing Sun cutaway setup, introducing a more boiling lava-like feel through judicious use of animated fractals and grads.
Overall I’ve worked on 8 episodes with a handful of shots in each show. After all that dedication to spheres in space I am now supervising the VFX on one of the last shows for this series!
More geeky details and videos for both shows to come!
Not so long ago I worked at Lola Post, London, on another documentary hosted by Richard Hammond. Similar to the Journey to The Centre of The Planet and Bottom of The Ocean shows I worked on some time back, this entailed a heck of a lot of vfx.
The concept is that we see the constituent parts of scaled-down planets and the solar system being brought together in a large space over the Nevada desert. In order for Hammond to be able to present things at the necessary altitude, he is up at the top of a 2 mile high tower, which is obviously not real for various reasons. Nor is the desert much of the time. Or Hammond.
My input on the show was working on dust and sand particle systems. I was working on 2 sequences of shots. I will warn you now that some of this will get technical.
The first sequence shows a large swirling cloud of high-silica sand and iron. This includes a shot which was to become my baby for a month or two. It pulls out from Hammond at the top of the tower, back through the dust cloud swirling around him, then really far back so we see the entire 2km wide cloud in the context of the landscape around it. The whole shot is 30 seconds long.
The second sequence of shots shows the formation of Jupiter out of a large swirling disc of matter. Jupiter itself attracts dust inwards, which swirls as it approaches.
A few challenges presented themselves quite early on. One was creating particle systems in Softimage’s ICE that behaved correctly, especially when it came to dust orbiting Jupiter as the whole system itself swirls around the protosun. The initial swirling round the protosun was solved using a handy ICE compound that Lola have kicking about on their server, but if you use that twice in an ICE tree it is only evaluated once as it sets the velocity using an execute node, effectively overriding the new velocity value for each particle, rather than passing that out so it can be added to the previous velocity.
The solution to this was to break apart the compound. Integrating new nodes, including some out of a Move Towards Goal node, meant that I was able to make a new compound that I could proudly label Swirl Towards Goal. It sets the goal, then outputs a velocity which can be added to the velocity from the previous swirling compound higher up the tree. It even has sliders for distance falloff, swirl speed, and weight.
The most challenging aspect of this project was actually rendering. The swirling dust in each of my shots is made up of about 4 different clouds of particles. One alone has 60 million particles in it.
Enter Exocortex Fury, the fabled point renderer that was to save our bacon. Aside from one fluffy cloud pass per shot, rendered as a simple Mental Ray job on a separate lower detail cache, each cloud pass was rendered with Fury. Unlike traditional particle renderers that use CPU to render, Fury is a point renderer which can take advantage of the raw power of graphics cards. The upside is a far faster render compared to traditional methods, and done correctly it is beautiful. To speed things up further, particles which were offscreen were deleted so Fury wouldn’t consider them at all. Downsides are that it can flicker or buzz if you get the particle replication settings wrong and it has no verbose output to tell you quite how far it is through rendering. Between us dust monkeys many hours were spent waiting for Fury to do something or crash.
Adding to the complications was the scale of the main scene itself. The tower is rendered in Arnold, a renderer that works best when using one Softimage unit per metre. Unfortunately the huge scene scale caused problems elsewhere. In a couple of shots the camera is so high off the ground that mathematical rounding errors were causing the translation to wobble. Also, as particles, especially Fury-rendered ones, prefer to be in a small scene to a gigantic scene for similar mathematical reasons, they weren’t rendering correctly, if at all. The particles were in their own scenes for loading speed and memory overhead purposes, but in order to fix these issues, the whole system was 1/5 of the main scene scale and offset in such a way that it was closer to the scene origin yet would composite on top of the tower renders perfectly.
Update! The CCTV-9 channel branding, including this ident, recently won a Gold for Best Channel Branding at the PromaxBDA awards in Singapore!
I was called back in to work at Lola in London for this Chinese TV channel ident for CCTV-9 Documentary. Only 2 of us worked on this shot: myself and Tim Zaccheo, head of 3D at Lola.
The ident sees a waterfall coming down the side of a cubic mountain. The camera pulls back down a valley with scenery akin to the Guilin area of China, then out into space to reveal that the Earth is indeed cubic. CCTV have a cubic theme, so this makes sense in context. Thanks to the real-world scale of Terragen and the existing workflow at Lola, Tim was able to come up with a camera move that once imported into Terragen matched perfectly with the Softimage scene. The Earth’s textures and even the clouds lined up perfectly in both sections allowing a seamless blend.
My part in this was embellishing the initially blocked out Terragen scene with the necessary details to make it look like the Guilin mountains. A challenge there was that Terragen is great for pointy Alpine style mountains dusted with snow. That is easy out of the box. Guilin mountains are almost bell jar in shape, carpeted in trees with rocky cliffs here and there. The valleys between have been eroded away by rivers, leaving behind relatively flat farming land.
The solution to this was a variety of painted map shaders. Although this allows flexibility and great detail when it comes to controlling displacements, they’re best replaced with actual textures if possible, else the rendering gets very intense. In this case it wasn’t really an option. The painted maps were used to define areas of low and high ground, plus to define where the river goes and to control where the farmland appeared.
As there is quite so much foliage in the area there needed to be a solution that didn’t rely entirely on populations of tree objects. In come the procedural trees. This is essentially a series of overlaid displacement textures that build up to create the cauliflower head look to the trees. Similarly, the farming land was achieved using a tiled texture of fields and a few trees distrbuted along hedgerows. It’s very easy in a procedural program like Terragen to forget that a bitmap texturing approach is still a valid method and often faster.
Something that took a while to figure out was the cubic mountain at the start. The cube was initially displaced using a square displacement map with a falloff around the edges, plus an area eroded away at the front. The stoney displacements were then layered on to this, taking the new normals into account, rather than throwing everything up vertically as is the default. It was then eroded in various directions using extra displacement maps.
The waterfall was Tim’s baby, done entirely in Softimage’s ICE using fairly straight forward techniques, but along with some coloured mattes it all came together nicely in the comp.
There’s no sound on the video above by the way. I’ll replace it with one with audio once I’ve located it.
For the past few months I’ve been working at Lola Post, London, on Mankind, soon to be shown on the History channel both here in the UK and the USA.
I worked on quite a few sequences, 30 shots in total. Most of these involved creating projectiles of differing sorts, predominantly arrows; People firing arrows, being shot by arrows, and avoiding arrows while simultaneously cheating the whole archer deal by using guns. All arrows in the sequence above are CG.
As with many documentaries, many shots on Mankind were illustrative map shots, presented as full scale Earth scenes and as full CG shots they were subject to much change. Luckily, the flexibility of CGI makes it easy to work outside the boundaries of reality and to change one’s mind.
A few of the shots I worked on involved creating digital sets. Firstly I created an aqueduct for a sequence of shots with Caesar in. This was a case of tracking shots, matching on set details and extending upwards.
The trickiest shot was a bullet time shot, first in the sequence above, showing an Irish navvy unwittingly getting a little too close to a tunnel blast within the Appelacians. The original footage was green screen with the actor effectively sitting on a green pole with the camera moving around him. This introduced a wobble but was significantly easier and cheaper than a timeslice rig. As the footage was ramped up and down as well as being slow mo, getting rid of the wobble was high priority and after many tests it was eventually solved with simple yet nifty 3d camera trickery.
To smooth out the wobble, I followed a suggestion of Lola’s MD, Grahame. Having tracked the raw footage in PFTrack I projected that original footage through the camera in Softimage onto a card, positioned where the actor should be. That way the actor stayed in the same place in 3d space whilst I moved my new 3d camera around him.
The entire environment in that shot is a 3d set I threw together out of multiple particle instances of the same handful of rock models.
Most of the other shots were relatively straight forward, the exception being another bullet time shot, this one actually being one of the first bullets ever fired! The footage for the start of the shot was different to that of the end, so although the start had lots of people thrusting spears and poles in a smokey landscape, the end was completely clear of people and smoke, plus the target dummy was way too near. To solve this I made a new 3d gun, texturing it with various camera projected textures from the original footage, then made a new background out of a humongous psd stitched together out of footage and photos. In the end none of the original footage is being used as footage, more as texturing inspiration! It’s a really long shot so I split it in the sequence above.
All the work I did on this show bar the Earth-scale shots was rendered using Arnold. This has an advantage over Mental Ray of being a fast method of getting realistic lighting complete with indirect light bouncing. The quality is superb. To me, Mental Ray is much more flexible, but Arnold trumps it for speed between initial light placement and realistic render. I’m very glad I’ve forced myself to learn it.
A few of the aforementioned Earth-scale map shots are shown below.
On the Recent Work page, and indeed right here, is a video of a few of the shots I worked on for Orbit: Earth’s Extraordinary journey.
The first and last shots featured are both from the same ‘journey’ setup that was used for many other shots too. The setup featured many different elements on their own passes, each passed into its own part of a Nuke composition. As the project progressed, both the 3d scene and the Nuke script needed subtle reworking.
The second shot is a pair of emFluid particle systems, whereas the third is a simple enough ICE simulation in Softimage. The particles in those two shots were rendered with beta versions of Exorcortex’s Fury rendering system which loads the particles onto the graphics card, rendering them in OpenGL. Without Fury the second shot would have been particularly time-consuming to render. It contains millions of particles and took many many hours to cache out.