Dark Side of the Sun

Thursday, December 22, 2011

Prometheus 1st trailer

The 1st official trailer for Prometheus, the movie I am working on, has been released. Watch it big and with sound!:



I am enjoying tremendously working on this film. I had never been this excited on any of the projects I have worked on previously. And this is not to belittle those projects, but rather to show how strongly I feel for Prometheus. I had the great chance to work in a Ridley Scott movie before, on Robin Hood, so I feel incredibly fortunate and happy to have that chance a second time, and this time in such a special film.

Can't wait to see this one in the cinema!

Sunday, July 31, 2011

fxpodcast: Greg Butler on Harry Potter 7 Part 2

In this fxpodcast, Greg Butler, MPC VFX Supervisor, discusses the work we did on the last Harry Potter movie:

More specifically, in minutes 5:30 to 14:00, he talks about the FX work done by my team: creating the river of fire and fire creatures in the Room of Requirements sequence.

Sunday, July 24, 2011

New FX Reel 2010

My new FX Reel, updated with work done during 2010 for:

- Ridley Scott's "Robin Hood"
- "Chronicles of Narnia 3: Voyage of the Dawntreader"


FX Reel 2010 - Mayec Rancel from Mayec Rancel on Vimeo.

Thursday, July 14, 2011

John Carter trailer

What I am working on these days:

Monday, May 2, 2011

Python: Hamiltonian Cycles in a graph

This is a follow-up to this other post , where I presented a python program to find Eulerian paths in a graph (more specifically, the iconic "Bridges of Königsberg" one, although the program could easily be adapted to any other arbitrary graph).

A few weeks ago, I made a variation on that, and wrote a program to explore all possible paths in a graph, to try to find any existing Hamiltonian cycles. The purpose of this was to solve a mathematical challenge by spanish newspaper "El País".

This is the graph presented for the challenge, and solved by this program:


In a graph (a series of vertices connected by edges), the basic difference between an Eulerian path and a Hamiltonian one, is that the Eulerian path is one that walks through every edge exactly once, whereas the Hamiltonian walks through every vertex exactly once. Also, the difference between a path and a cycle is simply that the cycle ends at the same place where it started, whereas the path can start and end in two different features of the graph. So, although the problems are different, the apparent similarity is what made the adaptation of the code fairly easy. The main difference was considering the additional condition of seeing the path end where it started, for a positive result.

The solution, by the way is... that there is no solution. There are no possible Hamiltonian cycles in that graph, although there are several Hamiltonian paths (not starting/ending in the same vertex).

If anyone is interested, you can find the python code here:

hamiltonianCycle.py

This could easily be adapted for any other graph, by modifying all the "Zones" and "Bridges" declarations at the beginning of the code.

Can you think of any interesting other uses/variations for this program?

Sunday, May 1, 2011

new trailer: Harry Potter and the Deathly Hallows pt.2

This new trailer came out last Wednesday, showing some more of the work we have been doing for this last episode in the series. Congratulations to all the friends and colleagues, in all facilities, who are now giving their sweat and blood to finish this one off with full on awesomeness! I'm dying to see this one on the big screen...

Thursday, February 3, 2011

making of: The Right Hand of Doom

I made this step-by-step making of my digital painting The Right Hand of Doom.

(click image to enlarge)

Sunday, January 16, 2011

Python: brute-force solution to Bridges of Königsberg

Yesterday I read about the mathematical problem of the Seven Bridges of Königsberg. Since Leonard Euler already solved the problem a while ago (or rather, assured us that there is no possible solution to it), through elegant and brilliant thinking and analysis (and by the same, initiating the discipline of Graph Theory, what a man!), I set myself a much more insignificant challenge: My obsessive pastime of the afternoon has been writing a python program which would find all possible paths of the Bridges of Königsberg problem, using only one recursive function. I know, I am such a geek! Get over it. I have.

My goals with this were simply:
  • To have an excuse to shake the dust off my programming/scripting (haven't written any code in a while).
  • To get some practice with recursivity, as this seemed like a perfect case to apply it.
Accidentally, along the way, I have learned some bits about things like Game Theory, Graph Theory and Decision Trees.

An interesting information that came out of this is that my brute-force approach says there are 372 different legal paths. Much less than I expected, which explains why it solves so fast. I would be interested in hearing if anyone reading this knows a way to calculate this number.

I haven't seen any examples of  a similar code online (probably because it's pretty useless), so I thought I'd share it, in case anyone is interested:

konigsbergBridges.py

And here is a diagram of the naming convention that my program uses for land zones & bridges:


It would be very easy to adapt this code to any other similar problem, with an arbitrary number of land zones and bridges connecting them, by changing the initial variable declarations. But with a high number of bridges, I am sure the program would hit some technical limits like the recursion limit, or would need to run for a long time.