Last Online: 2/14/08
johann deneux
Career Summary
Graduated from Ecole Superieure d'Ingenieur de Luminy in Marseilles, France in 2001 with a degree in software engineering.
Defended my PhD thesis about parameterized timed systems in Uppsala, Sweden in 2006.
Currently working in the field of formal methods to provide verification software for railway interlockings. The idea here is to write programs that analyze software controlling railway signals and find bugs that manual inspection may have missed.
Some of the achievements of which I'm the proudest include adding support for force-feedback into the Linux driver for I-Force devices (mostly racing wheels).
During my free-time, I'm acting as the lead (the only, actually) programmer in the Top10 project. The other half of the team is the artist modelling vehicles and tracks, using measurements made on the ground.
Defended my PhD thesis about parameterized timed systems in Uppsala, Sweden in 2006.
Currently working in the field of formal methods to provide verification software for railway interlockings. The idea here is to write programs that analyze software controlling railway signals and find bugs that manual inspection may have missed.
Some of the achievements of which I'm the proudest include adding support for force-feedback into the Linux driver for I-Force devices (mostly racing wheels).
During my free-time, I'm acting as the lead (the only, actually) programmer in the Top10 project. The other half of the team is the artist modelling vehicles and tracks, using measurements made on the ground.
An idea...
There is one thing in America's Army I love: When new maps are released, you get the excitement to find good spots before others do. For a little while, the smart guy with bad aiming skills (me) has the advantage on the FPS junkie capable of executing perfect sequences of running, shooting, flashing...
In FPS games played in closed-quarters such as America's Army, there are players who have been playing on maps for so long that they know all good spots and winning tactics in such detail that realism and fun take a hit. Consider for instance grenade spamming, where players do artillery shots using grenade launchers, throwing grenades at the other team on the other side of the map immediately after spawning.
The simple solution to that problem is random spawn points, but that does not solve the problem of having experienced players rush blindly towards strategic places, knowing enemies are too far away to pose a threat.
I would suggest taking a step further. The entire map should be randomized.
Now, having completely random maps is no fun. Level designers put a lot of effort into crafting maps carefully. See for instance the simple but effective linear design of Bridge Crossing. I think I have an idea allowing to design maps using constraints, allowing level designers to define the essence of a map, while leaving enough margin for the computer to generate a completely new map.
What was a good spot in one because it provided good cover and overlooked strategic places in one map may now be too exposed, or a truck may now provided cover in an area previously exposed.
I need to think a bit more on that, but I think I'll be making a prototype soon...
In FPS games played in closed-quarters such as America's Army, there are players who have been playing on maps for so long that they know all good spots and winning tactics in such detail that realism and fun take a hit. Consider for instance grenade spamming, where players do artillery shots using grenade launchers, throwing grenades at the other team on the other side of the map immediately after spawning.
The simple solution to that problem is random spawn points, but that does not solve the problem of having experienced players rush blindly towards strategic places, knowing enemies are too far away to pose a threat.
I would suggest taking a step further. The entire map should be randomized.
Now, having completely random maps is no fun. Level designers put a lot of effort into crafting maps carefully. See for instance the simple but effective linear design of Bridge Crossing. I think I have an idea allowing to design maps using constraints, allowing level designers to define the essence of a map, while leaving enough margin for the computer to generate a completely new map.
What was a good spot in one because it provided good cover and overlooked strategic places in one map may now be too exposed, or a truck may now provided cover in an area previously exposed.
I need to think a bit more on that, but I think I'll be making a prototype soon...
