Fergal Costello

I recently wanted to try my hand at some more low poly modeling. This time I wanted to try a character, but I still think I have some ways to go before I build anything fully organic. So I opted instead for something with hard edges – I figured that would be a good place to start. As I’ve mentioned in previous posts, I am new to the world of art. Hence I decided to build from someone else’s concept before I embark on drawing up my own.

I’m a big fan of Glen Southern’s work. If you get a chance, check him out! It was also his video tutorial series that made learning Silo easy, and I can’t recommend these enough if you’re new to the world of 3D.

I noticed that Glen had a low polygon robot character up on his site, so I decided to try as best as I could to mimic his design. Below are two screen shots of my attempt to replicate his work.

Apart from a slight variation in the torso, I decided to add opposable thumbs to the robot’s hands (Just to appease the engineer in me). I haven’t UV unwrapped the character yet, but I hope to in the future. I’d also like to attempt rigging this character, as I think that process is something I should devote some time to learning.

In all, this took the guts of an evening to do, and I am really happy with how it turned out. I’d like to thank Glen for graciously allowing me to post my replication of his work up here.

Fergal

One of the hypotheses of my research is that notions of difficulty can be captured from abstract games using heuristics, and these heuristics will hold when carried over into more concrete implementations. The problem with tackling a topic like this, is that I need to be able to evaluate a user’s experience on the implementation and then check their correlation against the output of a heuristic set. As such, I’m required to build my own game that can support the requirements of capturing these experiences. This post outlines early tests I’ve made into developing the game prototype to do this.

This first thing that needs to be considered is the game environment. I’ve used Silo to create a simple prototype environment, which you can see below.

I then wrote an XNA content pipeline extension that takes .sia files and parses them into my own run-time format. In the following screen shot, we see the same environment at run-time in my own engine, and the capsule at the center of the screen is the player’s bounding volume.

The place holder buttons at the bottom of the screenshot above are used for camera interaction, namely the ability to rotate the camera left and right. Pinch is used to zoom in on the player, but I’m not sure if this will make it into the final version of the game. The cancel button is another place holder control, which I envision will be used to allow the player to cancel an action they’ve invoked. The white lines drawn on the floor regions are the nodes of a navigation mesh, which will be used for path finding. When the user taps the screen, a ray is sent into the scene and hit detection tests are performed against the scene graph. If the ray intersects with one of the nodes in the navigation mesh, then the player’s character is informed of the interaction and steering behaviours are invoked to steer the capsule to that location on the screen. These features are seen in the following video.

http://vimeo.com/31648652

I will deal with the actual mechanics of the game in another post, as I don’t want to bloat this one. Also, as this prototype evolves I will post details of any new features that I code into it in separate posts.

Credit: Many thanks to Cian Walsh for doing up the icons for the radial dial menu shown in the video above.

One of the things I’ve been working on as of late, is parsing .bvh (motion capture) files and viewing them in a game engine. To achieve this I used C# and XNA, as I wanted to get the animations onto my Windows Phone 7 and into my own game engine. Below is a video that shows blending between two animations (a walk cycle with a rifle, and an injured walk cycle). Post processing is also demonstrated, with the head joint’s orientation being updated to get it to track a moving target in the scene. The bounding frustum highlights the character’s field of view.

http://vimeo.com/31648898

As this is my first attempt at full motion capture support, there is still plenty of work to do. I need to include retargeting at the pipeline stage, so that these animations can be attached to skinned meshes. Eventually, I would also like to implement blend trees and maybe some proper inverse kinematics support so that the character’s weapon can also be aimed at the target too.

I’ve never been much of an artist!

Recently however, I decided to try my hand at some simple low polygon projects. For these I chose to use Nevercenter’s Silo. I chose it because of its simplicity. Its intuitive UI means that most of your time goes to modeling, and less time goes to drilling through menus or learning shortcuts (although they help). It is also cheap and very powerful for a modeler.

My first project was to try and build a space ship for an RTS concept, similar to Relic’s Homeworld. It features a large prominent laser at its bow, a split hull, guide rails, and large engines at the stern. Since this was just meant to be a concept model, I am pretty happy with how it turned out, and have decided to just leave it as is.

The second project I wanted to work on was a helicopter model, for a mobile phone game I was considering making, which would have been along the lines of the old Desert Strike games with 3D elements. I hope to some day get back to finishing this model completely, by adding a gun turret, fixing the tail section and adding some texturing.

That’s it for now, but I hope to try more of these projects in the future and will update as I do.

Fergal