Steve Swink, a game designer at Flashbang Studios behind breakout indie titles like Raptor Safari and Jetpack Brontosaurus, lays out a unique vision in Game Feel: A Game Designer’s Guide to Virtual Sensation. The sensation that Swink is describing is not the thematic or emotional feel of a particular genre or plot. It’s the impulse that makes you lean forward when Mario makes a leap of faith. It’s the rush of defeating a difficult boss. It’s the reason some games are tight and exciting, while others are sluggish or confusing.

Approaching the subject as a blend of art and psychology, Swink splits the text into sections that identify the concept of game feel, define metrics to quantify both the hard numbers and subjective impressions of evaluating it, and finally apply that model to a handful of case studies picked from the lineage of brilliant game design.

Swink states in his introduction that his book is intended for a broad spectrum of readers, including both faculty and students of game design. In this effort he mostly succeeds, approaching some complex problems without delving too deeply into technical minutia. At worst, he occasionally repeats himself to drive a concept home. His detailing of input devices can be skipped by anyone familiar with basic electronics, but remains helpful as much of the later analysis requires a firm grasp of analog user input.

The greatest strength of Swink’s treatise on design is that it fills a gap between traditional discussions on aesthetics and usability: the former rarely touches on the practical problems of interactivity, while the latter usually aspires to a perfect and effortless model that overlooks the thrill of challenge and progression. Citing a classic philosophy that games should be easy to learn but difficult to master, Game Feel pursues the ideal of intuitive controls and captivating feedback that lead to the sweet spot of gameplay known as flow.

Game Feel is available at Amazon, get it.

Finished assembling the sculpture on Tuesday with a coat of paint on the frame. After designing some templates, I cut and stretched the fabric much more reliably, with less fraying and irregular overlap. I still need to clean up the outer edges, but otherwise this part of the project is done. Two more pictures after the jump.

The senior show for undergraduate Time Arts majors is coming up! I finally finished the promotional materials after weeks of struggling with the limits of 3D rendering to print resolutions. This poster design, partially created in 3D, was simple enough to render out at sizes considered “large” by digital standards. However, properly converting digital designs to print requires massive files that dwarf their digital counterparts. My target was a render that would create a 24 by 36 inch poster at 600dpi. This equates to a 14,400 x 21,600 pixel image, which is apparently beyond the capabilities of even the super-powered workstations in our render lab. After consulting with my professor and dialing down some of the settings, we were able to render a 24×36 graphic at 300 dpi, which is still a high quality print resolution. We printed some posters yesterday, and put in an order today for some 36×72 vinyl banners that will be displayed at the show. I can’t wait to see them.

For more information about the show, check out NineBitsTurnNeon.com

After several failed starts, I finally have the surface material planned out in an effective way. Initially I intended to stretch the fabric over a wireframe that preserved the hard flat edges of my design, then treat it with a chemical stiffener. But after failing to get anything resembling a hard edge using tape and wire, I stumbled onto the pleasing natural curves that appear when the surface is stretched tightly. Using staples, I was able to apply the surface in an unobtrusive, non-permanent way.

I tried out three fabrics as a potential surface material–Pellon JAS Interfacing, Pellon Easy-Knit Fusible, and a generic white chiffon. The JAS Interfacing was cheap, thin, and responded well to the stiffening liquid, but shared more properties with paper than cloth and tore easily under stress. The chiffon was durable, received light well, and had a very tight knit but at $8/yd was too expensive and difficult to work with. I ended up going with the Easy-Knit Fusible, which is slightly off-white but has a very stretchy, strong weave that creates a tight projection surface without bankrupting me in the process.

To reinforce the surface and ensure it retained its form, I planned on chemically treating it with a fabric stiffener. I conducted a spot test with commercial stiffener versus a mixture of Elmer’s glue and water (another method) on all three fabrics, and results were relatively positive for the commercial stiffener. However, during the test I had applied it to saturation against a flat surface. When I finally finished stretching the cloth over the wood frame, I tried to apply the liquid onto the freestanding cloth, with dripped and streaked horribly. I ended up peeling the ruined surface off and trying AGAIN, this time without the stiffener.

After a series of poorly measured cuts and a bit of cursing, I was able to create a serviceable prototype for the final surface design.  I marked and removed the cloth from the form, and am now working up a cutting pattern that will ensure clean, identical pieces for each of the six arms of the sculpture.

The School of Visual Arts at NIU recently acquired a 3D printer, which creates basic prototypes of digital models.  There are a number of different technologies that achieve this; our machine uses a special powder that is applied one very thin layer at a time.  A printer head moves across the powder, dispensing a glue binder in a cross-section of the model.  This way, the 3D structure is gradually built out of slices of bound powder.

Last semester, I designed this figure for an animation course.  After posing it against a wall for support, we exported the model as an STL file and sent it to the printer.  About 4-5 hours later, the print was complete and ready for excavation.  I’m really happy with how this turned out, it’s a kick to see something that once existed only in a computer turned into a real, tangible form.

I decided to use 3/4 inch square dowel for the frame, as it is easy for a novice to manipulate and provides rigidity without excessive weight.  After measuring and cutting the wood with a handsaw to ensure precision (my jigsaw sucks), I assembled the pieces with standard hardware and steel brackets, using a generic sealant to reinforce the joints.

To join the sections together, I cut and drilled a 1/16 inch steel plate, and will bolt the support arms to it for the final assembly.  It all feels very medieval right now, like I’m building a catapult in my basement.

I haven’t weighed it all out yet, but I’m hoping the hardware won’t make the entire thing too heavy to hang.  On paper the measurements seemed banal, but at 60 inches across and over 30 inches deep, it’s going to be pretty stunning.  I still haven’t decided exactly how to execute the surface material, but hopefully I can meet with the professor of the fibers/papermaking department and see what she thinks.

Here are two more crappy images of the axle, near and far.  Stay tuned.

The sculpture design is based on a variation of kite design, using the traditional forms and angles to create a metaphor for the fragility and beauty of life. The collective form is reminiscent of an ice crystal, again drawing parallels to unstable but elegant architecture. This early mockup was used to illustrate the ultimate goal for the physical structure, as a star-shaped frame covered with semi-transparent material.

The target size is 60 inches in diameter, to be suspended in air.

The game features amazing illustration work by Jason Zielinski.

My focus will be working with other developers to continue creating new content and features for this game, ahead of other exciting releases. Of course, I won’t be able to discuss what I’m working on in detail, but stay tuned for upcoming releases and updates on my other independent projects.

My eternal gratitude goes out to Colin Godsey, my friend and colleague who connected me with this amazing opportunity. Thanks also to my new boss, Christian, for giving me this shot.

I was asked to create a state machine for an assessment that has a few simple rules, basically creating a self-determined “creature” that had a few basic abilities. Building on the initial ruleset, I decided to develop it into an automata simulation.

Check it out here.

The creatures have four attributes: thirst, hunger, fatigue, and boredom. They are capable of idling, walking, running, eating, drinking, sleeping, and with user interaction they can play. Clicking with the mouse will cause nearby creatures to react.

Each state has a transition pre-state that occurs before it. The color of the character will blend from a color representing his previous state to the color representing his new state.

- Initial values of the four attributes are generated randomly when the SWF is loaded.
- If no state is queued, it will idle. Idling causes its thirst, hunger, fatigue, and boredom to increase.
- If it gets too bored, it will either run or walk (50% chance).
- Walking increases all attributes except boredom, which decreases slightly. If it gets too thirsty, hungry, or tired it will switch to that action.
- Running increases all attributes except boredom, which decreases. If it drops below 50% boredom or above 100% fatigue, it will eat or sleep (50% chance).
- Playing causes all attributes to increase except boredom, which decreases significantly. It will not stop playing until the user releases the mouse button.
- Eating causes all attributes to decrease except thirst, which increases significantly. It will drink after eating.
- Drinking causes all attributes to decrease. It will idle after drinking.
- Sleeping causes fatigue and boredom to decrease, thirst and hunger increase. It will idle after sleeping.

I had to disable the real-time stats for the web version, otherwise it gets too choppy. But in a standalone player is runs nice and smooth. If you want the source code just shoot me an email.