The Pixel Crush

-------------------------------------------|Digital Animation & Game Criticism|-------------------------------------------

Thursday, 31 January 2013

Global Game Jam 2013: Einsam

I recently attended a game jam in Bristol which is an event where lots of people get together for a short period of time, form teams, and create games, videogames. This one was a 48 hour game jam (note how the event is measured in hours not days, because each hour is precious) and myself and Nigel Kitts rocked up to try our hand(s) at game development. There was about 50 of us, some professional, some students, some with no experience at all. The atmosphere was enthusiastic, not especially competitive, and sleep deprived after a short while.

The theme we were all set was the sound of a heartbeat, just the sound. So our team (whom Nigel had made acquaintance with on the door before I arrived) set about coming up with ideas. We were joined at this point by a guy from radio 4 who was interested in sitting in on this early stage, him and other BBC staff were observing and offering creative input in an open minded attempt to get some idea of how games could be more relevant to the BBC's output. This went in a number of different directions and to our credit no game was derivative of another (at the game jam that is, we totally ripped of Amnesia: Dark Descent, crossed it with Dead Space and then made it 2.5D), each had its individual style and mechanics.

Some concept art from one of the talented members of our team:

The Athena Suit
As our coder pointed out, it was nice to have an actual asset workflow that went all the way from concept to 3D model and rig, not that it was flawless mind you.

Nigel did a commendable job modelling the character, unwrapping it, and texturing the the diffuse map, while I churned out a few environmental assets before helping out on the character sculpt and texture to create normal and AO maps.

Due to both my severe lack of rigging skills and a general ignorance of how to use Unity, our game engine of choice, the animation on the main character didn't quite turn out as expected. We started with a really simple set up that was no more than different parts of the model parented to each other in an appropriate hierarchy, and then a global movement control wacked on top.

Not only is our game's protagonist a girl, but also a camp black stereotype?
This was getting some good results and was very fast to make and use, I made all the animations in the videos below using it. But we couldnt get this stuff to import correctly into Unity, not using Maya files, nor the FBX format. Turns out they were there, just a few metres out of view, but we only discovered that much later.
Hero Pose
We didn't have too much of a problem getting the character into Unity and looking sweet, but movement was another thing altogether.
Unity Real-Time Renders

Mental Ray Render, 1m 30s

Mental Ray Render II, 1m 30s

Inspired by the Barn Owl apparently.
I hypothesised that maybe Unity required a proper rig with skinning and a bone structure (see the setup that tested that theory in the video below) so I set about constructing one for the first time in 3 years and probably only the 3rd time ever. Positioning the bones wasn't a problem (though they were hardly precision, but we were short on time, it was already day two) but I had to look up a number of things like setup up IKs for the legs and arms and had external help figuring out why parent constraints were screwing up the hierarchy.


I learnt a lot blasting through that rigging process, I even enjoyed painting the weight from the bones onto the mesh, seeing the character come to life was satisfying and I can now appreciate the appeal of rigging. Its made me much more confident that I could actually create and rig my own character from scratch which is a pretty exciting prospect seeing as its been a gaping whole in my knowledge of the CG workflow for ages.




Im tempted to go back and finish the rig etc but it'd mean re-doing all the animations on the new rig so instead what I'd like to do is create a small environment, and a character. Then I could go through the whole process of porting this scene into maybe a couple of different game engines, including Unity, in order to being myself up to speed on realtime awesomeness.


Here's the global game jam website. And here's where you can play the version of the game we submitted (its been improved significantly now):


Wednesday, 9 January 2013

The Pixel Cobbler

Organic shapes have always stumped me a little when it comes to modelling them in CG, so in making the roller derby skating boot the reidell colour lab has been invaluable for providing reference. You can pick a boot model number and then view it from a couple of angles, picking colours for different areas of the boot. Its effectively an interactive catalogue, but perfect for judging and matching the proportions of the boot to its real counterpart. There's a bit of in-medias-res blogging for you, now for the set up:

I've finished all my work at Aardman now, I returned there for a brief two weeks to help out on a music video and then spend a week with the marvellous Gavin Strange hammering out as much as we could on his exciting new Dew IP, creating characters and overall look in 3D from Gav's sweet designs. Here he is giving the resulting presentation.

Since then I've been back in falmouth working on a couple of things, one has been a rhino character for a game (for another time dear reader), and the other has been creating animation for the esteemed and ingenious Jennifer Rollason for her roller derby documentary final major project. And that is the set up. So I started by creating this boot, which may or may not be used in a credits sequence, but will definitely be used as a title effect for introducing interviewees, and as a means of creating transitions between scenes.

I started developing the materials and shaders alongside the modelling itself, basically got excited and skipped ahead a bit, before having to go back and finish the modelling. I knew I wanted the model to hold up to close examination, new standards and all that, so I left the threading to a bump map but made the ringlets their own geometry. Top tip: the slide edge tool, edit mesh/slide edge tool. This was great when spacing out the edge loops on the toe as it allows you to slide edges whilst maintaining the form of the model, learnt that one recently thanks to Rich.
I created some laces, more laboriously that I would have liked, by threading a CV curve through the ringlets and then extruding the lace along the curve. This created some dodgy bending and clipping but I could sort most of it just by tweaking the curve connected to the extruded mesh before deleting the history. The texture is just a simple cloth Maya texture.
You can be really sneaky with UV space for stuff like this by overlapping UVs for objects that share the same material, this way they can use the same area of a texture and save you texturing time later. Also the transfer mesh attributes tool is invaluable because that way I can unwrap one wheel, then give its UVs to all the other wheels, saving me both having to unwrap as I go, and the time and effort of doing each one individually.
I had some weird issues applying different materials to the inside of the wheel, it would apply randomly to stuff that I hadn't selected, I figured it out in the end. As is the case with so much Maya stuff- its just a case of learning the bug in the software, and then remembering to anticipate it. I love the soft bubblegum look of the wheels here, it was too soft, like wax or similar, I tightened up the radius of the scattering and got this effect:
With glossiness maps applied to the metal I was pretty much done at this point, I tried adding bump to the metal too but it was a horrible effect. I wish a had textured the edges of the leather strips/straps differently as it would have been nice to distinguish between wrinkled surface leather and the noisier interior of the material.
Lastly I wove the Kernow Rollers into the side of the skate and called it a day, this was just a matter of adding it to the diffuse and bump textures. Something exciting I found out about the mental ray ambient occlusion pass: it takes bump mapping into account, and helps pick out all those details even though they aren't physical geometry. Much more useful than what I usually do with the ambient occlusion shader piped into a surface shader stuck on a render layer.

Here is the final turn around:

And a spare render:

That's enough pink for one post. Rhino post next time, something which Luke is doing me a massive favour (paid favour) and rigging for me. Then I'll be animating it for a game that the refined and knowledgeable Chris Underwood is developing. No Pixel Propaganda this time! I must be getting out of practice, or less interesting, or less abreast of stuff only I care about. I've vowed to stop posting at 2am because no one reads them at that time, this shall be posted at a sensible time for sensible people.

Congratulations to whoever the 50,000th person to read the Pixel Crush was, you are my favourite reader so far.

Sunday, 6 January 2013

The Self Loathing Genre

Spec Ops: The Line
Fiction is used as a method of communicating ideas and emotions. So when I play a game, if I'm paying any attention I'm looking for that idea, that emotion. It has been twice in a row now that I've played and found only a mirror, but it hasn't so much made me look at myself in the mirror and make some worthwhile examination of character. What I've found instead is the game looking at itself and conveying themes to the effect of: "ugh, I'm horrible...why do you love me again?" Like some self loathing lover. The first time it was Spec Ops: The Line, and I thought (rather self righteously) "At last! My loathing for the first person shooter genre vindicated." A game that examined the gameplay conventions and fraudulent narrative framing of player actions in the genre, and subverted them.
The Mirror
This wasn't so much a revelation to me as a reassurance that AAA game development had grown some self awareness, and turned it into a commercial product. Though that didn't change the fact that I had played through the entirety of a mechanically uninteresting game in order to have something I already knew affirmed, though engaging character performances and linear narrative did go some way to easing my path to the game's conclusion.
Ripe With Meaning
When the idea that the fiction conveys is (trying to refrain from using the word merely) merely a piece of criticism, then that fiction really needs to justify its form as fiction at all when it might be able to better make its point as an essay. I mean, if some information can be said more clearly using writing and gains nothing from the indirectness of story or gameplay then why not just outright tell me rather than leading me through an elaborate labyrinth of reasoning before presenting the key argument? Far Cry 3 I'm looking at you, kind of sideways.
New Perspectives in Far Cry 3
Far Cry 3 should be celebrated for its ambition to be more, more than what most games aspire to convey, for attempting to challenge the existing flaws in the genre that players have grown lazy and comfortable with.
But its so flawed in the way it does this that all of it feels stupid. If subversion was the point, then don't you need that one strand of sanity amongst all the insanity? Something for the player to grab hold of and trust, so that the crazy stuff actually has a frame of reference?
This is not normal...
I feel like better use of the player's time would have been to play off the game's excellently designed combat and movement systems, against the narrative need to save the protagonist's friends. This was apparently one of the creative goals: can the player resist the freedom of the island in favour of saving Jason's friends? They're are all pretty much empty vacuous clichés barely worth saving regardless so its not the strongest of arguments to begin with. But then Far Cry 3 treats time no differently to any other game, and there are in fact no consequences, as far as I can tell, for gallivanting off doing trivial rubbish for hours. Why not have characters begin to die off if Jason doesn't turn up at all? Have them suffer longer at the hands of their captors should Jason arrive late. That way the consequences are part of the systems of the game.
The Clichés Assemble
I finished the story, made my hollow binary narrative decision, and then was allowed to stay on the island regardless. I held a gun in my hands, stood in the empty road and looked in the cool tropical moonlight. I felt really empty. I don't know what Jason was supposed to be feeling, having been so vocal with his endorsement of my horrible actions throughout the game he was suddenly speechless. The narrative framework had been removed and for a minute, and we were wrong footed. But then down the road came a jeep. We snuck into the grasses- bow drawn, and waited.
Playing Bows and Arrows
It was only now that the fake impetus of the story had expired that we could fully acknowledge that playing felt good, good enough that we needed no excuse or a purpose. I felt just a little bit tainted by the self loathing that Far Cry 3 had taught me through the seduction of its agency and systemic empowerment. I felt just a little bit used. And in that sense, there was one fibre of thematic resonance to be found. That, just as Citra exploits Jason to conquer the island, Far Cry 3 exploits the player to make an example of the genre, without giving us a choice.

You could have just written me an essay, maybe.

Pixel Propaganda

The writer has been interviewed extensively defending his work and I really sympathise with him. What he did was bold and flawed. It deserves recognition for ambition if nothing else.

Jonathan Blow wrote a fantastic piece over Christmas about it taking him 15 years to recover from burning out over work, and how he's finally inspired again.

A nice turn around on the the can games be art debate (that never dies). Can art be games?

Uplifting piece about videogames and self harm.