As seen above, the animation rigs are as finished as they are going to be so I am progressing towards actual animation.
The protracted nightmare that is weight painting. Each bone in the character exerts a certain amount of influence on the body surrounding it. Because the computer is too much of an idiot to approximate this in a reasonable manner, these influence/skin weights must be painted in by hand, often per individual vertex.
And they are textured. It seems rather pointless right now as they are largely defined by solid blocks of monochrome, but it will make sense once they start getting dirtied up.
Work has started on the genuine article, bona fide player characters. The first hurdle is constructing the actual mesh, based off of earlier 2D designs. As can be seen, these are as good as done, though a few tweaks while likely happen to their proportions. These objects are inert as of now, and will need bones and skinning, after which they can be animated and implemented into the system developed previously.
- UV Mapping (prepare the characters for texturing)
- Texturing (Three iterations, see Temporality)
- Skinning (Create skeleton and confer skin weights)
After which comes the animation.
Some marked improvements today, all building towards proper character, finally replacing the phallic stand-ins. A breakdown:
- Complex, blending animation rig has been prepared
- Right stick now has a button-press functionality that resets head rotation, eliminating orientation confusion
- Blending rig now able to smoothly transition between walking and running depending on the intensity of the input
- Rough designs for animation skeletons made for both characters
As we slowly eke our way forward towards the more definite player characters, some thoughts towards input schemes and rage of motion, and ‘expression’. The original prototype was geared towards the Xbox One controller because I had a couple of those lying around and they are good. Now, some months later, I can now say with measured certainty that these controllers will also feature in the final product. I have considered other forms of motion, like the Kinect or the keyboard, both of which present several problems not present in the XBONE: mostly those of personal representation. The Kinect, while wonderful for many wholly theoretical reasons, is ultimately about yourself, the meatspace vessel controlling the machine. The point of Exit Pursued By A Bear has been proven to be puppetry, which requires a certain disconect between puppet and puppeteer for it not be acting, and it requires, above all, precision. The XBONE presents this in spades, and is a good Rosetta stone for standard game interaction which cannot be dispensed with if this game is to achieve any of its design goals.
Which brings us to the question of what the controller is actually to control. I have drawn a rough sketch of what should at the very least be possible. The trick, and problem inherent in the design, is that in wishing for the players to have as much control as possible over what they want to portray, means arming them with as vague a range of movement as possible, rather than a shitload of specific interactions (like MMO’s tend to do in their emote animation range). Plus, they have to be mapped properly and intuitively to the interface. For now, the sketch is as follows:
Left Stick – Spatial translation (walking), possibly upgraded to pressure sensitive running/walking
Right Stick – Head rotation, as already implemented, possibly to be augmented with snapback to avoid confusion
A – Jump (platformer staple, can also denote anger or excitement)
X -‘interact’ (push?) Uncertain about this one, is an openly violent gesture/interaction necessary. It would seem to be a bit too much on the nose of regular gaming, but a certain dramatic element can’t be denied.
B – An expression or gesture, not sure what
Y – An expression or gesture, not sure what
LB – Raise left arm (for any number of things, waving, denoting presence, etc)
RB – Raise right arm (for any number of things, waving, denoting presence, etc)
LT – Sit down
RT – Lie down
LT + RT – Dead/asleep
If this game is to feature or imply some sort of narrative structure, there has to be a measure of Temporality. Most importantly, there has to be an Ending. An ending retroactively affirms that there has been a middle, and therefore a beginning. If the game does not ‘end’, no part of it can be considered a concerted stab at a story. The three-act structure, while sometimes (unfairly) maligned, has been around for thousands of years for a good reason: It works well. But even if it wouldn’t, it is a structure which audiences have been blasted with for centuries and is practically hardwired into our brains as the quintessential ‘story’, so not employing this structure would be idiotic.
This would confer some structural integrity to the experience, but could conceivably construed as three separate instances or levels. Therefore I surmise it would be best if the three levels (or acts) build off each other in a recognizable way, possibly being three variations of the same level, therefore denoting the passage of time.
Lastly, this reminds me of the earlier character sketches I did where white dirt or blood kept cropping up on their bodies, because I liked the way it looked. It broke up the monochromous monotony of their schemes. But, if this dirt (or damage or wear) would build up during play, it could also denote passage of time, as well as convince an audience of the fragility and temporality of the characters themselves, in short, sporting a banal sense of growth.
- The game will feature three levels structured together like a three-act setup: A beginning, a middle and an end.
- The levels will be strung together thematically and design-wise to present a recognizable whole.
- The characters will subtly become dirtier throughout the game, denoting mortality and the passage of time.