Professional Project

3rd Year Uni Module: working with a multi-discipline team to create a game for a live client brief

[pxlbpp-blog]

Second poster, as it was requested by the team to have another after it seems they liked the first one.

It was also pointed out, when they noticed that the first one fit perfectly as my Tumblr avatar, that it’d be ideal for badges and stickers, since it was small, nicely shaped, and fit into a circle. So I made sure it stuck to the same format of a circular pose into a square border, in order that it could be compatible with a layout for stickers or badges.

I also wanted to avoid it just being entirely the same as the fire one, which was somewhat monochrome and a lot softer. I didn’t want it to just feel like a pointless other pose, and wanted it to be notably different from the other. Since there’s only so many poses I can put her into that’d make sense (especially to a circle shape) I thought to put more focus onto WHY they’re special.

The fire one is hot, monochrome and glowy, being the lava/armour bird. This one is the shiny, special combo bird, so I went with a multicoloured, intense glitter factor.

If/when I get around to doing different ones for the individual seasons, they’ll be softer and have less glow in general, being that they’re the “normal” skins, and will focus more on their primary season colour.

[pxlbpp-blog]

To go along with Alissa’s environment of super cliche’d and metal mentalness, we got shown a screenshot of the textures/shaders mid-loading in game, and the bird’s textures had broken up a bit, giving the bird a metal/armoured look that we all basically loved.

I thought it’d be good to throw together a unique skin for the bird at last.

Unique in theme it is not, but it isn’t just a palette swap like the other textures. The colours are in other places more appropriate to the new fire and armoured look, and the metal texture has texturing and reflections/glossiness on it with the fire a degree of opacity. It’s still limited and could obviously be taken a lot further, but I thought it’d be a good opportunity to get like a proof of concept done, again, to go along with the environmental idea of some foresight and long-term planning for the game. This would be a good way to showcase what we could do with the bird, and that there is a lot of potential for advancement since everything until now is pretty much just flats.

Also, I realised that we have lots of proof of concept and pretty environmental images, but basically everything for the bird is concept or technical imagery, and nothing like a poster or show-off. So thought this would be a fun excuse to create one for it.

[pxlbpp-blog]

A different concept environment. Alissa and I have been asked to get a couple prepared for the Dare interview on Monday to show some foresight into what other environments/concepts might be available, as well as what kinds of things we could put together as monetised packages.

The two concepts we were given were a highly cliche’d (to contrast the original brief) canyon with metal and skulls and blood and stuff (Alissa wanted to do that) and a Mayan architecture style one, to fit into our story.

I went with an advanced version of the progression we’d shown, with the new world of man actually building up, rather than the previous world falling apart. To this, I went with an increasing number of trees and skybox assets (rather than empty skies and falling debris) and a river, as well as intact buildings and such. 

[pxlbpp-blog]

Peer Evaluation Form

Done.

[pxlbpp-blog]

Progress

Got invited for an interview with Dare. Four days before deadlines. Cool.

Anyways, got some transition animations done. Got the bird swooping slightly to the side as a visual aid for when the player moves left/right, and a transform/flap rolly-polly animation for the bird to do between seasons, within which she will change colour dramatically to suit the new season.

[pxlbpp-blog]

Same as before, but in groups~

[pxlbpp-blog]

We’re putting together our demo video application for DARE, so we thought it’d be good to have some more animated and exciting characters to go over the animation, rather than just static images and such, or us blethering into a camera boringly. Photos of the real people, at the end still of course.

[pxlbpp-blog]

Seasonal reskins for the bird, according to the concepts we originally planned. I realised that the way I’d textured them allowed for this really easily.

[pxlbpp-blog]

A victory flapabout of the bird. I was aiming specifically for something extravagant and flourishy, something that’d clearly stand out from teh rest of the animations (which would largely be loopable, gliding etc) and figured it wouldn’t hurt to be something that’d be nice to whack into the presentation.

Bit of a mess because I’ve tried to do it quickly (and with no rig, yes, I know..) but think it looks fairly smooth for how weird a thing it is to do. Also, there’s a fair bit of made-upness since I don’t know many birds in real life that do a barrel roll, so I had to kinda just figure that bit out.

Anyhoo.

[pxlbpp-blog]

Was asked to create a position for the bird that’s perched, for before the actual level begins. I figure  it might be used on a menu screen or something while selecting music and such, so I figure a good way to approach this would be to not only make it loop, but try and add some character clue as to what’s happening. 

In the game, the world is blowing up as the MFFB is flying away to safety, so I thought the menu screen beforehand could be that moment beforehand where MFFB knows something is about to happen, but not sure what yet. I also thought it’d be a nice touch to have the tail swinging in a clock-like fashion, just for some suggestion.

Again, feels a bit blank and dry, but I just don’t have more time to put more in.

[pxlbpp-blog]

Flight demo: 8 second flight loop, repeated with a turnaround.

My non-rig seems to be working fine for me. Pretty sure decent animators or lecturers would shout at me for doing a stupid rig like this, but I think it was worth it considering what I need to get done and in my timeframe. There’s been no complications, and there’s no need to animate particularly many things, like complex walk cycles with multiple limbs and stuff.

That said, I’m only so-so about it so far, as it doesn’t look that great since I’ve been trying to bone, weight-map and animate it in the space of a day since I can’t afford to put time into this tomorrow. 

I spent a lot of time watching birds fly around for reasearch and was aiming for more of a gliding/cruising look, rather than big heavy flaps, as I view her as quite a large bird that wouldn’t flap her wings constantly. That said, I think I’m still missing a bit of weight from her, and with all the extra dangly and pokley bits, I’m missing some follow-through and anticipation.

Overall though, I’d say it’s...acceptable, for the time being? :/

[pxlbpp-blog]

Weight Mapping. Went pretty well, not much to weight map after the wings, and a bit of mirror mode cleared up the rest of it. Easiest weight mapping process to date.

Now just to do some cleanup tests and then can get on to the hard bit.

[pxlbpp-blog]

Got the bone structure in place, and all the bones named, with an idea of how I plan on doing things. It’s a pretty stupid idea, but it seems like it’d be easy and quick, with little risk of rig breaks down the line. After all, there isn’t actually much animating to do, and I’m already spending too much time on this module this week. We’ll see how it goes.

[pxlbpp-blog]

Placing the bone structure, thinking about how best to position the bones in the wings for an effect that wouldn’t be stiff, but also wouldn’t be overly complex as I’m becoming time conscious.

I asked a classmate who has made a lot of 3D winged characters for some advice and was given this suggestion drawing which led onto a discussion about how to actually move and rig the bones themselves. I’m thinking of leaving the wings to FK/manual movement, and IKs for the lengthy tail. I’d especially want to manually rotate the shoulders to create as wide a flapping motion as I see to be best at the time.

[pxlbpp-blog]

Broken path segment for the ends of each season path. Did it fairly quickly, and felt it would look nicer than just a flat/blank end to the path.

[pxlbpp-blog]

After some feedback from my lecturers regarding my workload, after the absence of our other artist, they’ve given me a clearer priority list now. Rather than spending so much time on the environment, which wasn’t originally my focus, I’ve been told I can leave it largely as it is the now and work with placeholders, to come back later if time permits.

So, it’s time to get back to my main focus, the character, and get ahead to actually creating a finished product that’s animated and ready for the presentation.

Since we’re not really spending any time on complicated textures, I’ve just applied some flat textures directly to the poly faces and shifted around some vertices in order to sharpen the colours up a bit (they looked very square and blocky, a lot more so than now).

I originally drew colour choices for all the seasons as well as Quetzlcombo because at the time I wasn’t sure if we would be changing the textures or lighting to suit. Since there’s no time for that anymore, I figured I had to just pick one, and picking a single season seemed weird when the seasons would change, so I went with the original Quetzlcombo concept and followed that as closely as possible. Think it came out looking alright actually.

Ready to rig.

[pxlbpp-blog]

Was asked to create a loading screen. Not wanting to spend too much time on it, as I’m focussed on getting other modules caught up with if possible, I decided to keep it as simple as possible. In the name of avoiding it becoming boring (since I was going to animate it, rather than leave it static) I decided to make a few of them.

Went with this as another excuse to clue in the player that something else is going on here, and that there’s more to it, with zoomed in wall carvings and text. Mostly unreadable even though they are actual clues, so made one of the screens in the cycle inverted with the clues readable and the loading text (in the same place) unreadable.