microtonalmatt - Tumblr blog

microtonalmatt · 8 months

Text

Redesigning Aife

Hey so it's been a while since I did or said anything with/about Beeknighted. After the launch of the demo disc, I updated the game feverishly for a couple weeks, but I got into some dire financial straits and was looking for more work for a long while. I also took a break from Beeknighted to work on Star Crucible, which was only intended to be a one-month quick little project for a game jam that has spiraled out of control. It's still in the back pocket but I need to let it simmer for a while I refine some of its ideas.

I'm now working on a third project, this time something I'm telling myself I'll actually get done in a short time (I know, I know). But I just couldn't stop thinking about things I wanted to do with Beeknighted. I've been writing characters, chunks of the main quest, designing weapons and items, all just on paper because I haven't really had a chance to properly work on it for a while. But I thought I shouldn't set aside both Beeknighted and Star Crucible without at least dropping a crumb or two, so I've been working on the redesign of Aife, the bee.

First off, why the redesign?

The bee was the first full 3D model I had ever created, textured, and animated. Period. It had...a lot of problems and beginner mistakes. And I didn't just wake up one day and bust out a perfect model of what was in my head. Version 1.0 of the bee was...quite frankly horrifying.

Yeah, it...it was bad. It was made in 3 days for a game jam using youtube tutorials for a guide (which was not the fault of the tutorials at all, simply user error). When I decided to develop the game further, I set about retexturing the model, thinking I could save the geometry. I really just cleaned up the face and body.

But this didn't fix the underlying faults with the model. The rig was just plain bad. There were weird issues with weights where if you bent the character too far forward the head would shrink, the hair couldn't animate well, there are no shoulder bones so no shoulder movement, lots of issues all around. Those weren't easily fixed when I already had made around 40 animations using the rig.

There were also issues with the body geometry. The arms/hands/legs/feet/wings were all fine, but the head, thorax, and abdomen, to get technical, were all imperfect. There's weird shading issues you can see on the body that are a result of bad topology. A failure to properly triangulate the abdomen caused it to look weird at some angles. And the head is just poorly modeled overall. It uses a lot more triangles than necessary and doesn't achieve the intended design goal. Heads are hard, faces even more so. She doesn't even have a nose.

Even if you can excuse the mediocre rigging and amateur geometry, you can't excuse the biggest problem of all: my crime against video game collision shapes:

I shouldn't even be allowed to touch a computer after this atrocity.

It was really important to me that the collision represent the shape of the character for the combat to feel precise. But this disaster meant platforming is all over the place. Not to mention if the character gets horizontally wedged between two walls, there's no escape because the character can't turn around. I had thought "Spyro did it [an oddly shaped character], so can I" but I am not a seasoned Insomniac developer. Even so, I don't know how I was allowed to sneak in this egregious sin against video games for so long.

So in short, these changes had been a long time coming.

The goals of the redesign:

First and foremost, the bee had to stand more upright (or else more horizontal, not diagonal) so that I could use a regular upright collision capsule. At the same time, I wanted to maintain as many bee elements as possible.

The abdomen had to be a bit smaller and more out of the way so the player could always see where the feet were landing for platforming purposes. This was a bigger problem than you'd think.

The character had to appear older but still look small. This was simply a mistake in the old design that the face looked too young by the nature of me struggling to draw faces + the bad geometry. The bangs were meant to hang down more but the model didn't allow for it.

Details needed to be crisper, bolder, and chunkier. When I played the demo on my CRT, I noticed the thinner points of limbs threatened to get lost on screen. N64 characters tend to be noticeably more chunky to compensate.

My inspirations for the initial character came from a range of sources. Obviously with the game being inspired by N64 graphics, I picked inspirational references from contemporary characters, as well as characters from media that had been created/released in the run-up to the N64 that might have inspired later works. I wanted the bee to look like a mix between a silent JRPG protagonist (or, y'know, Link wouldn't be a bad pick) and classic fantasy/horror anime female protagonists (the ones that all have that same long hairstyle, black/dark hair, poofy bangs in the front).

My moodboard included Yuuko from Valis, Utena from Revolutionary Girl Utena, Popful Mail, Aika and Fina from Skies of Arcadia, a handful of leading ladies from late 80s/early 90s horror OVAs (like the kind Madhouse was putting out at the time, but usually worse), and some Fire Emblem characters.

The N64 doesn't really have a lot of 3D "anime" characters to use for references. You've got the oft-maligned Quest 64, some of the bosses from the Goemon games (all the npcs are very chibi), and some of the more anime-inspired characters from the Zelda games and a handful from Megaman 64, though Megaman's art style and way of 3D modelling is fairly unique in its own right.

I suppose I should emphasize that while the main character and some of the side characters are anime-inspired, the point was not to make an anime game per se. I chose the look because of its potential to create expressive and amusing faces because I wanted the bee to have as much character as, say, Link from Wind Waker, who is constantly emoting about anything and everything.

The New Version

Aife's new model sports a totally remade head and body. While using the old limbs, I thickened them up a bit for visibility while still trying to maintain that bug-like spindliness. The abdomen is slightly smaller and the geometry is simplified. I simply didn't need that many triangles. She stands more upright, and the head is in a more natural place above the shoulders, and the human-ish proportions are better. The redone body uses vertex colors, rather than textures, but leaves the option open to texture new costumes. I also redesigned the bangs to drape lower on the face but not get in the way of the expression. The old bangs made the character look too childish, imo. All of the textures have been redrawn, simplified, and cleaned up. She now has a nose.

Despite all of the improvements and added details, the model weighs in at 100 fewer triangles than the old model, for a total of 672 tris.

I'm much more confident that this new Aife will be better suited to the highly animated gameplay required of her, and can't wait to dive into the rigging/animating process. But that's all for now!

#n64 #anime #character design #3d model #retro anime #godot engine

7 notes · View notes

microtonalmatt · 8 months

Text

Introducing a hostile and mysterious world

I'm still getting used to this site, so I'm not entirely sure what to use it for. Right now, my plan is to share progress on my games and music, of course, but the blog format entices me to share more detailed thoughts on specific subjects into the ether, whether or not anyone cares to read.

I have been watching a lot of classic science fiction TV, partly as research for my own science fiction game, Star Crucible. I never really watched TVshows as a kid and that pattern followed me into college. In 2017 I watched Star Trek: The Next Generation for the first time and found out I loved it. A few years ago, I began sampling other shows. I watched a bit of Deep Space 9, Voyager, Enterprise, Babylon 5, and Stargate. The show I found myself hooked by the most was, probably to the dismay of many of its detractors, Star Trek Voyager.

Of the many Star Trek series, Voyager falls in a weird spot where its name is spoken in the same breath with the original series, TNG, and DS9 as the core and "best" Trek, yet it is frequently disparaged as terrible, though perhaps not as much as Enterprise. Voyager has some really bad episodes, it's true, but it also has some incredible highs, the cast is pretty well fleshed out and balanced, and there is a lot of humanity that I think Voyager explores better than other Trek shows.

The premise and long term goals of the show are very well laid out in the pilot episode: the starship Voyager is flung deep into a part of space far away from home and the federation and must travel back to familiar space. The problem however, is not finding the way home, as their sensors are more than capable of identifying where they are and where they want to go. The issue is the travel time required is unfeasible, and thus they must reach out to unknown alien cultures for assistance to make the journey. The Voyager crew must embrace new technology and share some of their own to survive, frequently in conflict with Trek's prime directive. And, in addition, half of Voyager's crew is made up of a dissenting faction of rebels that initially do not trust the military hierarchy of the federation. By the end of the pilot, it is very clear who the characters are, what their predicament is, and what they need to do, and how they plan to approach achieving their goals, and the series makes regular strides towards getting them home.

Just recently, I watched all of Farscape, a show with a similar premise which I enjoyed immensely. John Crichton and his motley crew of escaped alien prisoner friends must find their way home while fleeing the watchful eye of the Peacekeepers and other hostile aliens. The trouble is, through either a lack of initial planning or purposeful concealment or both, the introduction of Farscape's alien side of the galaxy is somewhat muddled.

Farscape's pilot focuses on getting the proverbial fish out of the proverbial water. John is an astronaut on Earth around 2000 and ends up in an accident that sends his solo-piloted module through a wormhole in the middle of a space battle. His craft accidentally collides with a small fighter craft and then is hauled into a larger organic ship, where he meets its alien pilot, three escaped prisoners, and a captured Peacekeeper. All of these characters, to one extent or another, want to find their way home. However, it's not initially obvious why this is a challenge for anyone other than John, because the level of technology available to this crew is not evident, and the prisoners are concealing their true goals.

Their ship, Moya, is a living creature called a leviathan, it has no weapons or means to defend itself, its starburst seems unreliable and random, and despite all of the aliens in the cast coming from space-faring cultures, none of them seem to have any idea where they are or any way of pinpointing what direction to travel aside from "away from Peacekeepers" into the "uncharted territories." At one point John asks to see a starchart in order to try to find a point of reference for where he is, and he is met with blank stares.

Sci-fi regularly breaks the laws of physics and realism for its narratives to work. However, it's hard to fathom how a part of the galaxy so rife with a multitude of diverse space-faring alien cultures constantly interacting can survive in space without a comprehensive mapping system for their neck of the woods. It took me nearly until the end of the first season to realize that the show was never going to make a serious attempt at establishing technological boundaries or relative location, both which I felt were essential to the progression of the overarching story. Granted, the true appeal of this show is its wide variety of interesting and weird aliens and the gradually developing troubled romance between John and the Peacekeeper Aeryn rather than a Trek-level dedication to worldbuilding.

I wanted to watch Farscape for inspiration to create aliens for Star Crucible, but I found myself thinking of Beeknighted instead. My current demo for the game does not yet establish much of the world and the player bee doesn't have a strong objective beyond "survive," so the subject has been on my mind. Much of the world map will be traversed by an airship-equivalent vessel called the Skepscog (literally a Beehive ("Skeps") ship ("cog")). How do I firmly establish the limitations and expectations of the world when the humans are using bows and arrows and bees are flying in skyships?

In the coming revised demo, Aife, the bee, will have the chance to meet a variety of other bee characters and learn about their place in the world. There are larger and more dangerous forces at work, the conflict between the humans and the troll, for instance, which will remain unknown for the player to discover. But I hope to better mesh the opening conflict with a more solid introduction to the world to give the story more weight and depth for those looking for it as I do.

Farscape is a harem isekai.

1 note · View note