#difficult things

(reminder that if anybody wants me to stop sending these asks at any time, just tell me and I will stop!)

what is the thing that you are the most passionate about in life and why? what sparked your interest in it? is their a goal with it you want to reach, and are you actively working on your passion, or is it just something you know a lot about?

The dignity of a human individual. I lost my freedom to make decisions and even think independently for a solid 10 years, and combined with my near brushes and continual fights with death and medical emergencies and disabilities, I believe that people need to be able to enjoy and meditate on their lives fully. I love seeing people in motion because I can't move much, and whenever they come to me to talk about their feelings, I feel really trusted. I want to respect other people and give voice to the things they know but can't find the words for. The glory of God is a human fully alive, to quote Irenaus. Oddly enough, I'm also very passionate for other people's right to make wrong decisions, especially when they concern their own bodies and destinies. For this reason, I've become interested in ethics as separate from theology, and made the very hard journey to have to admit to myself that even if abortion is wrong (no position taken here: my Christian followers mostly could beat up on me if I said it wasnt, and my non-Christian etc ones if I said it was), I need it to be legal, and mostly because so many women are experiencing life-threatening health complications due to lack of access. I truly thought before Roe was overturned that states should have the right to ban abortion and that it was exaggerating to say women would be denied care for miscarriages etc. But now enough women have died or had to have their reproductive organs removed after days without care or been denied even ultrasounds to figure out what's going on when they're bleeding during pregnancy because providers are afraid of being sued on either side because of choosing or not choosing to follow up with a procedure . . . anyway, there are so many other reasons. . . . 6-7 reasons generally are involved in any conclusion I reach . . . And yeah, I am super afraid that people are going to make me feel bad about whatever conclusion I currently am working through, but, yeah, being alone and having been trained to use logic with my conscience from my earliest memories, I think about ethics all day long, whether it's embodied in me, art, or other people, and I care so much about helping other people to be safe.

There is more to it than that. When we build game systems, one common method of differentiating the work is "things we probably will change" (e.g. things we plan to tune based on feedback) and "things we probably won't change" (e.g. the way a story character or environment looks). We build the things we expect to change such that it is easy to make those changes. The things that we don't expect to change are typically not built with changes in mind, because we assume we will lock them in and they'll be completely finished.

As you might expect, numerical balance values like 10% higher stun duration, 8% less HP, or an extra 5% critical hit rate are often built to be easy to change because we expect to have a lot of things in the game that can change them. A magic blessing spell, a potion that provides a temporary boost, a passive racial ability, or a piece of gear all might provide different modifiers to a character's health total and we're probably going to want to tune most of those to make them feel right for their power level and the player experience. Since we expect designers to modify these values a lot, the engineers build the system to make it easy to set and modify those values.

Animations are typically not built to be easy to modify because we don't need to do it all that often. Most of the time, the character performs the animation and we call it good. We might tweak and modify it leading up to launch, but they are generally locked in once we finalize them. Most games do not need to tune a particular animation beyond being able to scale the animation playback speed to fit within a certain number of frames. Tuning would generally require the animator to make a small change, the designer to play with it and decide if it is good enough, and then make changes if it isn't good enough. This cycle can take a significant amount of time, and we often have [huge numbers of animations] we need to complete. This kind of iteration often takes too long to do feasibly.

There are exceptions to this, of course. Fighting games, for example, tend to make animation changes regularly because of the way they are built. In addition to damage and health, fighting game strength are all about the character animation start-up, the active duration, and recovery period. A particular move can be buffed by shortening its recovery time or nerfed by increasing a move's startup duration. Because the designers need to make changes to these animations regularly, the engineers build the animation system to be tunable so that the designer doesn't need to keep calling on the animators, and the animators can instead focus their time on creating new animation suites for new characters.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Those arguments don't really hold water at all. Games are more than just the sum of their parts. Making all of the pieces work together cohesively is a huge amount of work. Every game is a complex web of features and content. Any given feature is dependent on a bevy of other systems and features and any failures can easily cascade into catastrophe. One of my acquaintances works on the Call of Duty franchise - she told me that they often have difficulty porting features between the single and multiplayer versions of the game due to the way the technology is set up. It can take days of engineering and scripting to bring a feature from DMZ to regular multiplayer or another game mode.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

The concept of physicality doesn’t actually exist in graphics. There’s no such thing as a solid object in CG - only the visual representation of something. CG models don’t have mass, skin, weight, internal organs, muscle, fat, or actually anything of substance. They’re essentially hollow 3D wireframe skeletons with a bunch of 2D textures stretched over the outside and a bunch of lighting calculations to determine what they should look like. This means that there is no such thing as “clothing” on top of skin, there is only different parts of the same model. The character’s clothing is just as much a part of the body as the skin or the hair. A CG model can’t normally remove clothes because there are no clothes to remove - the part that looks like clothing is actually just another part of the model’s body. Even when the CG model’s “clothing” animates, it’s not a separate piece - it’s the same thing that we do to animate arms, legs, hair, fingers, eyes, or feet. The model’s clothing has the same animation skeleton driving it as the rest of the body.

In order to simulate removing of clothing, a lot of additional work actually has to be done. This is like trying to choreograph a very precise fight scene between actors and get them to “interact” but way more precise and complicated because it’s in 3D and they cannot touch each other. Instead of building just a character model, we would need to create and animate at least two separate models - the jacket and the jacketless character. Then we’d need to animate the jacket on the character such that it inherited the proper motion from the character and the entire process of the jacket being taken off by the character. This means synchronizing two separate and precise animations on two separate models that are located at different positions in 3D space, all with zero physicality. Both motions - the character and the jacket - must be precisely synchronized in order for it to work. If there’s any timing, angular, or spatial differences, things won’t look right. 

As you may have gathered, this is an awful lot of work for one single shot in one cutscene. Instead of doing it the hard way, we often use visual tricks like having two models - a jacketed model and an unjacketed model. Then we can have the character walk offscreen, show a close-up of a separate jacket model being put somewhere, and then model swap the character to an unjacketed version instead for the rest of the scene. Doing this kind of off-camera swap is actually significantly less work than it is showing the jacket-wearing character take the jacket off. Characters physically and visibly touching other characters or objects realistically in CG is one of the most expensive things in CG. This is why the entire game dev world was floored when we saw Ellie and Dina kiss in the Last of Us 2 E3 trailer - making that look good is really really hard.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr