Last week’s estimates proved to be overly optimistic. The animation system, which I had hoped and expected would only take two or three days, ended up absorbing all of my effort this week. The good news is that it all basically works now. It doesn’t work especially well yet, but it’s starting to look about as good as it did with the old animation system, and hopefully will situate me well for making some other subtle improvements.
Polish is usually something that comes at the end of the project, but I find myself dedicating a lot of work and energy now to making sure that these relatively primitive prototype animations work as well as possible. This is in the interests of having to re-do as little work as possible in the long run, though I suppose it may also be the case that my priorities are just out of whack. Now, most of the animations so far look really solid: A few edge cases look ugly right now, but those problems can mostly be easily fixed by using existing animations to transition out of things more smoothly. However, I’m becoming dissatisfied with one of the air attacks: The falling attack, to be precise.
Here’s the animation as it is now:
Now that’s all well and good as far as it goes. I actually like the way the falling attack looks and feels. My concerns are two-fold, and tied to each other: First, the recovery looks really shitty right now. It transitions directly into the standing animation with no recovery frames. This attack would actually transition very naturally into the crouching attack recovery frames, but then the player would be unexpectedly locked into a crouch at the end of every air attack, which would probably be kind of annoying. Secondly, it feels very weird to be locked into a single attack for the entire descent, regardless of height. Both of these are issues of recovering from the attack elegantly, and both I think have a common solution: Add a standard air melee attack, with a more predictable recovery frame, and make the descending attack a special input of down+attack. This will remove the frustration of locking the controls, this will make it transition naturally into the crouching attack recovery frames.
The one major down-side of this decision is, of course, that it means I need to create another attack animation. I think it’s worthwhile, but I must admit it’s a bit disappointing after thinking of myself as being basically done with chapter 1 character animations.
So here’s what I’m thinking for this week: First, spend a day or two polishing up the current animations, making sure they transition nicely and that there aren’t any unexpected problems. Also spend some time in here going over the collision code, which has gotten glitchier since I refactored it, and try to make it perfect. At that point, I can start in on either the new attack animation or creating and testing an entity manager. Hopefully, again, if all goes smoothly with animation and collision, then everything will be in place to create a short video showcasing the character animations and environment system, such as it is thus far.
Not as much progress as I wanted out of this week, but progress still. As long as I keep moving forward, eventually I will find I have arrived at my destination.