Greetings, medievalists!
Let's talk about the camera, one more time.
You'll probably like the fact that we'll be ditching coyote time with the next update. Game development is a constant process of evolution and iteration and sometimes we have to try one solution before we figure out the next step of problem solving. Previously, we've explained the new changes that we were introducing to the camera system, but after playing with it some more and tightening some of the behavior, it was evident that coyote time will not be needed. How did we get there?
So, we were trying to figure out the system that would take all those specific scenarios that we mentioned in the last MMT - how should the camera handle small terrain elevation & demotion? how should it behave with beams? etc. We had to revisit our raycast.
In Unity, a raycast is a way to detect objects along an invisible line (or "ray") that is projected into the scene from a specific point and direction. It’s commonly used in gameplay mechanics like detecting what the player is looking at, shooting, collision detection, or AI vision. For the purpose of our camera - that's the center of the screen. If you turn on "camera target visuals" in Camera Options, you'll see exactly that (white dot).
Note how the camera goes up/down depending if the center encounters elevation/hole.
But now we'll introduce 13 raycasts to camera detection. Here is how that we'll work in theory (note that you will not be seeing 13 raycasts in-game and this is just for demonstration purposes):
So, now the camera jumping will happen when 13 raycasters are on the new ground level. What this means is that you'll essentially be able to pass through elevations, walls and holes shaped as 3m circles (diameter of 4 voxels) and the camera will not jump.
There are some other tricks, too. For instance, let's say that you want to jump from layer 5 to layer 4, but the terrain there is uneven - there are some holes and/or elevated parts, because the land can be terraformed. In those cases, the camera will calculate existing raycasters and at least 8 out of those 13 should be at the same level in order for the camera to ease onto the new level.
We’ll be introducing a couple more conditions to avoid unnecessary jumping. The camera will not be offset by beams anymore. In addition to that, walls that are displayed in half (due to being at .5 level) will not trigger camera offset.
Layer changing will still be done via Ctrl + mouse scroll, while Camera layer locking and scrolling will be done via Shift + mouse scroll.
We’re aware that many of you think that keeping the camera locked by default should be a go to solution. We were experimenting with that, but keep in mind that we want this game to feel welcoming for both old and new players. People that never played our game before were a bit confused by the layout in that case (the results were similar to the solution 2 in MMT 43). It added unnecessary friction.
That being said, we think that this multi raycast system fixes a lot of problems, while camera locking will provide additional quality of life improvement to those that seek it.
We can’t wait for you to try the new camera and let us know how it feels. There are more MMTs to come, about other, long awaited features. Talk to you about that next week. Until then…
Stay medieval!