Implementing Classic Resident Evil design

Hey there friends! As the title suggests, I'm curious if anyone has a guide/can point me in the right direction for something that may have instruction or a guide to using fixed camera angles on pre-rendered graphics? I'm wanting to do small demo of something like Resident Evil, and while I've seen a post here and there, they're in excess of 5 years old, and not filled with quite enough detail to really get going. I'm still finding my way, but was curious to know if anyone had a general idea of how to begin this, or if GMS was the right software for undertaking something like this?

To my knowledge, A LOT of the 2D principles of collision, actions, etc. carry over, but I'm getting a bit stuck on the 3D/fixed camera part.
 

Yal

šŸ§ *penguin noises*
GMC Elder
I think a good inspiration source for implementation details would be Ocarina of Time, since we've gotten a peek at how it looks behind the scenes:

(timestamped, jump to around 12:15 if it doesn't work)

So what we see here is that there exists a 3D environment, it's just drawn behind the prerendered texture layer so we normally won't see it. Why is it there? I can come up with a few potential reasons:
  • Used for testing the area
  • Used as a guide for what to pre-render and then also again to fit it to the in-game environment
  • Used to hide moving objects when they go behind 3D objects (as in a real-life greenscreen effect) - this tech wasn't used in OoT but it could be useful in Game Maker since shaders could deal with chroma-keying pretty easily (and you overall have less control over draw order and depth sorting, so mixing flat and 3D drawing in the way they did it in OoT could get a bit messy).
 
Thank you for that! Had no idea Ocarina had that same concept! Do you know, as far as something like this video, where the images are prerendered but have collision boundaries, and switch cameras, how that implementation would work? I haver blender and am decent enough with it.


Really any part of the video works.
 
Thank you for that! Had no idea Ocarina had that same concept! Do you know, as far as something like this video, where the images are prerendered but have collision boundaries, and switch cameras, how that implementation would work? I haver blender and am decent enough with it.
No idea exactly how Resident Evil does it, but here's how I would approach the problem.
  • Create the basics of a 3D movement system where all obstacles, walls, etc. are represented by extremely rudimentary textureless geometry
  • Have the camera be its own object that is stationary and only changes position when the player activates certain triggers (e.g. walking off-camera, cutscene)
  • Before the next step, have the game be 100% finishable and feature-complete
  • Make everything invisible except the player, enemies, and interactable objects
  • Create the pre-rendered backdrops for every necessary scene
  • Create multiple layers for scenes that have sections that can display in front of objects. Adjust the depth per-object so that they appear either above or below the object depending on where they are in the scene
 
Top