Hello everyone, I would need help with the pre rendered backgrounds.
Pretty much I would like to realize a small survival horror project similar to the first Resident Evil with the pre rendered backgrounds, could someone give me a hand on how to place the pre rendered background and if it takes depth? I would really appreciate it
1 - if it can be done in lightmass then you can just fix the camera and have everything else rendered in Unreal.
Pros: more or less seamless integration, shadows lights fog etc…
Cons: Lightmass is not Vray.
2 - Pre render outside Unreal and match characters to backgrounds with referenced objects imported into unreal for the environment + camera settings/angle to match perspective etc…
Pros: you can have unlimited control over quality of the backgrounds.
Cons: Matching characters/moving objects to that quality in order to help make the integration seamless, this inlcudes shadows and lights and atmospherics. will prove to be more challenging.
animating environment such as flickering lights or other moving objects will also be a second step of complication.
I don’t understand why you want to do this if you don’t know how?
Using pre-rendered background is not so simple as you expect.
Anyway you need to make some normal level with objects in 3D space and then render it to texture.
And you need to create some invisible collision of your level for your characters, and it should correspond to background.
your camera must use some predefined path for movement, because it must show scene according to that pre-rendered background.
This will be a really big piece of manual work.
Pre-rendered backgrounds used in old games because of performance. Old game engines and hardware can’t perform full 3D scene well.
For now you can create 3D scene with skeletal characters, static meshes and particles even for some old cell phones…
I have always been passionate about the old survival horror especially I love the first Resident Evil and have always been fascinated by the pre rendered background and fixed cameras, that’s why I would love to realize something like that.
In fact I would love to learn this method, now you know the precise reason.
Ok, I’ve got it.
I’ve completed RE3:Nemesis for a 2 times. But I’ve completed RE4 for a 4 times ))
For camera movement you should learn something about splines, So you need to setup many splines in your level to make camera movement along them.
For a pre-rendered backgrounds - start from SceneCaptureCube actor. It will render your 3D scene into cubemap. You can save it to static texture by pressing RMB -> create static texture or something like that.
Then you need some Sphere mesh in your level. You will map that static cubemap on it by specific material.
No harm said, but I think you need to go to the very basics of learning the engine first before anything else, what he said are the very basics. It’s good to have a goal so start a small step at a time, maybe with tutorials out there.
In addition to the above info, you can start by setting up your entire scene in your 3d app plus camera animations, then import them in. For instance camera tilts left slowly then pans right. In the game you can link this animation to when your character goes left or runs to the right etc…
I recommend not using prerendered Backgrounds alá Resident Evil, Dino Crisis, Final Fantasy 7 etc.
The Backgrounds in these games have Z information for overlaps when 3D Objects are further away.Splitting the backgrounds into several layers need a lot of practice, because you have to find the rigth Z Values for this.
The backgrounds also don’t scale. If you have a higher Resolution than the Background you’ll see scaling artifacts. Also what happens if you want to change the environment a bit because you found a bug? You need to change several backgrounds that include this area (e.g. multiple cameras show the same area)
I recommend using a standard Level with 3D Objects and fixed cameras instead. It’s much easier to adjust.
Yes, there will be such complications, however using standard workflow in UE with fixed camera means he’ll probably be stuck using Lightmass and the effect he is looking for may not be achieved that way since ‘maybe’ depending on the style and level of detail he is looking for, a certain aesthetic would be better achieved with offline renderers or even a paint over after a render is done. To find a middle ground he could import the objects as they are in UE but use Offline renderer to bake everything in as texture, then use an unlit material on those objects in UE.
Another technique would be to use projection mapping (commonly used in VFX), this way he’ll retain depth as well.
Later match everything to that for the characters in the engine.
In what sense?
Anyway guys thank you for the advice, Sincerely i would love to learn this thing of the seabed pre rendered because I am fascinated so much, but in fact I understand that it is very but very complex and so you are right on this side.
Don’t worry I have a nostalgic sense about this sort of game type myself, if you have the passion for it keep it up. Honestly it is not far fetched to do it right. there are just so many ways to go about it.
I have tried to achieve the same effect as in the PSX games, using only unity though with somewhat success. The technique ive used should apply to Unreal if you are able to change the rendering order or Z-depth for your objects: 1. characters and npc’s in front 2. static backgrounds
In Unity shaders are used to change the z-depth so that requires to do some script work, if Unreal has a way to script shaders or change them then that could work also.
Here is a thread about that: http://answers.unity3d.com/questions/173752/rendering-order-revisited.html
Would be cool to see more psx style games!