I was wondering that when you have a huge scene and if you make any changes into lighting or anything which makes you rebuild lighting which becomes heavy for machine especially when you have a huge scene like Elemental Demo, ParticleEffects or Infiltrator demo …so what I was thinking UE4 must invent a new tool (Build Selected Area Only tool) which will provide build options to selected area only or it could be like providing a one wireframe box wherever you put it only that particular area will build…the reason is when you make changes to particular area why you would like to spend build time on area that you are not looking at…apart from that building huge scene like ParticleEffects, Elemental Demo takes huge time sooo I guess this tool must be there so if we make any changes to particular area we can see results in less time.
That wouldn’t work, the lighting in that area is affected by much more than what’s in that area and the only way to know what things will affect it would be to build the lighting for the entire thing.
I guess I should put it in this way …when I say rendering particular area I don’t mean to build it for Play in Editor I know it won’t work for that but just to take a look when you are editing a particular part of your level simply whatever is in screen …for that purpose that tool should be there or it should be like Maya whatever is in the camera view it should render it so we can see the rendered image and then saving it to compare it for another render image. THE WHOLE PURPOSE OF THIS IDEA OR TOOL WOULD BE JUST TO TAKE A LOOK OF CHANGES THAT YOU MAKE TO A PARTICULAR AREA instead building whole scene just render it as image to see how it looks (it just should be like Maya render output window) its gonna be helpful ALOT CAUSE EVERYBODY DOESN’T HAVE Nvidia 980 GTX or high end computer. I am using Quadro 2000 and it works pretty awesome but still I think it will be good from CPU loadline point of view.
Which are you looking for–rendering a region of the camera image or building a region of the level? Those are completely different things. There’s little use for the ability to render a camera region(like Maya) even with a low-end graphics card it renders the full image very quickly since the engine is optimized for that purpose. The thing that takes a long time is baking the lighting which you can’t do by a level region as you were illustrating in the first post.
I guess you are saying level Streaming when you say building a region of the level…but thats totally different obviously I am talking about the changes that we make in particular part of your level (which is obviously gonna be in camera view) no matter its lighting or its color or it could be anything regarding changes but to see it in final build quality all I wanna say why do I need to build the whole scene I could just render that image and see it as image (Camera Region) that will save lot of time …I am putting all these point because when I was building “Particle Effects” project from unreal marketplace it was taking long time sooo definitely rendering camera region to take a look how that particular part looks in final view will be helpful for sure…another point is sometimes you might put your Qualitysettings to medium or low (not all the time epic when scene is heavy) so at that time you can render image to see how it finally looks but as image.
Forget that wireframe that I have shown in above image but imagine only that scene which is in that camera view and instead building whole scene just render that camera view…cause obviously I am working on that area which is in camera view only. This is just gonna be helpful thats all I know
No, not level streaming, that’s completely different.
What you want is to avoid rebuilding the lighting when you change an object, and you’re getting confused with building lighting vs. the camera image. When you render something in Maya for instance it only cares about what the camera sees and it has lighting rendering that work with that. But for games that doesn’t go fast enough so you have to find other ways of rendering high-quality lighting and it does that by baking the lighting to lightmaps beforehand. When you bake lighting what the camera sees is irrelevant to it building the lighting. When you change an object in the scene the lighting for that object and the things around it can be affected by many things and it can’t figure out what those things are until it does the lighting calculations.
