Blender Scene Creation

[Glendora Starr]

InBlender i am creating my islands and all the static objects with the island (Trees, rocks, folliage, buildings, etc.) and exporting this island & static objects together in 1 FBX file. Then in Unity3d I have my scene already done I don't have to perform the scene creation there, i can focus on developing the interactable objects.

What I understand is that when I export multiple islands that are reusing the same tree for example, in Unity3D even though these trees are the same they are in different fbx's. But this only affects project size afaik?

It depends what you are making. Unity has a lot of tools, specifically environment creation tools that are much faster and more powerful to use in Unity rather than import from blender, especially for laying out a scene and prototyping.

So I have a basic model that I created in blender, joined the objects together then exported the model as a GLTF. This model I am planning on creating multiple duplicates with it being the exact same. No differences just different locations. So I am not sure if clone is better for me in this scenario because I am not really changing anything with the duplicate.

I went through and created my full on project locally and then looked at the scene debug layer and noticed I had 175 draw calls and im only rendering 11 objects on the scene, so I started down the route of instancing or cloning.

Why dont i just make my level/scene fully in blender and then export it fully over to the game engine? Blender can hold large scenes in blender anyway and i see that you can export a level from a game engine to blender in full anyway (example is synty studio maps).

Why are designers creating the assets in a 3d model programme and then modular them, only to export them all and put them back in a game engine. Isnt that wasting time? Why not just create it all in the 3d programme then export it all to the game engine to then move onto the next steps that the game engine does.

But you can still edit it in the game engine. Creators could create the level in the 3d editor, then export it as already laid out into the game engine, then after testing it they can move around the assets, as the models will still be separated anyway.

In regards to my other question. In a game engine you can create a plane/cube and make into a wall/floor etc. Is it better to bring in the floor/wall from the 3d programme or create in game engine. Are they the same in terms of optimisation or better to bring it in from 3d programme.

Hi this doesnt answer the question. Please do be polite. Im asking to all you professionals on here that I inspire and simply asking a question that i dont know. There isnt need to be rude and say i forgot the question mark. Apologies for that but i think we all get what i am asking. Ill await another persons response but thanks for your feedback.

I see speed designs of whole scenes done in blender and in unity/unreal. Only difference is I see unreal and unity they are making in blender or maya then exporting all assets over and having to re position them. Where in blender they are moved around and placed all ready and edited in the same 3d editor. So im simply asking why not create it all, move and edit it all in the 3d programme to then export it all over when ready, then in the game engine do the game engine part?

There is no difference in asset creation and exporting full scenes but there is time wasted if not carefully thought out as you would not want to create a complete forest as a scene and have many duplicates that would be exported when a set of tree assets would be more efficient.

Below is a good example of someone doing all the work in a 3d programme then exporting it to UE for example. All seperate meshes incase in the engine it is needed for changes and the assets can be their own of course. But making it all in 3d programme, all assets and level then bringing it over when needed for the game engine part.

The very first step for the creation of environments is the reference collection along with the concept production. For this project here I decided to recreate the city of Utrecht in the Netherlands, so my first step was to look around on the web for some image references to get a good idea of the environment.

Once we have the base scene ready, we can focus on the fourth step of the environment creation, the set dressing of the scene where I placed some other models to add a further level of realism, so I start to add trees, bricks, stairs, ivy, and decals to fill up the scene and bring a bit of storytelling.

At this point, the scene should look already good, but there are still two key steps to consider.

An important step is the lighting of the scene. Here I placed several point lights and direct lights to define the mood of the scene and enlighten certain elements of the scene. Once I was satisfied with the result, I focused on the latest step of the process, the render settings.

The latest step of the process is the render settings with post-process, where I did the colour correction. I added some effects, vignette and lens distortion and made sure everything was looking the proper way. I did also some tests to see if there were so glitches. Last but not least, once I was satisfied with it, I recorded a simple camera movement and started to render it.

Pietro Chiovaro 3D Artist from Palermo, Italy. He has always been a movie lover and videogames enthusiast. In the past years, he has been focusing his studies on the development and creation of assets, characters and environments. To offer a complete set of skills, he has also developed his competences in art direction, game design and direction.

Working as a freelancer, he developed deep experience in many aspects of CG production from architecture works, to VFX effects and games development.

3D Expert of the 3D World & 3D Artist Magazine since 2015, every month Pietro shares his knowledge and helps other artists, giving answers to their questions and showing them the solution to their issues.

3a8082e126