Hi, I see some amazing characters modeled and rendered in Blender. I was wondering if they look as good when imported to Unity i.e. for the models to look as good as in Unity, should Unity be linked to Blender's rendering engine or would Unity's rendering engine be sufficient?
For example, a character with a lot of fur that is superbly rendered in Blender be imported to the same quality into Unity as well?
Unity can't be 'linked' to Blender's render engine.
An image rendered in Blender with cycles can be imported into Unity.
The model and the shaders, textures, etc., can be imported to Unity, and then plugged in to the relevant slots.
One thing to bear in mind, Unity is a game engine, therefore a real time renderer, similar to Eevee, whereas Blender has Cycles, a physically-based path tracer, for realistic looking renders. Real time renderers estimate the way light interacts with objects for speed of rendering.
So you will not get the same quality render in Unity.
For a character with a lot of fur in Blender, will be using the hair option in the physics settings. This cannot be exported to Unity. The hair would need to be converted to a mesh object first.
You can create a fur shader in Unity, with the how-to knowledge. Also there are lots of fur/hair shaders available in Unity's asset store.
In conclusion:
No they would not look as good in Unity.
Thank you AAdrian for the explanation. A follow up question is: would it be the same case if I were to use the Blender's Game Engine instead of Unity i.e. if I stay "within" Blender for not just modeling, texturing, and rigging, but also to drive interactivity that is normally done in Unity, would I get to keep the amazing looking characters while defining their behaviors using Blender's Game Engine (I honestly have no idea how capable Blender's Game Engine is but just as a thought!).
Blender no longer has a game engine.
You can animate actions in Blender, then export the animation into Unity, then have the animation play with a button press or joystick movement. However this doesn't solve the difference with the render engine.
The answer I think you are looking for is with the texture maps. To have amazing looking characters in Unity, you need to have a good quality game ready model, and probably at least 4k texture maps, preferably 8k.
The model will need to be a low poly mesh with normals maps baked from a high poly (sculpt)version, normal map again at least 4k.
There is a lot to consider, and work involved, in producing good quality game ready models.
Any real time application has to take the necessary short-cuts to render things in real time. That being said, there is a lot of technical wizardry you can learn to get some astoundingly good looking real time renders. But a game engine fundamentally cannot spend minutes calculating hair particle positions and volumetric lighting data for a single frame. It only gets 1/60th part of a second to do its calculations. It simply isn't possible in that amount of time to calculate a simulation of all the physics and optics that are happening in the real world, but again, you can get some pretty believable approximations if you know what you're doing.
The application you're using doesn't alter that, be it Blender or Unity or any other game engine. What specifically you could get away with would depend entirely on the nature of the scene, so there isn't a definite answer to the question. You certainly can render creatures with fur in a game engine, and you can do it well, but not as precisely as you could do it if you were giving cycles hours of render time to calculate everything.
