So I was just wondering what would be the amount of faces when you create a characters or building or items(sofa,desk,gun)
I am new to blender and I am just confused.
I followed someones tutorials and he has like 300k faces.
Could someone explain me what is important and what is the amount that should get when creating the things I listed above.
Thank you.
These polycounts usually are from people trying to make something look smooth, and just cranking up the subdivide modifier until it looks like what they are after…
Pretty much you need as little faces to make something look smooth… some areas (planar faces in particular) you don’t need a high poly count on.
That clock, should never have been 300,000 faces… that is just ridiculous. If this clock was in the background for one of my shots, it would be max 300 faces. But that being said, if there were some extreme close up shots needed of the clock that its understandable (think avengers credits type close up, not like the render that is provided)
Moved from “General Forums > Blender and CG Discussions” to “Support > Modeling”
Yeah, sorry about that.
Didn’t know where to post it.
Thank you for the answer.
These ore object dependent. While “object” being the task it is made for. If your going for still renders you main focus will be the highlight of your render (the main character on a vanity shot). So this character “could” have a high poly count. And if the background has some scenery or props, ideally these probs should be more optimized meaning a lower poly count as well as more simple materials. if there is depth of field , and they are going to be blurred then lower the poly count even more.
For animation the same principles apply but with some optimization controls. When it comes to animation if the main subject is animated first and foremost thing must be good deformation, so using a high poly count is not a priority. The second most important thing in animation is render time. For efficient rendering and meeting a deadline, you must use optimum number of faces. Because each object in the scene is put in a geometry pool and from this pool all the mesh info is used. the bigger the pool the longer the processing time.
So if your model looks good with 10k polys dont bother going for a 100k model it serves no purpose.
If your model deforms good and render time checks out then youre fine.
The face counts for game models are a different issue. Normally definition of a “face” can be used for a triangle, if you mean a 4 sided face be sure you are meaning a polygonal face (poly) when talking about gaming. Since games are rendered in real-time the conditions for animations apply here. The mesh must have good deformations if its animated , then must have the most possible optimized mesh and must play fast on real-time playback.
In modern day pc-console gaming the average model of a player character can vary between 12k - 20k triangles. On games which draw more characters on screen this number drops to half (6k-10k). If your making a game like for example total war-shogun2 , mount and blade or dota2 then this number is halved once more.(3k-5k). But these are just examples to give you an idea of the numbers, there are always exceptions to some models and to some games.And dont forget these are triangles not polys. The reason for that is every model you put on a game engine is converted to triangles for the graphics card to process it faster.
Thank you for the answer. I am using blender for gaming, right now all I want to do is something really simplistic but in a few months I am going to need to create more complex characters for more serious 3D games.
I find the gaming part really interesting and if you would like to explain more it would be great.
Idk why but I mostly like to read through peoples replies then look for documentations.
—snip----
