Render mesh settings

Hi guys,

my firm is thinking about getting new hardware for Rhino. And now I am reading through the Rhino v6 WIP NVIDIA Quadro Test: Graphic Card Test
I am trying to compare the testing files with the ones we mostly use and don’t understand or can’t find - render mesh: 10 million polygons approx.

Thanks a lot in advance.

You can probably create one easily, make a sphere in shaded mode, change your meshing settings to something relatively fine, and then look in Properties to see how many polygons it has. Then just array the spheres in sufficient numbers so that the total polygons reaches 10m.

I am not sure, if I understood the last part of your advice.
I did it as follows: I drew a sphere as you said and then checked out the details. It consists of 10.352 polygons.
Then I checked out the number of meshes, that our drawings usually have and multiplied both.
So 10.532 polygons x 8.298 meshes = 61.471.584 = approx. 60.000.000 polygons in total.

Because when I arrayed the spheres, marked them and then checked out the details, instead of getting a total number, I just got a list of each sphere.

Once you’ve created your test scene use the command _PolygonCount, it’ll give you the total amount of polys.

Note that for huge scenes with recurring objects it makes sense to consider the usage of block instances. It’ll still be the same amount of polygons on screen, but rendering those to the screen will be much more optimized. Especially the scene I post would benefit from it a lot, both in file size and rendering.

Thank you for your response.
So I marked my hole drawing (meshes, extrusions, surf. etc.) and used the command _PolygonCount. It consists of approx. 1.7 million polygons.
When I only mark the meshes I get 1.5 million polygons. So if it says render mesh, does’t it mean only the meshes? Or the other objects as well?

The other objects as well. Polysurfaces, extrusions and so on are all converted to render meshes for drawing purposes (in Rendered, Shaded and Raytraced mode at least).

edit: as an aside on the benefits of block instances:

The file is only 15MB in size, but on screen there are a little over 73million polygons. Compared to a file with 68million polygons without the (heavy) use of block instances at close to 2GB in size:


But that of course depends on your scene, I understand if you can’t use these (: (in which case ignore this part of the post).

Well, I assumed this was just for testing video card performance with a certain number of polygons… As you have one object with about 10K polygons, having 1,000 of them will make around 10 million polygons total in the scene. So a 10 x 10 x 10 array of the spheres should do it.

That makes sense!
Thank you!