Hi guys,
(@dale @wim)
it seems strange to me that Rhino isn’t using all cores to (render)mesh houndreds of objects at the same time. Are there any good reason for this?
Cheers.
Update: I did a count on the file and it is 69.000 objects 
But that makes it even more important to use the whole CPU 
(I am bugtracking renderspeed for a customer, the file is a 175MB (savesmall) Ship with lots of stuff. Some bad objects, but those isolatet meshed quickly. File in mm and default mesh settings)
There is some commented out code in core Rhino which shows that we did try to use all of the cores in the background meshing operation. Maybe @jeff remembers why this is not enabled. I could guess that there were some stability issues, but that’s only a guess.
1 Like
I added it years ago…but the demand just wasn’t there… Rhino saves the render meshes to the file most of the time…unless you purposely select “Save Small”… which means that Rhino having to re-mesh “everything” is a rare case… Either when you run “ClearAllMeshes”, or change the mesh settings, or as I said “SaveSmall” … but once the meshes are created, having to do it all again is not a typical case… So I commented the whole chunk of code out…and haven’t really heard anything since… until now.
I’ll take a look at it again…but will most likely hide it behind a test command rather than just let if fly freely out in the real world 
-J
2 Likes
fwiw I can give one pretty common use case for people working with rendering: when you select multiple objects and use object properties > custom mesh to adjust meshes for them.
Yes, that too JD … but again, once it’s done it’s done… Unless you’re constantly going in and doing that over and over again…
Note: I’m not going to turn this on in 6.x … it will be a test command in 7.x. I just don’t want to add any kind of instability potential to V6 at this point.
-J
2 Likes
Hi Jeff,
FWIW we are working with large files shared over GoogleDrive
Saving Small is substantially reducing upload and download times so it would be a welcome addition to have all cores creating the meshes.
Also importing files ( dwg iges step) all require mesing and having that sped up would really reduce waiting times.
-Willem
There is some commented out code in core Rhino which shows that we did try to use all of the cores in the background meshing operation. Maybe @jeff remembers why this is not enabled. I could guess that there were some stability issues, but that’s only a guess.