Announcement

Collapse
No announcement yet.

Display Speed of Quads vs. Triangles

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Display Speed of Quads vs. Triangles

    some time ago a friend told me that max is much faster when displaying triangles then quads.

    what he told me is that when using a meshsmooth all the triangles are converted to quads and graphicards are slower in displaying quads then triangles.

    he told me that i have to put a mesh select modifier on top of the stack, that will convert the quads back to triangles.

    i could confirm this behavior with max 4.
    my viewport was much faster when doing this.

    my question is, is this solved with max 6?
    should be easy to tweak the viewport output for the graphicard, not?

  • #2
    No, this is still the issue, and most likely will be for a long time. Its not about quads and triangles, its about poly object class and mesh object class editing. Poly objects contain and preserve a huge ammount of information compared to the ediatable mesh, hence the slowdown.
    Seems this information is then cached or something similar when adding a mesh select on top.
    As to why the meshsmooth modifier acts the same way as the poly object, I haven't the foggiest. True, it does store information about the different levels of iteration, but if one doesn't utilize the sub object controlls in the meshsmooth modifier, it shouldn't really take them into account.
    Signing out,
    Christian

    Comment


    • #3
      displaying triangles is much more easy for the graphics card. try using directx9 or maxtreme mode with triangle strips.
      incredible speed but a lot of "edge garbage".
      Marc Lorenz
      ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
      www.marclorenz.com
      www.facebook.com/marclorenzvisualization

      Comment


      • #4
        unfortunaly, DISPLAYING is not the only problem. another one is MANIPULATING polygons. A simple meshsmoothed box with some iterations will noticeably reduce performance, and it's not due to videocard, since all modifications go through CPU, but because of poorly optimized algorythms, that were not changed for years. For examle, in XSI you can subdivide a detailed character and then work with source cage without any lags in smoothed mesh.

        Comment

        Working...
        X