Home   Help Search Login Register  

Author Topic: Polygon Count  (Read 4673 times)

0 Members and 3 Guests are viewing this topic.

Offline Planck

  • Honoured
  • Former Staff
  • ****
  • I'm never wrong ....I'm just not always right !
Re:Polygon Count
« Reply #15 on: 01 Feb 2003, 19:47:24 »
Actually Polygon means 'many sided', so, a polygon is a triangle, square, rectangle, hexagon octagon, dodecahedron, duodecahedron, etc etc.

Any plane face with more than two sides in fact.


Planck
I know a little about a lot, and a lot about a little.

Offline KTottE

  • Former Staff
  • ****
Re:Polygon Count
« Reply #16 on: 01 Feb 2003, 22:38:53 »
Yes, but speaking in computer graphic terms, a polygon has three or four points/vertexes (sides), because more than that would be extremely difficult to program.
"Life is not a journey to the grave with the intention of arriving safely in a pretty and well preserved body, but rather to skid in broadside, thoroughly used up, totally worn out, and loudly proclaiming 'WOW What a Ride!'"

earl_laamanen

  • Guest
Re:Polygon Count
« Reply #17 on: 01 Feb 2003, 23:06:07 »
Regardless of what O2 says - I think video card performance is always based on triangles - and I always give poly counts based on # of triangles, not 02's numbers.  3dsmax does the same thing - they call it a 'poly count' but it's really a triangle count.

If someone can correct me, that would be great!  Then I could tell you that my weapon poly counts are much lower...

suchey

  • Guest
Re:Polygon Count
« Reply #18 on: 04 Feb 2003, 15:42:10 »
I would be interested in polly count and how it effects performance as well...I propose that the only way to come to a standard is to do independant testing.  I think the best way to accomplish this would be to pick 1 addon from each category, weapon, soldier, vehicle as the standard text subject...then a variety of people use the addon and record framerates under the following conditions:

addon placed in singular spot on a map across all tests
1 unit added (besides self)
increase units by 5 up to an arbitrary max
report framerates and hardware used across the test

This is the only way we will be able to come to some sort of real value.  This would wind up giving us a good idea of performance as it relates to user hardware and a good point for basing polly count on.  The more testing done the better the resulting value.

Offline KTottE

  • Former Staff
  • ****
Re:Polygon Count
« Reply #19 on: 04 Feb 2003, 19:18:45 »
Regardless of what O2 says - I think video card performance is always based on triangles - and I always give poly counts based on # of triangles, not 02's numbers.  3dsmax does the same thing - they call it a 'poly count' but it's really a triangle count.

If someone can correct me, that would be great!  Then I could tell you that my weapon poly counts are much lower...

Yes, graphic card performance is measured in number of drawn triangles/second, but they do state that it is triangles.

I.E they don't say "We can draw X million polygons per second" they say "We can draw X million triangles per second".

Most of the time, in both games and 3D applications, a triangle is the smallest piece available (some can have a single dot/vertice/vertex/point as well) and then there's the square polygon, which is made up out of two triangles.

As it's really diverse, there's no "one way" of doing it.
My personal recommendation however is always look at the largest number, I.E the triangles, and then you won't have to redo the whole thing due to high polycount.

"Life is not a journey to the grave with the intention of arriving safely in a pretty and well preserved body, but rather to skid in broadside, thoroughly used up, totally worn out, and loudly proclaiming 'WOW What a Ride!'"