I have noticed that the farther you can see into the distance the more inaccurate the depth buffer is, i.e an object might be 'behind' another but will show through. I was wondering if there is any standard for a frustum length. I am thinking of using 500 to 1000, before I was using 15000, lol. Ok thanks.
I haven't had this, Then again, I haven't been opertaing in
these distances yet. How can you see an object that far away
of being infront of or behind an object if the difference is small?
--
with 100% accuracy the closer objects cull the objects farthest away.
I heard OpenGL's depth buffer can be inaccurate at times. You may want to look into using a BSP Tree instead.
I don't want to do BSP for this simple game Mr. Wizard. I think I am going to end up scaling everything down so all of the depth values produced are accurate. I wrote a worldcraft .map loader (it reads in all of the planes) but the math to find the 3 plane intersection scared me and I gave up (for now). I also don't know how to read .wad texture files. Eventually though I would like to build the .map loader into a BSP compiler that performs constructive solid geometery (adjacent convex polygons combined), reads the .wad texture files, and the hardest part of all the BSP partitioning. I already 'know' how most of it is done, but I know actually getting it to work will be quite difficult.
how many bits did you allocate to the depth buffer in your pixel format?
> I was wondering if there is any standard for a frustum length.
depends on scaling and how far you want to see...
usually depending again on scaling i see 1(usually used for 2D) to 100.
also look into the glPolygonOffset function i hear its often used to alliviate the problems your having..
Last edited by no-one; 02-26-2003 at 09:31 PM.
ADVISORY: This users posts are rated CP-MA, for Mature Audiences only.
originally only 16 bits but i'm experimenting with more. Is there a limit.EDIT: using 32bits instead of 16 seems to have helped.
EDIT1:
I don't entirely get this, could someone explain it to me in english?
This is from MSDN. I understand how to use this(glEnable(GL_POLYGON_OFFSET) then call glPolygonOffset() but I don't understand what it does exactlY)
When GL_POLYGON_OFFSET is enabled, each fragment's depth value will be offset after it is interpolated from the depth values of the appropriate vertices. The value of the offset is factor * Äz + r * units, where Äz is a measurement of the change in depth relative to the screen area of the polygon, and r is the smallest value that is guaranteed to produce a resolvable offset for a given implementation. The offset is added before the depth test is performed and before the value is written into the depth buffer.
Last edited by Silvercord; 02-27-2003 at 12:03 PM.
If it starts to freak out at high variable numbers, Then why not
just divide them all through 2?
--
heres a pretty comprehensive explaination...
http://www.opengl.org/developers/faq...ygonoffset.htm
and i recommend always using a 32 bit depth buffer.
Last edited by no-one; 02-27-2003 at 04:30 PM.
ADVISORY: This users posts are rated CP-MA, for Mature Audiences only.
