Artur,
It is a long time since I studied what is being done with hidden surface removal but,
if your card is doing a simple scanline hidden surface removal then your program seems
like a
worst-case nightmare for it!
If the algorithm sorts polygons into z order and they don't intersect (in z) then it
should be no problem.
However if they all occupy the same coordinates (in particular depth) then they all
intersect for most
of the polygons (15/16th) and for most of the screen/view!
Let's assume your graphics card is doing a scanline hidden-surface removal, your
screen resolutuion is 1024x1024 and we have 100% overlap (nearly, in your case) and
they fill your view - "every square took entire screen". Then we are making the
renderer do
hidden surface checks 64 times per pixel to see which polygon is in front. This would
be over 64,000,000 operations.
So it may not be fill rate bound but compute bound. I suppose it would depend on your
card a lot.
Have you tried to run it at both very low resolution setting and a very high rez
setting? There should be a big difference.
Chris
Artur Biesiadowski wrote:
> I wanted to display a plane composed of 8x8 adjacent textured squares. I
> was surprised to see the performance - about 20 fps (only 64 textured
> quads there, no lights) !!!
>
> I've checked the program - turned out the texture, same problem. Then I
> discovered that by mistake I have not places squares in correct places -
> they are 16x16 units wide and I placed them in 1 distance instead of 16
> distance. This meant that they overlapped a lot. When I corrected this,
> fps jumped up to 100+fps.
>
> This solved my problem, but I would like to know what happened ? Is
> drawing polygons in same position really so bad ? Or was that just the
> problem of fill rate (every square took entire screen, so I suppose that
> it could be redrawn 64 times) ? But shouldn't graphic card detect that
> this area was already painted ?
>
> Artur
>
> ===========================================================================
> To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
> of the message "signoff JAVA3D-INTEREST". For general help, send email to
> [EMAIL PROTECTED] and include in the body of the message "help".
----------------------------------------------------------------------------
----------------------------------------------------------------------------
This Email may contain confidential and/or privileged information and is
intended solely for the addressee(s) named. If you have received this
information in error, or are advised that you have been posted this Email by
accident, please notify the sender by return Email, do not redistribute it,
delete the Email and keep no copies.
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JAVA3D-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
