Re: Incorrect perfly depth complexity calculation

New Message Reply Date view Thread view Subject view Author view

Sharon Clay (src++at++rose.asd.sgi.com)
Wed, 22 May 1996 09:09:19 -0700

+>---- On May 22, 1:12pm, Thomas Meier wrote:
> Subject: Incorrect perfly depth complexity calculation
->From guest++at++holodeck.csd.sgi.com Wed May 22 05:57:29 1996
->Date: Wed, 22 May 1996 13:12:55 +0200 (MET DST)
->From: Thomas Meier <thomas++at++yin>
->Subject: Incorrect perfly depth complexity calculation
->To: SGI Performer Maillist <info-performer++at++sgi.com>
->
->Has anybody already experienced that the average depth complexity
->calculation used by perfly (Performer 2.0) is probably not correct?
->
->We looked into the avg.d.c. in a real world database used for driving
->simulation at different locations using a lot of dynamic models.
->At some places the avg.d.c. seems to be calculated correctly.
->But everytime you look at parts of the database where the local depth
->complexity is (definetly) greater than 7 you get strange results. It looks
->to me as if the local depth complexities are calculated modulo 8 (?!)

The depth complexity calculations make use of the counts kept in the
stencil buffer. By default, you have only 3 bits (this is because on
some low end machines the stencil bits are subtraced from the Z buffer).
Perfly is using only 3 bits.
If you have a scene with higher depth complexity, you'll want to
use additional bits - but the most we use is 4.
You can put in your window configuration callback (in generic.c),
before opening the
window:
        pfStatsHwAttr(PFSTATSHW_FILL_DCBITS, 4).

Four bits is the most we use. You might also look at the
/usr/share/Performer/src/pguide/libpr/fillstats.c example that
does this as well.

src.

-- 
-----{-----{---++at++   -----{----{---++at++   -----{----{---++at++   -----{----{---++at++
Sharon Rose Clay (Fischler) - Silicon Graphics, Advanced Systems Dev.
src++at++sgi.com  (415) 933 - 1002  FAX: (415) 965 - 2658  MS 8U-590
-----{-----{---++at++   -----{----{---++at++   -----{----{---++at++   -----{----{---++at++


=======================================================================
List Archives, FAQ, FTP:  http://www.sgi.com/Technology/Performer.html
            Submissions:  info-performer++at++sgi.com
        Admin. requests:  info-performer-request++at++sgi.com

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:52:55 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.