RE: screen clear

New Message Reply Date view Thread view Subject view Author view

Gullen Jim (Jim.Gullen++at++GSC.GTE.Com)
Fri, 11 Apr 1997 09:03:00 -0700


Bryan Wasileski writes,

> "... From the
> results, it appears that I can clear up to 22 channels at 60Hz and beyond
> that it drops to 30Hz. This test was performed with a 2cpu R10k, 1 RM-7."

Assuming your 22 channels are same-sized, non-overlapping and cover
an entire 1280x1024 screen,

    1) The clear time per channel would be

        CT1 = 16.7/ n1;
                 n1= number of times you can clear 1 chan in 16.7 msec

    2) The overhead per channel would be

        OH1 = CT2 - CT1 or OH1 = CT22 - CT1
                CT2 = 16.7/n2
                CT22 = 16.7/n22
                n2 = number of times you can clear 2 chans in 16.7 msec
                n22 = number of times you can clear 22 chans in 16.7 msec

I strongly suggest you employ "eyeball" timing instead of the stats
display
for this type of "gross" timing. Just put the viewpoint into a heading
loop of 5
deg/loop and manually increment the number of clears until ...

Of other interest the clear times on our RE2 run 4-6 msec (normal) down
to
2.1 msec (fast tag) for 1 channel spanning an entire 1280x1024 screen.

Jim Gullen
GTE Government Systems
1805 W. Drake Dr.
Tempe, Az, 85283
voice: 602/777-1724; email: jim.gullen++at++gsc.gte.com
=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
            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 - 17:55:03 PDT

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