home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!spool.mu.edu!sgiblab!nec-gw!nec-tyo!wnoc-tyo-news!scslwide!wsgw!wsservra!onoe
- From: hasty@netcom.com (Amancio Hasty Jr)
- Newsgroups: fj.mail-lists.x-window
- Subject: question: xbench's screencopy and expose events
- Message-ID: <1993Jan2.081816.6128@sm.sony.co.jp>
- Date: 2 Jan 93 08:18:16 GMT
- Sender: onoe@sm.sony.co.jp (Atsushi Onoe)
- Distribution: fj
- Organization: Netcom Online Communications Services (408-241-9760 login: guest)
- Lines: 43
- Approved: michael@sm.sony.co.jp
-
- Date: 2 Jan 93 06:05:13 GMT
- Message-Id: <1993Jan2.060513.17119@netcom.com>
- Newsgroups: comp.windows.x
- Sender: xpert-request@expo.lcs.mit.edu
-
- Hi,
-
- I am working on an X server for S3 chipsets for 386bsd and Linux and
- am trying to determine if the xbench screencopy test should cause the
- X server to generate NoExposure events. The NoExposure events seems
- to cause xbench to grow considerably in size thus skewing my performance
- benchmarks.
-
- If I set noexposures delivery in the screen setup routine the problem
- of xbench growing in size goes away. The following modification was
- done to xbench's screen copy setup routine:
-
- screencopy_setup(dpy, win, dummy)
- Display *dpy;
- Window win;
- {
- int screen = DefaultScreen(dpy);
-
- myGC = XCreateGC(dpy, win, 0L, NULL);
- if (! myGC) return 1;
- XSetForeground(dpy, myGC, WhitePixel(dpy, screen));
- XSetBackground(dpy, myGC, BlackPixel(dpy, screen));
- >>added XSetGraphicsExposures(dpy, myGC, False);
-
-
- The following xbench results were generated without the above patch.
- So far I am getting 89k stones at 45Mhz interlace and 77xstones at 72Mhz with
- my Actix GraphicEngine32 (S3 801) ISA card . Cpu is a 486/50Mhz 256k cache
- with 8MB of memory :-)
-
- Thank you in advance,
- Amancio Hasty
-
-
- --
- Amancio Hasty |
- Home: (415) 495-3046 | ftp-site depository of all my work:
- e-mail hasty@netcom.com | sunvis.rtpnc.epa.gov:/pub/386bsd/incoming
-