SGIFanLongTime wrote:
Thanks for the link I actually have not seen it.
I see they mention MGRAS alot for the SOLID/HIGH/MAXIMPACT cards... Just curious where the O2 fits into all of this.
Nicoya's page talks about IMPACT only.
See
http://dart.ncsa.uiuc.edu/slevy/o2-stereo/
and
http://niri.ncsa.uiuc.edu/slevy/o2-stereo/highres.html
for information on applying the process to the O2.
I am currently in the process of creating a 1440x900_60 mode for my O2. The monitor I have is a Fujitsu W19-1 wide TFT. My initial attempt resulted in the monitor thinking it was in 1152x864, so even though there
were
1440 by 900 pixels the screen did a pretty bad job of it. I've just checked the X.org modeline that works on my Linux box with this monitor and will shortly be booting up the O2 to try again. I imagine I would be posting any results to the Usenet comp.sys.sgi.
something
newsgroup, though, being slightly old school...
Update... It was a complete no-brainer with Stuart Levy's instructions.
--- el snippo --- el snippo --- el snippo --- el snippo ---
% cat /usr/gfx/ucode/vfc/vfs/1440x900_60.vfs
#define ACTIVE_LINES 900
#define ACTIVE_PIXELS 1440
#define FPS 62
#define EXTRA_HBP 0
#include "TweakedBlockSync.vfs"
--- el snippo --- el snippo --- el snippo --- el snippo ---
That's all! Really! With FPS 60 the monitor just croaked, but I checked the Linux modeline I had, which resulted in a 62 Hz frame rate, so I upped that here too, the hsync freq went to 58.something kHz from the original 56, and the flat panel cooperates now!
It won't work in combination with the hardware accum buffer as the odyssey driver doesn't seem to like framebuffer + accum buffer + system buffer to approach 100MB but thats already the case with 2*1600*1200..
I also have stuff like 1920x1154_2398r2 and 1920x1154_30f ... I suppose the 1920x1154_5994 is 59.95 herz refresh ? There's a 1920x1154_50 in here too. The 1920x1200_50 is the very last one displayed before you get the 2@ lists ... maybe os revision ? This is 6.5.30.
...