Detail description of my 800x600 mode problem.

Detail description of my 800x600 mode problem.

Post by Daniel Newcom » Sat, 11 Feb 1995 23:23:38

To summarize my problem:
        Under X, in 800x600x256 mode, the screen does not take up all of the
display. It does something like:
     |             |
     |             |

Where the outer box is my display and the #'s are the part of the
screen that X uses.  In 1024x768 and 640x480, X uses the whole screen.
Also, under Windoze, OS/2, and even the SVGAlib, 800x600x256 uses the
whole screen.  So basically, X in this mode is the only thing that does

My Hardware:
        A Diamond Speedstar Plus (ET4000 w/ 1M of memory)
        Gateway2000 CrystalScan 1024NI monitor
        (Running Linux with XFree86 3.1)

What I have tried:
        I ran xf86Config to generate my XF86Config.  This worked okay.  Mind
you, I'm not new to X, and have manually done the creation of this file
before using the database entries for ET4000 w/ 1M, and VESA standard
        I then ran X to make sure everything worked. No problems (except the
problem I am now talking about :)   I then read the readme's that came
with X, and one of them said to "unsquish" the screen vertically, you
adjust the last number of the vertical timing settings by an amount,
and adjust the middle to vertical timing settings by half that amount.
I tried this.  It did NOTHING!  I then only adjusted the last setting.
This allowed me to move the image up and down the screen, but it still
only used about 66% of the display space.
        At the Atlanta Linux Enthusiast's meeting last night, someone
mentioned I should try a different clock mode.  Normally, 800x600 is
running with dot-clock 50.  I fired up modegen (a utility that runs
under sc) and fiddled around with the setups.  Modegen allows you to
enter your monitors vertical and horizontal timings, and the dot clock
you want to use.  It will then generate a resolution and monitor
timings for you.  I have always had great success with this program,
and every mode I have generated has worked.  So I put in 44.9, my next
lowest dot-clock.  I also tried 65, the next highest.  None of them
came close to 800x600, even when playing with the vert/horiz timing

This morning I had an idea.  Since the SVGAlib works fine in 800x600, I
ran a GIF viewer in that mode.  Came up fine...the image used all of
the screen.  I then switched to a different virtual console and ran
        sleep 3;vgaset
and in the 3 second grace period, switched back to the virtual console
with the GIF.  After 5 seconds, I exited the image viewer, and killed
the vgaset process.  Switching back to that VC, vgaset had pulled out
the timings for the 800x600 mode that SVGAlib was using.   I then
looked in the config.et4000 files for the monitor timings.
        Based on all that research, I came up with:
        Vertical Freq: 72.2
        Horizontal Freq: 48.3
        Timings 800 xxx xxx xxx 600 602 604 666  (something close)

The freq's are (naturally) within my monitor's specs.  The only thing I
am missing now is the dot-clock.  Anyone know how to find out what dot
clock a card is in?

Anyway, I put these timings into the XF86Config, and used a dot clock
of 50.  I also tried 44 and 65, but X told me I didn't have the
Horizontal Freq. that those modes would require.

I then fired up X, used Ctrl-Alt-+ to change to that new mode, and
guess what:  It still only uses 66% of the screen, even though under
SVGAlib, the same timings use the whole screen.

I am about at my wit's end on this one.  Logically, the above is
bullshit.  It should have worked.   Does anyone have any ideas,
pointers, or any of those nice white jackets that buckle in the back.


Dan A. Newcombe           Clayton State College       Computing Services

- Higher is lower, and less is like more, She never felt anything
  like this before...             (Marillion, The Party)


1. detailed description of linux socket filter and sniffing problem

well, a detailed description of the thing which i wanted to ask you is:

1) when i start sniffing i attach a bpf code to the socket by means of
setsockopt system call. (as given in the filter.txt file in the docs of

2) then i do some reads on the socket and now i want to change the
filtering criteria and so the bpfcode which was attached to the socket
also needs to be changed.

3) for that i need to again call the setsockopt to attach the new

4) but the necessity is that as soon as i find the need to change the
criteria, i need to grab all the packets after that till the time i
have not attached the new code (altough this will be few of

5) so i remove the filter expression by means of setsockopt
SO_DETACH_FILTER and then regenrate the bpfcode for the nw expression.

6) now i do a attach operation on the socket by means of setsockopt and
so attach the new filter.

7) now here comes my real question. All the packets which were buffered
by the kernel after the detach operation and then the attach operation
(as in point 6), will they be filtered on the basis of the new filter
expression and then passed to the user process or will they be passed
as it is (i.e. without filtering) and only those packets which were
received after the second attach operation will be filtered on the
basis of the new expression ?

(sorry if i am unclear this time also..i will mail u the program next
if i am unclear this time too.)


Image by

Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at
Please read the FAQ at

2. IBM DB2 7.1 for Linux download

3. problems with TERM109 (detailed description)

4. IPFilter IP Forwarding broken offten

5. Detailed description of network trafic

6. conexant softmodem with ndiswrapper?

7. looking for lp.c source or detailed description

8. certificates for HTTPS

9. detailed description/instructions for using `format`

10. Help - looking for detail description on signal_pending(current)

11. Q: Detailed description of lp and lpr

12. Detailed description of (free)OpenServer packages

13. detailed description of ports-system?