How to config your vedio subsystem for X386 (was Re: X386 1.1)

[Chin Fang]

In article <1991Mar7.014948.18583 at leland.Stanford.EDU> fangchin at leland.Stanford.EDU (Chin Fang) writes:
>     
>                  And you got 944x708 to boot.  Not bad at all!  You 
>                  can even improve it further to put it into almost 80 Hz
>                  by using the fact that your monitor in horz. direction
>                  can sync at lower frequency then 55 khz.  Please take
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

OOPS! This should be written "often can sync at higher frequency than 55khz"
Got it reversed.  Sorry! In addition, 80 Hz is probably very hard to reach.
76Hz is a more likely value.


In fact, if say the highest timing available for a board is 65Mhz, then to
raise refresh rate, one has to pay the price of lower resolution.  Unless 
a higher clock rate is available.  DRAMs usually impose wait states if their
recharge rate is not high enough.  The only card that I have data sheet so
far that can do higher than 70Mhz is Everex VRAM ViewPoint ET4000, it's     
equipped with one meg VRAM instead of DRAM, but quite pricy.  List price is
like $600+.  However, with VRAM's more streamlined data io, it should be a
very high performance SVGA card.  (My Sun Sparc's vedio adapter employes 
one meg VRAM and AMD's vedio chipset)


Vedio 7 is having a similar product now, but I haven't got any data sheet yet.


Finally, multiscanning monitor vedio bandwidth is in Mhz range.  But it has
nothing to do with config X386.  High vedio bandwidth allows higher fidelity
if images.  The reason, simply stated, is that monitor employes electronic
signals to represent images. Such signals contain many different components
which are of different frequencies, many of these are easily in high Mhz
range and usually are important to the fine details of your display. So a 
board vedio bandwidth allows most such signals coming thru without being 
distorted thus fidelity (a general term here) can be preserved very well.

The details of such are in the domain of Signal Processing.


However, this is again, just part of the display game.  One has to consider
dot pix size, misconvergence of electron beams, corner compansations, and
the type of screen coating materials used etc.  They have to be considered
all together.  Hmm...I am digressing ...Stop!

Thanks to the input from gwes at wjh12.harvard.edu and the very careful review
of witr at rwwa.com.


Regards,

Chin Fang
Mechanical Engineering Department
Stanford University
fangchin at leland.stanford.edu