RE: Poll preliminary results...

From: Clay Cowgill <ClayC_at_diamondmm.com>
Date: Fri Aug 21 1998 - 19:57:04 EDT

> This is what I was talking about with the capacitor thing. You can do
> the big
> jumps (256 x 256) instead of the smaller ones (1024x1024) if you have
> some type
> of lowpass filter to give the beam enough time to display a line
> between points.
>
So something integrator like to "connect the dots" so to speak?

> This filter can be a R/C on the output of the op amp, it can be the
> slew rate of
> the op-amp itself, maybe the speed of the yoke will keep you from
> being able to
> jump to a new spot instantaneously, allowing the drawing of a line
> between the
> two points.
>
I had thought about that a little, but convinced myself that it would be
particularly linear response on the screen... I can mess around with
it...

> It should be easy to hang a capacitor somewhere, or make a tweak here
> or there
> to make a predetermined pattern look nice.
>
Yup. Did plenty of that to get the timing right for the raster one.
(Raster video is very unforgiving-- one clock cycle off and you've got a
lovely "jaggy" to deal with. There's some *weird* assembler code in the
raster monitor tester... ;-)

> But if your going to do that, why even have DAC's? Just point to the
> four
> corners in the right order and let the trace "charged" to the next
> position
>
I had thought of just using a couple few resistors as a cheap D/A and
letting the beam sweep to the new location, but I wasn't sure it'd sweep
in a straight line...

> I'll bet the X/Y testers are even simpler than the PIC thing you're
> talking
> about.
>
Simpler is relative. I'll trade more complicated software if I can make
the board and/or parts count smaller.

> With just a few shift registers and an R/C network you could probably
> create a
> real simple pattern generator by simple turning off and on a resistor
> attached
> to a capacitor and using the voltage on the capacitor to drive the X
> (or Y)
> axis, as long as you timed everything right.
>
Yeah, it's nice to have a couple different settings depending on what
you're working on though... A vector tester would mostly be a debug and
convergence tool I'd think. (Since most stuff like linearity and
whatnot is adjusted on the game PCB...)

> It would be even easier if you replaced the shift registers with a
> PIC, but
> either way, chances are you won't need the DAC's, just good timing
> routines.
>
I'm inclined to do whatever's easy. ;-) Right now PICs sound easy...
not like there's a ton of people needing these, I'm reluctant to do
anything really "sneaky" with it...

> Didn't someone do something similar to this just using a Sound Card?
> I'll bet
> if you setup the sample rate properly (which would set the lowpass
> filter
> accordingly) you could write some simple software to test an X/Y
> monitor.
>
Yep, had a nice little rotating cube on the screen... The point of my
raster monitor tester is that it's about 1"x2" and powered off a 9V
battery so it's easy to keep in a pocket or something for quick checks
and adjustments inside a cabinet (instead of on the bench). The quality
is good enough for bench work, but it's just kinda simplistic for that.
(There's one button and a DPDT center-off switch. Flipping the switch
to either side turns it on, with either positive or negative sync--
depending on which side you flip it to. The button selects one of 8
test displays.)
> >Yeah, I was just seeing if I could weasel out of the extra
> transistors
> >and discretes. Actually, with surface mount 2n390x's it doesn't
> really
> >take up much room to have full color...
>
> Actually when I said switches, I meant "Switches" those toggle things
> ;^). I
> was talking about the tester constantly generating a pattern for all
> three guns
> (one B&W signal driving all three guns), and then having switch to
> turn off each
> gun separately.
>
Ah, I follow you. I though you meant three color output stages with one
switch each. Did you mean one "monochrome" output stage with three
switches off of it? My only concern there was that I have no clue how
much current a color input on the monitor will draw from me...

-Clay
Received on Fri Aug 21 18:57:28 1998

This archive was generated by hypermail 2.1.8 : Thu Jul 31 2003 - 23:00:57 EDT