Re: Why not just reproduce the second sight card?



zwsdotcom@xxxxxxxxx wrote:
On Mar 23, 5:54 pm, "Michael J. Mahon" <mjma...@xxxxxxx> wrote:

Give me a quick BASIC program to demo the color/hires effects that
sucked on the hardware you've tried and I'll take a video of the
On the "Cheese box" that I tried first, the colors in *any* graphic
were "cheesy" and garish in comparison to their appearance on a

I don't know exactly what an NTSC Apple II is supposed to look like in
color - we only had green screens in school, and besides the PAL II
behaves a bit differently from the NTSC models. However I followed
your advice on cam settings and pushed the exposure down as far as it
would go, and got these pictures:

<http://www.larwe.com/temp/40_bw.jpg>
40 col, b&w (in S-video mode)
<http://www.larwe.com/temp/40_color.jpg>
same screen with the box in NTSC mode, showing color
<http://www.larwe.com/temp/80_bw.jpg>
80col Appleworks demo screen with box in S-video mode

The pics are fairly representative. For 40_color.jpg, there is
actually more color smearing in real life than you see here, and in
all cases the jpgs are slightly blurrier than real life.

After examining your pictures, they look pretty good. The 80-
column text is a problem because of limited bandwidth (note, for
example, the difficulty of distinguishing a lower case m from n).

I think I'd find this pretty fatiguing for program development.

On the other hand, its artifact color seems pretty authentic, though
some complex game graphics would probably be needed to confirm that.
(I'm not much of a gamer, so I'm not the one to recommend those that
show subtle color in NTSC.)

Anyway, for the price I can't complain. Have you ever tried putting
the II onto the NTSC input of an LCD TV set, btw?

Yes, and the results were too blurry to support 80-column text.

text was readable, it was as if each horizontal scan line was
"squirming", since any vertical line in a character was jumping

Maybe the difference between yours and mine might be that yours was a
_scanline_ converter and mine is, I believe, a _framebuffer_
converter.

Same difference. To do 15kHz to 31kHz scan conversion, you need to
store two scan lines: one to receive the sampled data at 15kHz and
one to be regenerated twice at 31kHz. There is no advantage to storing
more than two scan lines (a whole frame, for example) unless you also
want to increase the frame rate above 60Hz. (BTW, if you're tricky,
you can get by with just a 1.5 line buffer, but memory is cheap. ;-)

adjusted a couple of the OSD parameters and that went away.
I wonder if that is allowing you to trim the sampling clock...?

This particular unit doesn't explicitly let you do that,
unfortunately; I've had some that do, but this one is fully automatic.
The other thing it is missing is sizing controls, which means that the
screen doesn't fill the visible area of the monitor (can be rectified
by adjusting the monitor's size controls).

Yes, its a shame that "overscan compensation" is not a standard
feature on most resamplers--including digital monitors. ;-(

S-video does not impose an arbitrarily low bandwidth on the luminance
signal, so that path can lead to sharp 80-column characters.

The good thing is, the S-video, component and NTSC all come in through
one (proprietary, octopus-style adapters supplied) mini-DIN port. And
the intelligent rascals made sure that the composite pin is mapped to
the same plug that carries the luma signal when using s-video mode. So
switching between "hires, no color" and "color, slightly fuzzier" is
just a matter of a menu setting on the box, no cable switchery
necessary. And it even has a remote :)

Cool!

Actually, switching off R and B is the standard way of simulating
a green phosphor on a color monitor, but its usually done with a

Of course, but I mean I don't want to wire anything to the box nor
modify the monitor, it's a bit extreme. I'll make do with white text.
Although now I come to think of it, the color balance/temperature
settings in the monitor will also let me make the text green.

And a monitor cable with a switch would do it too.

monitor for long enough that I'm quite used to it--and using all
colors tends to make the shadow mask pitch less of an issue, too.

All my monitors [in active service] are LCDs, so that's not much of an
issue here.

True.

That your converter can apparently phase lock to the Apple II 14MHz
dot clock is a huge advantage, and apparently relatively rare.

It's probably just a matter of where they set the limits in the
firmware of the converter. If I had to hazard a guess I'd say that the
box (which is marketed specifically as a game-console-to-VGA
converter) was tested with some of those various wacky Chinese clone
consoles which probably output all sorts of slightly weird rates, and
they kept the lockable range wide open to accommodate them. Could be
something as silly as the firmware developer happening to have a VTech
Apple clone at home, even :)

;-)

They seem a little harder to find now--though there are many
related products, I'd be loath to try anything but the exact

I see two on eBay, one at $4.99 with free shipping and the other at
$9.99 :)

That's what I get for searching the web instead of eBay. ;-)

-michael

******** Note new website URL ********

NadaNet and AppleCrate II for Apple II parallel computing!
Home page: http://home.comcast.net/~mjmahon/

"The wastebasket is our most important design
tool--and it's seriously underused."
.



Relevant Pages

  • Re: Why not just reproduce the second sight card?
    ... color - we only had green screens in school, ... behaves a bit differently from the NTSC models. ... 80col Appleworks demo screen with box in S-video mode ... screen doesn't fill the visible area of the monitor (can be rectified ...
    (comp.sys.apple2)
  • Triple multiheaded CDE works :) except when logging out and back in ??
    ... monitor setup working. ... the remander spread across all three screens. ... # This loads the DBE extension module. ... Identifier "Keyboard1" ...
    (comp.unix.solaris)
  • X refresh problem (XF86config-4 attached)
    ... I'm trying to lower the refresh rate of my monitor, ... EndSection ... # Identifier and driver ... the way multiple screens are organised. ...
    (alt.os.linux)
  • Re: NTSC stops to exist on Apple //e?
    ... What if NTSC monitor and television become obsolete? ... How can we overcome NTSC ... what if the Apple II becomes obsolete? ...
    (comp.sys.apple2)
  • Re: Graphics Card problem
    ... Thanks for this Paul - I tried every different screen resolution that exists ... monitor into the VGA port. ... I recognise it's not the 'best and latest' card available, ... Problem - the TV monitor shows all the startup screens exactly as on my ...
    (microsoft.public.windowsxp.hardware)