Re: Second Sight

[mjmahon@aol.com (Michael J. Mahon)]

Sheldon Simms wrote:

>I've been thinking about this and I don't see much point in
>making *just* another VGA card for the Apple II. I have a
>better idea...
>
>Consider a second sight card, it's basically a small computer
>with a VGA output. The Z180 processor watches the Apple II
>bus and emulates writes to Apple II video memory in it's own
>VGA video memory. One problem with the second sight card is
>that the Z180 is barely powerful enough to do this. Another
>problem is that it is difficult to correctly implement things
>like scan line interrupts.

Yes, the Second Sight was more a "concept card" than
a practical addition to an Apple II.

This is a generic problem with aftermarket "extensions"
to popular platforms like the Apple II.  Unless the extension
sells to a significant share of the installed base, it becomes
a "minority platform" within the larger community.

At best, this carves out a sustainable niche market and
generates an ecosystem.  At worst, it divides the market
and hastens the demise of the main platform's ecosystem
without creating one of its own.

I would put the versions of Appleworks after 3.0 in this
category.  Claris marketed AW 3.0 aggressively, even
offering their "amnest upgrade" program to roll the AW
market to 3.0.  When Quality introduced 4.0 and up,
the market was weak, the effort was low-profile, and,
though they were great extensions of one of the great
platforms, their lateness to the market doomed them to
low penetration, and they had the net effect of further
fragmenting the Appleworks market, including all the
add-ons.  These later versions did not cause many who
did not already use AW to start, and probably played a
role in many deciding to leave Appleworks for some other
platform (since their version was no longer "great" and
they didn't see a groundswell of migration to the "latest,
greatest" versions).

The business of making "upgrades" is a tricky one,
particularly to a mature platform.  The ultimate effect
may be quite contrary to the one that is intended.

Of course, none of this applies to the Apple II at this
point, since it is sufficiently out of the mainstream, even
of hobbyists, that it no longer constitutes a "market" in
any meaningful economic sense.

But you will still have the problem that it must deliver
_great_ results with almost all existing software, since
precious little new software will be written to exploit it.
This rules out perhaps the most exciting extension
possibilities, like increasing (at least) the vertical
resolution of the Apple II display.

>What happens if you replace the Z180 with a more powerful
>processor, like an Intel 80200 XScale processor at 600Mhz for
>example? Then you suddenly have a lot more than a VGA video
>card, you have an accelererator card too: All you have to do
>is write a 65816 emulation in assembly language, and use that
>to replace the hardware 65816. I'd be willing to bet that the
>effective speed would average at least the equivalent of a
>50Mhz 65816.

Sounds right.

>Video emulation could be done at full XScale speed because the
>Apple II bus would be removed from the video emulation process
>and since there would be some extra memory, you would
>effectively "cache" the entire Apple II memory space - i.e.: no
>external memory would be required at all. Emulating fill modes,
>scan line interrupts, and so on would be easy, since the card
>would control the scan lines itself. In fact, there would be no
>need for the emulated Apple II scan lines to correspond to real
>hardware VGA scan lines at all, the entire raster scanning
>process can be emulated. In short, one card would replace
>three: Zip/Transwarp, Second Sight, Memory Expasion.

Cool.  This is the kind of capability (and cost-effectiveness)
that market successes are made of--if there _was_ an Apple II
market. (But doesn't it sound just a little like the Apple II is only
a peripheral for this card?  ;-)

>The software is fairly simple, I have already done something
>similar in my Apple IIe emulator. The biggest problem is the
>price. Without some kind of quantity, it would probably be too
>expensive. By looking around on the net, I've found following
>prices:
>
>Intel 80200 600Mhz                     $65.91
>Intel 80312 (PCI, DRAM controller)     $57.67
>Asilant 65550 VGA Controller           $28.57
>32 Mb PC100 SDRAM                      $5.00
>                                     ---------
>                                      $157.15
>
>for the main components. Of course, these are mostly prices
>for single component purchases, and there might be some cheaper
>replacements for the 80312 and the 65550 but if you add glue
>components, analog components, circuit board, manufacturing,
>etc. It will add up quickly.

True.  This sounds like a $300 board unless it's purely a labor
of love--in which case it might be only $230.  ;-)

>The last problem is power. I have be assuming an XScale processor
>because you get a lot of computing power for very low power
>requirements. Still, it is a 600Mhz processor and the card might
>end up being too much of a power hog for an Apple II. For people
>who are a little more knowledgeable about the electrical side of
>things than I am, the 80200 at 600Mhz has a maximum current
>requirement of about 600mA and dissipates 1W with a core
>voltage of 1.3V and a periphery voltage of 3.3V.

One watt isn't much of a problem, but I'm guessing that all the
rest adds up to at least another watt.  Voltage levels are not a
problem either, since very compact DC-DC converters can be
used.  I haven't checked ZipGS power consumption, but I'll
bet that it alone is more than two watts!

-michael

Check out amazing quality 8-bit Apple sound on my
Home page:  http://members.aol.com/MJMahon/