Re: Modern A2 successor

[Roy and/or Janet Miller <millers@inetnebr.com>]

**** **** wrote:

> You are all worried about the wrong things. First off, no one here ever
> thought
>
> #1 I want to be rich.

John, I never thought that. Besides, the machine that was being discussed
originally was a suggestion by Arkain, not by you. The only references that I've
made to being rich was that Bill didn't expect to see that happen from his
projects. Ok?

>
>
> and then moved on to
>
> #2 I know, I'll make a successor to the Apple//

Second verse, same as the first. Arkain was discussing a successor to the Apple
//. All that I mentioned was what I would want on a successor machine. And then
I mentioned, after getting his permission, the project that my friend wishes to
start AFTER he finishes the ethernet card that he started researching last
summer. The PPC card that he has in mind started because he worked on my ZipGSX,
increasing the cache for me, and soldiering in a socket so I can try different
crystals with my 14 mhz 65816. As he worked with it, he realized: 1) the odds
were long that he would be able to buy one, 2) that using a PPC chip that was
inexpensive would allow much greater speed and power, without having to rewrite
system or application software. I think that perhaps the two posts that I had
made have been conflated in your mind: you talked about PCI slots in reply to
the post about his PPC card. Hope this helps to get the two separate.

>
>
> Anyone can see it wouldn't work like that, and what's more, no one suggested
> anything like that at all. It could work, but it needs more than cynicism.

It isn't cynical to say that there isn't and almost certainly won't be a market
for a new Apple II. It's realism. The most that one could hope for is to offer
an upgrade, such as Apple offered IIe owners when the IIGS came out. Plus a few
new systems could be sold. I wish it were otherwise, I really do.

>
> It needs to be a simple enough circuit board, that it can be soldered by
> hand by someone of intermediate skill.

Such a project takes you back to 1975. If that's what you have in mind, well, I
hope things work well for you. But I think you could count on fewer than 100
people who would buy such.

> It needs a cpu(s) that isn't as high
> end as a pIII or PPC like everyone is suggesting. It's the same problem Woz
> faced then, that we face now. How can I build my own computer, if the only
> cpu's costs hundreds? Well, he found one that was alot cheaper.

Woz looked for power without the price. The 6502 was much cheaper than the 6800
that he was interested in, and nearly as powerful. In addition, it could use
6800 series support chips. Please note that my friend's turbo board would be
based on a 603 or 604 PPC, not the G3 or G4; power, without the price.

> It needs a
> simple architecture, one that encourages you to build your own expansion
> cards.

Sure, if you wish to create a new motherboard. Go for it. Have fun. What I don't
understand is your negative attitude to another good idea - keep the bus and
motherboard you have, but make a quantum leap forward in the computational power
of the cpu. And avoid as many bottlenecks as you can, so that the overall speed
of the machine is increased. If you dropped a 486 turbo card in an XT, you would
want to do the same thing - try to avoid the XT's bus and memory as much as
possible. Microsoft did something like that in the mid `80s with their Mach 20
card: 286, one meg of RAM, and daughter card's that controlled hard drives and
floppy drives. It also had a port for the mouse that they had at that time - I
think it was called smartport, but I really am not sure. The point is to bring
something as out of date as a PC/XT or II up to the early to mid 90s, you either
give it a whole new motherboard, or you find ways to use the old equipment while
working around it's weaknesses.

BTW - if you were to create a new motherboard, see if Mike Westerfield can ROM
his GSoft BASIC for you.

> That rules out PCI, AGP, Nubus, Microchannel... perhaps ISA is the
> only one that fits the bill, but let's face it, isa sucks ass.

The 1984 AT bus still moves more data than Woz's 1976 bus.

> I say an
> extended applebus is the way to go, as was suggested earlier.

And how fast, not just how wide, and you push it? And if you increase it's
speed, will it still work with legacy cards? The advantage to a turbo card,
again, is that you already have the motherboard and slots, and can use legacy
cards. The disadvantage, of course, is that you have to have a IIGS (unless a
IIe version could be created - and then there is the ROM problem), otherwise the
card is useless: it would recycle old machines, but not create new machines.
Which is also an advantage.

> Then, we
> already have some cards to get us started. We leave everything but the bare
> minimum off the motherboard, but give it plenty of slots.

Passive backplane? Like my Kaypro 16 (PC XT clone)?

> Heck, design the
> applebus slots so that you have daughtercards in mind, that way, 7 slots
> isn't limiting at all (not that it was in the first place). Power supplys we
> can mod from junk computers, or maybe even use them unmodded. Keep things
> simple enough, and this can concievably be all homemade

Use old 486 cases, and you'll have about twice the power of the IIGS PS. And
internal drive bays.

> (doubtful: that
> means keeping the mainboard and rest 2-layered, and would prolly mean
> 4-layer minimum). All of this gives us a brand new computer, but does it
> give us the apple4, so to speak? Dunno. Certainly not like things are now,
> but if there was community consensus, and if it weren't just another pc
> wannabe...

Well, John, if I were to do a new II (yeah, in my dreams... :-) I wouldn't want
to be a pc wannabe, but then, the PC was an Apple II wannabe in many, many ways.
I'd keep some of the II slots (I'd probably ditch slots 3-6, and like the //c
(and IIGS) put their functions on the motherboard, the physical space I'd give
over to two or three PCI slots so that cards created for PC and Mac could be
used in the II, let those platforms pay for creating and making the cards, which
is why I'd follow Apple's lead in moving from SCSI to IDE, the drives are cheap,
fast, large, and plentiful, and why I'd use some kind of VGA, since VGA monitors
(used) run about $15. Since there isn't a 32 bit version of the 6502 family, I'd
chose on the basis of price between a Pentium, PPC, or Alpha. Power, without the
price. Of course, without Apple's permission, the machine is dead, because you'd
have to have the IIGS ROM, and while I love the 8 bit Apples (I own at least one
of every model except the original II and the black II), without the ability to
run IIGS software I think the machine would be unable to gather enough momentum
to see new software, even on the cottage industry level, which is what any Apple
II projects can only hope to grow up to be. I don't think any such machine will
ever come to be, but if I were pushing it, that's the direction it would go in.
What would Woz do?

Roy

>
>
> John