Re: Apple II-Or Come on people, It's a 20 year old computer

[pausch@saafNOSPAM.se (Paul Schlyter)]

In article <39543DC5.A22EE5E3@wanadoo.nl>,
ruud.dingemans <ruud.dingemans@wanadoo.nl> wrote:
 
> Paul Schlyter wrote:

>>> So? Does that make any fundamental difference? Is having an x86
>>> processor or an ISA bus (which no same computer manufacturer uses
>>> anymore, btw) more important in computer design than having a GUI?
>> 
>> Yes -- it's a much more fundamental property of a computer: it
>> determines what software you can run and what hardware add-on stuff
>> you can attach.  As you perhaps are aware of, there was a vast number
>> of ISA cards to choose among.
> 
> So? There also were a lot of Nubus cards,
 
Were?  Where did they go?   <g>
 
> but that's hardly a determining factor in computer use.
 
It's not?  Well, if you have requirements beyond the standard vanilla
requirements, then a big selection means a greater chance of finding
what you want.  Don't you agree about this?
 
> Neither is the cpu type. You're looking at it from an *engineering*
> point of view, not from a *user's* point of view
 
FYI: an engineer might be a computer user too.....
 
> - and the user, that's us, the people the computer is ultimately
> meant for. In personal computer design, the definite determining factor.
 
So you're telling me that when I want a computer, I should choose one
that you want and not one that I want?
 
Do you choose computers the same way -- to fullfill the needs of others
rather than the needs of yourself?  I think not!
 
> To us, to the user, it's not relevant what cpu is in there as long as
> its design is competitive (and if it's not, it will play no part in the
> industry). But having a gui or no gui, is definitely more relevant.
 
If WIMP interfaces are so important to you, then by all means choose
one and use it!  It's there on many computers -- so what are you
complaining about?  However, note that different users have different
needs, and all users shouldn't be "straight-jacketed" into the same
user interface, regardless of what they want.
 
> If you want to change your cpu type, just go out and hack one in. This
> is really getting silly. A non-GUI design like the original ibm pc does
> not a usable mouse/gui computer make. Besides, we were talking about
> *existing* designs and their impact on computer history, not about
> theoretical options that never were.
 
WIMP interfaces do exist, in several incarnations by now.  If you want
one, then just choose one -- OK?
 
> You brought up the issue that it's possible to use a cli on a Windows
> machine. Well, it's possible, but most people sure don't know how to use
> it anymore these days. Everybody mouses. So, that kinda shows which machine
> was more influential to the users in the end: the Lisa, or the commandline
> ibm pc.
 
Very few people uses a Lisa though....  <evil grin>.  But if you want to talk
about the original inspiration: the GUI originated at Xerox, which you've
already admitted.  Thus the starting point was not the Lisa computer.
 
>> It's not "a lot" more -- 8 MHz isn't even twice as much as 4.77 MHz.
> 
> So, a 40 per cent speed increase

LEARN TO CALCULATE, will you?   4.77 plus 40% = 6.68 MHz, not 8 MHz...

> means nothing to you it seems.
> Let's see how you'd react if your own machine suddenly was 40 per
> cent slower. It *does* amount to a significant difference, I'd say.
 
Only if you're doing CPU intensive stuff which takes a long time: from
several seconds to hours and up.
 
Once, may years ago, I replaced the 8088 on my homebuilt XT with the
NEC V20 -- this should increase the CPU performance by about 25-30%.
But that difference was in practice hardly noticeable.  A 68% speed
increase would be more noticeable, but only on CPU intenstive stuff.
 
Also note that we're comparing different CPU's here, and then a simple
clock speed says very little.  Compare e.g. the 1 MHz 6502 with the
2 MHz Z-80, as in the Apple II with a SoftCard.  Your naive approach
would conclude that the Z-80 would run its programs about twice as
fast, when in fact the 2 MHz Z-80 actually often ran things a little
slower than the 1 MHz 6502.

And an X% speed increase of the CPU does NOT imply an X% overall speed
increase of your application!  That's very application dependent.
Most everyday applications tend to be I/O bound, i.e. they're not that
sensitive to moderat echanges in CPU speed.
 
Finally: if you're making so much noise about a mere 40% difference
in execution speed, then your #1 priority ought to be to throw out
any WIMP interface, because such a thing will eat a lot of CPU
cycles, making your application run much slower.  And you should at
least agree that a CLI has one advantage over a GUI: it's much more
efficient on CPU cycles, leaving more cycles to your application.
And if your application uses so few cycles that this doesn't matter,
then a mere 40% decrease in clock frequency won't matter either.
 
> Wrong again. The original IBM AT ran at 6 Mhz (look it up).
 
The speed was switchable though -- "hi speed" was 8 MHz.
 
> This is just silly. First you argue that the Lisa was such a slow
> machine (which it purely was because Motorola could only deliver 5
> MHz 68k's and not 8 Mhz) and now you're telling that the speedup
> to 8 Mhz didn't make any difference in driving the GUI
 
It wouldn't have made a big difference.  The reason the Lisa was so
slow was that it loaded that 5 (or 8) MHz CPU with a CPU-heavy WIMP
interface -- plus of course that the only programming language
available for the Lisa -- Lisa Pascal -- used a P-code approach which
meant a running program had to have its pseudo-code interpreted.
That slows down performance a lot, and was definitely not feasible
back in those days (todays Java running on 500 MHz to 1 GHz COU's is
naturally a quite different story)
 
> (well let me tell you, on the Mac it sure did: the complaints about
> slowness were generally gone. (I have magazines from that era to
> prove that).
 
You don't think that may be because the GUI then had been moved into
ROM's, instead of being read off disk?
 
> Anyhow, from 5 to 8 MHz does matter.
 
If you seriously think so, then throwing out the GUI will matter even
more...
 
>> Wrong -- the most important factor has been affordability.
> 
> In that case, the IBM PC was *NOWHERE NEAR* the godfater of personal
> computers, since it was *DARN* expensive for its time. It was known
> for it (again, read the old computer mags).  Clones like the CBM PC-10
> were a lot cheaper, and they needed to be without the ibm label.
 
Quite true -- the clones, not the IBM original, made the PC affordable.
My first PC was a home-built clone.
 
If clones had appeared for the Macintoy, this would have made the Mac
affordable sooner too.  Now, the Macintoy only got affordable when
forced to by the competition from affordable PC's.  Thus, also
Macintoy users did benefit from the PC clone market.
 
> I don't care who outsold who.
 
But the manufacturers do -- and Apple Computer Inc did.....
 
> I care about which design had the most influence on the personal computer
> of today. And today, personal computers look a lot more like the
> mouse/gui-driven Lisa than the commandline ibm pc, and that's that.
 
That's your religious belief, and since you appear unable to have a
rational discussion about this, it's really pointless to have a holy
war about this.
 
>>> Now you're nitpicking. They had to cut SOMETHING down to avoid a
>>> Lisa-type price tag.
>> 
>> Sure -- but please don't call that an "extension".
> 
> The Mac *in general*  *was* an extension of Lisa technology. I never
> said it was an improvement in every respect.
 
So far you've only been able to point to two improvements: its
significantly lower price, and its marginally higher CPU clock speed.
In all other respect it was a cut-down Lisa - right?
 
>> In my book, a stripped-down "mini" version is *not* an extension....
> 
> This stripped-down mini was a lot faster than the original though,
 
Could you please attach some figures to "a lot faster"?  How much
faster?  2x?  5x?  10x?  100x?  1000x?
 
> and more affordable, the argument you've been waving about a lot.
 
"Less overpriced" is a more appropriate term for the first Macintoy
versions.  Onle a decade or so later did they get really affordable,
due to competition from the Pc clone market.
 
> This is too silly to argue. With "the world" I generally mean "the
> computer world", of course. I'm not talking about the basic structure
> of the universe!
 
You were only talking about a small subset of the human world...
 
> The Wimp interface did change the world, in a way. No but's or if's.
 
It did, in a way, change a small subset of the human world, yes.  That's
quite different from "changing the world".
 
Unless you consider every infinetisemal influence to be "changing the
world": in that case, each and every decision we all make every day
are "chaning the world", because the world would have been somewhat
different if we had made another decision.
 
> Irrelevant. It reached the consumer market, if only in small numbers
> (initially; it got better later on).
 
It reached the consumer market mostly through ads....
 
> Irrelevant again. There only were about six Apollo's that actually
> reached the moon, but that doesn't mean the Apollo team didn't get
> there first, or didn't have any historical impact.
 
But it DOES mean that moon trips never reached the consumer market,
don't you agree about that?
 
Btw, if numbers are that irrelevant: the Lisa didn't create the first
GUI - Xerox did.  It was only implemented on a few computers, and it
never got out of their labs, but you think such things are
irrelevant.  Xerox was first, period.
 
> It was a revolution because just about everybody copied it's radically
> different design and because THEN the gui was shipped on just about
> every computer, it did change the world as we live in now. Otherwise
> we'd all still be using cli-commands.
 
...and doing what we wanted to do on the computer that way.  A marginal
difference: life is so much more than just GUI vs CLI...
 
>> It's not fundamental at all -- it's merely a matter of appearance.
> 
> Like I said, try putting your granny behind a commandline IBM PC and a
> Lisa and try to have her use it within the hour, and then tell me there's
> no fundamental difference.
 
Paul Guertin did that (see his post) --- and he found no big
difference.  It's mostly a cultural thing.
 
> That was your personal priority set. Doesn't have a lot to do with
> historical impact.
 
...and then we get into those numbers you sao are irrelevant... <g>
 
 
>> You mean the Lisa run barely faster than that deradful IBM pee-see?
> 
> Irrelevant.

Then why did you make so much fuzz about it's CPU clock rate?

> It introduced the gui to the world,
 
NO!!!!   Xerox did -- they were FIRST!  Numbers don't matter, as you
claimed above...
 
> that's what important - not that it was somewhat slow for its day.
> Nobody remembers the Lisa for its performance, but people
> today want one like crazy. That has a reason, and performance is not it.
 
If so, why is the Lisa not sitll on the market?  Any machine which "people
want like crazy" would be a very profitable machine to sell, don't you
think so?  So why aren't Apple still selling it?
 
>> ..and the Mac had zero.  Do you still think the number is unimportant?
> 
> Of course. The Mac was a success even without slots, remember?
 
It was a success as a niche computer, yes -- for people with only
basic needs, and no wish to expand later.  Like that grandma....
 
>> It coudldn't be -- the 68K FPU (the 68881) wasn't even available when
>> Lisa was released.
> 
> Sigh... Read up. I have a real Lisa here, and if I say there's a socket
> for a copro, better believe it. I never said its type was a 881.
 
Perhaps you could elaborate on WHAT kind of copro it was, then?  There
are so many coprocessors out there....
 
>> Perhaps you are, but I'm not.  Why ignore those other PC's?
> 
> Because your argument was that the *IBM PC* was the design modern
> computing works like today.
 
It was -- the clones copied the IBM PC design, and most of today's
computers are a legacy of that design.  That's a *fact*, which isn't
changed by adding on WIMP interface software later.
 
> Maybe. But that's just a matter of *volume*. Not of design type.
 
Volume matters to availability.  Your old grandma, for instance,
could never buy that Xerox machine with the first GUI -- because the
volume of manufacture was too small, and thus the price was too high.
Many senior citizens have a limited budget, and are thus quite
dependent on computers being cheap, if they are to afford one.
 
Volume also matters in another way: it reflects people's actual
choice when they really bought their computers.  Why didn't most
computer owners buy the Macintoy, if it was such a superior computer
as you claim?
 
>> Perhaps you were -- I wasn't.  I've never owned one single "true-blue"
>> pee-see.
> 
> Yeah, well, I have one. And I know how this thing is/was built. And what
> role it really played in computing history - a minor one. The *clones*
> made the ibm noteworthy; the copies, not the design or the computer itself.
 
The PC clones also forced Apple to lower their price on the Macintoy,
which enabled many old grandma's to buy one....
 
>> Perhaps you were -- I wasn't.
> 
> Yes you were. You brought the *IBM PC* up as the originator of modern
> computing.

No I didn't!  I brought it up as the mother of most modern personal
computERS !  And that's a fact -- look up the sales figures of PC's
versus Mac's versus any other architecture of your choice.

> Not the Compaq, not the other clones, not the Dells or
> Shitecs, but the IBM.
> 
> I quote:
> 
>>>>>>> Many people frown on this, but "the grandmother of most modern
>>>>>>> computers" happens to be the IBM PC.
> 
> Q.e.d.
 
Well, the clone makers certainly didn't clone the price of the
original IBM PC --- thus your earlier remark about the IBM PC
being expensive is irrelevant: most PC buyers didn't buy the IBM
PC, but some clone instead.
 
>> Why disregard the evolution which was spawned by the clone market?
>> A lot of today's features in the PC computers were NOT spawned by
>> IBM.
> 
> Indeed. Like the GUI, the mouse, the expansion slots, the microprocessor,
> the floppy disk, the Internet, etc.
 
You're confusing things a lot here -- do you seriously think you'll get
your own Internet if you buy a PC clone?
 
Please keep the hardware design separate from the applications.  You
do know the difference, don't you?
 
> So what role did the IBM PC *really* play? It was a second-rate rushjob
> in a technical sense, it was expensive, and it got copied because it
> said IBM on the label.  That's it.
 
And it spawned the clone market (against IBM's intentions, of course,
but it nevertheless happened), which significantly lowered the price
of personal computers.
 
It also spawned a standard floppy disk format, which helped to enable
easy data exchange a lot!  Even the Machintoy later changed to a floppy
format which, on the lowest level, was compatible to the PC disk format.
This enabled PC's and Mac's to read one another's floppies with only
software add-ons needed -- no hardware add-ons was needed.
 
> Well, that's my point: Amiga's Workbench already reached most of that
> level in 1985, and it still was better in a lot of ways. (And yes, I
> do have an original Amiga 1000 to compare.)
 
So why aren't Amigas outselling PC's today?  You said it was better....
 
> Besides, the 'pre-emptive multitasking'
 
No need to add quotes here -- it was genuine pre-emtpive multitasking
between different DOS boxes.  Windows applications used cooperative
multitasking though.
 
> of Windows 3.1 is nothing short of ghastly compared to the original
> 1985 Amiga! And it needs four times the memory to do something, too.
 
If the Amigas were that superior, why aren't they outselling the
PC's today?
 
> Oh yes you did. You said in your original argument that modern computers
> are descendants of the *commandline* IBM PC,
 
No.  I said "IBM PC", not "*commandline* IBM PC" -- the latter is your
twisted quote of what I wrote.  And I referred to the hardware design,
not which application you happened to be running at the moment.
 
> and resemble it more than the Lisa. Which is wrong. Q.E.D.
 
Q.E.D. what?  Where's your proof?
 
I have some proof for you: pick an old IBM PC application, and an old
Lisa application, and try to run it on a modern PC.  Which one do you
think will have the greatest chance to run successfully?   <evil grin>
 
> The IBM PC was a commandline computer. Q.E.D.
 
If you ran such software on it, yes.  And so was the Lisa, if you
e.g. ran CP/M-68K on it....
 
> II. Most personal computers today are not, they're Mac or Windows
> (and if they have the capability, this is barely used in comparison
> to mouse use). Q.e.d. III.
 
You're talking about software here.  ANY computer can be a CLI or a
GUI computer -- it's merely a matter of which software you choose to
run.
 
> Small wonder, since the Mac did not actually evolve into a Pentium III,
> thank goodness.
 
Yet you claim the P-III evolved from the Lisa through the Mac.  Could
you please stop make such obviously erroneous claims?
 
>> I can't even change the processor -- a PowerPC processor won't fit in
>> that P-III socket.
> 
> I never said it needs to be a PowerPC. Ever heard of Cyrix, or AMD, or
> Transmeta?
 
They're all x86 clones.....
 
> That last part is important, because it means that the IBM PC was not an
> "open" design at all.
 
IBM published curcuit diagrams of the motherboard plus an assembly listing
of the ROM BIOS of both the PC, XT and AT.  And you claim this was not
an open design?
 
> Yes, of course I agree.
 
Fine -- then you must agree that the PC clone market was a Good Thing,
and that the absence if a similar clone market for the Mac was a Bad Thing.
 
> Yeah, but that doesn't mean the IBM PC was anything special in computing
> terms.
 
Did anyone ever claim it was?
 
> The Lisa was.
 
No it wasn't.  It was only remarkable, for its time, in the user interface
terms.  But the actual computing was quite non-remarkable.
 
> In case of the Lisa, its introduction onto the mass market meant a
> change in the way computers were built.
 
No -- the hardware of the Lisa wasn't that remarkable.  The software
was built differently, and aimed more at appearance than performance.
 
> In case of the IBM PC, it just meant more of (basically) the
> same were built, and the original design of it didn't bring anything
> really new in the first place.
 
Again, you're re-iterating things we already agree upon.....
 
>> So the Lisa and Alto then become icons, which the masses perhaps can
>> admire but never afford ????   I do think a computer which people can
>> afford is more useful, even if it's not a technological milestone.
> 
> We were talking (originally) of which computer has had more historical
> influence. Well, the IBM PC and the Lisa and Alto qualify in this respect.
> But the IBM didn't really change computers in the sense that it changed
> the way we use computers today (it only got cloned);
 
This cloning meant one change: computers got significantly cheaper.
That alone meant a change in teh way people sued computers: they were
now using them for more trivial tasks, where the former higher costs
couldn't be justified, but the current lower cost could.  That matters
too.
 
> the other ones did. So it's pretty obvious which computers
> have had more impact both historically and technically.
 
Why do you ignore the economical impact of computers?  If they hadn't
become cheaper, they would never have become as widespread, no matter
what fancy GUI it may have been having!!!!
 

-- 
----------------------------------------------------------------
Paul Schlyter,  Swedish Amateur Astronomer's Society (SAAF)
Grev Turegatan 40,  S-114 38 Stockholm,  SWEDEN
e-mail:  pausch at saaf dot se   or    paul.schlyter at ausys dot se
WWW:     http://hotel04.ausys.se/pausch    http://welcome.to/pausch