Re: Steve Jobs demos Macintosh in 1984

Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:

Woody <usenet@xxxxxxxxxxxxxx> wrote:

Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:

Woody <usenet@xxxxxxxxxxxxxx> wrote:

Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:

Woody <usenet@xxxxxxxxxxxxxx> wrote:

Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:

Woody <usenet@xxxxxxxxxxxxxx> wrote:
[snip]

I would have difficulty doing it with that equipment, therefore
I find it impressive

Not hugely so - I don't know how they did it in 1962, but the
transducer needed to make machine singing was first created in the
1920s.

huh? why does the invention of one part matter to the complexity of
making something happen? The concept of a lens was invented a long time
before the hubble telescope, does that make it unimpressive?

But the /first/ telescope was invented very shortly after lens making
was sorted out properly. Well, the first *recorded* telescope, I should
say. The ancient Greeks knew a thing or two about curved mirrors and I
wouldn't be surprised if it turns out that someone like Archimedes had
come up with a telescope centuries before current reckoning on the birth
of the telescope.

But aside from that, you are setting up an invalid comparison as far as
I can tell - you cite the most sophisticated telescope of its day,
created centuries after the first telescope, and you do so as if it is
in some way comparable to the first effort in a new field.

So are you saying that a transducer was invented in 1920 and no-one used
it for anything for 42 years?

Neither, of course.

Read the links I provided, please, read what I wrote, and stop being
absurd.

Why do I bother when you distort what I say, ignore the information
links I give you, and just stick to your tiny little world in which no
new information may enter in case it disturbs your cosy view of the
world?

Can you *** off with the abuse and try to keep a civil tounge in your
head for a change please?


The first singing computer was 1962 - apparently. The technology
required to make such a thing existed in 1948, just about. All the
parts were there in 1948, if anyone had thought to put them together.

I dont believe that is true.

So what? You've made no attempt to find out about the technology,
you're just relying on `what you think you already know'.

I already know about transducers thanks, so yes.

I took the trouble to point you at vocoder information. You have chosen
to ignore it. Learn, damnit! Learn about vocoders and you will see
that I am right.

ok, I read the page and it is considerably less than I already know
(other than names - I know the technology, not the celebs). is there
something in particular I was supposed to gain from that?

Have you bothered do to so yet? Please don't make any further comment
until you've learnt something, please.

I have read it but I didn't learn anything. In fact I would say it was
rather light on information as it didn't cover much about modern
vocoding which we use every day in mobile phones.


I don't believe you could have done what
they did in 1962 in 1948, there just wasn't the memory for it.

So what?

I said that the technology to make a singing computer existed in 1948,
just about.

Now engage your brain and think again.

For sure not much computer memory had been made in 1948 - but it could
have been.

Could it? A war had just finished and resources were very scarce. I
assume you mean technically it could have been made if it was the one
research of human kind, but it wasn't.

All they needed to do was make more mag drums, more
Williams-Kilburn storage tubes. And it wouldn't have taken a huge lot
more than they had to start with to make a singing computer back then.
It doesn't, not if you use a vocoder.

How would the vocoder make it simpler in this case? OK, if they were
using modern vocoder methods that weren't created then, but not analog
vocoding.

But of course you don't know how vocoders work, so you didn't know that.

Except I do know how they work. You seem rather clueless though.,


(The original computer, the 1948 Manchester Baby, *could* have been used
to draw pictures - using the CRT that mirrored the memory.

Yes, far easier than singing.

That is not the point.

If it is not the point, why did you bring it up?

And in any case, I suspect that persuading it to
sing /using a vocoder/ would require less tricky programming than
drawing interesting pictures on screen under software control, given
that the `screen' was a reflection of RAM contents.

ok, can you explain how then as I would imagine it was very tricky. Can
you explain how using 1948 technology you would do it other than your
suspicions?

I think the problem is that you have no idea how a vocoder works.

I think the problem is that you seem to have very little clue how either
a 1948 computer, memory or a vocoder works.
But I am happy to admit I am wrong if you can demonstrate a grasp of any
of the technology to the point where it can be used other than an
armwaving 'yes, use a vocoder, that will solve everything'

Do you get my point now?

I got your point straight away,

Since you don't know how a vocoder works, you still haven't got it.

I am just disagreeing with it. It *was*
impressive.

Be as emphatic as you like: I see no reason to be that impressed. What
do *YOU* think was impressive about it?

The fact that it is hard. With the equipment of the day I would have
found it hard to do, which lets face it means you would find it
impossible.

People doing things I find hard to do, I find impressive when they do. I
know it is a simple thing, but I do. People impress me all the time with
the things they can do.

No, that doesn't mean it was impossible to do in 1961 or
maybe 1960, but that doesn't make it any less impressive.

Tell me why you are impressed by the feat.

Just did.

I know that you want to
shout and shout and shout that it's impressive until no-one else dares
to contradict you, but how about you try some *reasoning*, based on
addressing the facts that I've pointed out to you. And then explain you
reasoning.

Just did. I don't do the shouting like you, my reasoning is much more
simple.
You haven't pointed any *facts* out to me, you pointed to a basic
vocoder page on a wiki site, said that solves every problem (without
explaining how) and assume that for some reason I didn't understand
school level vocoding.

Before you do that, please learn about vocoders.

Tell you what, you try that as you seem to know nothing.

The fact that someone got a computer to sing
14 years later is not particularly impressive.

OK, we will have to disagree there then as I find it impressive. Far
more impressive than the Mac in 1984 or the bbc in 1980 for example.

Yes, but that's only because you're technologically ignorant

hahaha - coming from you that is the funniest thing I have heard! you
can't even make the technology you have work!

But it's not *very* impressive, either way. It was just a case of
putting together the existing parts and telling them to do their thing.

So what can you demonstrate that is impressive that isn't a case of
putting together the existing parts?

Inventing the parts from whole cloth after working out what you wanted
to do, and then creating the whole system to do what you want to do.

Implementing a technology by putting parts together /as soon as it
becomes practical to do so/ is moderately impressive, but less so than
the above.

<http://en.wikipedia.org/wiki/Voder#Analog_vocoders>

(although according to that article, the first really musical vocoder
wasn't created until 1970 - by Walter Carlos, not Wendy Carlos as
claimed; because Wendy Carlos didn't exist until 1972).

Again, the fact that someone had done something hardly stops something
else being doable.

Huh?

My point is that it's not hugely impressive to do something decades
after it first becomes possible.

Well, I am impressed. you can be as impressed as you want!

But why are you impressed? It seems to me that it's only because you're
ignorant of what was do-able, ignorant of the technology that was
around.

[snip]

You
had no access to state of the art computer music gear - you probably had
no concept of what it was like.

I had a full concept of what it was like,

You've convinced me that you had none.

I had a synthesiser and I
could program it. OK, my synth wasn't the most advanced but it worked
well and I made music with it (and an old reel to reel tape recorder).

Like I said, no idea about the state of the art stuff. Teenagers with
synths in their bedrooms didn't have a clue about the state of the art.

What a load of crap. Teenagers with an interest in things have more of a
clue of the state of the art than most other people, in fact, mostly an
obsession.

The BBC Micro was inspiring
for music when you did that - as I showed with one of the links I
provided.

Well, yes, anything was inspiring if you connect it to the right things.
However, there was no particularly decent BBC software for music,

<http://en.wikipedia.org/wiki/Bbc_micro#Use_in_the_entertainment_industr
y>

indicates otherwise.

No, it indicates there was some software, it doesn't indicate there was
anything decent.

Point is, it was used /professionally/ - that implies it was better than
what teenagers played with in their bedrooms.

No it doesn't. Many things used professionally are worse than things
teenagers have in their bedrooms. Just becasue something is used
professionally doesn't mean it has to be the best thing going.

It may indicate it cost more although not always.

Why ignore that crucial point?

Because it is flawed.

Probably not a lot of hope for a ZX81, no matter what you did with
it...

Well, if you connected it to the same equipment and gave it some memory,
I don't see why it should be that much of a problem.

You want problems? Okay: connecting a ZX81 to `the same equipment and
some memory' meant plugging everything in via the one wobbly edge
connector on the back. This did not provide for reliability.

if you were going to add it on, you could solder it.

But if you do that, you have the problem of not being able to unplug it
conveniently.

Why would you need to? I soldered things to mine and never had a need to
unplug it.

To permit a ZX81 to use the same stuff as a BBC Micro, you'd have to
implement the same interfaces, since the ZX81 had none beyond the system
bus exposed at the back, a UHF modulator for video output, and a
cassette port.

but it had the system bus, the ultimate interface.

<sigh> And the least convenient of the lot.

Why is convenience an issue all of a sudden?

The BBC Micro had the
system bus available externally so you can't argue that the Sinclair
machines had anything the Beeb did not.

Yes it did. It had a means to access the non-maskable interrupt and take
control of the machine.

But that bus was rarely needed
because of the shed-load of other interfaces that was provided.

The spectrum had a lot of interfaces available for it.
I prefered the BBC to the spectrum - I had both. The BBC was easier to
use for programming and basic interfaces.

Doing so would not be trivial - I don't see any way a
Beeb Tube interface could have been added to a ZX81, for example.

Why would you want to?

For the obvious reasons, obviously.

What a daft answer, when someone asks what a tube interface was for and
you say its obvious?
I had a bbc, I never used the tube interface.

What sort of a question is that?
You add a second CPU so that you can do things faster, or do things that
otherwise cannot be done. You want CP/M on a Beeb? No worries, plug in
a Z80. You want to develop Archimedes software before the first Arc has
been built? No worries, plug in an ARM (Acorn did that, and sold 'em to
the public). You want a super-fast normal Beeb? No worries, plug in a
6502 (common). You want to play with a 68000? No worries, plug one in.
And I did know a chap who had in fact done that.

I know a chap with a PPC in an amiga as a second processor. Doesn't mean
that most people want to do it, or ever did.

If you wanted to run CP/M on a machine it was probably cheaper to get a
CP/M machine. Want to play with a 68000, by an ST - also probably
cheaper.

No, you had a full z80 system bus and interface lines, which gave you
everything you wanted in an easy to use form.

This is some kind of joke, yes?

No.

`Everything you wanted in an easy to use form'. Yeah, right. So you
can just plug your analoge joystick and light pen and disc drive and
printer and network and second CPU et al into the system bus and it all
just works *AND* you've got interfaces to spare because you've got the
hardware and software required to make it work all just there, somehow
appearing by magic?

No, but you could do it cheaper than you could do it on the beeb.

Not
without an awful lot of work; the Beeb needed a custom chip for that job
(designed by Steve Furber, who gave it a nasty race condition that could
be triggered by odd operation, the silly sod).

The beeb had much older technology (more of) with a better operating
system.

Huh? No, the Beeb was technologically more advanced than all other 8
bit micros ever.

What, because you say so? What are its techological advances then?

Are you mad, or what? What technological advances do you think the
Spectrum had over the BBC Micro? I'm intrigued.

The processor? The integrated logic array chips. The BBC was an old
processor and standard TTL logic chips. There was no techological
advance brought in by the BBC.

Even then, you'd be missing the far greater processing speed of the BBC
Micro and also the support hardware that sat back of the interfaces and
did a lot of the work to run things.

I think you are reading that wrong, the zx81 has a far greater
processing speed to the bbc micro.

Erm, you're totally wrong.

Because you say so?

I've got a BBC Micro and I've got a ZX81. I had both in 1982 (possibly
1981; well, the ZX81 turned up in that year for sure) and I've got both
now. I like them both a lot.


The ZX81 is a pathetic sluggard by comparison to the BBC Micro.

So you have done benchmarks have you??

Remember that the 6502 was a proto-RISC CPU and didn't have the clumsy
four-clock-tick machine cycle implemented in the Z80. 6502s were cheap
and they were very fast for their clock speed, compared to the
competition.

Okay, so maybe the 6502 instruction set was a bit funky round the edges
- so? It's still a blinder of a CPU and you can work round the
`missing' instructions without any bother.

I have programmed both. I don't know what your 'Proto-risc CPU' means -
I haven't heard it before, but if you mean a precursor to risc, then
they both are, as instructions hadn't got complex enough to reduce them!

Okay, if you soldered your add-ons in place, that'd work okay and people
did do that. But I could get my ZX81 to crash quite reliably without
any hardware wobble problems being involved - the `OS' and/or Basic was
buggy as far as I could tell. Could have been some sort of hardware
fault, I suppose.

And the spectrum
probably had better music software than the BBC did.

Why claim that? I don't see any reason to make an assumption like that.
BBC Micros had greater potential all round than Spectrums.

So? who is talking about potential? I am talking about music software.

<shrug> So who used a Spectrum for professional music?

Who knows - I am not as interested in celebrities as you must be,

I've come
across a Beeb being used for professional music.

I haven't.

That ends the argument
unless you can come up with references to people using Speccies for
proper professional work, not just 6th form college rock bands.

Because you say so?

I'd assume
that Beebs had better music software myself.

You can assume what you want, I owned both.

You don't seem to have known much about what was available, though.

I certainly don't know everything available. As you know these groups
using professional bbc music software can you point me to it, and I am
happy to go and learn.

After all, it was easier
to connect them to external hardware and they had all-round better
abilities at everything. Spectrums were technically inferior to BBC
Micros in all respects, bar none[1].

Apart from speed, memory and expandability

Huh? Are you completely mental? Lost *all* your marbles?

Speed and expandabilty of the BBC Micro are hugely greater than that of
the Spectrum. The Speccy's an unexpandable slug by comparison - the
superiority of the speed and expandability of the BBC Micro over the
Spectrum is so vast that the comparison is absurd.

Clearly not.

In both cases, it's like comparing a Morris Marina (it's okay, it'll get
you there) with a Mercedes (it's nice, it'll get you there quickly and
in style and comfort).

Can you stop using analogies, we aren't allowed to use them when talking
to you about things (as then you 'don't understand them').

Now let's look at memory. This is more complex. By the end, BBC Micros
were coming with 128K memory fitted as standard (32K ROM, 96K RAM),
while Speccies only had 64K memory fitted at the end of development (16K
ROM, 48K RAM).

Althoguh the speccies could access all of them.

Both machines could use 64K addressable memory space.

The BBC Micro Model B came with all 64K fully populated - and could have
more paged ROMs added, giving access to vast memory (4 16K ROMs max with
standard hardware, 15 or 16 ROMs max with a simple plug-in card that
fits inside the case. Yeah, I should look up the ROM limit but I can't
be bothered).

The BBC Micro could have a plug-in card added giving paged RAM, thus
permitting the machine to store screen data in `shadow RAM', leaving all
32K of system RAM for the user (and OS, but not screen).

This meant you could have super graphics *and* big programs.

I don't think anyone did ever sell anything like that for Speccys.

Because noone needed to. The BBC for most people could barely muster 20k
of memory, whereas the spectrum had 48k - 6k screen & 2k system, so
double the memory.

Beebs had the hardware and OS required to make it really, really easy to
add shadow RAM, and it was a commonplace fitment for BBC Micros in
normal use.

The BBC Micro Mk 2 (Master series) had shadow RAM as standard, so I
recall.

I know nothing about the master or beyond. By then the 16 bit machines
were available and technology outside educational establishments moved
on.

Did anyone use a Spectrum for professional music work? Because they
used Beebs for professional music work - as Wikipedia says.

Back in those days, you mostly wrote your own software, remember?

I know, I did, on both.

You don't seem to know much about them, though.

[1] Although yes, you could have 48K RAM available in one page to a
Spectrum but only 32K RAM available in one page on a Beeb.

of which you couldn't use 32K in one go.

No, you get it all in one go. What are you on about? Of course the OS
used a bit of RAM as it does on all computers.

So how can you use all of it - switch the screen off? use the system?

The Spectrum had the problem that because it used its single memory page
for everything, you had to put your screen display inside that 48K.

That isn't a problem.

Shadow RAM on a BBC Micro (optional extra for early models, standard
fitment for the Master series) meant that you didn't have to put up with
most of your available RAM being used for screen display if you had a
BBC Micro.

you still needed your memory for the screen, and it still had to be
accessible. So if you wanted to make a screen with a similar colour and
resolution to the spectrum you needed to devote 20k of your 32k to it.
If you have shadow ram over your rom, you have less space for your
application.

So the BBC Micro might only have let the user see 32K RAM at one time,
but the user actually had more *usuable* RAM available than on a Speccy,
in the final development.

Huh? you lost the ability to add? the spectrum screen is 6k, so the
spectrum has over 40k available to the user *at all times*.

Even without paged RAM, you could use Mode 7 display on a BBC Micro and
reduce video RAM consumption to 1K.

Yay - great graphics.
Then you could get to just 10k less than a spectrum.

If you are going to use paged ram on a beeb you could have used it on a
spectrum, and as the processor was much faster, you wouldn't lose as
much.

Erm, what? Totally wrong on all counts.

The BBC Micro was built to use paged ROMs, and so it was easy to use
that existing hardware and OS support for paged memory to add paged RAM
hardware inside the case - which was a standard fitment in the BBC
Master series (the BBC Micro Mk 2)

Again, not refering to the master series, this is about the BBC. or we
would have to bring in the spectrum 128 etc.

and a standard add-on widely
purchased and just plugged in to the BBC Micro. No such add-on existed
for the Spectrum that I know of.

well, I had paged rom, but not ram. In fact it was slightly less memory
I had as I had the disk interface, so lucky to clear 10k of memory free.

The Spectrum had no support for paged memory in either hardware or
software. Nor did it have the room inside the case. Paged RAM was
common in BBC Micros and was a factory fitment from the Mk 2; I've never
heard of anyone using paged RAM (or paged ROMs) in a Spectrum.

Well, if you are insisting on using the master as well, how about the
spectrum 128? that used paged ram.

A 4MHz Z80 is no faster than a 2MHz 6502 (I reckon maybe a bit slower
unless you use the mirror registers)

Mirror registers? Can you describe these 'mirror registers' on the z80
for me please?
I have never heard of mirror registers and have studied the architecture
of both the 6502 and z80. In fact I think I probably have the
manufacturers manuals for both processors (unless I threw them out).

, and the BBC Micro's architecture
was much faster than any other 8 bit micro. ISTR a double rate clock so
that doing the video stuff didn't slow down processing, or something
like that. There were other things, too - do you want me to go into
them? The 6522s and so on...

Feel free, go into them.

Agreed that the spectrum had a problem in its lower 16k of ram having to
share with the screen.

6502s are very efficient CPUs - check out their timings compared to the
Z80. The Z80 has a clumsy four-tick machine cycle its got to run
through to execute anything. 6502s get things done at least twice as
fast, clock tick for clock tick, so a 2MHz 6502 can beat a 4MHz Z80 (and
does, when the one is in a BBC Micro and the other is in a Sinclair
box).

You could also add a second CPU to a BBC Micro and get a warp speed 8
bit micro. The Spectrum didn't have the interface to permit that to be
done conveniently. So no-one did it, while lots of people added second
CPUs to BBC Micros.

'Lots of people'? I never knew anyone with one. can you define lots?


--
Woody

www.alienrat.com
.

Loading