Text 5667, 103 rader
Skriven 2010-03-06 12:10:53 av Michiel van der Vlist (2:280/5555)
     Kommentar till en text av Bo Simonsen (2:236/100.0)
Ärende: Windows Update Fix
==========================
Hello Bo,

On Tuesday March 02 2010 19:36, you wrote to me:

 MvdV>> Indeed, I have lived in interesting times. Saw the first computers,
 MvdV>> saw the first man walk on the moon, witnessed the birth of quark
 MvdV>> and inflation theories, etc, etc. All very interesting.

 BS> When I reach your age, I probably say the same..

Since things are going at an accelerating pace, maybe even more so. Nothing
much happened between 1200 and 1300 with regard to technology. Quite a lot
happened between 1800 and 1900. And between 1900 and 2000 the world changed
completely.

 BS> Since there was hardly any cell phones when I was a child, and
 BS> definitely no internet available to the "normal" person. And no
 BS> laptop. :)

And who knows what the next 50 years will bring? I think it is almost certain
we will have robots that pass the Turing test. You will live to see it, I just
may. We may contact extra terrestrial intelligence. I do not think we will ever
break the light barrier, but who knows, we just may. You may live to see it.

 MvdV>> It was actually a microcoded processor. The instruction set
 MvdV>> presented on the assembler level was not the bottom level. Below
 MvdV>> it was yet another level of microcode. A very limited set of
 MvdV>> instructions that were stored in a diode matrix that formed the
 MvdV>> micro programme that defined the instruction set.

 BS> Ah, all processors today contains microcode I guess.. However at that
 BS> time it was'nt a fact I guess.

The unique thing about the Multi-8 was that the micro code was accessible to
the user. It came with a full description of the microcode instruction set.

 MvdV>> In theory it was possible to rewrite (and rewire) the
 MvdV>> microprogramme to change the macro instruction set.

 BS> Cool, I hope that such things are read-only today ;)

The microcode was stored in a diode matrix. A printed circuit some 30x30 cm.
Each diode represented one bit. To change it one needed a soldering iron.

 MvdV>> As it was the macro instruction set was basically 8 bit, Its
 MvdV>> addressing space was 8 bit, but it provided for dealing with
 MvdV>> variable length registers of 1, 2, 3 or 4 bytes.

 BS> I would survive with an 8bit computer today, doing math with long
 BS> integers would just take some more time...

Even a one bit machine can do that. Turing proved it... ;-)

 MvdV>> Another interesting thing was that because of the magnetic core
 MvdV>> memory, one did not have to reload after a shut down. The content
 MvdV>> of memory was preserved. So when properly shut down (manually or
 MvdV>> by the power fail interrupt), once power came up, it would just
 MvdV>> carry one where it was shut down without having to reload
 MvdV>> anything.

 BS> Cool! With today's electronic building blocks it's hardly possible to
 BS> create such a construction.

OTOH, that 12K core memory had the size of three today's complete mother
boards...

 BS> Make sense, a lot of languages constructed at that time, seemed
 BS> really BASIC-like.. I've not programmed in any of them, but I think
 BS> of ALGOL as an example, am I right?

Nope. Algol is not at all like Basic. I did some programming in Algol. It is
more like Pascal. I would even say, Algol was a precursor of Pascal.

 BS> It was probably constructed before the first BASIC interpreters.. So
 BS> what came first, the chicken or the egg?

Algol was a compiled language. It dates from before the time that the user had
direct access to the computer via a terminal. When I wrote in Algol, you handed
the deck of cards to the operator would put them in the reader. Some time later
(often hours) you were handed over the printed output. Batch jobs only.

 MvdV>> It has been over a decade since I last fired up my universal
 MvdV>> cross assembler, so it would take a while to pick it up again. It
 MvdV>> is not like swimming, one DOES forget...

 BS> Indeed, what I think is the most difficult to remember is the number
 BS> of the interrupts.

Most of those 8 bitters had only three interrupts. The maskable interrupt, the
non-maskable interrupt and the software interrupt.

 BS> The way you program assembler, is probably like swimming or riding a
 BS> bike.

No, it is not. One never forgets how to swim or how to ride a bike. But I
*have* forgotten how to start up the assembler. I would have to look it up
again.


Cheers, Michiel

--- GoldED+/W32-MINGW 1.1.5-b20070503
 * Origin: http://www.vlist.org (2:280/5555)