Maker Pro
Maker Pro

Blast from the past... Z80!

N

Nobody

Jan 1, 1970
0
there's some people doing that with microcontrollers now.

how was the display driven? was it a 8 bit SIPO shift register

It was a bidirectional shift register (it was also used for save/load
to/from cassette). But for video, it was used PISO ;)
 
D

David L. Jones

Jan 1, 1970
0
Was not Atmel for sale or sometng?

Yes, Microchip (PIC) tried to buy them. The GFC stepped in and ruined
the deal though, not to mention that Atmel put up a fight.

Dave.
 
Anthony said:
??? You must be mistaken; I've always been a PIC advocate. Now I'm in love
with ARM.

[Blush]. My sincere apologies Anthony!.
I'm totally intrigued now as to who the hell made the AVR case. I'll
have to check back.
Apologies again.
 
J

JosephKK

Jan 1, 1970
0
The variable length in the X86 is mostly the result of the extra bytes
specifying details about what is to be done. Consider the ADD
immediate to memory or register as an example.

The upper 6 bits of the first byte gives you what is being done. The
two lower give you the width of the ALU action and the immediate
operand.

The later bytes give the addressing details. This is how it is for
pretty much all of the longer ones.
Then there is also the base register prefixes, with up to 3 prefixes
possible. Nor were the address/register/mode bitfield positions and
widths 100% consistent, they could even appear in different octets
(bytes).
I was around and did read about it. This is why I pointed out that
the breaking apart of the load/store and the ALU operations was not
the only thing under consideration. Early in the RISC development
they went back to the fixed instruction length. The PDP-8 was fixed
instruction length and was created before instruction sets got complex
enough to need reducing.
Do you also remember such things as writable control stores?
[...]
Load/store ops get separated from ALU
ops and RMW ops are unheard of.  That is the *key* difference
between RISC and CISC.

Not really key.  The original main objective was a short pipeline and
single cycle execution once the operands were available.  This is what
rules out RMW opcodes.

That really was a sub goal to get to the goal of getting greater speed
by having more efficient use of transistors. Transistors were not
considered free so any that were used for an ASCII adjust for divide
couldn't be used for something else that really mattered.
Close enough.
 
M

MooseFET

Jan 1, 1970
0
On Sat, 31 Jan 2009 19:21:07 -0800 (PST), MooseFET





Then there is also the base register prefixes, with up to 3 prefixes
possible.  Nor were the address/register/mode bitfield positions and
widths 100% consistent, they could even appear in different octets
(bytes).

Lets also not forget the REP prefixes. They are a strange duck in
that they work a lot like a decrement and conditional jump was added
after the string instruction.


[...]
Do you also remember such things as writable control stores?

Yes.

For those who don't know, things like the IBM 360 had a control store
that had to be loaded with the micro code before the thing would have
its defined instruction set.

I am also old enough to have worked on the design of bit slice
computers, printed out my listings on a teletype, input my code on
paper tape, and write programs on punch cards that were sent away to
the computer.

[...]
 
K

krw

Jan 1, 1970
0
[email protected]>, [email protected]
says...>
Then there is also the base register prefixes, with up to 3 prefixes
possible.  Nor were the address/register/mode bitfield positions and
widths 100% consistent, they could even appear in different octets
(bytes).

Lets also not forget the REP prefixes. They are a strange duck in
that they work a lot like a decrement and conditional jump was added
after the string instruction.
[...]
Do you also remember such things as writable control stores?

Yes.

For those who don't know, things like the IBM 360 had a control store
that had to be loaded with the micro code before the thing would have
its defined instruction set.

Which IBM 360? "Things like the IBM 360" is a meaningless
statement, when talking about hardware (implementation). "IBM
360" signifies an architecture not an implementation. Each model's
implementation was quite different from the others in the series.
I am also old enough to have worked on the design of bit slice
computers, printed out my listings on a teletype, input my code on
paper tape, and write programs on punch cards that were sent away to
the computer.

[...]
 
L

LM

Jan 1, 1970
0
I put together a Z80 microcomputer the other day.  Real simple:
processor, clock oscillator, NVRAM holding program and data, pair of
Brings fond memories. Who sells those chips these days. Like 808* and
Z8*.
Last time I checked distributor didn't even reqogbize those chips.

Someday I hopefully have time to play with those chips.
 
A

Archimedes' Lever

Jan 1, 1970
0
Brings fond memories. Who sells those chips these days. Like 808* and
Z8*.
Last time I checked distributor didn't even reqogbize those chips.


All kinds of chips that you think are gone are likely still available.

I would be wary of your 'distributor' if he comes back with crap
responses like that.
Someday I hopefully have time to play with those chips.

Didn't you just say you can't get them? D'oh! Time is not even in that
formula.
 
Have a look at the HD64180. It is a Hitachi chip with a Z80 core. It
also has a memory management unit which allows it to access 1MB of
memory. It also has a number of other upgrades such as serial ports
and timers. I have designed a few projects with it. And we still
have many intruments in production that use the chip. Though many of
our newer products use the H8.
 
J

JosephKK

Jan 1, 1970
0
Lets also not forget the REP prefixes. They are a strange duck in
that they work a lot like a decrement and conditional jump was added
after the string instruction.
Not only that i know two other architectures that have such prefix /
instructions. There may be more.
[...]
Do you also remember such things as writable control stores?

Yes.

For those who don't know, things like the IBM 360 had a control store
that had to be loaded with the micro code before the thing would have
its defined instruction set.

I am also old enough to have worked on the design of bit slice
computers, printed out my listings on a teletype, input my code on
paper tape, and write programs on punch cards that were sent away to
the computer.

[...]

You ain't the only one. Never got to build my bit slice machine
though. Strange, it influenced the design of many machines along the
way.
 
M

MooseFET

Jan 1, 1970
0
[....]
I am also old enough to have worked on the design of bit slice
computers, printed out my listings on a teletype, input my code on
paper tape, and write programs on punch cards that were sent away to
the computer.

You ain't the only one.  Never got to build my bit slice machine
though.  Strange, it influenced the design of many machines along the
way.

*my* bit slice didn't get built. Somebody else's did. I also
consulted to someone building a 3rd one that I can't go into.

I also did a machine that was a lot like a bit slice that didn't
really have an ALU. It was a system that needed to read a bunch of
bytes from memory and rearrange the nibbles quickly. The idea of
using a registered PROM and a counter to make logic signals has popped
up many times.
 
M

MooseFET

Jan 1, 1970
0
MooseFET wrote: [... computer stuff ...]

...and you were from Florida with the name of "chad"...

I know someone called Chad. He didn't like hearing about all the
hanging chad.

BTW: The term chad got attached because a guy named Chadless invented
one that only cut 3 sided and folded the bit of paper over. People
mistook it for being a name for one that doesn't make chad and thuse
the little bits must be chad.
 
K

krw

Jan 1, 1970
0
Hm... I remember that the microcode for the S360/30 was suppied as
CCROS (silver(?)-coated mylar(?) capacitor arrays) and the S360/40
used TROS (transformer "loops" embedded in thin plastic strips).
You could modify the CCROS cards with an 026/029 keypunch, but you
needed a special leather-hole-punch-like tool to "program" the
TROS strips.

WCS (writeable, loaded from floppy) was present on the S370/158
and possibly the S360/67, but I'm sure there were other wild and
wonderful variants. <grin!>

The 360/75 (and the /195, IIRC) was hardwired. No microcode at
all.
Ah, the joys of nostalgia. <grin!>

History lessons. ;-)
 
J

JosephKK

Jan 1, 1970
0
[....]
I am also old enough to have worked on the design of bit slice
computers, printed out my listings on a teletype, input my code on
paper tape, and write programs on punch cards that were sent away to
the computer.

You ain't the only one.  Never got to build my bit slice machine
though.  Strange, it influenced the design of many machines along the
way.

*my* bit slice didn't get built. Somebody else's did. I also
consulted to someone building a 3rd one that I can't go into.

I also did a machine that was a lot like a bit slice that didn't
really have an ALU. It was a system that needed to read a bunch of
bytes from memory and rearrange the nibbles quickly. The idea of
using a registered PROM and a counter to make logic signals has popped
up many times.
Ahhh. Primitive state machines. Another useful tool that has been
around a while. Some of the current tools for state machines help you
set up multiple interacting state machines. Now that can be fun.
 
M

MooseFET

Jan 1, 1970
0
[....]
I am also old enough to have worked on the design of bit slice
computers, printed out my listings on a teletype, input my code on
paper tape, and write programs on punch cards that were sent away to
the computer.
[...]
You ain't the only one.  Never got to build my bit slice machine
though.  Strange, it influenced the design of many machines along the
way.
*my* bit slice didn't get built.  Somebody else's did.  I also
consulted to someone building a 3rd one that I can't go into.
I also did a machine that was a lot like a bit slice that didn't
really have an ALU.  It was a system that needed to read a bunch of
bytes from memory and rearrange the nibbles quickly.  The idea of
using a registered PROM and a counter to make logic signals has popped
up many times.

Ahhh.  Primitive state machines.  Another useful tool that has been
around a while.  Some of the current tools for state machines help you
set up multiple interacting state machines.  Now  that can be fun.

The project I am working on has multiple state machines in the micro.
They still make a great way to arrange the problem to be solved.

Some years back, these guys showed up trying to sell a state machine
based product development tool for products. It was a very clever
scam. Among their claims was that their product output C code that
was 100% if your design was correct in their tool. When I got into
the details, I discovered that to develop you wrote stuff like:

if (some expression) tra-la-tra-la {
Some code to do tra-la-tra-la
}

they then ran using that code. When it came time to output, they just
had to strip out the "tra-la-tra-la". I spent a weekend and made a
more useful tool. Their's required you to basically write your
application before you could show what it did to marketing. Mine made
a very quick fake up of the user interface with a lot less code.
 
R

Rich Grise

Jan 1, 1970
0
Looks like Zilog is still alive:

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=269-3898-ND

They have some nice modern chips, too.

http://www.zilog.com

Maybe i hit them at a bad time, but it is the worst pig dog slow site
i have seen for months. I have about 1.5 Mb/s download, but it took
over a minute to load, and was crap slow afterwards. That got my
attention, negatively.[/QUOTE]

These days, wouldn't you just plop a Z80 core into an FPGA or whatever?

Cheers!
Rich
 
K

krw

Jan 1, 1970
0
These days, wouldn't you just plop a Z80 core into an FPGA or whatever?

No. I wouldn't waste the transistors on a Z80. ;-)

FPGAs still cost real money and there are far better solutions. I
also wouldn't use an unsupported tool chain.
 
Top