🚀 go-pugleaf

I think the Z80 is definitely regarded as more powerful. It was found in more 
expensive systems, and it was a modified version of the 8080 with extra features.

John Savard

On Wed, 06 Jun 2018 17:17:46 -0400, Andreas Kohlbach
<a...@spamfence.net> wrote:

>Back in the late 70s and mid 80s you could chose between a Z80 and 6502,
>when it came to 8-bit home computers. Sure there where some micros using
>a M6809 (almost 16-bit) and some 8080 CPUs. But most of the micros then
>ran a 6502 or Zilog Z80.
>
>Which one is more powerful, not regarding custom chips taking load off the
>CPU, like early "GPUs" chips in the Commodore 64?
>
>I say the Z80 was. It could pair some 8bit registers to 16bit
>entities. It also already had some kind of primitives like the later 8086
>and all future microprocessors had the 6502 lacked, speeding up things
>and made programming in assembly easier.

     I am inclined to agree.  I did a bit of programming both, and it
seemed that more instructions were needed to make a 6502 go.

Sincerely,

Gene Wirchenko

On 07/06/2018 05:48, Gene Wirchenko wrote:
> ...  and it
> seemed that more instructions were needed to make a 6502 go.

... as might be said for any RISC architecture today!  :-)

On Wed, 06 Jun 2018 17:17:46 -0400
Andreas Kohlbach <a...@spamfence.net> wrote:

> Back in the late 70s and mid 80s you could chose between a Z80 and 6502,
> when it came to 8-bit home computers. Sure there where some micros using
> a M6809 (almost 16-bit) and some 8080 CPUs. But most of the micros then
> ran a 6502 or Zilog Z80.
> 
> Which one is more powerful, not regarding custom chips taking load off the
> CPU, like early "GPUs" chips in the Commodore 64?

	That was a regular argument in the late 70s to mid 80s, almost as
common as emacs/vi. There really wasn't enough in it to matter, the Z80 was
nicer to program the 6502 had some efficient tricks. Both made decent 8 bit
micros.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:\>WIN                                     | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
> 	That was a regular argument in the late 70s to mid 80s, almost as
> common as emacs/vi. 


Never could understand the quasi-religious fervour about them,
for EDLIN provided all that was ever necessary in an editor.

The time spent editing a program is infinitesimally insignificant
compared to the time that a progam may exist.

Gareth's Downstairs Computer <headstone255.but.n...@yahoo.com> writes:
>On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
>> 	That was a regular argument in the late 70s to mid 80s, almost as
>> common as emacs/vi. 
>
>
>Never could understand the quasi-religious fervour about them,
>for EDLIN provided all that was ever necessary in an editor.

EDLIN is no better than a keypunch and a deck of cards.

Gareth's Downstairs Computer
<headstone255.but.n...@yahoo.com> wrote:
> On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
>> That was a regular argument in the late 70s to mid 80s, almost as
>> common as emacs/vi. 
> 
> 
> Never could understand the quasi-religious fervour about them,
> for EDLIN provided all that was ever necessary in an editor.
> 
> The time spent editing a program is infinitesimally insignificant
> compared to the time that a progam may exist.
> 
> 

Sure, butbthe time spent editing is a large percentage of the time the
programmer spends working on it. Afterwards it may exist for years without
being touched, while the programmer has moved on to working with other
code. Besides vi vs. emacs is a useless argument - ISPF is far and away the
best editor.

-- 
Pete

On 7 Jun 2018 20:23:30 GMT
Huge <Huge@nowhere.much.invalid> wrote:

> (Actually, the proper answer is that they're all crap.)

	Yep but some are more crap than others.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:\>WIN                                     | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

On Thu, 07 Jun 2018 14:52:34 -0400, Peter Flass wrote:

> Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
>> On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
>>> That was a regular argument in the late 70s to mid 80s, almost as
>>> common as emacs/vi.
>> 
>> 
>> Never could understand the quasi-religious fervour about them, for
>> EDLIN provided all that was ever necessary in an editor.
>> 
>> The time spent editing a program is infinitesimally insignificant
>> compared to the time that a progam may exist.
>> 
>> 
>> 
> Sure, butbthe time spent editing is a large percentage of the time the
> programmer spends working on it. Afterwards it may exist for years
> without being touched, while the programmer has moved on to working with
> other code. Besides vi vs. emacs is a useless argument - ISPF is far and
> away the best editor.

Get the latest editor book:

 https://www.michaelwlucas.com/tools/ed

Despite the publication date, this is a real book.



-- 
Using UNIX since v6 (1975)...

Use the BIG mirror service in the UK:
 http://www.mirrorservice.org

Peter Flass <peter...@yahoo.com> writes:

> Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
>> On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
>>> That was a regular argument in the late 70s to mid 80s, almost as
>>> common as emacs/vi. 
>> 
>> 
>> Never could understand the quasi-religious fervour about them,
>> for EDLIN provided all that was ever necessary in an editor.
>> 
>> The time spent editing a program is infinitesimally insignificant
>> compared to the time that a progam may exist.
>
> Sure, butbthe time spent editing is a large percentage of the time the
> programmer spends working on it. Afterwards it may exist for years without
> being touched, while the programmer has moved on to working with other
> code. Besides vi vs. emacs is a useless argument - ISPF is far and away the
> best editor.

I considered myself an expert ISPF user (including the editor).
Sorry having to hit enter just didn't cut it for me.
Only 7 colors?
Emacs offers so much more.

-- 
Dan Espen

Huge <Huge@nowhere.much.invalid> writes:

> On 2018-06-07, Peter Flass <peter...@yahoo.com> wrote:
>
> [18 lines snipped]
>
>> Besides vi vs. emacs is a useless argument - ISPF is far and away the
>> best editor.
>
> BURN THE HERETIC!!!
>
> (Actually, the proper answer is that they're all crap.)

I fail to see how an editor that allows insane amounts of customization
can be crap.

-- 
Dan Espen

On Wed, 6 Jun 2018 17:54:58 -0700 (PDT), Quadibloc wrote:
>
> I think the Z80 is definitely regarded as more powerful. It was found in more 
> expensive systems, and it was a modified version of the 8080 with extra features.

The ZX 81 and ZX Spectrum were less powerful computers. At least when
compared to the Commodore 64 and other 8bit computers on the market at
that time, using a 6502 CPU. Due to the advanced GPU of for example the
Commodore 64.

Still I believe the Z80 was more powerful than the 6502, disregarding
customs chips for a certain computer.
-- 
Andreas

My random toughts and comments
https://news-commentaries.blogspot.com/

Dan Espen <dan1...@gmail.com> writes:
> I considered myself an expert ISPF user (including the editor).
> Sorry having to hit enter just didn't cut it for me.
> Only 7 colors?
> Emacs offers so much more.

after the 23june1969 unbundling announcement and starting to charge for
(applicaton) software, se services, etc ... POK had hard time adjusting
to the rules, monthly charge had to cover original development as well
as ongoing development and support. unbundling posts
http://www.garlic.com/~lynn/submain.html#unbundle

JES2 NJE couldn't come up with business case ... at any price, market
was somewhat price sensitive, as they raised the (forecasted) product
price (to try and make up the shortfall), the number of (forecasted)
customers dropped. What they eventually hit on was announcing JES2 NJE &
VM370 VNET as "joint" products ... where the revenue from VNET was used
to make up the NJE shortfall. hasp, jes, nje, etc posts
http://www.garlic.com/~lynn/submain.html#hasp

JES2 had tried $300/month, $600/month, & $1200/month for NJE ... but
forecasted customers were dropping faster than the price was going up.
VNET easily met its business case at $30/month ... they announced joint
product at $600/month (with effectively nearly all the money going to
NJE).

ISPF had similar problem coming up with business case ...  but by the
early 80s, they had relaxed the rules so the products just had to be in
the same organization. They moved vm370 performance products into the
same business unit as ISPF ... reducing the staff to three people
... and used the VM370 products revenue to underwrite the cost of the
200 people in the ISPF group.
https://en.wikipedia.org/wiki/ISPF

some misc. recent posts mentioning ISPF:
http://www.garlic.com/~lynn/2017c.html#2 ISPF (was Fujitsu Mainframe Vs IBM mainframe)
http://www.garlic.com/~lynn/2017c.html#82 Great mainframe history(?)
http://www.garlic.com/~lynn/2017e.html#25 [CM] What was your first home computer?
http://www.garlic.com/~lynn/2017g.html#34 Programmers Who Use Spaces Paid More
http://www.garlic.com/~lynn/2017i.html#23 progress in e-mail, such as AOL
http://www.garlic.com/~lynn/2018d.html#48 IPCS, DUMPRX, 3092, EREP

more (editor) drift, trying to promote RED which had quite a bit of
internal use and maturity at the time Endicott decided to develop XEDIT.
Endicott eventually responded it was the RED author's fault that RED was
so much better than XEDIT, and therefor it should be RED author's
responsibility to bring XEDIT up to RED level.
http://www.garlic.com/~lynn/2006u.html#email790606
in this (long winded) post along with other old RED/XEDIT email from the
period (and other subjects)
http://www.garlic.com/~lynn/2006u.html#26

-- 
virtualization experience starting Jan1968, online at home since Mar1970

In article <87po13w...@usenet.ankman.de>,
Andreas Kohlbach  <a...@spamfence.net> wrote:
>Back in the late 70s and mid 80s you could chose between a Z80 and 6502,
>when it came to 8-bit home computers. Sure there where some micros using
>a M6809 (almost 16-bit) and some 8080 CPUs. But most of the micros then
>ran a 6502 or Zilog Z80.
>
>Which one is more powerful, not regarding custom chips taking load off the
>CPU, like early "GPUs" chips in the Commodore 64?
>
>I say the Z80 was. It could pair some 8bit registers to 16bit
>entities. It also already had some kind of primitives like the later 8086
>and all future microprocessors had the 6502 lacked, speeding up things
>and made programming in assembly easier.

>From my own practical 1st hand experience of writing shed-loads of both
6502 and Z80 code, I'd say that typical systems were about the same -
however a typical 6502 was 1Mhz and a typical Z80 was 4Mhz, so clock
for clock the 6502 was faster.

I preferd the 6502 though and still use it to this day. (at 14Mhz now though)

Gordon

On Thu, 07 Jun 2018 23:38:06 GMT, Dan Espen <dan1...@gmail.com> wrote:

> Huge <Huge@nowhere.much.invalid> writes:
> 
>> On 2018-06-07, Peter Flass <peter...@yahoo.com> wrote:
>>
>> [18 lines snipped]
>>
>>> Besides vi vs. emacs is a useless argument - ISPF is far and away the
>>> best editor.
>>
>> BURN THE HERETIC!!!
>>
>> (Actually, the proper answer is that they're all crap.)
> 
> I fail to see how an editor that allows insane amounts of customization
> can be crap.
> 

It's the tragedy of the spreadsheet (abuse as a database);

You are lost in a maze of twisty passages, all slightly different.

-- 
Bah, and indeed, Humbug.

On Thu, 07 Jun 2018 21:31:40 GMT, Bob Eager <news...@eager.cx> wrote:

> On Thu, 07 Jun 2018 14:52:34 -0400, Peter Flass wrote:
> 
>> Gareth's Downstairs Computer
>> <headstone255.but.n...@yahoo.com> wrote:
>>> On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
>>>> That was a regular argument in the late 70s to mid 80s, almost as
>>>> common as emacs/vi.
>>> 
>>> 
>>> Never could understand the quasi-religious fervour about them, for
>>> EDLIN provided all that was ever necessary in an editor.
>>> 
>>> The time spent editing a program is infinitesimally insignificant
>>> compared to the time that a progam may exist.
>>> 
>>> 
>>> 
>> Sure, butbthe time spent editing is a large percentage of the time the
>> programmer spends working on it. Afterwards it may exist for years
>> without being touched, while the programmer has moved on to working 
with
>> other code. Besides vi vs. emacs is a useless argument - ISPF is far 
and
>> away the best editor.
> 
> Get the latest editor book:
> 
>  https://www.michaelwlucas.com/tools/ed
> 
> Despite the publication date, this is a real book.
> 
> 
> 

Here's a review:
https://dan.langille.org/2018/04/01/book-review-ed-mastery/
TL;DR - it's crap. So says it's author.

-- 
Bah, and indeed, Humbug.

On Thu, 07 Jun 2018 20:19:21 -0400, Andreas Kohlbach <a...@spamfence.net> wrote:
> On Wed, 6 Jun 2018 17:54:58 -0700 (PDT), Quadibloc wrote:
>>
>> I think the Z80 is definitely regarded as more powerful. It was found in more 
>> expensive systems, and it was a modified version of the 8080 with extra features.
>
> The ZX 81 and ZX Spectrum were less powerful computers. At least when
> compared to the Commodore 64 and other 8bit computers on the market at
> that time, using a 6502 CPU. Due to the advanced GPU of for example the
> Commodore 64.

Yep the 6502 based micros tended to have more support chips than stuff
like the ZX81 and Speccie. I am not sure I would call the VIC-II chip a
GPU but it was far more featureful than the Speccie's ULA.

Technically the C64 had a 6510 (integrated 6bit IO but in all other
senses a 6502). It also had the VIC-II, a 6581 (for sound) and 2 6526s
for IO.

The Spectrum managed with just a Z80 and the ULA. 

-- 
Andy Leighton => an...@azaal.plus.com
"We demand rigidly defined areas of doubt and uncertainty!"
   - Douglas Adams

On 08/06/2018 11:32, Andy Leighton wrote:
> 
> The Spectrum managed with just a Z80 and the ULA.

The ULA was made by Ferranti at Gem Mill, Oldham.

(I used to work at an electronics company that supplied
some of the test equipment.)

On Fri, 8 Jun 2018 12:16:51 +0100
Gareth's Downstairs Computer
<headstone255.but.n...@yahoo.com> wrote:

> On 08/06/2018 11:32, Andy Leighton wrote:
> > 
> > The Spectrum managed with just a Z80 and the ULA.
> 
> The ULA was made by Ferranti at Gem Mill, Oldham.

	Ferranti ULAs and the way some people read the spec sheets (this
many gates, up to this speed ... means we can have nearly all of them
going this fast right!) broke the Newbrain (Grundy eventually put the
TTL prototype into production), put pyjama stripes on Beebons (until the
"case support" was added) and caused many rude words to be uttered by
Sinclair engineers.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:\>WIN                                     | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

On Thu, 07 Jun 2018 19:38:06 -0400
Dan Espen <dan1...@gmail.com> wrote:

> I fail to see how an editor that allows insane amounts of customization
> can be crap.

	If you wind up spending more time customising it than editing then
that's not good.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:\>WIN                                     | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

On 08/06/2018 13:38, Ahem A Rivet's Shot wrote:
> On Fri, 8 Jun 2018 12:16:51 +0100
> Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
> 
>> On 08/06/2018 11:32, Andy Leighton wrote:
>>>
>>> The Spectrum managed with just a Z80 and the ULA.
>>
>> The ULA was made by Ferranti at Gem Mill, Oldham.
> 
> 	Ferranti ULAs and the way some people read the spec sheets (this
> many gates, up to this speed ... means we can have nearly all of them
> going this fast right!) broke the Newbrain (Grundy eventually put the
> TTL prototype into production), put pyjama stripes on Beebons (until the
> "case support" was added) and caused many rude words to be uttered by
> Sinclair engineers.
> 

They were going absolutely f(er)rantic?

Ahem A Rivet's Shot <ste...@eircom.net> writes:

> On Thu, 07 Jun 2018 19:38:06 -0400
> Dan Espen <dan1...@gmail.com> wrote:
>
>> I fail to see how an editor that allows insane amounts of customization
>> can be crap.
>
> 	If you wind up spending more time customising it than editing then
> that's not good.

Since you don't, no issue there.

-- 
Dan Espen

On 2018-06-08, Gareth's Downstairs Computer <headstone255.but.n...@yahoo.com> wrote:

> The ULA was made by Ferranti at Gem Mill, Oldham.

I worked there ... <http://www.jusme.com/perm/gem-mill-demo.jpg>

I also worked here ... <http://www.jusme.com/perm/man05-demo.jpg>

Muhahahaha!


-- 
Ian

"Tamahome!!!" - "Miaka!!!"

On 2018-06-08, Andreas Kohlbach <a...@spamfence.net> wrote:

> On Thu, 07 Jun 2018 14:50:45 GMT, Scott Lurndal wrote:
>>
>> Gareth's Downstairs Computer
>> <headstone255.but.n...@yahoo.com> writes:
>>
>>> On 07/06/2018 11:39, Ahem A Rivet's Shot wrote:
>>>
>>>> 	That was a regular argument in the late 70s to mid 80s, almost as
>>>> common as emacs/vi. 
>>>
>>> Never could understand the quasi-religious fervour about them,
>>> for EDLIN provided all that was ever necessary in an editor.
>>
>> EDLIN is no better than a keypunch and a deck of cards.
>
> *LOL* Good one.

I found edlin similar enough to CP/M's ed that I could quickly start
using it, but different enough that I would frequently get bitten.

-- 
/~\  cgibbs@kltpzyxm.invalid (Charlie Gibbs)
\ /  I'm really at ac.dekanfrus if you read it the right way.
 X   Top-posted messages will probably be ignored.  See RFC1855.
/ \  Fight low-contrast text in web pages!  http://contrastrebellion.com

On Fri, 8 Jun 2018 08:10:09 -0000 (UTC), Gordon Henderson wrote:
>
> In article <87po13w...@usenet.ankman.de>,
> Andreas Kohlbach  <a...@spamfence.net> wrote:
>>Back in the late 70s and mid 80s you could chose between a Z80 and 6502,
>>when it came to 8-bit home computers. Sure there where some micros using
>>a M6809 (almost 16-bit) and some 8080 CPUs. But most of the micros then
>>ran a 6502 or Zilog Z80.
>>
>>Which one is more powerful, not regarding custom chips taking load off the
>>CPU, like early "GPUs" chips in the Commodore 64?
>>
>>I say the Z80 was. It could pair some 8bit registers to 16bit
>>entities. It also already had some kind of primitives like the later 8086
>>and all future microprocessors had the 6502 lacked, speeding up things
>>and made programming in assembly easier.
>
> From my own practical 1st hand experience of writing shed-loads of both
> 6502 and Z80 code, I'd say that typical systems were about the same -
> however a typical 6502 was 1Mhz and a typical Z80 was 4Mhz, so clock
> for clock the 6502 was faster.

You mean the clock was faster for the Z80?

What about primitives the Z80 supported, if you can call them primitives?

If you want to copy 1000 bytes from $1000 to $1005 with LDIR:

LD     HL, $1000
LD     DE, $1005
LD     BC, 1000
LDIR

It would be more complex if you wanted to achieve this with a 6502.

> I preferd the 6502 though and still use it to this day. (at 14Mhz now though)

I like it old school. No 6502 should ever run faster than 1 MHz. ;-)
-- 
Andreas

https://news-commentaries.blogspot.com/2018/06/the-zilog-z80-is-8-bit-microprocessor.html

On Fri, 08 Jun 2018 05:32:41 -0500, Andy Leighton wrote:
>
> On Thu, 07 Jun 2018 20:19:21 -0400, Andreas Kohlbach <a...@spamfence.net> wrote:
>> On Wed, 6 Jun 2018 17:54:58 -0700 (PDT), Quadibloc wrote:
>>>
>>> I think the Z80 is definitely regarded as more powerful. It was found in more 
>>> expensive systems, and it was a modified version of the 8080 with extra features.
>>
>> The ZX 81 and ZX Spectrum were less powerful computers. At least when
>> compared to the Commodore 64 and other 8bit computers on the market at
>> that time, using a 6502 CPU. Due to the advanced GPU of for example the
>> Commodore 64.
>
> Yep the 6502 based micros tended to have more support chips than stuff
> like the ZX81 and Speccie. I am not sure I would call the VIC-II chip a
> GPU but it was far more featureful than the Speccie's ULA.

The VIC-II had sprites, could do soft scrolling, thus took load off the
CPU. You would just set a new value for a register if the VIC-II to move
an object. On a Sinclair Spectrum you would need to move a bunch of data
around and redraw the screen by software. Not sure though if the term GPU
I used applies for the VIC-II though.
-- 
Andreas

My random toughts and comments
https://news-commentaries.blogspot.com/

On Friday, June 8, 2018 at 12:30:07 PM UTC-7, Andreas Kohlbach wrote:
> [...]
> I like it old school. No 6502 should ever run faster than 1 MHz. ;-)
> -- 
> Andreas
> 
> https://news-commentaries.blogspot.com/2018/06/the-zilog-z80-is-8-bit-microprocessor.html

Maybe we all have a different point of view regarding the definition
 of "powerful".  In the hands of an expert, either one could produce
 impressive results.  The Z-80 had better machine code density, and
 the 6502 had a better CPI ratio.  I personally prefer 6502 assembly
 language over Z-80, but that may be because I learned 6502 first.  

In my own experiences, MS-BASIC on the 1.023 MHz Apple II+ with 40-bit
 floats was visibly faster than MS-BASIC on the 1.774 MHz TRS-80 M1 L2
 with 32-bit floats, but that's not really the same as comparing native
 machine code performance.  Maybe Mr. Gates spent more effort on the
 6502 version of the interpreter, or was too busy to optimize the older
 8080 version for the Z-80.

Mike B.

On Saturday, June 9, 2018 at 9:59:53 PM UTC-6, barr...@yahoo.com wrote:

>  Maybe we all have a different point of view regarding the definition
>  of "powerful".

That could be.

But the Z-80 had extended instructions that other 8-bit microprocessors did not: 
block copy, byte search, 16-bit arithmetic instructions.

While the Z-80 had a lot of fancy features, though, the 6809, which came along 
later, would have been significantly more powerful yet - as it had a hardware 
*multiply* instruction, even if it wasn't crammed with as many bells and 
whistles as the Z-80, most programs would run considerably faster on the 6809 as 
a result of that.

John Savard

Andreas Kohlbach wrote:

> Back in the late 70s and mid 80s you could chose between a Z80 and 6502,
> when it came to 8-bit home computers. Sure there where some micros using
> a M6809 (almost 16-bit) and some 8080 CPUs. But most of the micros then
> ran a 6502 or Zilog Z80.
> 
> Which one is more powerful, not regarding custom chips taking load off the
> CPU, like early "GPUs" chips in the Commodore 64?
> 
> I say the Z80 was. It could pair some 8bit registers to 16bit
> entities. It also already had some kind of primitives like the later 8086
> and all future microprocessors had the 6502 lacked, speeding up things
> and made programming in assembly easier.
Yes, the Z80 had a lot of extra features related to handling 16-bit values.
You could do some 16-bit arithmetic, it had 3 16-bit index registers that 
could be used as pointers to memory.

One instruction group I used a lot was the INIR/OUTIR group.  You could set 
one of the 16-bit registers as a memory pointer, and one as a counter, and 
then execute one of these instructions.  By manipulating the WAIT signal, 
you could have it transfer an arbitrary number of bytes from one I/O 
register to a block of memory.  It was essentially a CPU-driven version of 
DMA, with no extra hardware needed.  I used it in mag tape controls, SCSI 
disk interfaces and a bunch of other places.  No other 8-bit CPU had a 
feature like this.

Jon

Quadibloc <jsa...@ecn.ab.ca> wrote:
> On Saturday, June 9, 2018 at 9:59:53 PM UTC-6, barr...@yahoo.com wrote:
> 
>> Maybe we all have a different point of view regarding the definition
>> of "powerful".
> 
> That could be.
> 
> But the Z-80 had extended instructions that other 8-bit microprocessors did not: 
> block copy, byte search, 16-bit arithmetic instructions.
> 
> While the Z-80 had a lot of fancy features, though, the 6809, which came along 
> later, would have been significantly more powerful yet - as it had a hardware 
> *multiply* instruction, even if it wasn't crammed with as many bells and 
> whistles as the Z-80, most programs would run considerably faster on the 6809 as 
> a result of that.

After a lot of research the 6809 was my pick. It's too bad it didn't catch
on, and of course in a few years the whole 8-bit thing was moot.

-- 
Pete

<barry...@yahoo.com> wrote:
> On Friday, June 8, 2018 at 12:30:07 PM UTC-7, Andreas Kohlbach wrote:
>> [...]
>> I like it old school. No 6502 should ever run faster than 1 MHz. ;-)
>> -- 
>> Andreas
>> 
>> https://news-commentaries.blogspot.com/2018/06/the-zilog-z80-is-8-bit-microprocessor.html
> 
> Maybe we all have a different point of view regarding the definition
>  of "powerful".  In the hands of an expert, either one could produce
>  impressive results.  The Z-80 had better machine code density, and
>  the 6502 had a better CPI ratio.  I personally prefer 6502 assembly
>  language over Z-80, but that may be because I learned 6502 first.  

I can't recall the details, but my recollection is that the 6502
instruction set was missing a couple of instructions I considered
important, vis-a-vis other microprocessors.

-- 
Pete

On Sunday, June 10, 2018 at 4:18:55 AM UTC-7, Peter Flass wrote:
> [...]
> 
> After a lot of research the 6809 was my pick. It's too bad it didn't catch
> on, and of course in a few years the whole 8-bit thing was moot.
> 
> -- 
> Pete

Moot?  Today there are billions of small devices quietly executing 6502
 machine code, and I am willing to speculate that the Z-80 isn't far off
 that mark.

The 6809 was very sweet.  Its descendants are even sweeter.

https://www.nxp.com/docs/en/reference-manual/S12CPUV2.pdf

Mike B.

On Sunday, June 10, 2018 at 8:22:27 AM UTC-6, barr...@yahoo.com wrote:
 
> Moot?  Today there are billions of small devices quietly executing 6502
>  machine code, and I am willing to speculate that the Z-80 isn't far off
>  that mark.

> The 6809 was very sweet.  Its descendants are even sweeter.

That is all very well, but certainly given the expense of a keyboard, a mouse, a 
monitor, a power supply, and varioius ancilliary circuits, if one is going to 
plunk down the money for a computer to sit down at, and use for playing games, 
spreadsheet calculations, surfing the web, and so on, it's worth the money to 
invest in something more powerful than an 8-bit microprocessor. In fact, not 
only don't they make desktop computers or laptops with 8-bit processors these 
days, even *smartphones* are 32 or 64 bits, with hardware floating-point.

Thus, while 8-bit processors do indeed carry out much useful work, they do so 
while hidden away, their instruction sets only seen by people working for OEMs 
behind the scenes.

John Savard

Of course, thinking of the Motorola 6809 has reminded me that in the days of the 
old 8-bit processors, AMD, today giving Intel a run for its money thanks to 
Ryzen, was for a time the maker of the most powerful chip out there - although 
it wasn't a microprocessor, only a coprocessor.

Remember the Am9512? Or even the Am9511A. But the Am9512 would be the one that 
made me genuinely excited, since it handled double precision, not just single 
precision.

However, that was when the IEEE 754 standard already was being worked on, so the 
8086 and 8087 were out there in the time of the 9512, whereas the 9511 was 
available back in the days of 8-bit micros (although some early 16-bit ones, 
like the 9900 from TI, were probably around too).

John Savard

On Sat, 9 Jun 2018 21:06:28 -0700 (PDT), Quadibloc wrote:
>
> On Saturday, June 9, 2018 at 9:59:53 PM UTC-6, barr...@yahoo.com wrote:
>
>>  Maybe we all have a different point of view regarding the definition
>>  of "powerful".
>
> That could be.
>
> But the Z-80 had extended instructions that other 8-bit microprocessors did not: 
> block copy, byte search, 16-bit arithmetic instructions.

Isn't  that something you call "primitives"? When reading through some
Intel 8086 the term "primitives" was mentioned when it came to such functions.

> While the Z-80 had a lot of fancy features, though, the 6809, which came along 
> later, would have been significantly more powerful yet - as it had a hardware 
> *multiply* instruction, even if it wasn't crammed with as many bells and 
> whistles as the Z-80, most programs would run considerably faster on the 6809 as 
> a result of that.

Isn't a 6809 considered an 8-bit CPU with a lot of 16-bit features? May
be like a 68000 is a 16-bit CPU with some 32-bit features?
-- 
Andreas

My random toughts and comments
https://news-commentaries.blogspot.com/

On Sun, 10 Jun 2018 17:33:59 -0400
Andreas Kohlbach <a...@spamfence.net> wrote:

> Isn't a 6809 considered an 8-bit CPU with a lot of 16-bit features? May
> be like a 68000 is a 16-bit CPU with some 32-bit features?

	The Motorola engineer who came to talk to us about the 6809 and
actually talked about the 68000, which had come off the company
confidential list so recently that the documents still had the security
legends, gave the strong impression that the 6809 was deliberate practice
for the 68000 which is what they wanted to do but weren't ready for.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:\>WIN                                     | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

On Mon, 11 Jun 2018 08:00:27 +0100, Ahem A Rivet's Shot <ste...@eircom.net> wrote:
> On Sun, 10 Jun 2018 17:33:59 -0400
> Andreas Kohlbach <a...@spamfence.net> wrote:
>
>> Isn't a 6809 considered an 8-bit CPU with a lot of 16-bit features? May
>> be like a 68000 is a 16-bit CPU with some 32-bit features?
>
> 	The Motorola engineer who came to talk to us about the 6809 and
> actually talked about the 68000, which had come off the company
> confidential list so recently that the documents still had the security
> legends, gave the strong impression that the 6809 was deliberate practice
> for the 68000 which is what they wanted to do but weren't ready for.

I am not so sure about that. The 6809 seems like an outgrowth of the
6800. Take for example the registers - the 6809 just had A and B (like
the 6800) but could also treat them together as a 16bit register (like
the Z80's BC,DE,HL), whereas the 68K had 8 32bit general purpose 
registers. That's a big difference in design philosophy.

-- 
Andy Leighton => an...@azaal.plus.com
"We demand rigidly defined areas of doubt and uncertainty!"
   - Douglas Adams

On Monday, June 11, 2018 at 2:20:19 AM UTC-6, Andy Leighton wrote:
> On Mon, 11 Jun 2018 08:00:27 +0100, Ahem A Rivet's Shot <ste...@eircom.net> wrote:

> > 	The Motorola engineer who came to talk to us about the 6809 and
> > actually talked about the 68000, which had come off the company
> > confidential list so recently that the documents still had the security
> > legends, gave the strong impression that the 6809 was deliberate practice
> > for the 68000 which is what they wanted to do but weren't ready for.

> I am not so sure about that. The 6809 seems like an outgrowth of the
> 6800. Take for example the registers - the 6809 just had A and B (like
> the 6800) but could also treat them together as a 16bit register (like
> the Z80's BC,DE,HL), whereas the 68K had 8 32bit general purpose 
> registers. That's a big difference in design philosophy.

That's definitely true - the 6809 was an improved 6800.

But since it did have a hardware multiply, designing it did still give some 
experience relevant to making the 68000, I would think.

John Savard

On Mon, 11 Jun 2018, Quadibloc wrote:

> On Monday, June 11, 2018 at 2:20:19 AM UTC-6, Andy Leighton wrote:
>> On Mon, 11 Jun 2018 08:00:27 +0100, Ahem A Rivet's Shot <ste...@eircom.net> wrote:
>
>>> 	The Motorola engineer who came to talk to us about the 6809 and
>>> actually talked about the 68000, which had come off the company
>>> confidential list so recently that the documents still had the security
>>> legends, gave the strong impression that the 6809 was deliberate practice
>>> for the 68000 which is what they wanted to do but weren't ready for.
>
>> I am not so sure about that. The 6809 seems like an outgrowth of the
>> 6800. Take for example the registers - the 6809 just had A and B (like
>> the 6800) but could also treat them together as a 16bit register (like
>> the Z80's BC,DE,HL), whereas the 68K had 8 32bit general purpose
>> registers. That's a big difference in design philosophy.
>
> That's definitely true - the 6809 was an improved 6800.
>
> But since it did have a hardware multiply, designing it did still give some
> experience relevant to making the 68000, I would think.
>
But I thought the two were almost concurrent in design.

There was a three part article in Byte about how they came to the 6809's 
architecture, and it was by studying how the 6800 was used.  It was also 
intended to be useful by a higher level language, the two stacks hint at 
that.  They didn't take the 6800 and add things on, they studied the 6800 
and improved on it.

But given the time, the 68000 was probably in design around the same time. 
And the 68000 is not an extension of the 6909, it's a big shift to a 16bit 
processor.

It's not unlike the Apple IIGS, which has that CPU that can be shifted to 
a 16bit CPU rather than a 6502.  It came out after the Macintosh has come 
out, a different path rather than an intermediate step.

We heard about the 6809 and the 68000 about the same time, if I recall 
properly.

   Michael

On Sunday, June 10, 2018 at 12:06:28 AM UTC-4, Quadibloc wrote:
> On Saturday, June 9, 2018 at 9:59:53 PM UTC-6, barr...@yahoo.com wrote:
> 
> >  Maybe we all have a different point of view regarding the definition
> >  of "powerful".
> 
> That could be.
> 
> But the Z-80 had extended instructions that other 8-bit microprocessors did not: 
> block copy, byte search, 16-bit arithmetic instructions.
> 
> While the Z-80 had a lot of fancy features, though, the 6809, which came along 
> later, would have been significantly more powerful yet - as it had a hardware 
> *multiply* instruction, even if it wasn't crammed with as many bells and 
> whistles as the Z-80, most programs would run considerably faster on the 6809 as 
> a result of that.

Excellent points.   Asking which is "more powerful" requires context
to make the question proper.  So many questions in computers are
answered by  "In Depends".  This has gone back to since when computers
were first invented--different computers were designed to run best
with different applications.   To this day, while a great many computers
are general purpose, there still is a need for certain specialty
machines for specialty applications, such as the supercomputers
used at Los Alamos.

On Sunday, June 10, 2018 at 12:15:10 AM UTC-4, Jon Elson wrote:
> Andreas Kohlbach wrote:
> 
> > Back in the late 70s and mid 80s you could chose between a Z80 and 6502,
> > when it came to 8-bit home computers. Sure there where some micros using
> > a M6809 (almost 16-bit) and some 8080 CPUs. But most of the micros then
> > ran a 6502 or Zilog Z80.
> > 
> > Which one is more powerful, not regarding custom chips taking load off the
> > CPU, like early "GPUs" chips in the Commodore 64?
> > 
> > I say the Z80 was. It could pair some 8bit registers to 16bit
> > entities. It also already had some kind of primitives like the later 8086
> > and all future microprocessors had the 6502 lacked, speeding up things
> > and made programming in assembly easier.
> Yes, the Z80 had a lot of extra features related to handling 16-bit values.
> You could do some 16-bit arithmetic, it had 3 16-bit index registers that 
> could be used as pointers to memory.
> 
> One instruction group I used a lot was the INIR/OUTIR group.  You could set 
> one of the 16-bit registers as a memory pointer, and one as a counter, and 
> then execute one of these instructions.  By manipulating the WAIT signal, 
> you could have it transfer an arbitrary number of bytes from one I/O 
> register to a block of memory.  It was essentially a CPU-driven version of 
> DMA, with no extra hardware needed.  I used it in mag tape controls, SCSI 
> disk interfaces and a bunch of other places.  No other 8-bit CPU had a 
> feature like this.

In reading the detailed descriptions above the innards of the 
respective chips, the manual for the IBM 603 electronic calculator
comes to mind.  This manual, available on the IBM website, describes,
in detail, the circuitry and functionality of the machine.  Worth a
look.  (The 603 was IBM's first mass produced electronic calculator,
a predecessor to the very popular 604.)

https://www.ibm.com/ibm/files/D165357D44082B98/us__en_us__ibm100__603calc__603_manual.pdf