9 Apr
2021
9 Apr
'21
5:34 p.m.
On 09/04/2021 11:12, emanuel stiebler wrote: >
You're comparing a z80 SBC running CP/M? Or are you thinking of 68000 SBCs?
Z80 CP/M machines were still competitive in 1981-1983 (Osborne, Kaypro)
I've never seen a 68k SBC. Have I missed out
something along the way? Is there a community for 68k SBC's? Kind regards, Andrew
Well, Rob Pike designed one: http://doc.cat-v.org/bell_labs/blit/
I guess the original hacker scene for the 68K was around Hal Hardenberg’s newsletter:
https://en.wikipedia.org/wiki/DTACK_Grounded
The ready-made 68K SBC’s only arrived 1984-1985:
https://en.wikipedia.org/wiki/Sinclair_QL (I think Linus Torvalds owned one)
https://en.wikipedia.org/wiki/Atari_ST
https://en.wikipedia.org/wiki/Macintosh_128K
https://en.wikipedia.org/wiki/Amiga_1000
All these machines are rather similar at the hardware level - 68K processor, RAM shared
between CPU and display. Only the Amiga had a (simple) hardware GPU.
What set the SUN-1 apart was its MMU, which none of the above have.
What influenced the timing was probably that Motorola made the 68K more affordable by the
mid-80’s.
Paul
9 Apr
9 Apr
7:01 p.m.
On Fri, Apr 9, 2021 at 11:35 AM Paul Ruizendaal via TUHS <
tuhs(a)minnie.tuhs.org> wrote:
There is an active community around DIY 68k SBCs these days. Some
representative examples:
https://www.eejournal.com/article/wallowing-in-68k-nostalgia/
https://www.ist-schlau.de
https://www.bigmessowires.com/category/68katy/
https://github.com/74hc595/68k-nano
http://mc68k.blogspot.com/2012_10_01_archive.html
There are even a couple of fairly advanced 68030 design floating around:
https://www.retrobrewcomputers.org/doku.php?id=boards:sbc:gryphon_68030:sta…
https://www.retrobrewcomputers.org/doku.php?id=boards:ecb:kiss-68030:start
(I have a soft spot for 68k.)
- Dan C.
Well, Rob Pike designed one: http://doc.cat-v.org/bell_labs/blit/
On 09/04/2021
11:12, emanuel stiebler wrote: > You're comparing a z80
SBC running CP/M?
Or are you thinking of 68000 SBCs?
Z80 CP/M machines were still competitive in 1981-1983 (Osborne, Kaypro) I've
never seen a 68k SBC. Have I missed out something along the way? Is
there a
community for 68k SBC's? Kind regards, Andrew
I guess the original hacker scene for the 68K was around Hal Hardenberg’s
newsletter: https://en.wikipedia.org/wiki/DTACK_Grounded
The ready-made 68K SBC’s only arrived 1984-1985:
https://en.wikipedia.org/wiki/Sinclair_QL (I think Linus Torvalds owned
one)
https://en.wikipedia.org/wiki/Atari_ST
https://en.wikipedia.org/wiki/Macintosh_128K
https://en.wikipedia.org/wiki/Amiga_1000
All these machines are rather similar at the hardware level - 68K
processor, RAM shared between CPU and display. Only the Amiga had a
(simple) hardware GPU.
What set the SUN-1 apart was its MMU, which none of the above have.
What influenced the timing was probably that Motorola made the 68K more
affordable by the mid-80’s.
Paul
7:20 p.m.
When Digital Systems Research Lab started in 1984 after the implosion of PARC CSL, the
first machine we built was a 68000 (68010?) version of the Firefly multiprocessor. We
were able to score some MicroVAX II chips soon enough that we redesigned using those,
which was, ahem, more politically astute at Digital.
Only a few 68K versions were built. The Firefly supported a Unix/Ultrix system call
interface but otherwise used unrelated software.
(Funny story about close(2). The initial version raised a Modula-2 signal when you tried
to close an already closed file, which was very slow. The OS folks, unused to Unix, had
no idea that was something you do all the time.)
Regarding the SUN-1 design, I had heard a rumor that it was designed using TTL “typical”
propagation delays rather than worst case, and as a result was fairly flakey. This caused
me to not join sun <very> early since Eric Schmidt had the office next to me. One
of my many life mistakes.
-Larry
On 2021, Apr 9, at 1:01 PM, Dan Cross
<crossd(a)gmail.com> wrote:
On Fri, Apr 9, 2021 at 11:35 AM Paul Ruizendaal via TUHS <tuhs(a)minnie.tuhs.org
<mailto:tuhs@minnie.tuhs.org>> wrote:
On 09/04/2021 11:12, emanuel stiebler wrote: >
You're comparing a z80 SBC running CP/M? Or are you thinking of 68000 SBCs?
Z80 CP/M machines were still competitive in 1981-1983 (Osborne, Kaypro)
I've never seen a 68k SBC. Have I missed out
something along the way? Is there a community for 68k SBC's? Kind regards, Andrew
There is an active community around DIY 68k SBCs these days. Some representative
examples:
https://www.eejournal.com/article/wallowing-in-68k-nostalgia/
<https://www.eejournal.com/article/wallowing-in-68k-nostalgia/>
https://www.ist-schlau.de <https://www.ist-schlau.de/>
https://www.bigmessowires.com/category/68katy/
<https://www.bigmessowires.com/category/68katy/>
https://github.com/74hc595/68k-nano <https://github.com/74hc595/68k-nano>
http://mc68k.blogspot.com/2012_10_01_archive.html
<http://mc68k.blogspot.com/2012_10_01_archive.html>
There are even a couple of fairly advanced 68030 design floating around:
https://www.retrobrewcomputers.org/doku.php?id=boards:sbc:gryphon_68030:sta…
<https://www.retrobrewcomputers.org/doku.php?id=boards:sbc:gryphon_68030:start>
https://www.retrobrewcomputers.org/doku.php?id=boards:ecb:kiss-68030:start
<https://www.retrobrewcomputers.org/doku.php?id=boards:ecb:kiss-68030:start>
(I have a soft spot for 68k.)
- Dan C.
Well, Rob Pike designed one: http://doc.cat-v.org/bell_labs/blit/
<http://doc.cat-v.org/bell_labs/blit/>
I guess the original hacker scene for the 68K was around Hal Hardenberg’s newsletter:
https://en.wikipedia.org/wiki/DTACK_Grounded
<https://en.wikipedia.org/wiki/DTACK_Grounded>
The ready-made 68K SBC’s only arrived 1984-1985:
https://en.wikipedia.org/wiki/Sinclair_QL
<https://en.wikipedia.org/wiki/Sinclair_QL> (I think Linus Torvalds owned one)
https://en.wikipedia.org/wiki/Atari_ST <https://en.wikipedia.org/wiki/Atari_ST>
https://en.wikipedia.org/wiki/Macintosh_128K
<https://en.wikipedia.org/wiki/Macintosh_128K>
https://en.wikipedia.org/wiki/Amiga_1000
<https://en.wikipedia.org/wiki/Amiga_1000>
All these machines are rather similar at the hardware level - 68K processor, RAM shared
between CPU and display. Only the Amiga had a (simple) hardware GPU.
What set the SUN-1 apart was its MMU, which none of the above have.
What influenced the timing was probably that Motorola made the 68K more affordable by the
mid-80’s.
Paul
10 Apr
10 Apr
12:28 a.m.
On Fri, Apr 9, 2021 at 12:33 PM Jon Steinhart <jon(a)fourwinds.com> wrote:
Lawrence Stewart writes:
Even into the 2000s. I had a 6-month long war with one of the hardware guys
for a time collection ISA card he did. It worked great, the driver worked
great. Life was good. We shipped product. 5 years later, the customer comes
back and wants a dozen more. So, we got new parts and 4 of the 6 new cards
were flakey, 2 were good. Fingers pointed at the device driver, etc. Long
months of intermittent troubleshooting continued for 5 months. During this
time I build an ISA bus trace card, showed the traces were good and the
flakiness was the result of bad data coming back from the card. At which
point they brought in a different hardware guy to look at things. He
discovered the first hardware guy had built an async circuit with typical
delay patterns. One of the parts we used was rated at 200ns, but parts from
the flakey board worked at 50ns. Turns out the manufacturer substituted a
faster part, so the 'typical' delay propagation worked for this async
circuit, but the faster response time would corrupt data from time to time.
The design was tweaked to be synchronous with a latch, and the unmodified
driver worked perfectly then...
Fun times that...
Warner
Regarding the SUN-1 design, I had heard a rumor
that it was designed
using TTL
“typical” propagation delays rather than worst
case, and as a result was
fairly
flakey.
It's astonishing how common a practice that was back then.
5:16 a.m.
On Fri, 9 Apr 2021, Paul Ruizendaal via TUHS wrote:
Z80 CP/M machines were still competitive in 1981-1983
(Osborne, Kaypro)
And the Aussie Microbee... Wonderful machine, and easily hacked upon.
For example, you could expand the memory by soldering several chips on top
of each other and addressing the CS* line via bank-switching.
-- Dave
2:06 p.m.
On 10 Apr 2021, at 13:17, Dave Horsfall
<dave(a)horsfall.org> wrote:
On Fri, 9 Apr 2021, Paul Ruizendaal via TUHS wrote:
6116 static RAM meant no mucking about with DRAM refresh either.
And the single switch upgrade from 2MHz to 4MHz.
Happy days.
But I never tried to get a Unix on it. UZI or Fuzix might work?
d
Z80 CP/M machines were still competitive in
1981-1983 (Osborne, Kaypro)
And the Aussie Microbee... Wonderful machine, and easily hacked upon.
For example, you could expand the memory by soldering several chips on top of each other
and addressing the CS* line via bank-switching.
13 Apr
13 Apr
11:57 p.m.
On Sat, 10 Apr 2021, David Arnold wrote:
[ Microbee ]
6116 static RAM meant no mucking about with DRAM
refresh either.
Yep, that helped a lot; slower, but who cared? My memory is fading now,
but I recall that the Z-80 had a "refresh" pin to tell any attached
dynamic RAM to refresh itself. The Z-80 was my favourite chip :-)
Happy days.
But I never tried to get a Unix on it. UZI or Fuzix might work?
I toyed with the idea of Minix or LSX, but it would have to be stripped
back and I didn't think that the Z-80 was up to it, even though I had the
128KB bank-switched model. With the Hi-Tech C compiler I did get a number
of simple Unix programs to run, and even found a copy of CP/M UUCP (which
was overlaid to to hell and back).
I did have a copy of Concurrent CP/M, but never tried it.
-- Dave
14 Apr
14 Apr
12:30 a.m.
On Apr 13, 2021, at 2:57 PM, Dave Horsfall <dave(a)horsfall.org> wrote:
Cromemco Inc sold Cromix, a "unix like os" for their Z80 based System
Three in 1979. It had many of the unix commands but some had longer
names and no UUCP. Written from scratch AFAIK. I believe you can still
find a copy online and run it under a Z80 emulator.
Happy days.
But I never tried to get a Unix on it. UZI or Fuzix might work?
I toyed with the idea of Minix or LSX, but it would have to be stripped back and I
didn't think that the Z-80 was up to it, even though I had the 128KB bank-switched
model. With the Hi-Tech C compiler I did get a number of simple Unix programs to run, and
even found a copy of CP/M UUCP (which was overlaid to to hell and back). 
15 Apr
15 Apr
7:01 a.m.
On Sat, 10 Apr 2021, Dave Horsfall wrote:
On Fri, 9 Apr 2021, Paul Ruizendaal via TUHS wrote:
That worked on the old Radio Shack (Tandy) Color Computer 2 as well.
Until this moment I didn't know it had been demonstrated on any other
architecture.
The Operating System OS-9[1] Level One would detect this and use the
bank-switched memory if it was available. Presumably it kept identical
copies of itself in each bank as the entire address space switched.
Microware OS-9 was *nix-like in look and feel although it was very
different internally I think. OS-9 still exists today.
I started with OS-9 and so found Unix a comfortable environment when I
transitioned over.
[1] Which should not be confused with any operating system running on a
Mac. That's another story.
Rob
Z80 CP/M machines were still competitive in
1981-1983 (Osborne, Kaypro)
And the Aussie Microbee... Wonderful machine, and easily hacked upon.
For example, you could expand the memory by soldering several chips on top of
each other and addressing the CS* line via bank-switching. 
16 Apr
16 Apr
3:17 a.m.
Robert Brockway <robert(a)timetraveller.org> writes:
On Sat, 10 Apr 2021, Dave Horsfall wrote:
I did a lot with OS-9 too, both Level One on the Color Computer 2 and
Level Two on the Color Computer 3. The CC3 had a very primitive memory
manager, no faulting, but would allow 8k chunks from up to a 512k pool
of memory to be mapped into the 64k address space of the 6809. There
was a C compiler, probably K&R based or a bit before for OS-9. I ported
a number of the BSD utilities. I also worked on a implementation of
UUCP and ran a UUCP node and proper domain for email using UUNET as the
provider. I received email and a bit of Usenet. I wrote a clone of rn
to read Usenet on the CC3 with OS-9 Level Two. The block diagram for
6809 OS-9 was very simular to V[small number] Unix, with some notable
differences. OS-9 is a microkernel probably being the biggest thing and
6809 OS-9 is all written in assembly. There was a login program that
you could attach to a serial port and actually login with a username and
password and such. Lots of fun and somewhat Unix like in a lot of
ways. There was also a 68000 version of OS-9 Level One that I saw
once. I understand that it may have been mechanically translated from
the 6809 version. It ran pretty much exactly in the same way.
--
Brad Spencer - brad(a)anduin.eldar.org - KC8VKS - http://anduin.eldar.org
On Fri, 9 Apr 2021, Paul Ruizendaal via TUHS
wrote:
That worked on the old Radio Shack (Tandy) Color Computer 2 as well.
Until this moment I didn't know it had been demonstrated on any other
architecture.
The Operating System OS-9[1] Level One would detect this and use the
bank-switched memory if it was available. Presumably it kept identical
copies of itself in each bank as the entire address space switched.
Microware OS-9 was *nix-like in look and feel although it was very
different internally I think. OS-9 still exists today.
I started with OS-9 and so found Unix a comfortable environment when I
transitioned over.
[1] Which should not be confused with any operating system running on a
Mac. That's another story.
Rob Z80 CP/M machines were still competitive in
1981-1983 (Osborne, Kaypro)
And the Aussie Microbee... Wonderful machine, and easily hacked upon.
For example, you could expand the memory by soldering several chips on top of
each other and addressing the CS* line via bank-switching.
1311
days inactive
1318
days old
10 comments
10 participants
participants (10)
-
Bakul Shah -
Brad Spencer -
Dan Cross -
Dave Horsfall -
David Arnold -
Jon Steinhart -
Lawrence Stewart -
Paul Ruizendaal -
Robert Brockway -
Warner Losh