20 Jun
2026
20 Jun
'26
2:51 p.m.
Yes, I did early development on both. At least the interface to the OS was Pascal. I
still have the early documentation buried in a box somewhere.
Sent from my iPhone
On Jun 20, 2026, at 2:02 PM, Fred Cisin via cctalk
<cctalk(a)classiccmp.org> wrote:
On Sat, 20 Jun 2026, ben via cctalk wrote:
PL/1 and C are the only two high level languages
a operating system is written in, that I can think of that are well documented, and easily
found on the WEB.
I heard that all of the Lisa OS, and much of Macintosh, were written in Pascal.
(obviously other than some low-level drivers and performance critical routines?)
Is that correct?
http://pascal.hansotten.com/apple-lisa-pascal/
When did Macintosh development switch to C?
--
Grumpy Ol' Fred cisin(a)xenosoft.com 
22 Jun
22 Jun
5:35 a.m.
New subject: What constitutes a dead computer language
> On Sat, 20 Jun 2026, ben via cctalk wrote:
>> PL/1 and C are the only two high level languages a operating system is written
in, that I can think of that are well documented, and easily found on the WEB.
That clearly is not even close to true. There is ALGOL (or rather its dialect ESPOL) used
by Burroughs. There's CYBIL, used by CDC. There's BLISS, used by DEC. And if
you include embedded RTOS, there's FORTH.
Just a couple I can think of in 30 seconds.
Did anyone use FORTRAN? I do remember IBM wrote the OS/360 FORTRAN (H) compiler in
FORTRAN.
paul
5:56 a.m.
New subject: What constitutes a dead computer language
On 6/22/26 09:35, Paul Koning via cctalk wrote:
Did anyone use FORTRAN? I do remember IBM wrote the OS/360 FORTRAN (H) compiler in
FORTRAN.
PRIMOS, the OS used by Prime Computers was partially written in FORTRAN,
at least in early versions. I think they migrated to something else later.
Although I didn't work with the machines myself, I remember seeing my
colleagues digging in FORTRAN code chasing system problems.
-- Gary
23 Jun
23 Jun
5:08 p.m.
New subject: What constitutes a dead computer language
On 6/22/2026 9:56 AM, Gary Grebus via cctalk wrote:
On 6/22/26 09:35, Paul Koning via cctalk wrote:
Primos was a conglomerate of a number of languages. Not including
C.
bill
(Who used to maintain a cluster of 850's as an NCO and later
as a beltway bandit.)
Did anyone use FORTRAN? I do remember IBM wrote the OS/360 FORTRAN
(H) compiler in FORTRAN.
PRIMOS, the OS used by Prime Computers was partially written in FORTRAN,
at least in early versions. I think they migrated to something else later.
Although I didn't work with the machines myself, I remember seeing my
colleagues digging in FORTRAN code chasing system problems. 
22 Jun
22 Jun
6:09 a.m.
New subject: What constitutes a dead computer language
On Mon, 22 Jun 2026, Paul Koning via cctalk wrote:
Prime computers were in FORTRAN.
Then there are many university projects - e.g. Edinburgh used Imp77.
Oxford used BCPL (OS6), UCSD Pascal, ...
Gordon
> On Sat, 20 Jun 2026, ben via cctalk wrote:
>> PL/1 and C are the only two high level languages a operating system is written
in, that I can think of that are well documented, and easily found on the WEB.
That clearly is not even close to true. There is ALGOL (or rather its
dialect ESPOL) used by Burroughs. There's CYBIL, used by CDC. There's
BLISS, used by DEC. And if you include embedded RTOS, there's FORTH.
Just a couple I can think of in 30 seconds.
Did anyone use FORTRAN? I do remember IBM wrote the OS/360 FORTRAN (H)
compiler in FORTRAN.
24 Jun
24 Jun
3:09 p.m.
New subject: What constitutes a dead computer language
On Mon, Jun 22, 2026 at 9:09 AM Gordon Henderson via cctalk <
cctalk(a)classiccmp.org> wrote:
Then there are many university projects - e.g.
Edinburgh used Imp77.
Oxford used BCPL (OS6), UCSD Pascal, ...
Indeed, and not just one system. EMAS (3 architectures), Deimos, Hamish
Dewar's PDP9/15 OS (not sure if it had a name), LEGOS, ISYS, APM OS,
Mouses, and even GUTS (Groeningen University Time-sharing System.)
Probably more that don't immediately spring to mind.
G
PS Imp, not just specifically Imp77 which is just one of many Imp compilers.
22 Jun
22 Jun
8:57 a.m.
New subject: What constitutes a dead computer language
On Mon, Jun 22, 2026 at 8:36 AM Paul Koning via cctalk <
cctalk(a)classiccmp.org> wrote:
Let's not forget HP's SPL used to write the HP 3000's MPE operating system,
on a machine that never had/needed an assembler.
G.
> On Sat, 20 Jun 2026, ben via cctalk wrote:
>> PL/1 and C are the only two high level languages a operating system is
written in, that I can think of that are well documented, and easily found
on the WEB.
That clearly is not even close to true. There is ALGOL (or rather its
dialect ESPOL) used by Burroughs. There's CYBIL, used by CDC. There's
BLISS, used by DEC. And if you include embedded RTOS, there's FORTH.
Just a couple I can think of in 30 seconds.
9:01 a.m.
New subject: What constitutes a dead computer language
Gavin Scott via cctalk [22.06.2026 18:57:44]:
On Mon, Jun 22, 2026 at 8:36 AM Paul Koning via cctalk
<
cctalk(a)classiccmp.org> wrote:
Let's not forget HP's SPL used to write the HP 3000's MPE operating system,
on a machine that never had/needed an assembler.
G.
Or Planck, used to write Norsk Data´s Sintran.
--
Hilsen Harald
> On Sat, 20 Jun 2026, ben via cctalk wrote:
>> PL/1 and C are the only two high level languages a operating system is
written in, that I can think of that are well documented, and easily found
on the WEB.
That clearly is not even close to true. There is ALGOL (or rather its
dialect ESPOL) used by Burroughs. There's CYBIL, used by CDC. There's
BLISS, used by DEC. And if you include embedded RTOS, there's FORTH.
Just a couple I can think of in 30 seconds. 
10:03 a.m.
New subject: What constitutes a dead computer language
On Mon, 2026-06-22 at 09:35 -0400, Paul Koning via cctalk wrote:
Did anyone use FORTRAN? I do remember IBM wrote the
OS/360 FORTRAN
(H) compiler in FORTRAN.
About forty years ago, FORTRAN was called "the universal assembly
language." People who invented new languages generated Fortran code
because it's easier than generating machine code, and your new language
would instantly run on every machine that had a Fortran compiler. Now
that's mostly done in C.
I wrote compilers for several "small" languages in Fortran. I use a
Pager's method LR parser generator written in Fortran. My 1401
Autocoder cross assembler is in Fortran.
10:15 a.m.
New subject: What constitutes a dead computer language
On Jun 22, 2026, at 2:03 PM, Van Snyder via cctalk
<cctalk(a)classiccmp.org> wrote:
On Mon, 2026-06-22 at 09:35 -0400, Paul Koning via cctalk wrote:
Wow. I've never heard of that, or seen it done. And I would certainly never have
considered doing so myself.
Then again, one of my DEC colleagues wrote a parser in COBOL, but that was more a
"watch this -- hold my beer" type of thing.
paul
Did anyone use FORTRAN? I do remember IBM wrote
the OS/360 FORTRAN
(H) compiler in FORTRAN.
About forty years ago, FORTRAN was called "the universal assembly
language." People who invented new languages generated Fortran code
because it's easier than generating machine code, and your new language
would instantly run on every machine that had a Fortran compiler. 
11:39 a.m.
New subject: What constitutes a dead computer language
On Mon, 22 Jun 2026, Paul Koning via cctalk wrote:
Wow. I've never heard of that, or seen it done.
And I would certainly never have considered doing so myself.
Then again, one of my DEC colleagues wrote a parser in COBOL, but that was more a
"watch this -- hold my beer" type of thing.
Wow
Was there also a wager involved?
It used to be traditional in graduate school compiler construction
classes, that one had to develop it until the compiler for a language was
written in that language and compiled in the final iteration of itself.
While being a traditional test of certain aspects of completeness, it
pointedly ignores the fact that any language has an "orientation" of its
own. Some languages are not optimized towards writing compilers, and will
require unnecessary amounts of work, and may result in a compiler that is
not as good as one written in a language oprimized for that sort of
programming.
Would a COBOL, FORTRAN or RPG compiler written in COBOL, FORTRAN or RPG be
as good as one written in a slightly lower level language? and how long
would somebody have to hold that beer? (room temperature, flat, and
stale?)
How many compilers written in the language that it compiles are as good
as one written in carefully crafted assembly language?
--
Grumpy Ol' Fred cisin(a)xenosoft.com
11:47 a.m.
New subject: What constitutes a dead computer language
On Jun 22, 2026, at 3:39 PM, Fred Cisin via cctalk
<cctalk(a)classiccmp.org> wrote:
On Mon, 22 Jun 2026, Paul Koning via cctalk wrote:
No, but he was a member of the COBOL compiler team so it was a case of pushing the limits
of what he worked on every day.
I also remember running into him late one night, in the middle of a debug session. I
asked what he was working on. Answer "packed decimal exponentiation". My
question "what is that used for?" Answer: "compound interest in COBOL
programming".
Oh yeah.
paul
Wow. I've never heard of that, or seen it
done. And I would certainly never have considered doing so myself.
Then again, one of my DEC colleagues wrote a parser in COBOL, but that was more a
"watch this -- hold my beer" type of thing.
Wow
Was there also a wager involved?
12:33 p.m.
New subject: BCD (Was: What constitutes a dead computer language
On Mon, 22 Jun 2026, Paul Koning via cctalk wrote:
I also remember running into him late one night, in
the middle of a
debug session. I asked what he was working on. Answer "packed decimal
exponentiation". My question "what is that used for?" Answer:
"compound interest in COBOL programming".
Oh yeah.
That's what it is for
In beginning high level language programming classes (FORTRAN, BASIC, C),
I often gave compound interest as an early exercise in loops (displaying
each iteration, not exponential direct calculation of end result, not
BCD)
But, I used BCD a little in my sales tax TSR.
And, I have often [mis]used x86 BCD instructions for display of numbers:
AAM as a substitute for dividing by 10 for 0-99 two decimal digit number
in an 8 bit register,
and DAA in converting a single hexadecimal digit into ASCII numeral: (notice ADC to avoid
JMPs)
AND AL, 0Fh
DAA
ADD AL, 0F0h
ADC AL, 40h
or:
AND AL, 0Fh
ADD AL, 0
ADC AL,28h
DAA
That is extendable to full content (4 hex digits) of 16 bit AX register:
PUSH CX ; in case CX is in use outside of this
MOV CX, 0404h ; CH and CL are independent; CL as a ROL loop, CH as a manual
count down
N1: ROL AX, CL
PUSH AX ;gonna get stepped on in display output
AND AL, 0Fh ;disply single hexadecimal digit
DAA
ADD AL, 0F0h
ADC AL, 40h
MOV AH, 0Eh ;displays char in AL at cursor position
INT 10h ; or save registers and use DL, and INT21h display
POP AX ;bring it back after display
DEC CH
JNZ N1
POP CX ; put it back
23 Jun
23 Jun
7:28 p.m.
New subject: BCD (Was: What constitutes a dead computer language
On Mon, Jun 22, 2026 at 4:33 PM Fred Cisin via cctalk <cctalk(a)classiccmp.org>
wrote:
Unfortunately (or not?) the x86 BCD instructions have been removed from
'64-bit mode' in x86-64. I assume those processors are fast enough
emulating BCD if required to keep COBOL users happy.
The PDP-11 CIS instructions include BCD types and are very
interesting/weird (e.g. EDIT instruction).
KJ
My question
"what is that used for?" Answer:
"compound interest in COBOL programming".
Oh yeah.
That's what it is for
24 Jun
24 Jun
6:22 a.m.
New subject: BCD (Was: What constitutes a dead computer language
On Jun 23, 2026, at 11:28 PM, Ken Seefried via cctalk
<cctalk(a)classiccmp.org> wrote:
On Mon, Jun 22, 2026 at 4:33 PM Fred Cisin via cctalk <cctalk(a)classiccmp.org>
wrote:
Unfortunately (or not?) the x86 BCD instructions have been removed from
'64-bit mode' in x86-64. I assume those processors are fast enough
emulating BCD if required to keep COBOL users happy.
The PDP-11 CIS instructions include BCD types and are very
interesting/weird (e.g. EDIT instruction).
I never looked at PDP-11 or VAX decimal instructions all that much, but from a superficial
look they seemed to be inspired by the analogous instructions of the IBM 360 series,
including (I'm pretty sure) the EDIT instruction.
paul
My
question "what is that used for?" Answer:
"compound interest in COBOL programming".
Oh yeah.
That's what it is for
22 Jun
22 Jun
1:56 p.m.
New subject: What constitutes a dead computer language
Burroughs ESPOL has been dead since the early 1980s. The B6700/B7700 MCP (OS) was
completely rewritten at that time in a new language, NEWP, which is somewhat MODULA-like.
It is still in use for the Unisys ClearPath MCP systems that have evolved from the
B6x00/B7x00 systems.
There are various stores about the name, one that it stands for No Executive Washroom
Privileges and other that when asked if the new compiler was ready yet, John McClintock,
the primary developer, would answer "newp!" This first is probably apocryphal;
the second probably isn't.
3:32 p.m.
New subject: What constitutes a dead computer language
On Jun 22, 2026, at 5:56 PM, paul.kimpel--- via cctalk
<cctalk(a)classiccmp.org> wrote:
Burroughs ESPOL has been dead since the early 1980s. The B6700/B7700 MCP (OS) was
completely rewritten at that time in a new language, NEWP, which is somewhat MODULA-like.
It is still in use for the Unisys ClearPath MCP systems that have evolved from the
B6x00/B7x00 systems.
I remember looking at various ALGOL dialects on the B6700 system back in 1975, when I was
trying to read DEC format tapes on the TUE Burroughs system. MCP doesn't make that
easy. I looked at DCALGOL, which didn't quite do the job. I see from the Unisys
ClearPath webpage that was mentioned that DCALGOL is still around. I don't think
ESPOL was suitable either, but that was mostly because it wasn't available in any way
shape or form for ordinary users.
I ended up using an OS/360 system instead, with EXCP mode I/O. Fun.
paul
3:34 p.m.
New subject: What constitutes a dead computer language
The AMD 2900 microcode assembler was written in FORTRAN. As well as
the Z80 assembler, 1982.
On Mon, Jun 22, 2026 at 2:56 PM paul.kimpel--- via cctalk
<cctalk(a)classiccmp.org> wrote:
Burroughs ESPOL has been dead since the early 1980s. The B6700/B7700 MCP (OS) was
completely rewritten at that time in a new language, NEWP, which is somewhat MODULA-like.
It is still in use for the Unisys ClearPath MCP systems that have evolved from the
B6x00/B7x00 systems.
There are various stores about the name, one that it stands for No Executive Washroom
Privileges and other that when asked if the new compiler was ready yet, John McClintock,
the primary developer, would answer "newp!" This first is probably apocryphal;
the second probably isn't.
23 Jun
23 Jun
6:55 a.m.
New subject: What constitutes a dead computer language
On 6/22/26 18:34, Jim Davis via cctalk wrote:
The AMD 2900 microcode assembler was written in
FORTRAN. As well as
the Z80 assembler, 1982.
I was building a machine for my own use around AMD 2903 bit
slices back in about 1981, planning to execute a superset of
the 360 instruction set. I wrote a microcode assembler in
Pascal.
Jon
11:32 a.m.
New subject: What constitutes a dead computer language
On 2026-06-23 8:55 a.m., Jon Elson via cctalk wrote:
On 6/22/26 18:34, Jim Davis via cctalk wrote:
Did you use, PASCAL for any special reason, or was it the only thing around?
Ben.
The AMD 2900 microcode assembler was written in
FORTRAN. As well as
the Z80 assembler, 1982.
I was building a machine for my own use around AMD 2903 bit slices back
in about 1981, planning to execute a superset of the 360 instruction
set. I wrote a microcode assembler in Pascal.
Jon
11:40 a.m.
New subject: What constitutes a dead computer language
On 6/23/26 14:32, ben via cctalk wrote:
On 2026-06-23 8:55 a.m., Jon Elson via cctalk wrote:
I actually LIKE Pascal! I had some version of Pascal on my
Z-80 CP/M system, and I used Pascal for all complicated
programs, like cross assemblers. Since I had already
written several cross assemblers in Pascal, I was able to
re-use a large chunk of the code and just write the
2903-specific parts.
Jon
On 6/22/26 18:34, Jim Davis via cctalk wrote:
Did you use, PASCAL for any special reason, or was it the
only thing around? The AMD 2900 microcode assembler was written in
FORTRAN.
As well as
the Z80 assembler, 1982.
I was building a machine for my own use around AMD 2903
bit slices back in about 1981, planning to execute a
superset of the 360 instruction set. I wrote a microcode
assembler in Pascal.
Jon
11:50 a.m.
New subject: What constitutes a dead computer language
Did you use,
PASCAL for any special reason, or was it the only thing around?
I actually LIKE Pascal!
12:08 p.m.
New subject: What constitutes a dead computer language
On Tue, 2026-06-23 at 12:50 -0700, Cameron Kaiser via cctalk wrote:
> I actually LIKE Pascal!
After I finished my first compiler class in the 1983 UC Santa Cruz
Summer Institute in Computer Science (taught by Frank De Remer and Tom
Pennello), a group in another section approached me to write a compiler
for a "little language" called MENDEL, for Menu Definition Language. It
was about putting boxes on a screen with response buttons, the kind of
thing that would be done today using HTML etc, which didn't exist yet.
Two guys in the group had worked for a year and produced nothing more
than some low-level run-time library, in Fortran for a VAX. I wrote the
MENDEL compiler in 90 days, in Pascal, using the framework I learned at
UC Santa Cruz. It generated FORTRAN 77 code. My group supervisor
reported it to the Section Manager to report to the Division as part of
the monthly Section Significan Events report. He suppressed it.
2:15 p.m.
New subject: What constitutes a dead computer language
On Tue, Jun 23, 2026 at 3:56 PM Cameron Kaiser via cctalk <
cctalk(a)classiccmp.org> wrote:
Turbo Pascal was my first language after FORTRAN. It was very nice on
CP/M. Then my freshman CS intro language was Pascal so I was ahead of the
game (tho I did most of it in C for other reasons). Pascal (suitably
extended) is a fine programming language.
and I even did some Modula-2 programming with MacMETH (worst name
> Did you use, PASCAL for any special reason,
or was it the only thing
around?
I actually LIKE Pascal!
I also enjoy Pascal. Any substantial program I wrote on DOS was in Turbo
Pascal,
ever).
The only Mac (OG MacOS 7) programming I ever did was with the Metrowerks M2
compiler. It was pretty great. You could even look at the Pascal examples
in "Inside Macintosh" and you were 90% of the way there. Metrowerks even
did a pretty good job at making M2 coroutines work on MacOS; I wasn't
pushing any envelopes but wrote some things with multiple coroutines
and didn't manage to crash things too often. Pretty wild considering the
state of the art at the time.
KJ
2:47 p.m.
New subject: What constitutes a dead computer language
On Tue, 2026-06-23 at 18:15 -0400, Ken Seefried via cctalk wrote:
The only Mac (OG MacOS 7) programming I ever did was
with the
Metrowerks M2
compiler.
One of my supervisors insisted I write an FTP server for a Mac IIci in
about 1988 using Common Lisp. I used Guy Steele's CLTL2 text. He had a
good sense of humor. The index item for "kludges" says 1-954. My
supervisor wanted me to use T or Scheme but we wouldn't find a
compiler.
27 Jun
27 Jun
2:15 p.m.
New subject: Need MicroVAX-II / Emulex controller help
Hello,
Still trying to figure out what is wrong. I got an Emulex
QD22 from a guy at a recent VCF meeting, and am trying to
resurrect my MicroVAX-II (KA630-AA). I have a BA23 cabinet
with a H9278-A backplane.
I have the KA630-AA in slot 1, and the QD21 in slot 2 - A
and B connectors. I don't have any grant continuity
jumpers, but I don't think I should need any in this
configuration. I am trying to configure the QD21 by running
the built-in diagnostic program in the QD21 EPROMS. I have
done the procedure dozens of times as described in the
Emulex manual, but it doesn't work. I am supposed to get an
0400 (hex) in the CSR, but instead I get 0200, and there is
no code loaded at location 80. I can't figure out what is
wrong - bad QD21, bad backplane, typo in the manual's
procedure or something else.
Any ideas on how to diagnose this are very welcome. The
going price on a QD21 seems to be over $200, which is
getting a bit expensive. Does anybody know where I can get
one cheaper in the US?
Thanks,
Jon
2:17 p.m.
New subject: Need MicroVAX-II / Emulex controller help
On 6/27/26 17:15, Jon Elson via cctalk wrote:
Hello,
Still trying to figure out what is wrong. I got an Emulex
QD22 from a guy at a recent VCF meeting, and am trying to
resurrect my MicroVAX-II (KA630-AA). I have a BA23
cabinet with a H9278-A backplane.
I have the KA630-AA in slot 1, and the QD21 in slot 2 - A
and B connectors. I don't have any grant continuity
jumpers, but I don't think I should need any in this
configuration. I am trying to configure the QD21 by
running the built-in diagnostic program in the QD21
EPROMS. I have done the procedure dozens of times as
described in the Emulex manual, but it doesn't work. I am
supposed to get an 0400 (hex) in the CSR, but instead I
get 0200, and there is no code loaded at location 80. I
can't figure out what is wrong - bad QD21, bad backplane,
typo in the manual's procedure or something else.
Any ideas on how to diagnose this are very welcome. The
going price on a QD21 seems to be over $200, which is
getting a bit expensive. Does anybody know where I can
get one cheaper in the US?
ACK - SORRY - that should be QD21!
2:28 p.m.
New subject: Emulex QD21 controller help
Oh my gosh! I just found a document that says units with
A-D firmware rev levels do NOT have the firmware resident
diagnostics!
That seems to be the likely cause of the problem. Does
anybody have the EPROM dump from an E65E or above firmware
of an Emulex QD21 ESDI disk controller? I have plenty of
128K EPROMS here to reprogram.
Thanks,
Jon
2:50 p.m.
New subject: Emulex QD21 controller help
On Sat, Jun 27, 2026 at 3:28 PM Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
Oh my gosh! I just found a document that says units with
A-D firmware rev levels do NOT have the firmware resident
diagnostics!
That seems to be the likely cause of the problem. Does
anybody have the EPROM dump from an E65E or above firmware
of an Emulex QD21 ESDI disk controller? I have plenty of
128K EPROMS here to reprogram.
I have dumps of E65G and E65J handy. Those are 32KB dumps so you need
a 256K EPROM, not a 128K EPROM. E65J is what I currently have
installed on all of my QD21 controllers.
I don't see any QD21 firmware dumps on Bitsavers currently.
2:57 p.m.
New subject: Emulex QD21 controller help
On Sat, Jun 27, 2026 at 3:50 PM Glen Slick <glen.slick(a)gmail.com> wrote:
On Sat, Jun 27, 2026 at 3:28 PM Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
The Rev G firmware has the following on board main menu:
========================================================
Emulex Corporation
Copyright All rights reserved
QD21 controller, firmware revision level IP address =
Option menu
1 - Format
2 - Format and verify
3 - Verify
4 - Read only test
5 - Data reliability test
6 - List known units
7 - Replace block
8 - Display Novram
9 - Edit / Load Novram
Enter option number:
The Rev J firmware has the following on board main menu:
========================================================
Firmware-Resident Diagnostic
Copyright (c) 1988 Emulex Corporation all rights reserved
QD21 controller, firmware revision level IP address =
Option menu
1 - Self Test Loop
2 - Format
3 - Verify
4 - Format and Verify
5 - Data Reliability Test
6 - Format, Verify, and Data Reliability Test
7 - Read Only Test
8 - List Known Units
9 - Replace Block
10 - Print RCT
11 - Display Novram
12 - Edit / Load Novram
Enter option number:
Oh my gosh! I just found a document that says units with
A-D firmware rev levels do NOT have the firmware resident
diagnostics!
That seems to be the likely cause of the problem. Does
anybody have the EPROM dump from an E65E or above firmware
of an Emulex QD21 ESDI disk controller? I have plenty of
128K EPROMS here to reprogram.
I have dumps of E65G and E65J handy. Those are 32KB dumps so you need
a 256K EPROM, not a 128K EPROM. E65J is what I currently have
installed on all of my QD21 controllers.
3:54 p.m.
New subject: Emulex QD21 controller help
On 6/27/26 17:50, Glen Slick via cctalk wrote:
Wow, thanks a BUNCH! I just loaded the E65J file to an
EPROM and the diag starts up just fine. Now, I need to
figure out what settings to use for this drive and I ought
to be able to boot VMS.
Thanks!
Jon
4:37 p.m.
New subject: Emulex QD21 controller help
On Sat, Jun 27, 2026 at 4:54 PM Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
On 6/27/26 17:50, Glen Slick via cctalk wrote:
Wow, thanks a BUNCH! I just loaded the E65J file to an
EPROM and the diag starts up just fine. Now, I need to
figure out what settings to use for this drive and I ought
to be able to boot VMS.
Nice. Thanks for the update. Hope you can figure out the correct
settings to match the drive you have.
-Glen
4:57 p.m.
New subject: Emulex QD21 controller help
Glen,
I have a QD21 that's waiting for its turn to get into a machine that still
needs some work, but thinking ahead, is there somewhere I can download your
firmware dumps from?
On Sat, Jun 27, 2026 at 5:37 PM Glen Slick via cctalk <cctalk(a)classiccmp.org>
wrote:
On Sat, Jun 27, 2026 at 4:54 PM Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
On 6/27/26 17:50, Glen Slick via cctalk wrote:
Wow, thanks a BUNCH! I just loaded the E65J file to an
EPROM and the diag starts up just fine. Now, I need to
figure out what settings to use for this drive and I ought
to be able to boot VMS.
Nice. Thanks for the update. Hope you can figure out the correct
settings to match the drive you have.
-Glen
15 Jul
15 Jul
7:17 a.m.
New subject: Emulex QD21 controller help
Well, I have gotten the diags to come up, after getting an
EPROM dump from Glen Slick. THANKS, Glen!
Now, the problem is to get the NOV RAM configured for these
drives.
I have two Fujitsu ESDI drives that should have working VMS
systems on them. I first tried coming up with my own drive
parameters, that didn't work. Then, I tried setting the
drive to type 2 (get parameters from drive automatically),
that didn't work either.
The small drive I have has some numbers written down on the
bad block list paper, I will try those but it doesn't look
like enough numbers.
Anyway, the drives I have are:
Fujitsu M2246E and Fujitsu M2263E. The manual does list the
parameters for the M2246E, I will have to try those. But, I
have no parameters for the M2263E. Does anybody know where
to get these config parameters for the M2263E? That is the
drive that was used most recently, so I would prefer to get
that one running.
Thanks much for any help!
Jon
11:55 a.m.
New subject: Emulex QD21 controller help
In the past I had a QD21 and ESDI drives. What controller were the
Fujitsu drives connected to originally? There maybe some kind of
proprietary format on them that the QD21 doesn't understand.
Does the QD21 see or sense the drives?
Doug
On 7/15/2026 11:17 AM, Jon Elson via cctalk wrote:
Well, I have gotten the diags to come up, after
getting an EPROM dump
from Glen Slick. THANKS, Glen!
Now, the problem is to get the NOV RAM configured for these drives.
I have two Fujitsu ESDI drives that should have working VMS systems on
them. I first tried coming up with my own drive parameters, that
didn't work. Then, I tried setting the drive to type 2 (get
parameters from drive automatically), that didn't work either.
The small drive I have has some numbers written down on the bad block
list paper, I will try those but it doesn't look like enough numbers.
Anyway, the drives I have are:
Fujitsu M2246E and Fujitsu M2263E. The manual does list the
parameters for the M2246E, I will have to try those. But, I have no
parameters for the M2263E. Does anybody know where to get these
config parameters for the M2263E? That is the drive that was used
most recently, so I would prefer to get that one running.
Thanks much for any help!
Jon
16 Jul
16 Jul
7:04 a.m.
New subject: uVAX-II new MSCP media?
Hello, all,
IO am trying to get my uVAX-II running for the VCFMW. I am
having issues getting my 40+ year old ESDI drives configured
with an Emulex QD21, but may eventually get that working.
But, that is not a likely long term solution. Is there a
Q-bus MSCP disk controller that takes something like SD
cards? I think I have seen some projects like that.
Anybody know?
I have 2 ESDI drives with VMS on them (last time they were
used) and I also have a tape container file of a VMS BACKUP
tape of my system from about 1993. What would be great is a
procedure to get the whole bootable file system off that
BACKUP file or the extracted files from that. All of these
files are now on a Linux system.
Anybody know how to do this?
Thanks,
Jon
7:59 a.m.
New subject: uVAX-II new MSCP media?
Q-bus disk options, using uSD or recent media, arguably include:
- RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
- Qbus SCSI master - SCSI bus - uSD or spinning rust SCSI disk
- a Qbone configured as a disk
All of those could use uSD as storage
Convering VMS BACKUP tapes to disk images is, if you are in luck, a job for simh
HtH; Martin
-----Original Message-----
From: Jon Elson via cctalk [mailto:cctalk@classiccmp.org]
Sent: 16 July 2026 15:04
To: Douglas Taylor via cctalk <cctalk(a)classiccmp.org>
Cc: Jon Elson <elson(a)pico-systems.com>
Subject: [cctalk] uVAX-II new MSCP media?
Hello, all,
IO am trying to get my uVAX-II running for the VCFMW. I am having issues getting my 40+
year old ESDI drives configured with an Emulex QD21, but may eventually get that working.
But, that is not a likely long term solution. Is there a Q-bus MSCP disk controller that
takes something like SD cards? I think I have seen some projects like that. Anybody know?
I have 2 ESDI drives with VMS on them (last time they were
used) and I also have a tape container file of a VMS BACKUP tape of my system from about
1993. What would be great is a procedure to get the whole bootable file system off that
BACKUP file or the extracted files from that. All of these files are now on a Linux
system.
Anybody know how to do this?
Thanks,
Jon
10:34 a.m.
New subject: uVAX-II new MSCP media?
On 7/16/26 12:18, emanuel stiebler via cctalk wrote:
On 2026-07-16 11:59, Martin Bishop via cctalk wrote:
Years ago, when I started, I used a 40 MB MFM disk with a
Viking Systems controller. 40 MB doesn't go real far with a
VMS system.
Jon
Q-bus disk options, using uSD or recent media,
arguably
include:
- RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
He has ~660MByte disks in the system. MFM won't work for
him :)
11:22 a.m.
New subject: uVAX-II new MSCP media?
You don't have SCSI in the machine?
I have a MicroVAX 3100 with a ZuluSCSI emulator in it. Works great.
It can emulate several drives at the same time.
- Peter
On Thu, Jul 16, 2026 at 2:49 PM Jon Elson via cctalk <cctalk(a)classiccmp.org>
wrote:
On 7/16/26 12:18, emanuel stiebler via cctalk wrote:
On 2026-07-16 11:59, Martin Bishop via cctalk
wrote:
Years ago, when I started, I used a 40 MB MFM disk with a
Viking Systems controller. 40 MB doesn't go real far with a
VMS system.
Jon
Q-bus disk options, using uSD or recent media,
arguably
include:
- RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
He has ~660MByte disks in the system. MFM won't work for
him :)
10:41 a.m.
New subject: uVAX-II new MSCP media?
You can fit a VMS systems onto an RD54 - I ran systems like that from the 80's into
the early 00's. Albeit with another RD54 for the user software, and another two
RD54's for analysis data. However, SCSI or a solid state solution is a markedly
better solution.
Martin
PS VMT pointer to 5volts.ch : not previously above my horizon
-----Original Message-----
From: emanuel stiebler [mailto:emu@e-bbes.com]
Sent: 16 July 2026 17:18
To: General Discussion: On-Topic and Off-Topic Posts <cctalk(a)classiccmp.org>
Cc: Martin Bishop <mjd.bishop(a)emeritus-solutions.com>
Subject: Re: [cctalk] Re: uVAX-II new MSCP media?
On 2026-07-16 11:59, Martin Bishop via cctalk wrote:
Q-bus disk options, using uSD or recent media,
arguably include:
- RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
He has ~660MByte disks in the system. MFM won't work for him :)
17 Jul
17 Jul
7 a.m.
New subject: uVAX-II new MSCP media?
Well, you can fit VMS onto an R(M,A, )80 drive with no problem, the
mighty 11/730 is living proof.
RD54's though are a lot less power hungry.
C
On 7/16/26 14:41, Martin Bishop via cctalk wrote:
You can fit a VMS systems onto an RD54 - I ran systems
like that from the 80's into the early 00's. Albeit with another RD54 for the
user software, and another two RD54's for analysis data. However, SCSI or a solid
state solution is a markedly better solution.
Martin
PS VMT pointer to 5volts.ch : not previously above my horizon
-----Original Message-----
From: emanuel stiebler [mailto:emu@e-bbes.com]
Sent: 16 July 2026 17:18
To: General Discussion: On-Topic and Off-Topic Posts <cctalk(a)classiccmp.org>
Cc: Martin Bishop <mjd.bishop(a)emeritus-solutions.com>
Subject: Re: [cctalk] Re: uVAX-II new MSCP media?
On 2026-07-16 11:59, Martin Bishop via cctalk wrote:
Q-bus disk options, using uSD or recent media,
arguably include:
- RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
He has ~660MByte
disks in the system. MFM won't work for him :) 
7:28 a.m.
New subject: uVAX-II new MSCP media?
How many RD54s are still alive? Mine (10) are all dead :-((
Am Fr., 17. Juli 2026 um 17:00 Uhr schrieb Chris Zach via cctalk <
cctalk(a)classiccmp.org>gt;:
Well, you can fit VMS onto an R(M,A, )80 drive with no
problem, the
mighty 11/730 is living proof.
RD54's though are a lot less power hungry.
C
On 7/16/26 14:41, Martin Bishop via cctalk wrote:
You can fit a VMS systems onto an RD54 - I ran
systems like that from
the 80's into the early 00's. Albeit with another RD54 for the user
software, and another two RD54's for analysis data. However, SCSI or a
solid state solution is a markedly better solution.
Martin
PS VMT pointer to 5volts.ch : not previously above my horizon
-----Original Message-----
From: emanuel stiebler [mailto:emu@e-bbes.com]
Sent: 16 July 2026 17:18
To: General Discussion: On-Topic and Off-Topic Posts <
cctalk(a)classiccmp.org>
Cc: Martin Bishop
<mjd.bishop(a)emeritus-solutions.com>
Subject: Re: [cctalk] Re: uVAX-II new MSCP media?
On 2026-07-16 11:59, Martin Bishop via cctalk wrote:
Q-bus disk options, using uSD or recent media,
arguably include:
- RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
He has ~660MByte disks in the system. MFM won't work for him :)
7:32 a.m.
New subject: uVAX-II new MSCP media?
On Fri, Jul 17, 2026 at 11:29 AM Hans-Ulrich Hölscher via cctalk
<cctalk(a)classiccmp.org> wrote:
How many RD54s are still alive? Mine (10) are all dead
:-((
I have several. I haven't tried powering any up recently. Some were
spares, and only two were in main use - a MicroVAX II, and a
VAXstation 2000. The VAXstation 2000 has been disassembled for a long
time - I hope to get it back together for a future VCF (after 2027. I
already have plans for next year).
-ethan
8:35 a.m.
New subject: uVAX-II new MSCP media?
I have several that work fine and one that is a primary disk on TITAN (a
11/73+ system) using the RQDX3.
It's dog-slow compared to the ESDI disk and controller in the 11/83+
(PMI) but to be honest firing up a lot of memory as cache gets the
system to reasonable performance levels. Still a bit slow.
C
On 7/17/26 11:28, Hans-Ulrich Hölscher via cctalk wrote:
How many RD54s are still alive? Mine (10) are all dead
:-((
Am Fr., 17. Juli 2026 um 17:00 Uhr schrieb Chris Zach via cctalk <
cctalk(a)classiccmp.org>gt;:
> Well, you can fit VMS onto an R(M,A, )80 drive with no problem, the
> mighty 11/730 is living proof.
>
> RD54's though are a lot less power hungry.
>
> C
>
> On 7/16/26 14:41, Martin Bishop via cctalk wrote:
>> You can fit a VMS systems onto an RD54 - I ran systems like that from
> the 80's into the early 00's. Albeit with another RD54 for the user
> software, and another two RD54's for analysis data. However, SCSI or a
> solid state solution is a markedly better solution.
>> Martin
>>
>> PS VMT pointer to 5volts.ch : not previously above my horizon
>>
>> -----Original Message-----
>> From: emanuel stiebler [mailto:emu@e-bbes.com]
>> Sent: 16 July 2026 17:18
>> To: General Discussion: On-Topic and Off-Topic Posts <
> cctalk(a)classiccmp.org>
>> Cc: Martin Bishop <mjd.bishop(a)emeritus-solutions.com>
>> Subject: Re: [cctalk] Re: uVAX-II new MSCP media?
>>
>> On 2026-07-16 11:59, Martin Bishop via cctalk wrote:
>>> Q-bus disk options, using uSD or recent media, arguably include:
>>> - RQDX3 - MSCP - David Gesswein's MFM emulator (or similar)
>> He has ~660MByte disks in the system. MFM won't work for him :)
10:34 a.m.
New subject: uVAX-II new MSCP media?
On Jul 17, 2026, at 11:28 AM, Hans-Ulrich Hölscher via
cctalk <cctalk(a)classiccmp.org> wrote:
How many RD54s are still alive? Mine (10) are all dead :-((
As many as you want if you emulate one with David Gesswein's emulator.
How much larger can an MFM drive actually be? I don't know, my use for that emulator
is in a Pro which can only handle 16 bit block numbers.
paul
11:19 a.m.
New subject: uVAX-II new MSCP media?
For a Pro the max was an RD53. I have a few of those running as well.
MFM can be big; depends on what your controller can support. Once you
get to ESDI it's a lot better; I run a half height CDC Wren at 330mb and
I have an 870mb full height drive if needed.
C
On 7/17/26 14:34, Paul Koning via cctalk wrote:
On Jul 17, 2026, at 11:28 AM, Hans-Ulrich
Hölscher via cctalk <cctalk(a)classiccmp.org> wrote:
How many RD54s are still alive? Mine (10) are all dead :-((
As many as you want if you emulate one with David Gesswein's emulator.
How much larger can an MFM drive actually be? I don't know, my use for that emulator
is in a Pro which can only handle 16 bit block numbers.
paul
7:30 a.m.
New subject: uVAX-II new MSCP media?
On Fri, Jul 17, 2026 at 11:00 AM Chris Zach via cctalk
<cctalk(a)classiccmp.org> wrote:
Well, you can fit VMS onto an R(M,A, )80 drive with no
problem, the
mighty 11/730 is living proof.
We ran VMS 5.0 on an 11/730 on the internal R80 for years (it was our
VMS 5.x object linking box for deliverables). It fits but is a little
cramped. Room for one copy of our product but not lots of room for
user files.
RD54's though are a lot less power hungry.
RD54s were fine through VMS 5.x. ISTR a buddy of mine put VMS
6.something on an RD54 for me and it was not really large enough. I
think he had to remove the help files.
For our main 11/750, we ran either 4.5 or 4.7 to the last of its days
in 1993 (I forget if we did that last update cycle or not). Our boot
volume was 1/2 of a Fujitsu SMD drive that emulated two RM03s, so DRA0
was 67MB. We might have shoved some system stuff on DRA1.
I'd have to check the SPD for VMS 6.x but it wouldn't surprise me if
200MB was the smallest supported system device.
-ethan
16 Jul
16 Jul
8:06 a.m.
New subject: uVAX-II new MSCP media?
On 2026-07-16 11:04, Jon Elson via cctalk wrote:
Hello, all,
But, that is not a likely long term solution. Is
there a Q-bus MSCP
disk controller that takes something like SD cards? I think I have seen
some projects like that. Anybody know?
https://retrocmp.com/projects/qbone/313-qbone-introduction
and a new one in development:
https://www.5volts.ch/pages/mscpemulator/010-introduction/
(you can follow it on Vintage Computer Forum)
I have 2 ESDI drives with VMS on them (last time they
were used) and I
also have a tape container file of a VMS BACKUP tape of my system from
about 1993. What would be great is a procedure to get the whole
bootable file system off that BACKUP file or the extracted files from
that. All of these files are now on a Linux system.
I think your drives are shot. If they click forever, and give errors,
fun ahead :(
Where are you located?
Cheers
5:26 p.m.
New subject: uVAX-II new MSCP media?
Jon;
The only solution I can think of that can be done quickly is: buy a UC07
from Ebay, there is one for $300, and get a BlueSCSI SD adapter and an
SD card. Getting a bootable VMS onto the SD card is something everyone
here can guide you through.
Doug
On 7/16/2026 11:04 AM, Jon Elson via cctalk wrote:
Hello, all,
IO am trying to get my uVAX-II running for the VCFMW. I am having
issues getting my 40+ year old ESDI drives configured with an Emulex
QD21, but may eventually get that working.
But, that is not a likely long term solution. Is there a Q-bus MSCP
disk controller that takes something like SD cards? I think I have
seen some projects like that. Anybody know?
I have 2 ESDI drives with VMS on them (last time they were used) and I
also have a tape container file of a VMS BACKUP tape of my system from
about 1993. What would be great is a procedure to get the whole
bootable file system off that BACKUP file or the extracted files from
that. All of these files are now on a Linux system.
Anybody know how to do this?
Thanks,
Jon
22 Jul
22 Jul
10:54 a.m.
New subject: uVAX-II new MSCP media?
On 7/16/26 20:26, Douglas Taylor via cctalk wrote:
Jon;
The only solution I can think of that can be done quickly
is: buy a UC07 from Ebay, there is one for $300, and get a
BlueSCSI SD adapter and an SD card. Getting a bootable
VMS onto the SD card is something everyone here can guide
you through.
OK, I have a UC07 and blueSCSI coming soon. I do have a
.BCK tape image that I extracted back in 2014 using a github
program : vmsbackup, with a few command line switches to
make it accept my container file.
Is there a way to use the vmsbackup program to move the
contents of the tape container file onto the SD card? That
would be really wonderful.
Thanks for any help,
Jon
30 Jul
30 Jul
9:26 a.m.
New subject: uVAX-II new MSCP media?
On 7/16/26 20:26, Douglas Taylor via cctalk wrote:
Jon;
The only solution I can think of that can be done quickly
is: buy a UC07 from Ebay, there is one for $300, and get a
BlueSCSI SD adapter and an SD card. Getting a bootable
VMS onto the SD card is something everyone here can guide
you through.
The UC07 and BlueSCSI are here. I'd like to put VMS 4.7 on
it and see if I can get it running.
Also, will there be a simple way to transfer files from the
unpacked VMS backup dump on my Linux system to the SD card?
Jon
1 Aug
1 Aug
3:52 p.m.
New subject: copy bootable VMS to BlueSCSI SD card
On 7/30/26 12:26, Jon Elson via cctalk wrote:
On 7/16/26 20:26, Douglas Taylor via cctalk wrote:
Douglas, or any one who knows how,
can you give me the procedure to do the above? I am running
short on time to get this all done before VCFMW.
Thanks a lot in advance.
Jon
Jon;
The only solution I can think of that can be done quickly
is: buy a UC07 from Ebay, there is one for $300, and get
a BlueSCSI SD adapter and an SD card. Getting a bootable
VMS onto the SD card is something everyone here can guide
you through.
The UC07 and BlueSCSI are here. I'd like to put VMS 4.7
on it and see if I can get it running.
Also, will there be a simple way to transfer files from
the unpacked VMS backup dump on my Linux system to the SD
card?
4:41 p.m.
New subject: serial comm help Linux
Well, everything is a fight!
The project I'm working on is to get a VMS MicroVAX system
working at VCFMW. The plan was to have serial comm on a
Linux laptop connected to the MicroVAX. I have the serial
comm working on a Dell desktop machine, so I just plugged
the same USB serial cable into the laptop, and NO COMM! I
plugged back into the desktop, and serial data works fine!
I'm totally mystified! The desktop is running an OLD Debian
kernel (7.11), the laptop is running a late Ubuntu (24.04.4).
I have no idea why the laptop doesn't communicate. I have
tried minicom, picocom and cutecom, same result on all of them.
Thanks for any help you can suggest.
Jon
6 p.m.
New subject: serial comm help Linux
Well the first thing to do is make sure the USB serial device gets
configured, when you plug it in, you should see the serial device listed
in the output from 'lsusb' and there should be messages about it in
'dmesg' (requires root authority). It is possible that the naming
convention for the tty device that get created is different between the
two versions of linux. The usual way is the device gets created
dynamically when you plug in the USB serial device. on my linux,
Devuan which is debian based but unencumbered by systemd, it creates a
device /dev/ttyUSB0 most modern distributions I have seen use names like
this.
Paul.
On 2026-08-01 21:41, Jon Elson via cctalk wrote:
Well, everything is a fight!
The project I'm working on is to get a VMS MicroVAX system working at
VCFMW. The plan was to have serial comm on a Linux laptop connected
to the MicroVAX. I have the serial comm working on a Dell desktop
machine, so I just plugged the same USB serial cable into the laptop,
and NO COMM! I plugged back into the desktop, and serial data works
fine! I'm totally mystified! The desktop is running an OLD Debian
kernel (7.11), the laptop is running a late Ubuntu (24.04.4).
I have no idea why the laptop doesn't communicate. I have tried
minicom, picocom and cutecom, same result on all of them.
Thanks for any help you can suggest.
Jon
11:27 p.m.
New subject: serial comm help Linux
To add to what Paul wrote:
Here is a oneliner to check for usb serial ports on a linux machine:
$ ls -l /dev/tty[ACU]*
(the initial '$' is the prompt, don't include it in the command)
run this before and after you plug in your usb serial device, and you
will find it - if it is detected.
If the device only works as user root, you need to create a file to
set ownership, group and permissions on the device when it is plugged
in. This is commonly done through udev / udevadm on linux.
On Sun, Aug 2, 2026 at 4:00 AM Paul Berger via cctalk
<cctalk(a)classiccmp.org> wrote:
Well the first thing to do is make sure the USB serial device gets
configured, when you plug it in, you should see the serial device listed
in the output from 'lsusb' and there should be messages about it in
'dmesg' (requires root authority). It is possible that the naming
convention for the tty device that get created is different between the
two versions of linux. The usual way is the device gets created
dynamically when you plug in the USB serial device. on my linux,
Devuan which is debian based but unencumbered by systemd, it creates a
device /dev/ttyUSB0 most modern distributions I have seen use names like
this.
Paul.
On 2026-08-01 21:41, Jon Elson via cctalk wrote:
> Well, everything is a fight!
>
> The project I'm working on is to get a VMS MicroVAX system working at
> VCFMW. The plan was to have serial comm on a Linux laptop connected
> to the MicroVAX. I have the serial comm working on a Dell desktop
> machine, so I just plugged the same USB serial cable into the laptop,
> and NO COMM! I plugged back into the desktop, and serial data works
> fine! I'm totally mystified! The desktop is running an OLD Debian
> kernel (7.11), the laptop is running a late Ubuntu (24.04.4).
>
> I have no idea why the laptop doesn't communicate. I have tried
> minicom, picocom and cutecom, same result on all of them.
>
> Thanks for any help you can suggest.
>
> Jon
>
--
mvh
Torfinn
2 Aug
2 Aug
7:20 a.m.
New subject: serial comm help Linux
I missed the beginning of this discussion but I am using USB serial ports (with a variety
of USB-serial devices) on Ubuntu everyday in my day job and the device files are
/dev/ttyUSB*. I connect with putty.
alan
On Aug 2, 2026, at 00:27, Torfinn Ingolfsen via cctalk
<cctalk(a)classiccmp.org> wrote:
To add to what Paul wrote:
Here is a oneliner to check for usb serial ports on a linux machine:
$ ls -l /dev/tty[ACU]*
(the initial '$' is the prompt, don't include it in the command)
run this before and after you plug in your usb serial device, and you
will find it - if it is detected.
If the device only works as user root, you need to create a file to
set ownership, group and permissions on the device when it is plugged
in. This is commonly done through udev / udevadm on linux.
On Sun, Aug 2, 2026 at 4:00 AM Paul Berger via
cctalk
<cctalk(a)classiccmp.org> wrote:
Well the first thing to do is make sure the USB serial device gets
configured, when you plug it in, you should see the serial device listed
in the output from 'lsusb' and there should be messages about it in
'dmesg' (requires root authority). It is possible that the naming
convention for the tty device that get created is different between the
two versions of linux. The usual way is the device gets created
dynamically when you plug in the USB serial device. on my linux,
Devuan which is debian based but unencumbered by systemd, it creates a
device /dev/ttyUSB0 most modern distributions I have seen use names like
this.
Paul.
> On 2026-08-01 21:41, Jon Elson via cctalk wrote:
> Well, everything is a fight!
>
> The project I'm working on is to get a VMS MicroVAX system working at
> VCFMW. The plan was to have serial comm on a Linux laptop connected
> to the MicroVAX. I have the serial comm working on a Dell desktop
> machine, so I just plugged the same USB serial cable into the laptop,
> and NO COMM! I plugged back into the desktop, and serial data works
> fine! I'm totally mystified! The desktop is running an OLD Debian
> kernel (7.11), the laptop is running a late Ubuntu (24.04.4).
>
> I have no idea why the laptop doesn't communicate. I have tried
> minicom, picocom and cutecom, same result on all of them.
>
> Thanks for any help you can suggest.
>
> Jon
>
--
mvh
Torfinn 
9:51 a.m.
New subject: serial comm help Linux
That is normal for devices using hardware such as FTDI, CH340 etc. which
have their own specific drivers. However for serial drivers running on a
microcontroller having it's own USB stack such as more modern Arduinos, the
serial device uses a generic USB class called Communications device class
(CDC) and specifically part of the Abstract Control Mode (ACM). This uses a
generic serial driver and gives rise to device nodes called /dev/ttyACMxx.
It's possible that eventually dedicated hardware will appear that also
conforms to this model.
On Sun, Aug 2, 2026 at 4:59 PM Alan Perry via cctalk <cctalk(a)classiccmp.org>
wrote:
I missed the beginning of this discussion but I am
using USB serial ports
(with a variety of USB-serial devices) on Ubuntu everyday in my day job and
the device files are /dev/ttyUSB*. I connect with putty.
alan
On Aug 2, 2026, at 00:27, Torfinn Ingolfsen via
cctalk <
cctalk(a)classiccmp.org> wrote:
To add to what Paul wrote:
Here is a oneliner to check for usb serial ports on a linux machine:
$ ls -l /dev/tty[ACU]*
(the initial '$' is the prompt, don't include it in the command)
run this before and after you plug in your usb serial device, and you
will find it - if it is detected.
If the device only works as user root, you need to create a file to
set ownership, group and permissions on the device when it is plugged
in. This is commonly done through udev / udevadm on linux.
On Sun, Aug 2, 2026 at 4:00 AM Paul Berger via
cctalk
<cctalk(a)classiccmp.org> wrote:
Well the first thing to do is make sure the USB serial device gets
configured, when you plug it in, you should see the serial device listed
in the output from 'lsusb' and there should be messages about it in
'dmesg' (requires root authority). It is possible that the naming
convention for the tty device that get created is different between the
two versions of linux. The usual way is the device gets created
dynamically when you plug in the USB serial device. on my linux,
Devuan which is debian based but unencumbered by systemd, it creates a
device /dev/ttyUSB0 most modern distributions I have seen use names like
this.
Paul.
> On 2026-08-01 21:41, Jon Elson via cctalk wrote:
> Well, everything is a fight!
>
> The project I'm working on is to get a VMS MicroVAX system working at
> VCFMW. The plan was to have serial comm on a Linux laptop connected
> to the MicroVAX. I have the serial comm working on a Dell desktop
> machine, so I just plugged the same USB serial cable into the laptop,
> and NO COMM! I plugged back into the desktop, and serial data works
> fine! I'm totally mystified! The desktop is running an OLD Debian
> kernel (7.11), the laptop is running a late Ubuntu (24.04.4).
>
> I have no idea why the laptop doesn't communicate. I have tried
> minicom, picocom and cutecom, same result on all of them.
>
> Thanks for any help you can suggest.
>
> Jon
>
--
mvh
Torfinn
3:38 a.m.
New subject: serial comm help Linux
On Sat, 2026-08-01 19:41:08 -0500, Jon Elson via cctalk <cctalk(a)classiccmp.org>
wrote:
Well, everything is a fight!
[...]
I have no idea why the laptop doesn't
communicate. I have tried
minicom, picocom and cutecom, same result on all of them.
Well, summing up:
* Check that the USB device is found at all by invoking `lsusb` as root
* Check that the serial device is found by calling `dmesg` and `ls -l /dev/ttyACM*
/dev/ttyUSB*`
* Ensure your user has permissions to access the TTY device node
* Ensure you have a corr3ct serial config (Ctrl-A O): Correct speed, bits per byte,
parity, stopbits. And disable all handshake (rts/cts, dtr/dsr, xon/xoff) initially
MfG, JBG
--
7:37 a.m.
New subject: serial comm help Linux
I'll add that the group ownership of your USB serial device matters.
"ls -l /dev/tty*" shows owner and group, and adding your user to that
group - usually "dialout" gives your account access without having to
fiddle around in udev.
This is for Ubuntu because it seems most common, but works in most
distributions:
https://askubuntu.com/questions/112568/how-do-i-allow-a-non-default-user-to…
Doc
PS - I typically use minicom, but ckermit is my fallback when I can't
get coherent output. It will do *exactly* what you tell it; no more &
no less.
On 8/2/26 06:38, Jan-Benedict Glaw via cctalk wrote:
On Sat, 2026-08-01 19:41:08 -0500, Jon Elson via
cctalk <cctalk(a)classiccmp.org> wrote:
Well, everything is a fight!
[...]
I have no idea why the laptop doesn't
communicate. I have tried
minicom, picocom and cutecom, same result on all of them.
Well, summing up:
* Check that the USB device is found at all by invoking `lsusb` as root
* Check that the serial device is found by calling `dmesg` and `ls -l /dev/ttyACM*
/dev/ttyUSB*`
* Ensure your user has permissions to access the TTY device node
* Ensure you have a corr3ct serial config (Ctrl-A O): Correct speed, bits per byte,
parity, stopbits. And disable all handshake (rts/cts, dtr/dsr, xon/xoff) initially 
1:53 p.m.
New subject: serial comm help Linux
On Sun, 2 Aug 2026, Doc Shipley via cctalk wrote:
PS - I typically use minicom, but ckermit is my
fallback when I can't get
coherent output. It will do *exactly* what you tell it; no more & no less.
I find `cu' the easiest to use, e.g.:
$ cu -l /dev/ttyUSB1 -s 115200 dir
$ cu -l /dev/ttyS1 -s 9600 dir
and that's it. If you need to disable hw flow control or modem control:
$ stty -F /dev/ttyUSB1 clocal -crtscts
$ stty -F /dev/ttyS1 clocal -crtscts
respectively, *after* starting `cu'.
Maciej
4:29 p.m.
New subject: serial comm/ KA630 POST error
The ground was the problem. For some CRAZY reason, pin 5 of
the 9-pin serial connector of the KA630-AA and cabinet kit
has a 1K Ohm resistance to system ground. I ran a clip lead
from the cabinet to pin 5 of the USB serial cable and then
the communication works.
But, now, for the POST I get 7..6..5..4..
?24 UNXINT
consistently. Now, I don't know what to do about that. It
used to do that once in a while, but now it is on every
power-up.
Anybody know what this means?
5:40 p.m.
New subject: serial comm/ KA630 POST error
On Sun, Aug 2, 2026, 6:44 PM Jon Elson via cctalk <cctalk(a)classiccmp.org
wrote:
The ground was the problem. For some CRAZY reason, pin 5 of
the 9-pin serial connector of the KA630-AA and cabinet kit
has a 1K Ohm resistance to system ground. I ran a clip lead
from the cabinet to pin 5 of the USB serial cable and then
the communication works.
But, now, for the POST I get 7..6..5..4..
?24 UNXINT
consistently. Now, I don't know what to do about that. It
used to do that once in a while, but now it is on every
power-up.
Anybody know what this means?
Unexpected Interrupt. I hit that once in the late 80s when i installed a
new nic. But it's been so long i don't recall what I did. I know in VMS I
had to do a SYSGEN, but i may have had to tell the boot roms... or maybe I
had to change the csr... it's been a long time though...
Warner
10 Aug
10 Aug
9:27 a.m.
New subject: serial comm/ KA630 POST error
On Sun, Aug 2, 2026 at 8:44 PM Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
The ground was the problem. For some CRAZY reason,
pin 5 of
the 9-pin serial connector of the KA630-AA and cabinet kit
has a 1K Ohm resistance to system ground.
I don't doubt that you had a problem with ground, but on a MicroVAX
console port, Pin 5 is not ground. Ground is Pin 7.
There are pinout similarities with a PC DE9 serial port, but AFAIK
it's not intentional.
1: Protective Ground
2: Transmited Data
3: Received Data
4: Request To Send (RTS)
5: Data Terminal Ready (DTR)
6: Data Set Ready (DSR)
7: Signal Ground
8: Shorted to pin 9 on MicroVAX and VAXstation 2000...
9: ...series systems, otherwise left floating.
BCC08 cable has 8-9 shorted because it was the standard console cable
back when "real" serial ports were on DB25s. I forget which machines
actually use it to detect if a serial console is attached.
-ethan
10:16 a.m.
New subject: serial comm/ KA630 POST error
On 8/10/26 12:27, Ethan Dicks via cctalk wrote:
On Sun, Aug 2, 2026 at 8:44 PM Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
Ah HA! This is different than the PC convention! I will
have to make up another cable. Hmm, I wonder if grounding
DTR is responsible for the ?24 UNXINT error during POST?
Thanks,
Jon
The ground was the problem. For some CRAZY
reason, pin 5 of
the 9-pin serial connector of the KA630-AA and cabinet kit
has a 1K Ohm resistance to system ground.
I don't doubt that you had a problem
with ground, but on a MicroVAX
console port, Pin 5 is not ground. Ground is Pin 7.
There are pinout similarities with a PC DE9 serial port, but AFAIK
it's not intentional.
1: Protective Ground
2: Transmited Data
3: Received Data
4: Request To Send (RTS)
5: Data Terminal Ready (DTR)
6: Data Set Ready (DSR)
7: Signal Ground
8: Shorted to pin 9 on MicroVAX and VAXstation 2000...
9: ...series systems, otherwise left floating.
2 Aug
2 Aug
4:32 p.m.
New subject: Sick KA630-AA (uVAX-II CPU)
I have a KA630-AA that used to work. While debugging other
problems, I tried plugging it into slot 8 of a BA23 card
cage, and it popped something. Does anybody know what parts
are damaged on the board? I know this exposed some parts to
+/- 12 V supplies.
Thanks in advance for any hints.
Jon
5:02 p.m.
New subject: Sick KA630-AA (uVAX-II CPU)
Probably the memory controller as it uses the CD bus extensively (and
not in a PMI way)
There's a thought: Plug a MV2 into a backplane with PMI memory boards,
see if it explodes.
Microvax I did not use PMI, correct?
C
On 8/2/26 20:32, Jon Elson via cctalk wrote:
I have a KA630-AA that used to work. While debugging
other problems,
I tried plugging it into slot 8 of a BA23 card cage, and it popped
something. Does anybody know what parts are damaged on the board? I
know this exposed some parts to +/- 12 V supplies.
Thanks in advance for any hints.
Jon
16 Jul
16 Jul
7:08 a.m.
New subject: Emulex QD21 controller help
On 7/15/26 14:55, Douglas Taylor via cctalk wrote:
In the past I had a QD21 and ESDI drives. What
controller
were the Fujitsu drives connected to originally? There
maybe some kind of proprietary format on them that the
QD21 doesn't understand.
They were used with a DIFFERENT QD21. I had to update
THIS
QD21 firmware to a rev level that had the field resident
diagnostics. I gave away my old QD21 at VCF a while ago.
Does the QD21 see or sense the drives?
Yes, when the uVAX tries to boot, the drives seek and
recalibrate several times and then I get a ??DSK ERR?? message.
I have scribbled notes of the config params on the smaller
disk, and I will try those next time.
Thanks,
Jon
15 Jul
15 Jul
7:26 a.m.
New subject: MicroVAX-II to Linux communications (minicom)
I am trying to get my old VMS system going for demo at VCFMW.
I got a USB-serial adapter, and it mostly works OK. I am
using minicom, and trying to get the Emulex firmware
resident diagnostics loaded on the uVAX. This requires
entering a BUNCH of ODT commands. I tried to put these in a
file, and then do an ASCII send file command in minicom, but
it gets garbled. I assumed that minicom was sending too
fast, so put in a bunch of delays on each character and on
linefeed. I did this in the T (terminal) menu. I can
clearly see these delays work when typing on the keyboard,
but they DON'T seem to work when sending the file, as the
file send doesn't take any longer to complete. Is there
another place to set delays that is effective for ASCII send?
Also, minicom doesn't seem to save these terminal (T)
settings when restarted.
Should I be using a different program?
Thanks for any advice!
Jon
8:16 a.m.
New subject: MicroVAX-II to Linux communications (minicom)
On 2026Jul 15,, at 8:26 AM, Jon Elson via cctalk <cctalk(a)classiccmp.org> wrote:
I am trying to get my old VMS system going for demo at VCFMW.
I got a USB-serial adapter, and it mostly works OK. I am using minicom, and trying to
get the Emulex firmware resident diagnostics loaded on the uVAX. This requires entering a
BUNCH of ODT commands. I tried to put these in a file, and then do an ASCII send file
command in minicom, but it gets garbled. I assumed that minicom was sending too fast, so
put in a bunch of delays on each character and on linefeed. I did this in the T
(terminal) menu. I can clearly see these delays work when typing on the keyboard, but
they DON'T seem to work when sending the file, as the file send doesn't take any
longer to complete. Is there another place to set delays that is effective for ASCII
send?
Also, minicom doesn't seem to save these terminal (T) settings when restarted.
Should I be using a different program?
This sounds like the inherent problem with USB-serial adapters.
The terminal program (minicom in your case) may well be adding the per-character delay,
but somewhere in the host-to-USB stream the characters are accumulated in a buffer/packet,
when the USB adapter finally gets them it sends them out to its serial port all at the
regular bps rate, sans the inter-character delay.
Or at least, I have had such a flow control problem in trying to emulate a paper-tape
reader (for downloading BASIC programs to an HP2116). Adding delays, adding TC_FLUSHES to
the terminal program, etc., were of no effect. In that situation, I was eventually able
to resolve it by adding some code to (my) terminal program to wait for the LF returned by
the receiving software (BASIC interpreter on the 2116) after sending the EOL CR, that is,
using the LF as an ACK.
If it might work for your situation, or if you want a hackable terminal program:
http://madrona.ca/e/scomm/index.html
16 Jul
16 Jul
12:54 p.m.
New subject: MicroVAX-II to Linux communications (minicom)
I'd do prep and all the heavy lifting with a PC that has a real serial
terminal, get it working correctly, and then tackle the USB2Serial issues.
Not all adapters work correctly and they are a headache to rely on.
b
On Wed, Jul 15, 2026 at 12:23 PM Brent Hilpert via cctalk <
cctalk(a)classiccmp.org> wrote:
On 2026Jul 15,, at 8:26 AM, Jon Elson via cctalk
<cctalk(a)classiccmp.org>
wrote:
I am trying to get my old VMS system going for demo at VCFMW.
I got a USB-serial adapter, and it mostly works OK. I am using minicom,
and trying to get the Emulex firmware resident diagnostics loaded on the
uVAX. This requires entering a BUNCH of ODT commands. I tried to put
these in a file, and then do an ASCII send file command in minicom, but it
gets garbled. I assumed that minicom was sending too fast, so put in a
bunch of delays on each character and on linefeed. I did this in the T
(terminal) menu. I can clearly see these delays work when typing on the
keyboard, but they DON'T seem to work when sending the file, as the file
send doesn't take any longer to complete. Is there another place to set
delays that is effective for ASCII send?
Also, minicom doesn't seem to save these terminal (T) settings when
restarted.
Should I be using a different program?
This sounds like the inherent problem with USB-serial adapters.
The terminal program (minicom in your case) may well be adding the
per-character delay, but somewhere in the host-to-USB stream the characters
are accumulated in a buffer/packet, when the USB adapter finally gets them
it sends them out to its serial port all at the regular bps rate, sans the
inter-character delay.
Or at least, I have had such a flow control problem in trying to emulate a
paper-tape reader (for downloading BASIC programs to an HP2116). Adding
delays, adding TC_FLUSHES to the terminal program, etc., were of no effect.
In that situation, I was eventually able to resolve it by adding some code
to (my) terminal program to wait for the LF returned by the receiving
software (BASIC interpreter on the 2116) after sending the EOL CR, that is,
using the LF as an ACK.
If it might work for your situation, or if you want a hackable terminal
program:
http://madrona.ca/e/scomm/index.html
4:46 p.m.
New subject: MicroVAX-II to Linux communications (minicom)
On 7/16/26 15:54, Bill Degnan via cctalk wrote:
I'd do prep and all the heavy lifting with a PC
that has a real serial
terminal, get it working correctly, and then tackle the USB2Serial issues.
Not all adapters work correctly and they are a headache to rely on.
Yes, I believe I saw the same issues with the native (on
motherboard) serial port on this Dell desktop machine. I
think the OS is too out of date to download other serial
apps. But, I can try with my laptop.
Thanks,
Jon
22 Jun
22 Jun
3:51 p.m.
New subject: What constitutes a dead computer language
Another dead language is FPL, Foxboro Process Language. It was a BASIC-like
multitasking/scheduling monitor for data acquisition and real time control that ran on the
PDP-11/15 in 1973. I scanned this sales brochure that describes the flavour of it:
https://archive.org/details/fox210minisystem
When I recovered the FOX system I found some paper tapes which I dumped a few years ago.
Unfortunately no FOXBORO labelled ones with the FPL development system per se, but one of
them was a complete 28KW core dump of the running system that quite possibly has the FPL
kernel along with the steelworks oxygen plant monitoring application on it. I've not
looked at it since then though, would be better left for someone who can knowledgeably
analyse them and see if there's anything interesting.
1
days inactive
52
days old
72 comments
36 participants
participants (36)
-
Adrian Godwin -
Alan Perry -
ben -
Bill Degnan -
Bill Gunshannon -
Brent Hilpert -
Cameron Kaiser -
Chris Zach -
Doc Shipley -
Douglas Taylor -
emanuel stiebler -
Ethan Dicks -
Fred Cisin -
Gary Grebus -
Gavin Scott -
Glen Slick -
Gordon Henderson -
Graham Toal -
Hans-Ulrich Hölscher -
Harald Arnesen -
Jan-Benedict Glaw -
Jim Davis -
Jon Elson -
Ken Seefried -
Maciej W. Rozycki -
Mark Green -
Martin Bishop -
Nadav Eiron -
Paul Berger -
Paul Koning -
paul.kimpel＠digm.com -
Peter Ekstrom -
steven＠malikoff.com steven＠malikoff.com -
Torfinn Ingolfsen -
Van Snyder -
Warner Losh