9 Sep
2025
9 Sep
'25
5:52 p.m.
I just love the PDP11's assembly language. I needed a super-tight
subroutine to print a 16-bit value as 6 octal digits. This is as tight as I
could make it, 16 words (including writing to the serial port, which takes
5 words). Can you beat it?
;*** Subroutine *************************
;Print a 16-bit value as 6 octal digits
;Calling Sequence:
; jsr PC,PROCT6
;On Entry:
; R2 = value to print
;Trashes R1,R2
;****************************************
PROCT6: mov #100030,R1 ;Digit loop ends when '1' lands in C
;..'30' makes it ASCII
sec ;All done when this is in C again
;Extract a digit and convert it to ASCII. Check for done.
1$: rol R2 ;Shift digit out of R2 & into R1
beq 3$ ;Return when done
rol r1 ;Build next octal digit
bcc 1$ ;Done when c = shift pattern bit
;Write digit to the serial port
2$: tstb @#CTXSTA ;Wait for transmitter (clears C)
bpl 2$
movb R1,@#CTXDAT ;Transmit now
;Next digit
mov #020006,R1 ;Digit ends when "2" lands in C.
;Printing ends when "sec" bit
;..leaves R2. '6' makes ASCII
br 1$
3$: rts PC
Martin E.
10 Sep
10 Sep
6:32 a.m.
On Sep 9, 2025, at 8:52 PM, Martin Eberhard via cctalk
<cctalk(a)classiccmp.org> wrote:
I just love the PDP11's assembly language. I needed a super-tight
subroutine to print a 16-bit value as 6 octal digits. This is as tight as I
could make it, 16 words (including writing to the serial port, which takes
5 words). Can you beat it?
;*** Subroutine *************************
;Print a 16-bit value as 6 octal digits
;Calling Sequence:
; jsr PC,PROCT6
;On Entry:
; R2 = value to print
;Trashes R1,R2
;****************************************
PROCT6: mov #100030,R1 ;Digit loop ends when '1' lands in C
;..'30' makes it ASCII
sec ;All done when this is in C again
;Extract a digit and convert it to ASCII. Check for done.
1$: rol R2 ;Shift digit out of R2 & into R1
beq 3$ ;Return when done
rol r1 ;Build next octal digit
bcc 1$ ;Done when c = shift pattern bit
;Write digit to the serial port
2$: tstb @#CTXSTA ;Wait for transmitter (clears C)
bpl 2$
movb R1,@#CTXDAT ;Transmit now
;Next digit
mov #020006,R1 ;Digit ends when "2" lands in C.
;Printing ends when "sec" bit
;..leaves R2. '6' makes ASCII
br 1$
3$: rts PC
Martin E.
Very nice. Here is the code used in the KMON (resident part of the keyboard monitor,
i.e., CLI) of RT-11 V2. I think it was created by Anton Chernoff.
OPRINT: MOV #30,R0 ;CONVERT WORD TO OCTAL AND PRINT IT
SEC
4$: ROL R2 ;DON'T TRY TO UNDERSTAND THIS ROUTINE
ROLB R0 ; JUST USE IT & LOVE IT
.TTYOUT
MOV #206,R0
5$: ASL R2 ;DONE YET ?
BEQ 6$ ;YES
ROLB R0
BCS 5$
BR 4$
6$: <exit>
Same approach but yours is shorter because it forms the next digit in one place rather
than two. Interesting that was missed in the earlier version.
paul
7:46 a.m.
On 9/10/25 08:32, Paul Koning via cctalk wrote:
> On Sep 9, 2025, at 8:52 PM, Martin Eberhard via
cctalk <cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant,
and did the very best with what you could do in 16 bits!
Jon
9:02 a.m.
On 2025-09-10 8:46 a.m., Jon Elson via cctalk wrote:
On 9/10/25 08:32, Paul Koning via cctalk wrote:
But even then you could see the hand writing on the wall for 16 bits.
> On Sep 9, 2025, at 8:52 PM, Martin Eberhard
via cctalk
> <cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant, and did the
very best with what you could do in 16 bits! Jon
Still waiting for the PDP K.
Ben.
10:34 a.m.
Yep, 16 bits, or even 32 seemed like a lot, If you're writing asm, It is.
But like you alluded to, it's not good enough for the stuff we expect today,
On Wed, Sep 10, 2025 at 9:26 AM ben via cctalk <cctalk(a)classiccmp.org>
wrote:
On 2025-09-10 8:46 a.m., Jon Elson via cctalk wrote:
On 9/10/25 08:32, Paul Koning via cctalk wrote:
But even then you could see the hand writing on the wall for 16 bits.
> On Sep 9, 2025, at 8:52 PM, Martin Eberhard
via cctalk
> <cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant, and did the
very best with what you could do in 16 bits!
Jon
Still waiting for the PDP K.
Ben.
1:35 p.m.
On Sep 10, 2025, at 1:34 PM, Jim Davis via cctalk
<cctalk(a)classiccmp.org> wrote:
Yep, 16 bits, or even 32 seemed like a lot, If you're writing asm, It is.
But like you alluded to, it's not good enough for the stuff we expect today,
That depends a lot on what sort of "stuff" you work with.
For various fun projects I use the Raspberry Pico, which is a 32 bit microcontroller with
1/4 or 1/2 MB of RAM, which is quite a lot. But a bunch of that work uses the PIO engines
-- basically programmable state machines that talk to the I/O pins. Those have 16 bit
instructions, and a 32 word program memory. No data RAM, just a couple of registers.
Quite powerful, though. One application is the DDCMP controller I published a while ago.
I'm thinking about making it talk Ethernet in software as the next exercise...
paul
1:45 p.m.
I did 8080 programming and had 16kb to play with.
I thought there would never be a program large enough that 1mb was needed. Hah
Sent from my iPhone
On Sep 10, 2025, at 13:41, Paul Koning via cctalk
<cctalk(a)classiccmp.org> wrote:
On Sep 10, 2025, at 1:34 PM, Jim Davis via cctalk
<cctalk(a)classiccmp.org> wrote:
Yep, 16 bits, or even 32 seemed like a lot, If you're writing asm, It is.
But like you alluded to, it's not good enough for the stuff we expect today,
That depends a lot on what sort of "stuff" you work with.
For various fun projects I use the Raspberry Pico, which is a 32 bit microcontroller with
1/4 or 1/2 MB of RAM, which is quite a lot. But a bunch of that work uses the PIO engines
-- basically programmable state machines that talk to the I/O pins. Those have 16 bit
instructions, and a 32 word program memory. No data RAM, just a couple of registers.
Quite powerful, though. One application is the DDCMP controller I published a while ago.
I'm thinking about making it talk Ethernet in software as the next exercise...
paul
5:56 p.m.
I've been doing embedded my whole career. ESP 32 is my retirement toy. Lots
of C, forth and ASM. I still can't write a python app.
On Wed, Sep 10, 2025 at 1:45 PM Wayne S <wayne.sudol(a)hotmail.com> wrote:
I did 8080 programming and had 16kb to play with.
I thought there would never be a program large enough that 1mb was needed.
Hah
Sent from my iPhone
On Sep 10, 2025, at 13:41, Paul Koning via cctalk
<cctalk(a)classiccmp.org>
wrote:
> On Sep 10, 2025, at 1:34 PM, Jim Davis via cctalk <
cctalk(a)classiccmp.org> wrote:
>
> Yep, 16 bits, or even 32 seemed like a lot, If you're writing asm, It
is.
> But like you alluded to, it's not good
enough for the stuff we expect
today,
That depends a lot on what sort of "stuff" you work with.
For various fun projects I use the Raspberry Pico, which is a 32 bit
microcontroller with 1/4 or 1/2 MB of RAM, which is quite a lot. But a
bunch of that work uses the PIO engines -- basically programmable state
machines that talk to the I/O pins. Those have 16 bit instructions, and a
32 word program memory. No data RAM, just a couple of registers. Quite
powerful, though. One application is the DDCMP controller I published a
while ago. I'm thinking about making it talk Ethernet in software as the
next exercise...
paul
11 Sep
11 Sep
3:10 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On 10/09/2025 18:34, Jim Davis via cctalk wrote:
Yep, 16 bits, or even 32 seemed like a lot, If
you're writing asm, It is.
But like you alluded to, it's not good enough for the stuff we expect today,
I believe it was Chuck Moore that said "If you can't do it in 8K it's
not worth doing", before designing the Novix chip with 32K of address space.
1:30 p.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
You can get an emulated 11/23 running 4.2 BSD for $3.00. That's crazy.
On Thu, Sep 11, 2025 at 4:11 AM Frank Leonhardt via cctalk <
cctalk(a)classiccmp.org> wrote:
On 10/09/2025 18:34, Jim Davis via cctalk wrote:
I believe it was Chuck Moore that said "If you can't do it in 8K it's
not worth doing", before designing the Novix chip with 32K of address
space.
Yep, 16 bits, or even 32 seemed like a lot, If
you're writing asm, It is.
But like you alluded to, it's not good enough for the stuff we expect
today,
5:44 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
The only machines I used that has blinky lights and switches was my altair
and the old PDP-8 at OMSI.
I have other junk, A Alpha micro with hex pad and a Tek 8002a with switches.
Their 11/45 had lights, but ran RSTS most of the time. Every 11/23 I used
was without front panel.
The machine I'm referring to is the ESP32, USB power, and no case, but you
could print one.
I can almost remember the PDP-8 dectape boot from memory, 50 years later.
Has it been that long?
On Fri, Sep 12, 2025 at 1:26 AM ben via cctalk <cctalk(a)classiccmp.org>
wrote:
On 2025-09-11 2:30 p.m., Jim Davis via cctalk wrote:
You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
What device? Power supple, case and front panel all must be extra. Ben.
6:46 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 12, 2025, at 8:44 AM, Jim Davis via cctalk
<cctalk(a)classiccmp.org> wrote:
...
Their 11/45 had lights, but ran RSTS most of the time.
I'm confused by the "but". 11/45s have lights and switches panels, and RSTS
would display a nice lights pattern on that when idle. Many versions could optionally do
a rather amazing lights pattern involving both the address and the data lights
("fancy nuljob").
paul
7:43 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
It's crazier than you think. 4.2 BSD doesn't run on the PDP-11.
But I can get 4.whatever running on an emulated VAX or 2.11 BSD
running on an emulated PDP-11 of my choice (I usually do 11/44
cause I had some and really liked them) for free.
bill
13 Sep
13 Sep
7:49 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill
9:35 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 13, 2025, at 10:49 AM, Bill Gunshannon via
cctalk <cctalk(a)classiccmp.org> wrote:
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11.
You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill
9:59 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On 2025-09-13 10:35 a.m., Milo Velimirović via cctalk wrote:
But can you still get a real PDP-11 replacement for driving vintage
hardware?
On Sep 13, 2025, at 10:49 AM, Bill Gunshannon via
cctalk <cctalk(a)classiccmp.org> wrote:
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11. You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill
1:21 p.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 13, 2025, at 12:59 PM, ben via cctalk
<cctalk(a)classiccmp.org> wrote:
On 2025-09-13 10:35 a.m., Milo Velimirović via cctalk wrote:
But can you still get a real PDP-11 replacement for driving vintage hardware?
Probably not. But I think John Wilson of Ersatz-11 fame created a Unibus (or QBus?)
interface for PCs so E-11 could drive real DEC peripherals.
Given suitable transceivers, it would not be all that hard to create a PDP-11 replacement
with Unibus and/or Qbus capability out of an FPGA. There are working PDP11 emulations for
FPGAs.
paul
On Sep
13, 2025, at 10:49 AM, Bill Gunshannon via cctalk <cctalk(a)classiccmp.org> wrote:
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11. You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill 
14 Sep
14 Sep
2:28 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
But can you still get a real PDP-11 replacement for
driving vintage hardware?
Probably not. But I think John Wilson of Ersatz-11 fame created a Unibus (or QBus?)
interface for PCs so E-11 could drive real DEC peripherals.
Given suitable transceivers, it would not be all that hard to create a PDP-11 replacement
with Unibus and/or Qbus capability out of an FPGA. There are working PDP11 emulations for
FPGAs.
Paul
I picked up a PC of a retired "DEC solutions" engineer.
He worked with the licensed Ersatz-11 of John Wilson.
In one of the PCs is that famous card to connect PDP-11 peripherals to a PC.
That card is not supported in the free version of Ersatz-11.
I have not yet looked at that PC, so I do not know whether it is QBUS or UNIBUS.
One of these days ...
Henk, PA8PDP
7:10 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sat, Sep 13, 2025 at 12:07 PM ben via cctalk <cctalk(a)classiccmp.org> wrote:
But can you still get a real PDP-11 replacement for
driving vintage
hardware?
Strobe Data is almost still around, but I think whoever's left may
just be in limited best-effort support mode for their remaining
customers. Osprey cards turn up occasionally and I *think* some may
have had luck getting licenses from them.
http://www.strobedata.com/home/products.html
3:39 p.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 13, 2025, at 12:59 PM, ben via cctalk
<cctalk(a)classiccmp.org> wrote:
On 2025-09-13 10:35 a.m., Milo Velimirović via cctalk wrote:
But can you still get a real PDP-11 replacement for driving vintage hardware?
Real pdp11s are somewhat scarce, but . . .
If you’ve got pdp11 UNIBUS or Q-Bus devices/controllers AND a backplane to plug them into,
but no pdp11 CPU, then a Unibone or Qbone is the thing you want. It will be as much (or as
little) of a CPU as you want.
On Sep
13, 2025, at 10:49 AM, Bill Gunshannon via cctalk <cctalk(a)classiccmp.org> wrote:
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11. You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill 
5:20 p.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 14, 2025, at 17:46, Milo Velimirović via cctalk
<cctalk(a)classiccmp.org> wrote:
The only cpu that the unibone and qbone emulate is a pdp-11/20 processor, (in addition to
several peripherals) last I knew. Nothing with memory management, for example. QBus
PDP-11 machines can be had (eBay has a couple, one looks like it probably has, one looks
like it probably has a power supply that needs work) and would be easier for most to deal
with as compared to most of the UNIBUS machines. Maybe you could sell John that old PDP
11 that I recall you obtaining from me decades ago. ;) of course it doesn’t have any
memory management either.
JRJ
On Sep 13, 2025, at 12:59 PM, ben via cctalk
<cctalk(a)classiccmp.org> wrote:
On 2025-09-13 10:35 a.m., Milo Velimirović via cctalk wrote:
But can you still get a real PDP-11 replacement for driving vintage hardware?
Real pdp11s are somewhat scarce, but . . .
If you’ve got pdp11 UNIBUS or Q-Bus devices/controllers AND a backplane to plug them
into, but no pdp11 CPU, then a Unibone or Qbone is the thing you want. It will be as much
(or as little) of a CPU as you want.
On Sep
13, 2025, at 10:49 AM, Bill Gunshannon via cctalk <cctalk(a)classiccmp.org> wrote:
> On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
>> You can get an emulated 11/23 running 4.2 BSD for $3.00. That's crazy.
>
> Can you tell me where? With that glaring error I am really
> interested in learning more. Google finds nothing with those
> terms.
>
> bill
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11.
13 Sep
13 Sep
1:15 p.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 13, 2025, at 12:35 PM, Milo Velimirović via
cctalk <cctalk(a)classiccmp.org> wrote:
I would think a Raspberry Pico, which goes for $4, is enough of a processor to run SIMH.
Not a large config I suppose, and given that it's a bare metal system the port would
be a bit of work. But it seems doable.
paul
On Sep 13, 2025, at 10:49 AM, Bill Gunshannon via
cctalk <cctalk(a)classiccmp.org> wrote:
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11. You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill 
14 Sep
14 Sep
8:07 a.m.
New subject: [List] Re: Fun with PDP11 Assembly Language
On Sep 13, 2025, at 4:15 PM, Paul Koning via cctalk
<cctalk(a)classiccmp.org> wrote:
SIMH seems to be a bit heavyweight for a Pico, especially in the memory department. I’ve
been working on a standalone emulator for a while. An RP2040 is capable of running an
18-bit system with around 64KW of memory while an RP2350 is capable of running a full
18-bit system with 124KW. Using a “Maker Pi Pico” board from cytron.io
<http://cytron.io/> (an additional $10) to provide an SPI-based microSD drive, I’ve
been able to fun RSX-11M V4.5 and RSX-11D V6.2. Performance is not great but for ~$15, I
can’t complain.
John.
On Sep 13, 2025, at 12:35 PM, Milo Velimirović
via cctalk <cctalk(a)classiccmp.org> wrote:
I would think a Raspberry Pico, which goes for $4, is enough of a processor to run SIMH.
Not a large config I suppose, and given that it's a bare metal system the port would
be a bit of work. But it seems doable.
paul
On Sep 13, 2025, at 10:49 AM, Bill Gunshannon via
cctalk <cctalk(a)classiccmp.org> wrote:
On 9/11/2025 4:30 PM, Jim Davis via cctalk wrote:
It has to be hyperbole.
No pdp11 ran 4.x, the best you can do is 2.11— which is still being patched and
maintained; see the pidp-11 mailing list. Probably not $3.00, but $30.00 will get you a
sytem on a DIP package that will emulate most anything. Spend a bit more for a Pi4 or Pi5
if you want to drive a PiDP-11.
You can get an emulated 11/23 running 4.2 BSD for
$3.00. That's crazy.
Can you tell me where? With that glaring error I am really
interested in learning more. Google finds nothing with those
terms.
bill
10 Sep
10 Sep
12:10 p.m.
On 9/10/25 11:02, ben via cctalk wrote:
On 2025-09-10 8:46 a.m., Jon Elson via cctalk wrote:
Oh, absolutely, the PDP-11 was the best 16-bit instruction
set I've ever seen, and I've seen quite a few. But, there
were substantial limitations. Then, I moved up to the VAX,
which was a real experience! Over the top instruction set,
but it kept all the greatness of the PDP-11, and solved the
address space issue.
Jon
On 9/10/25 08:32, Paul Koning via cctalk wrote:
But even then you could see the hand writing on the wall
for 16 bits. > On Sep 9, 2025, at 8:52 PM, Martin Eberhard
via cctalk
> <cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY
brilliant, and did the very best with what you could do
in 16 bits!
12:25 p.m.
On Sep 10, 2025, at 3:10 PM, Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
On 9/10/25 11:02, ben via cctalk wrote:
For a while, until programmers used it all up. This is the Microsoft effect: programs
will expand to consume all available CPU, memory, and storage.
paul
On 2025-09-10 8:46 a.m., Jon Elson via cctalk
wrote:
Oh, absolutely, the PDP-11 was the best 16-bit instruction set I've ever seen, and
I've seen quite a few. But, there were substantial limitations. Then, I moved up to
the VAX, which was a real experience! Over the top instruction set, but it kept all the
greatness of the PDP-11, and solved the address space issue. On 9/10/25 08:32, Paul Koning via cctalk wrote:
But even then you could see the hand writing on the wall for 16 bits. > On Sep 9, 2025, at 8:52 PM, Martin Eberhard
via cctalk <cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant, and did the very best with
what you could do in 16 bits!
1:18 p.m.
You’re a hardware guy, huh?
Sent from my iPhone
On Sep 10, 2025, at 12:31, Paul Koning via cctalk
<cctalk(a)classiccmp.org> wrote:
On Sep 10, 2025, at 3:10 PM, Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
For a while, until programmers used it all up. This is the Microsoft effect: programs
will expand to consume all available CPU, memory, and storage.
paul
On 9/10/25 11:02, ben via cctalk wrote:
On 2025-09-10 8:46 a.m., Jon Elson via cctalk wrote:
Oh, absolutely, the PDP-11 was the best 16-bit instruction set I've ever seen, and
I've seen quite a few. But, there were substantial limitations. Then, I moved up to
the VAX, which was a real experience! Over the top instruction set, but it kept all the
greatness of the PDP-11, and solved the address space issue.
On 9/10/25 08:32, Paul Koning via cctalk wrote:
>> On Sep 9, 2025, at 8:52 PM, Martin Eberhard via cctalk
<cctalk(a)classiccmp.org> wrote:
>>
>> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant, and did the very best with
what you could do in 16 bits!
But even then you could see the hand writing on the wall for 16 bits.
1:25 p.m.
Nope. But I learned programming back when machine memories were measured in k words --
like the college timesharing 11/20 with 28kW, or the physics department lab 11/20 with 8
kW (and an RC11 hard drive, 64 kW if I remember right). For that matter, I remember
squeezing CDC 6400 boot code into a 12 word "deadstart panel" and the secondary
boot into a 320 word disk sector.
Nowadays some of the machines I work with have a terabyte of RAM. Mindboggling.
paul
On Sep 10, 2025, at 4:18 PM, Wayne S
<wayne.sudol(a)hotmail.com> wrote:
You’re a hardware guy, huh?
Sent from my iPhone
> On Sep 10, 2025, at 12:31, Paul Koning via cctalk <cctalk(a)classiccmp.org>
wrote:
>
>
>
>> On Sep 10, 2025, at 3:10 PM, Jon Elson via cctalk <cctalk(a)classiccmp.org>
wrote:
>>
>>> On 9/10/25 11:02, ben via cctalk wrote:
>>> On 2025-09-10 8:46 a.m., Jon Elson via cctalk wrote:
>>>> On 9/10/25 08:32, Paul Koning via cctalk wrote:
>>>>>> On Sep 9, 2025, at 8:52 PM, Martin Eberhard via cctalk
<cctalk(a)classiccmp.org> wrote:
>>>>>>
>>>>>> I just love the PDP11's assembly language.
>>>>
>>>> Yes, the PDP-11 machine language was INCREDIBLY brilliant, and did the
very best with what you could do in 16 bits!
>>>
>>> But even then you could see the hand writing on the wall for 16 bits.
>>
>> Oh, absolutely, the PDP-11 was the best 16-bit instruction set I've ever
seen, and I've seen quite a few. But, there were substantial limitations. Then, I
moved up to the VAX, which was a real experience! Over the top instruction set, but it
kept all the greatness of the PDP-11, and solved the address space issue.
>
> For a while, until programmers used it all up. This is the Microsoft effect:
programs will expand to consume all available CPU, memory, and storage.
>
> paul
>
7:37 p.m.
On 2025-09-10 2:25 p.m., Paul Koning via cctalk wrote:
Nope. But I learned programming back when machine
memories were measured in k words -- like the college timesharing 11/20 with 28kW, or the
physics department lab 11/20 with 8 kW (and an RC11 hard drive, 64 kW if I remember
right). For that matter, I remember squeezing CDC 6400 boot code into a 12 word
"deadstart panel" and the secondary boot into a 320 word disk sector.
Nowadays some of the machines I work with have a terabyte of RAM. Mindboggling.
I find it more mind boggling that it runs with out memory errors.
paul
Ben.
9:18 a.m.
I was able to get over my PDP-11 depression after programming 6502's and
Z80's for 10 years.
I finally worked on the 68K in the late 80's, It felt like an old shoe.
On Wed, Sep 10, 2025 at 8:11 AM Jon Elson via cctalk <cctalk(a)classiccmp.org>
wrote:
On 9/10/25 08:32, Paul Koning via cctalk wrote:
> On Sep 9, 2025, at 8:52 PM, Martin Eberhard
via cctalk <
cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant,
and did the very best with what you could do in 16 bits!
Jon
7:51 p.m.
when the PDP 11 first came out, that was the first time I wanted a computer
for myself at home, the earlier ones that I knew of had too cockeyed and
limited instruction sets, after my programming experience being on
mainframes with CISC instructions. I actually thought about how much
sacrifice it would take to buy one.
--
--Carey
On 9/10/25 08:32, Paul Koning via cctalk wrote:
> On Sep 9, 2025, at 8:52 PM, Martin Eberhard
via cctalk <
cctalk(a)classiccmp.org> wrote:
>
> I just love the PDP11's assembly language.
Yes, the PDP-11 machine language was INCREDIBLY brilliant,
and did the very best with what you could do in 16 bits!
Jon
10:30 p.m.
On 2025-09-10 8:51 p.m., Carey Schug via cctalk wrote:
when the PDP 11 first came out, that was the first
time I wanted a computer
for myself at home, the earlier ones that I knew of had too cockeyed and
limited instruction sets, after my programming experience being on
mainframes with CISC instructions. I actually thought about how much
sacrifice it would take to buy one.
You still have cocky and limited instruction sets,
but $1.99 will get you a 16 bit chip micro controller in quantities of one.
I drooled at the Heath-kit H-11 when it came out,but when I had money
it was gone.
I just picked up a cheap Laser cutter/engraver and you have a tiny hand
sized usb box controlling it. To think, it was not that long ago it took
a whole rack box to do the the same thing.
Ben.
0
days inactive
5
days old
32 comments
14 participants
participants (14)
-
ben -
Bill Gunshannon -
Carey Schug -
Frank Leonhardt -
Gavin Scott -
Henk Gooijen -
Jay Jaeger -
Jim Davis -
johnforecast＠comcast.net -
Jon Elson -
Martin Eberhard -
Milo Velimirović -
Paul Koning -
Wayne S