-----Original Message-----
From: owner-hecnet at Update.UU.SE
[mailto:owner-hecnet at Update.UU.SE] On Behalf Of Johnny Billquist
Sent: Sunday, January 06, 2013 15:54
To: hecnet at Update.UU.SE
Cc: John Wilson
Subject: Re: [HECnet] DU11 vs. DUV11
On 2013-01-06 21:47, John Wilson wrote:
I still can't find where it says "8085A" in the 110/130 manual, but
page
B-3 gives I/O page timings and mentions an "I/O microprocessor" and
the fact that word I/O takes 90 usec but byte I/O takes 180
usec. So
that certainly explains the difference in the DUV driver. Plus the
manual specifically says that RSX11S is available for the
/110, so I
guess it was even supported! Sweet.
Also mentioned on page 4-6, but pretty much with the same
level of detail. Byte writes cause a read-modify-write. Time
of 180 usec. And there is an (unmentioned) 8-bit
microprocessor acting in between the
PDP-11 and any peripherials.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
I just found my VT103 LSI-11 Video Terminal User's Guide.
The beast looks like a VT100 but has dual, count them dual, TU58's in
the front and a 4x4 18-bit backplane in the back. Typical configuration
(slot-wise) is:
1 - LSI-11 Processor
2 - MSV11-DD memory
3 - DJV11-J (4-channel serial line interface)
Or
1 - LSI-11 Processor
2 - MVX11-A memory and serial in/out
The bus slots are numbered from 0-7.
I will look for my PDT documentation a little later. I am cleaning up
that part of the basement now anyway.
-Steve
You must be talking of the PDT-11/150 then. The /110 and /130 sat inside
a VT100 shell. Extremely similar to a VT103 (I actually never figured
out what the difference between a VT103 and a PDT-11/130 is.)
I may be grossly misinformed, but my understanding is that although the
/110 and /130 both live inside a VT100, they're *not* like VT103s, and
have no Q-bus but use a custom CPU board just like the /150 (I doubt it's
the exact same board, but a similar design).
I could have sworn the 110/130 manual on bitsavers (which is misfiled
away from the 150 manual -- I forget where) said so but on a quick skim I
can't find it. The I/O hardware has enough programming differences that
it can't be regular Q-bus stuff, although I suppose they could have made
a whole set of PDT-specific boards.
Yes, I think that's the case. I found a paragraph in the PDT 11/130 manual that states that the PDT family all have "bounded bus" systems that are completely buried in the PDT card, and an 8 bit processor actually talks to the IO devices. It also says that a word transfer to the IO system takes 90 microseconds, and a byte transfer takes 180 seconds, so that explains the logic behind the optimization.
I never thought of any of them running any type of RSX - I ll have to dust off one of my PDT-11/150s and see if I can get some sort o' RSX on it.
--
Lee K. Gleason N5ZMR
Control-G Consultants
lee.gleason at comcast.net
On 2013-01-06 21:47, John Wilson wrote:
I still can't find where it says "8085A" in the 110/130 manual, but page
B-3 gives I/O page timings and mentions an "I/O microprocessor" and the
fact that word I/O takes 90 usec but byte I/O takes 180 usec. So that
certainly explains the difference in the DUV driver. Plus the manual
specifically says that RSX11S is available for the /110, so I guess it
was even supported! Sweet.
Also mentioned on page 4-6, but pretty much with the same level of detail. Byte writes cause a read-modify-write. Time of 180 usec. And there is an (unmentioned) 8-bit microprocessor acting in between the PDP-11 and any peripherials.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
I still can't find where it says "8085A" in the 110/130 manual, but page
B-3 gives I/O page timings and mentions an "I/O microprocessor" and the
fact that word I/O takes 90 usec but byte I/O takes 180 usec. So that
certainly explains the difference in the DUV driver. Plus the manual
specifically says that RSX11S is available for the /110, so I guess it
was even supported! Sweet.
John Wilson
D Bit
On 2013-01-06 21:34, John Wilson wrote:
From: "Steve Davidson" <jeep at scshome.net>
The main target for the PDT-11 was RT-11. It was slow. The floppies
spent a great deal of time seeking.
Slow, but still a PDP-11! I had a /150 (plus VT62 and LA36) as my main
computer when I was a frosh at RPI ('84-'85) and it was quite a bit above
average compared to other personal computers at the time (C64, Apple ][,
TRS-80, Atari 800, but IBM PCs were still rare -- so mostly fun toys with
pretty graphics but slow and no disk space and usually no 80-column text).
With 60 KB and most of the floppy/TU58 driver up in ROM space its free
memory was at the very high end of MMU-less PDP-11s -- stuff fits in a PDT
that won't run on a "real" LSI-11.
I ran a PDP-8 at home back then. Boy, those RK05 were fast, and stored a lot. :-)
Also, PAL8 was ridiculously fast... MACREL was a bit worse... But still, compared to what most people played with...
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
On 2013-01-06 21:28, John Wilson wrote:
From: Johnny Billquist <bqt at softjar.se>
You must be talking of the PDT-11/150 then. The /110 and /130 sat inside
a VT100 shell. Extremely similar to a VT103 (I actually never figured
out what the difference between a VT103 and a PDT-11/130 is.)
I may be grossly misinformed, but my understanding is that although the
/110 and /130 both live inside a VT100, they're *not* like VT103s, and
have no Q-bus but use a custom CPU board just like the /150 (I doubt it's
the exact same board, but a similar design).
I could have sworn the 110/130 manual on bitsavers (which is misfiled
away from the 150 manual -- I forget where) said so but on a quick skim I
can't find it. The I/O hardware has enough programming differences that
it can't be regular Q-bus stuff, although I suppose they could have made
a whole set of PDT-specific boards.
You know, I had never even thought that thought, but now that you mention it, it makes total sense. I think you must be right.
All the PDT-11 use the same system board. It's not a Qbus at all, in any of them.
That definitely also makes it very different from the VT103.
Thanks for making me understand this. :-)
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
From: <Paul_Koning at Dell.com>
BTW, does anyone know why SIMH doesn't support booting from RC11?
I never miss a chance for a plug: Ersatz-11 has RC11 booting.
(The device name is DC:, so MOUNT/DISMOUNT/BOOT work as usual.)
John Wilson
D Bit
From: "Steve Davidson" <jeep at scshome.net>
The main target for the PDT-11 was RT-11. It was slow. The floppies
spent a great deal of time seeking.
Slow, but still a PDP-11! I had a /150 (plus VT62 and LA36) as my main
computer when I was a frosh at RPI ('84-'85) and it was quite a bit above
average compared to other personal computers at the time (C64, Apple ][,
TRS-80, Atari 800, but IBM PCs were still rare -- so mostly fun toys with
pretty graphics but slow and no disk space and usually no 80-column text).
With 60 KB and most of the floppy/TU58 driver up in ROM space its free
memory was at the very high end of MMU-less PDP-11s -- stuff fits in a PDT
that won't run on a "real" LSI-11.
John Wilson
D Bit
From: Johnny Billquist <bqt at softjar.se>
You must be talking of the PDT-11/150 then. The /110 and /130 sat inside
a VT100 shell. Extremely similar to a VT103 (I actually never figured
out what the difference between a VT103 and a PDT-11/130 is.)
I may be grossly misinformed, but my understanding is that although the
/110 and /130 both live inside a VT100, they're *not* like VT103s, and
have no Q-bus but use a custom CPU board just like the /150 (I doubt it's
the exact same board, but a similar design).
I could have sworn the 110/130 manual on bitsavers (which is misfiled
away from the 150 manual -- I forget where) said so but on a quick skim I
can't find it. The I/O hardware has enough programming differences that
it can't be regular Q-bus stuff, although I suppose they could have made
a whole set of PDT-specific boards.
John Wilson
D Bit
On 2013-01-06 21:22, Paul_Koning at Dell.com wrote:
On Jan 6, 2013, at 2:50 PM, Johnny Billquist wrote:
...
I'm even more curious what a PDT-11 optimization is doing in an RSX
driver...was there at one time an RSX product product planned for the
PDT family?
I'm not at all surprised to learn that RSX ran on the PDT-11.
RSX can run on pretty much any PDP-11, as long as you have atleast around 50K of ram. That's about the lower limit, I'd guess. No other hardware frills or features required
RT requires far less. In college, I ported RT to run on the physics department 11/20, which had 8 kW of memory. It required a port because for a disk it had an RC11, which RT did not support. Fortunately, one of the RT authors had come to work at our college, and he helped me a lot. I asked him "why no RC11 support". Answer: "there weren't any around DEC to test it on".
Yes. RT-11 can run on even smaller things, but RSX can really run on much less than most people believe.
Originally that machine ran DOS-11 V4, ugh. RT was a much better OS. We ran RT Basic on it, with mods I made to support all the lab equipment (DR11-A, AD-01, AD-11, KW11-P). Interrupt handlers written in Basic, fun...
BTW, does anyone know why SIMH doesn't support booting from RC11?
Noone have written the code for it? :-)
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
