23 Nov
2015
23 Nov
'15
7:10 a.m.
This is great news, Noel. Are you on keeping bits
and/or status for these
projects on github or anywhere else that might encourage participation and
contribution? I'd be glad to pitch in if I can. Actually, I (and I bet
others) would love to see an up to date wiki entry or somesuch listing
folks' retro projects like these and would be happy to create it if you'd
provide a couple of pointers.
I have the little code I've written so far in a private git server but
not on github. Eventually the plan is to do something like that but
we're pretty early in the process. Certainly nothing is running yet.
Even though I've set up a git server, I still have to learn how to
actually use git.
However, once we get a prototype doing something interesting, we were
talking about looking around for people interested in helping out.
We'll do a couple disk controllers but if someone wants to add others,
great. Especially if someone wants to add MSCP. We're happy to skip
that one ourselves.
But the real thing we'll want help with is circuit board design and
manufacturing. We can do easy things but I'd really like to use BGA
parts for the FPGA and memory chips so if someone has a handle on that,
we'd like to talk. A I said, our plan was to get a prototype working
before going in search of this kind of assistance. We're using an FPGA
module from ZTEX right now to sidestep the issue.
Is there a wiki out there that holds this kind of information? It seems
like a great idea and I'll put some information together on our project
if there is. Or maybe, now that news is spreading that we're doing
this, it's time to write up a web page describing what we're up to.
-Dave
23 Nov
23 Nov
9:11 a.m.
On Nov 23, 2015, at 10:10 AM, David Bridgham <dab
at froghouse.org> wrote:
...
However, once we get a prototype doing something interesting, we were
talking about looking around for people interested in helping out.
We'll do a couple disk controllers but if someone wants to add others,
great. Especially if someone wants to add MSCP. We're happy to skip
that one ourselves.
I can imagine. MSCP is a large effort.
For a classic/straightforward programming interface, the Massbus disks (RP04 and
successors) are a good choice. That will take you just over 500 MB, if you emulate the
layout of the RP07.
paul
9:17 a.m.
On 11/23/15 9:11 AM, Paul Koning wrote:
That's per-drive. Massbus allows for 8 drives per controller.
While, MSCP is interesting in that it's somewhat drive independent, it's
complex and it only really works with newer OS's.
TTFN - Guy
On Nov 23,
2015, at 10:10 AM, David Bridgham <dab at froghouse.org> wrote:
...
However, once we get a prototype doing something interesting, we were
talking about looking around for people interested in helping out.
We'll do a couple disk controllers but if someone wants to add others,
great. Especially if someone wants to add MSCP. We're happy to skip
that one ourselves.
I can imagine. MSCP is a large effort.
For a classic/straightforward programming interface, the Massbus disks (RP04 and
successors) are a good choice. That will take you just over 500 MB, if you emulate the
layout of the RP07. 
10 a.m.
On 2015-11-23 18:17, Guy Sotomayor wrote:
On 11/23/15 9:11 AM, Paul Koning wrote:
That's per-drive. Massbus allows for 8 drives per controller.
Right. But then you also need to remember that there are some slight
differences between different type of disks, meaning that in DEC
parlance, if you have both an RP06 and an RP07 (for example) on the same
massbus, it's called a mixed massbus, which not all OSes supported.
As far as I can tell, disks fall into two groups, as far as massbus
control is concerned. The RM02, RM03, RM05, RM80 and RP07 is one group.
The RP04, RP05, RP06 is another. A few register addresses between the
groups are the same, but the actual register at that address is
different. But if I remember right, it's registers that have to do with
error recovery, so potentially not something people would care about in
emulation anyway. But it still means there are different drivers in the
OS for them.
And of course, you also have the TM02/TM03 and TM78, which have yet
again different registers on the massbus.
On Nov
23, 2015, at 10:10 AM, David Bridgham <dab at froghouse.org> wrote:
...
However, once we get a prototype doing something interesting, we were
talking about looking around for people interested in helping out.
We'll do a couple disk controllers but if someone wants to add others,
great. Especially if someone wants to add MSCP. We're happy to skip
that one ourselves.
I can imagine. MSCP is a large effort.
For a classic/straightforward programming interface, the Massbus disks
(RP04 and successors) are a good choice. That will take you just over
500 MB, if you emulate the layout of the RP07. While, MSCP is interesting in that it's somewhat
drive independent, it's
complex and it only really works with newer OS's.
Well, "newer" in this case is sortof anything beyond the mid 80s. :-)
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
11:26 a.m.
On Nov 23, 2015, at 1:00 PM, Johnny Billquist <bqt
at Update.UU.SE> wrote:
On 2015-11-23 18:17, Guy Sotomayor wrote:
That sounds right.
RSTS/E supports mixed massbus, and supports RP07. At least in the sense of "it
definitely works". I don't think it shows up as supported in the documentation,
because as far as I remember the RP07 was not actually ever sold as a PDP11 option. But
it works just fine on a fast Massbus (one capable of supporting an RM03 rather than just
an RM02). In the RSTS/E development group, there was an RP07 which I think was used to
hold all the .LST files produced during system build.
On 11/23/15 9:11 AM, Paul Koning wrote:
That's per-drive. Massbus allows for 8 drives per controller.
Right. But then you also need to remember that there are some slight differences between
different type of disks, meaning that in DEC parlance, if you have both an RP06 and an
RP07 (for example) on the same massbus, it's called a mixed massbus, which not all
OSes supported.
As far as I can tell, disks fall into two groups, as far as massbus control is concerned.
The RM02, RM03, RM05, RM80 and RP07 is one group.
The RP04, RP05, RP06 is another. A few register addresses between the groups are the
same, but the actual register at that address is different. But if I remember right,
it's registers that have to do with error recovery, so potentially not something
people would care about in emulation anyway. But it still means there are different
drivers in the OS for them. On Nov
23, 2015, at 10:10 AM, David Bridgham <dab at froghouse.org> wrote:
...
However, once we get a prototype doing something interesting, we were
talking about looking around for people interested in helping out.
We'll do a couple disk controllers but if someone wants to add others,
great. Especially if someone wants to add MSCP. We're happy to skip
that one ourselves.
I can imagine. MSCP is a large effort.
For a classic/straightforward programming interface, the Massbus disks
(RP04 and successors) are a good choice. That will take you just over
500 MB, if you emulate the layout of the RP07.
And of course, you also have the TM02/TM03 and TM78, which have yet again different
registers on the massbus.
Yes. And mixing disk and tape on a massbus is something that I don't think was done
on PDP-11s. It certainly could have been done, and it was on VMS and/or TOPS if I
remember right.
While, MSCP is interesting in that it's
somewhat drive independent, it's
complex and it only really works with newer OS's.
Well, "newer" in this case is sortof anything beyond the mid 80s. :-)
11:37 a.m.
On 2015-11-23 20:26, Paul Koning wrote:
I can't remember if the RP07 shows up as supported on RSX, but I'm
pretty sure at least some documents mention it. Like I said, I did run
one for a few years. And yes, an RH70 is required. You also need to
jumper the RP07 for slow speed (there is documentation inside the RP07
on that, if people actually ever decide to try this).
Oh. It was done. RSX-11M-PLUS (but not 11M) supports it. For disks, I
don't think -11M even supports having RP and RM disks on the same massbus.
But -11M-PLUS can have both tapes and disks on the same massbus just
fine. And it can also deal with devices that are dual-ported on two
different massbuses, as long as they end up on the same system.
RSTS added MSCP support in V8, so if you want to run V7 you'll want to have a
traditional disk emulation.
Yeah. And unless I remember wrong, V8 came in 1986?
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 Nov 23, 2015, at 1:00 PM, Johnny Billquist
<bqt at Update.UU.SE> wrote:
On 2015-11-23 18:17, Guy Sotomayor wrote:
That sounds right.
RSTS/E supports mixed massbus, and supports RP07. At least in the sense of "it
definitely works". I don't think it shows up as supported in the documentation,
because as far as I remember the RP07 was not actually ever sold as a PDP11 option. But
it works just fine on a fast Massbus (one capable of supporting an RM03 rather than just
an RM02). In the RSTS/E development group, there was an RP07 which I think was used to
hold all the .LST files produced during system build.
On 11/23/15 9:11 AM, Paul Koning wrote:
Right. But then you also need to remember that there are some slight differences between
different type of disks, meaning that in DEC parlance, if you have both an RP06 and an
RP07 (for example) on the same massbus, it's called a mixed massbus, which not all
OSes supported.
As far as I can tell, disks fall into two groups, as far as massbus control is concerned.
The RM02, RM03, RM05, RM80 and RP07 is one group.
The RP04, RP05, RP06 is another. A few register addresses between the groups are the
same, but the actual register at that address is different. But if I remember right,
it's registers that have to do with error recovery, so potentially not something
people would care about in emulation anyway. But it still means there are different
drivers in the OS for them. > On Nov 23, 2015, at 10:10 AM, David Bridgham
<dab at froghouse.org> wrote:
>
> ...
> However, once we get a prototype doing something interesting, we were
> talking about looking around for people interested in helping out.
> We'll do a couple disk controllers but if someone wants to add others,
> great. Especially if someone wants to add MSCP. We're happy to skip
> that one ourselves.
I can imagine. MSCP is a large effort.
For a classic/straightforward programming interface, the Massbus disks
(RP04 and successors) are a good choice. That will take you just over
500 MB, if you emulate the layout of the RP07.
That's per-drive. Massbus allows for 8 drives per controller. And of course,
you also have the TM02/TM03 and TM78, which have yet again different registers on the
massbus.
Yes. And mixing disk and tape on a massbus is something that I don't think was done
on PDP-11s. It certainly could have been done, and it was on VMS and/or TOPS if I
remember right. While, MSCP is interesting in that it's somewhat
drive independent, it's
complex and it only really works with newer OS's.
Well, "newer" in this case is sortof anything beyond the mid 80s. :-)
12:02 p.m.
On Nov 23, 2015, at 2:37 PM, Johnny Billquist <bqt
at Update.UU.SE> wrote:
...
RSTS added MSCP support in V8, so if you want to run V7 you'll want to have a
traditional disk emulation.
Yeah. And unless I remember wrong, V8 came in 1986?
No, 1983. I remembered wrong, though; UDA50 support arrived in RSTS 7.2, in 1982.
A nice reference for RSTS history is the wonderfully fictional "RSTS 80th
anniversary" document. A copy of it lives here:
http://www.silverware.co.uk/rsts_80th_birthday.htm . I don't remember who wrote that;
it came out as part of the 20th anniversary celebration, in 1990. In other words, the
entries for 1990 and before are basically accurate; everything after that is a product of
some very nice imagination.
paul
While, MSCP is interesting in that it's somewhat
drive independent, it's
complex and it only really works with newer OS's.
Well, "newer" in this case is sortof anything beyond the mid 80s. :-)
12:26 p.m.
On 2015-11-23 21:02, Paul Koning wrote:
Ouch. I remember when I started playing on a PDP-11, it was an 11/70
running V7.1, and that was in 1982. We upgraded to 7.2 a year later or
so, and I remember we upgraded to 8.0, and that must have been 85 come
to think of it. But I guess we were a bit behind on the upgrades...
I remember that the school upgraded to 9.0 as well, but I had
essentially finished at that school then, which was 86. So 9.0 had to be
around 86-87?
Lots of bit errors in my memories from then. :-)
Things have changed... We were 4 schools sharing one 11/70, and we had
one RM03 as the disk. Normal disk quota was 20 blocks, but if you really
were playing around you could get 100 or 200 blocks of disk quota...
People wouldn't believe me if I said that today.
RUN $MONEY
On Nov 23, 2015, at 2:37 PM, Johnny Billquist
<bqt at Update.UU.SE> wrote:
...
Yeah. And unless I remember wrong, V8 came in 1986?
No, 1983. I remembered wrong, though; UDA50 support arrived in RSTS 7.2, in 1982.
>
While, MSCP is interesting in that it's somewhat drive independent, it's
> complex and it only really works with newer OS's.
Well, "newer" in this case is sortof anything beyond the mid 80s. :-)
RSTS added MSCP support in V8, so if you want to run V7 you'll want to have a
traditional disk emulation. A nice reference for RSTS history is the wonderfully
fictional "RSTS 80th anniversary" document. A copy of it lives here:
http://www.silverware.co.uk/rsts_80th_birthday.htm . I don't remember who wrote that;
it came out as part of the 20th anniversary celebration, in 1990. In other words, the
entries for 1990 and before are basically accurate; everything after that is a product of
some very nice imagination.
I've read it. Fun read indeed.
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
9:55 a.m.
On 2015-11-23 18:11, Paul Koning wrote:
Indeed. But you have code like the stuff in simh, which will tell you
how it is supposed to work.
On Nov 23, 2015, at 10:10 AM, David Bridgham
<dab at froghouse.org> wrote:
...
However, once we get a prototype doing something interesting, we were
talking about looking around for people interested in helping out.
We'll do a couple disk controllers but if someone wants to add others,
great. Especially if someone wants to add MSCP. We're happy to skip
that one ourselves.
I can imagine. MSCP is a large effort. For a classic/straightforward programming interface,
the Massbus disks (RP04 and successors) are a good choice. That will take you just over
500 MB, if you emulate the layout of the RP07.
Unfortunately, support for the RP07 is lacking in most OSes. I don't
know about RSTS/E, but in RSX, there are some odd holes in there,
suggesting that they never used them much in real life. I did get it to
work, but I had to write my own formatting module for FMT, among other
things.
Otherwise it's the RM05 at 256 Meg, if I remember right. But all those
disks are also expected to expose the actual geometry of the disk. While
fairly straight forward, it do expose you to a rather low level
interface, where you need to do a lot of calculations all the time.
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
3317
days inactive
3317
days old
8 comments
4 participants
participants (4)
-
bqt＠update.uu.se -
dab＠froghouse.org -
ggs＠shiresoft.com -
paulkoning＠comcast.net