Re: PDP-11/20 in Iowa (x3)

[Paul Anderson via cctalk]

If you are concerned about hex height boards or mos, you can always use a
BA11-KE expansion box.

On Mon, Mar 27, 2017 at 6:04 PM, Ethan Dicks via cctalk <
cctalk@classiccmp.org> wrote:

> On Mon, Mar 27, 2017 at 12:20 PM, Al Kossow via cctalk
>  wrote:
> > On 3/27/17 7:37 AM, Systems Glitch via cctalk wrote:
> >>> I don't know of MOS memory appearing in 11/20s as built by DEC, but
> there's no reason against it.
> >
> > There were 3rd party cards, Monolithic Systems, for example.
> > The problem with the 11/20 is that hex wide cards don't fit, the fans
> overlap the unibus jumper area.
> > If you ever see hex-wide cards with a big notch in one end, they were
> designed to work in the 11/15 11/20
>
> Exactly.  There are plenty of hex-height cards with MOS memory for
> Unibus (I have several for the 11/04 and/or 11/34 of a suitable size)
> but they would fit in an external BA11-K or BA11-L w/DD11-DK, not the
> CPU's BA11 because of the fans.  That's why I was wondering if DEC
> ever made quad-height MOS modules for Unibus or if all DEC-supplied
> 11/20s and 11/15s had core and only core.
>
> -ethan
>

Unibus/QBus transceivers and drivers - was Re: PDP-11/20 in Iowa (x3)

[Toby Thain via cctalk]

On 2017-03-27 6:46 PM, Ethan Dicks via cctalk wrote:

On Mon, Mar 27, 2017 at 6:35 PM, Steve Malikoff via cctalk
 wrote:

After seeing this modern memory board project for the PDP-8/e using NVRAMMs
http://www.tronola.com/html/ram_for_pdp-8e.html


I have a couple of those I need to solder up.


I wonder how much effort it would be to refactor the Omnibus logic to Unibus
to build a PDP-11 version, and thus have a quad size card for /15's and /20's?


Conceptually, it's not much different for Unibus.  There's at least
one project out there that's working toward this but not yet available
for purchase.

If I was in a real hurry, I could take a quad prototype board and roll
my own.  I happen to have a handful of DS8641s on hand from making
Unibus COMBOARDs 25 years ago.  The quantity of discussion on this
list about modern Unibus transceiver replacements is rather large and
merely mentioning it is likely to cause a flare-up.


It just begs for a conclusive summary. Perhaps those who are actively 
building stuff will eventually provide one (not necessarily on the list 
per se) :-)


--Toby




-ethan

Re: PDP-11/20 in Iowa (x3)

[Ethan Dicks via cctalk]

On Mon, Mar 27, 2017 at 12:20 PM, Al Kossow via cctalk
 wrote:
> On 3/27/17 7:37 AM, Systems Glitch via cctalk wrote:
>>> I don't know of MOS memory appearing in 11/20s as built by DEC, but there's 
>>> no reason against it.
>
> There were 3rd party cards, Monolithic Systems, for example.
> The problem with the 11/20 is that hex wide cards don't fit, the fans overlap 
> the unibus jumper area.
> If you ever see hex-wide cards with a big notch in one end, they were 
> designed to work in the 11/15 11/20

Exactly.  There are plenty of hex-height cards with MOS memory for
Unibus (I have several for the 11/04 and/or 11/34 of a suitable size)
but they would fit in an external BA11-K or BA11-L w/DD11-DK, not the
CPU's BA11 because of the fans.  That's why I was wondering if DEC
ever made quad-height MOS modules for Unibus or if all DEC-supplied
11/20s and 11/15s had core and only core.

-ethan

Re: PDP-11/20 in Iowa (x3)

[Guy Sotomayor Jr via cctalk]

Yes, the MEM11 is about 1/2 assembled at the moment (I just haven’t had the time
to finish it…you know WORK gets in the way).

TTFN - Guy

> On Mar 27, 2017, at 3:46 PM, Ethan Dicks via cctalk  
> wrote:
> 
> On Mon, Mar 27, 2017 at 6:35 PM, Steve Malikoff via cctalk
>  wrote:
>> After seeing this modern memory board project for the PDP-8/e using NVRAMMs
>> http://www.tronola.com/html/ram_for_pdp-8e.html
> 
> I have a couple of those I need to solder up.
> 
>> I wonder how much effort it would be to refactor the Omnibus logic to Unibus
>> to build a PDP-11 version, and thus have a quad size card for /15's and 
>> /20's?
> 
> Conceptually, it's not much different for Unibus.  There's at least
> one project out there that's working toward this but not yet available
> for purchase.
> 
> If I was in a real hurry, I could take a quad prototype board and roll
> my own.  I happen to have a handful of DS8641s on hand from making
> Unibus COMBOARDs 25 years ago.  The quantity of discussion on this
> list about modern Unibus transceiver replacements is rather large and
> merely mentioning it is likely to cause a flare-up.
> 
> -ethan

Re: PDP-11/20 in Iowa (x3)

[Ethan Dicks via cctalk]

On Mon, Mar 27, 2017 at 6:35 PM, Steve Malikoff via cctalk
 wrote:
> After seeing this modern memory board project for the PDP-8/e using NVRAMMs
> http://www.tronola.com/html/ram_for_pdp-8e.html

I have a couple of those I need to solder up.

> I wonder how much effort it would be to refactor the Omnibus logic to Unibus
> to build a PDP-11 version, and thus have a quad size card for /15's and /20's?

Conceptually, it's not much different for Unibus.  There's at least
one project out there that's working toward this but not yet available
for purchase.

If I was in a real hurry, I could take a quad prototype board and roll
my own.  I happen to have a handful of DS8641s on hand from making
Unibus COMBOARDs 25 years ago.  The quantity of discussion on this
list about modern Unibus transceiver replacements is rather large and
merely mentioning it is likely to cause a flare-up.

-ethan

Re: PDP-11/20 in Iowa (x3)

[Steve Malikoff via cctalk]

>> On Mar 27, 2017, at 9:15 AM, Ethan Dicks via cctalk  
>> wrote:
>> Was there ever a quad MOS memory board for the 11/20?  It would have
>> to be 4K or less, I'd expect, given the era and size of RAM.  I know
>> by a few years later, with the 11/04, MOS was becoming common.
>
> I don't know of MOS memory appearing in 11/20s as built by DEC, but there's 
> no reason against it.  It's just Unibus memory.  It wouldn't have to be 4k 
> either; if you had a 32kW memory card that should work just fine.
>
>   paul

After seeing this modern memory board project for the PDP-8/e using NVRAMMs
http://www.tronola.com/html/ram_for_pdp-8e.html
I wonder how much effort it would be to refactor the Omnibus logic to Unibus
to build a PDP-11 version, and thus have a quad size card for /15's and /20's?

Steve.

Re: PDP-11/20 in Iowa (x3)

[Mark Linimon via cctalk]

On Mon, Mar 27, 2017 at 09:15:27AM -0400, Ethan Dicks via cctalk wrote:
> I also saved what C-sized prints I could find, and the rest
> appear to be on Bitsavers.

I was given a pile of them many years ago.  If there is something missing,
let me know.  My project to declutter has stalled at the moment.

I was always hoping to get a machine to go along with the prints :-)  It
was the first "real" machine I learned from both a hardware and software
perspective.  Lo, those many years ago ...

mcl

Re: PDP-11/20 in Iowa (x3)

[ben via cctalk]

On 3/27/2017 3:32 AM, Philipp Hachtmann via cctalk wrote:



On 03/27/2017 08:05 AM, Paul Birkel via cctalk wrote:

http://www.vcfed.org/forum/showthread.php?56897-Came-across-three-PDP-11-20-

units-today


Haha, those low-brain comments!

User g4ugm: "have core memory... lower value"

I can offer high performance pdp11 QBus gear for core memory based
machines, no problem!

User Qbus: "Low performance version"
Year, low performance Of course! But it's no low performance version
of any other pdp11. Perhaps Qbus should put an Univerter in front of his
eyes to clear up his view :-)

The 11/20 without "/20" on the front panel is *THE* pdp11. The first
one. I consider it more valuable than $200-300. Even rusty.

:-)


Well for $200 you build a front panel, stick a PI in there and
run what ever CPU you like. :)
Ben.

Re: PDP-11/20 in Iowa (x3)

[Al Kossow via cctalk]

On 3/27/17 7:37 AM, Systems Glitch via cctalk wrote:
>> I don't know of MOS memory appearing in 11/20s as built by DEC, but there's 
>> no reason against it.

There were 3rd party cards, Monolithic Systems, for example.
The problem with the 11/20 is that hex wide cards don't fit, the fans overlap 
the unibus jumper area.
If you ever see hex-wide cards with a big notch in one end, they were designed 
to work in the 11/15 11/20

Re: PDP-11/20 in Iowa (x3)

[jim stephens via cctalk]

I have replied over on vcfed, but so far not sign that my activities are 
going anywhere.


If someone can help in contacting, I've got the resources to collect the 
system, but so far

no sign of what is going on anywhere but the original posts.
thanks
jim

On 3/27/2017 4:53 AM, Ben Sinclair via cctalk wrote:

I'm only a half hour from Boone. But, I can barely keep my 11/23 running,
so I don't think I would be a good owner for these. I offered to help the
poster if they want, as I don't want to see these scrapped either.

On Mon, Mar 27, 2017 at 1:05 AM, Paul Birkel via cctalk <
cctalk@classiccmp.org> wrote:


http://www.vcfed.org/forum/showthread.php?56897-Came-
across-three-PDP-11-20-
units-today



Need some lovin'.  Not near enough to here .



-

Re: PDP-11/20 in Iowa (x3)

[Systems Glitch via cctalk]

> I don't know of MOS memory appearing in 11/20s as built by DEC, but there's 
> no reason against it.  It's just Unibus memory.  It wouldn't have to be 4k 
> either; if you had a 32kW memory card that should work just fine.

Indeed, Unibus is Unibus, except when it's MUD :)

Thanks,
Jonathan

Re: PDP-11/20 in Iowa (x3)

[Paul Koning via cctalk]

> On Mar 27, 2017, at 9:15 AM, Ethan Dicks via cctalk  
> wrote:
> 
>> ...Haha, those low-brain comments!
>> 
>> User g4ugm: "have core memory... lower value"
> 
> Indeed... I'm still kind of annoyed that the 11/20 I rescued from the
> dumpster at work in 1984 got mildly "souveniered" before I got it - at
> least one core plane went to another employee who wanted a
> wall-hanger.
> 
> Was there ever a quad MOS memory board for the 11/20?  It would have
> to be 4K or less, I'd expect, given the era and size of RAM.  I know
> by a few years later, with the 11/04, MOS was becoming common.

I don't know of MOS memory appearing in 11/20s as built by DEC, but there's no 
reason against it.  It's just Unibus memory.  It wouldn't have to be 4k either; 
if you had a 32kW memory card that should work just fine.

paul

Re: PDP-11/20 in Iowa (x3)

[Ethan Dicks via cctalk]

On Mon, Mar 27, 2017 at 8:59 AM, Philipp Hachtmann via cctalk
 wrote:
> Hm, interesting point of view... The 11/20 is the simplest 11 as far as I
> know. No microcode that can go bad.

Yep.  Simple and spread across multiple cards.

> The core memory systems are quite reliable. And if there's a problem,
> they're not that difficult to fix.

I liken them to OMNIBUS core in terms of repair complexity.  You _can_
fix them, but the analog nature of the devices put some people off.
There are things to learn for certain.

> A pdp11/34, /44 or any QBUS machine can ever be as cool as a 11/20, 11/40,
> 11/45/, /70 and so on.

Agreed, though at least the 11/34 can have a programmer's panel.  Not
as cool as individual lights...

-ethan

Re: PDP-11/20 in Iowa (x3)

[Ethan Dicks via cctalk]

On Mon, Mar 27, 2017 at 5:32 AM, Philipp Hachtmann via cctalk
 wrote:
> On 03/27/2017 08:05 AM, Paul Birkel via cctalk wrote:
>> http://www.vcfed.org/forum/showthread.php?56897-Came-across-three-PDP-11-20-
>> units-today

Nice find, even if they are bit rough.

> Haha, those low-brain comments!
>
> User g4ugm: "have core memory... lower value"

Indeed... I'm still kind of annoyed that the 11/20 I rescued from the
dumpster at work in 1984 got mildly "souveniered" before I got it - at
least one core plane went to another employee who wanted a
wall-hanger.

Was there ever a quad MOS memory board for the 11/20?  It would have
to be 4K or less, I'd expect, given the era and size of RAM.  I know
by a few years later, with the 11/04, MOS was becoming common.
Obviously, despite the lack of supported examples in the configuration
guide, one could put some RAM in a DD11-CK or DD11-DK and hang it off
of the CPU chassis (as long as module order and grant priority isn't
violated) but I wonder if anyone _did_ that.  I would think the cost
of a 4K or larger MOS board and the relative cost of PDP-11s in
general in the mid-1970s, it would have been better to have just
retired the 11/20 and gone with a single-cabinet machine like an 11/04
or 11/34, so it probably would have been some college lab experiment
to use parts on hand rather than a commercial endeavor.  Obviously,
it's easy enough to make a modern quad-height Unibus memory board and
bypass debugging core problems, but that's back to the functional vs
"authentic" argument (I prefer authentic except where impractical or
unobtainable, like trying to find an RF11/RS11).

> The 11/20 without "/20" on the front panel is *THE* pdp11. The first one. I
> consider it more valuable than $200-300. Even rusty.

Right.  I'm sad mine says "/20" but only a little.  The real trick is
going to be buying a crate of Boxer fans and finding enough power to
run this thing.  When it was dumpstered, my old boss stripped the fans
and PSUs and tossed the rest out (which I saved).  So I have 3 BA11
boxes (including the intact front panel), all the backplanes, nearly
all the boards, but in the form of a very large and very heavy jigsaw
puzzle.  I also saved what C-sized prints I could find, and the rest
appear to be on Bitsavers.

I'd be interested in talking with whoever ends up with this haul (or
with anyone else who has an 11/20, really) to get some specific
pictures of the power wiring and fans so I can put mine back together.
Also, if anyone knows where to get a crate of 110V Boxer fans for
around $5/each, let me know (I see them on a regular basis for
$15-$20/unit, but I need at least 9).  Used at that price is fine, or
else I'll just have to pay full tick for new.

-ethan

Re: PDP-11/20 in Iowa (x3)

[JP Hindin via cctalk]

On Mon, 27 Mar 2017, Ben Sinclair via cctalk wrote:

I'm only a half hour from Boone. But, I can barely keep my 11/23 running,
so I don't think I would be a good owner for these. I offered to help the
poster if they want, as I don't want to see these scrapped either.


Hey Ben;

Apologies for replying direct, but I've got a two-fold query - firstly, I 
don't have a VCF account, and I've never felt the need to get one before - 
is there any chance you know of a way of contacting the original poster 
off the board? If not, I'll figure something out.


The second was I'm also in Iowa - I'm up in the Hampton area. I didn't 
realise there was anyone else around here (although the law of averages 
said there must).


Mostly wanted to say Hi to another classic machine owner in the area.

Bring on Spring, eh?

 - JP


On Mon, Mar 27, 2017 at 1:05 AM, Paul Birkel via cctalk <
cctalk@classiccmp.org> wrote:


http://www.vcfed.org/forum/showthread.php?56897-Came-
across-three-PDP-11-20-
units-today



Need some lovin'.  Not near enough to here .



-





--
Ben Sinclair
b...@bensinclair.com

Re: PDP-11/20 in Iowa (x3)

[Philipp Hachtmann via cctalk]

On 03/27/2017 12:03 PM, dave.g4...@gmail.com wrote:


On 03/27/2017 08:05 AM, Paul Birkel via cctalk wrote:

http://www.vcfed.org/forum/showthread.php?56897-Came-across-three-

PDP-

11-20-
units-today


Haha, those low-brain comments!

User g4ugm: "have core memory... lower value"




I have the impression that 11's with core in are harder to fix, so the pool
of buyers are smaller...


Hm, interesting point of view... The 11/20 is the simplest 11 as far as 
I know. No microcode that can go bad.
The core memory systems are quite reliable. And if there's a problem, 
they're not that difficult to fix.


For me it's important that the computer has core memory: It's a big part 
of the fun!


The sc memory systems are too close to modern hardware. Without 
blinkenlight panels or at least LEDs instead of incandescent bulbs.


Not to forget: More vintage -> More cool ( -> higher value)

A pdp11/34, /44 or any QBUS machine can ever be as cool as a 11/20, 
11/40, 11/45/, /70 and so on.
I somewhere have an 11/45 with full core memory (something over 200k). 
But the CPU has issues... :-(

Re: PDP-11/20 in Iowa (x3)

[Ben Sinclair via cctalk]

I'm only a half hour from Boone. But, I can barely keep my 11/23 running,
so I don't think I would be a good owner for these. I offered to help the
poster if they want, as I don't want to see these scrapped either.

On Mon, Mar 27, 2017 at 1:05 AM, Paul Birkel via cctalk <
cctalk@classiccmp.org> wrote:

> http://www.vcfed.org/forum/showthread.php?56897-Came-
> across-three-PDP-11-20-
> units-today
>
>
>
> Need some lovin'.  Not near enough to here .
>
>
>
> -
>
>


-- 
Ben Sinclair
b...@bensinclair.com

RE: PDP-11/20 in Iowa (x3)

[Dave Wade G4UGM via cctalk]

> 
> 
> 
> On 03/27/2017 08:05 AM, Paul Birkel via cctalk wrote:
> > http://www.vcfed.org/forum/showthread.php?56897-Came-across-three-
> PDP-
> > 11-20-
> > units-today
> 
> Haha, those low-brain comments!
> 
> User g4ugm: "have core memory... lower value"
> 


I have the impression that 11's with core in are harder to fix, so the pool
of buyers are smaller...


> I can offer high performance pdp11 QBus gear for core memory based
> machines, no problem!

Dave
G4UGM

Re: PDP-11/20 in Iowa (x3)

[Philipp Hachtmann via cctalk]

On 03/27/2017 08:05 AM, Paul Birkel via cctalk wrote:

http://www.vcfed.org/forum/showthread.php?56897-Came-across-three-PDP-11-20-
units-today


Haha, those low-brain comments!

User g4ugm: "have core memory... lower value"

I can offer high performance pdp11 QBus gear for core memory based 
machines, no problem!


User Qbus: "Low performance version"
Year, low performance Of course! But it's no low performance version 
of any other pdp11. Perhaps Qbus should put an Univerter in front of his 
eyes to clear up his view :-)


The 11/20 without "/20" on the front panel is *THE* pdp11. The first 
one. I consider it more valuable than $200-300. Even rusty.


:-)

Re: PDP-11/20 in Iowa (x3)

[Mark Linimon via cctalk]

On Mon, Mar 27, 2017 at 02:05:56AM -0400, Paul Birkel via cctalk wrote:
> Need some lovin'.  Not near enough to here .

I would go to Iowa.  Hell, it's not *that* far from Texas.

(OTOH I could only take one of them.)

Guess I'll have to sign up in the a.m. so I can see the pictures.

mcl

Re: PDP-11/20 in Iowa (x3)

[jim stephens via cctalk]

On 3/26/2017 11:50 PM, Paul Anderson via cctalk wrote:

I'm close and interested, but it won't let me reply
If you have a vcfed account private message the original poster.  I also 
replied have close in as well, and am very interested.


I replied on the thread, but it is moderated for members like me, I 
guess (don't know the rules on vcfed).  but the PM seemed to go

direct to the OP.

I suspect there is a lot of interest.
thanks
Jim

Re: PDP-11/20 in Iowa (x3)

[Paul Anderson via cctalk]

I'm close and interested, but it won't let me reply

On Mon, Mar 27, 2017 at 1:05 AM, Paul Birkel via cctalk <
cctalk@classiccmp.org> wrote:

> http://www.vcfed.org/forum/showthread.php?56897-Came-
> across-three-PDP-11-20-
> units-today
>
>
>
> Need some lovin'.  Not near enough to here .
>
>
>
> -
>
>