6 Aug
2008
6 Aug
'08
4:53 p.m.
General question for PDP-8 fans: I've seen references to someone having
created an RF08/RS08 replacement. Does anyone on the list know details
of such a project?
Thanks,
Rich
Rich Alderson RichA at vulcan.com
Server Engineer, PDPplanet Project (206) 342-2239
Vulcan, Inc., 505 5th Avenue S, Seattle, WA 98104 (206) 465-2916
cell
7 Aug
7 Aug
1:26 a.m.
From: "Rich Alderson"
General question for PDP-8 fans: I've seen
references to someone having
created an RF08/RS08 replacement. Does anyone on the list know details
of such a project?
I have a project like that. You can find Eagle drawings at
http://www.so-much-stuff.com/pdp8/cadlib.html.
Look for "RF08: RF08 Replacement" in the green scrolling region.
The idea was to order an 8"x11" panel containing a controller, two
drive pairs, and a light panel. The "drive pairs" are implemented
with pairs of SRAM chips, which taken together form a removable
512Kw battery backed store. Two such stores provide the full 1Mw
of storage. The 40 pin headers take BC08J cables or the equivalent,
to connect the controller to a Posibus PDP-8.
The design is my rework of C. Morris' DF32 replacement (which I
also now own the prototype of).
Like many of my projects, it was fun to think about the design,
but I haven't yet made the time to actually build and debug it.
Vince
N.B.: The 74LS151's in the write-protect section need to be real
TTL, and float their inputs high without pull-ups. Of course, if
you don't care about separate write-protect switches for each 32K,
you can just hard wire the inputs to the right values :-).
2:45 p.m.
From: Vincent Slyngstad
Sent: Wednesday, August 06, 2008 10:26 PM
From: "Rich Alderson"
> General question for PDP-8 fans: I've seen references to someone
having
> created an RF08/RS08 replacement. Does anyone on
the list know
details
> of such a project?
I have a project like that. You can find Eagle
drawings at
http://www.so-much-stuff.com/pdp8/cadlib.html.
Look for "RF08: RF08 Replacement" in the green scrolling region.
Thanks, Vince! That's the kind of thing I was looking for!
Rich
4:54 p.m.
From: "Rich Alderson" <RichA at vulcan.com>
Sent: Thursday, August 07, 2008 11:45 AM
From: Vincent Slyngstad
Sent: Wednesday, August 06, 2008 10:26 PM
Feel free to mail me offlist if you decide to pursue a project
like this. I'd love to hear how it goes!
Vince
I have a project like that. You can find Eagle
drawings at
http://www.so-much-stuff.com/pdp8/cadlib.html.
Look for "RF08: RF08 Replacement" in the green scrolling region.
Thanks, Vince! That's the kind of thing I was looking for! 
4:19 p.m.
On Wed, Aug 06, 2008 at 10:26:08PM -0700, Vincent Slyngstad wrote:
From: "Rich Alderson"
That sounds like a clever way to keep the costs per unit down.
I even have a few larger SRAMs (for a SBC6120 RAM disk board),
so if this project ever gets past the design phase, I'm past
the hump for the expensive components.
General question for PDP-8 fans: I've seen
references to someone having
created an RF08/RS08 replacement. Does anyone on the list know details
of such a project?
I have a project like that. You can find Eagle drawings at
http://www.so-much-stuff.com/pdp8/cadlib.html.
Look for "RF08: RF08 Replacement" in the green scrolling region.
The idea was to order an 8"x11" panel containing a controller, two
drive pairs, and a light panel. The "drive pairs" are implemented
with pairs of SRAM chips, which taken together form a removable
512Kw battery backed store. Two such stores provide the full 1Mw
of storage. Like many of my projects, it was fun to think about
the design,
but I haven't yet made the time to actually build and debug it.
One of these days, like when I'm home for more than six months at
a shot, I really want to shake down one of my Posibus boxes (i.e.,
watch it pass the entire suite of paper-tape diagnostics) and
hang some interesting hardware off of it.
Ever since I got that first PDP-8/L when I was in high-school, I've
been trying to get a OS/8-capable configuration. I've gotten as
far as an -8/L w/12kW (via a BM08) and PC8L high-speed papertape,
but no posibus mass storage. My negibus -8/i has 64kW of DF32,
but only 4kW of core (another long-term project is to get that up
to 8kW internally).
N.B.: The 74LS151's in the write-protect section
need to be real
TTL, and float their inputs high without pull-ups. Of course, if
you don't care about separate write-protect switches for each 32K,
you can just hard wire the inputs to the right values :-).
I don't remember seeing too many designs with floating TTL inputs.
Why no pullups? What's the advantage?
-ethan
--
Ethan Dicks, A-333-S Current South Pole Weather at 7-Aug-2008 at 20:10 Z
South Pole Station
PSC 468 Box 400 Temp -47.0 F (-43.9 C) Windchill -90.1 F (-67.8 C)
APO AP 96598 Wind 26.7 kts Grid 9 Barometer 681.3 mb (10579 ft)
Ethan.Dicks at usap.gov http://penguincentral.com/penguincentral.html
5:27 p.m.
From: "Ethan Dicks"
Ever since I got that first PDP-8/L when I was in
high-school, I've
been trying to get a OS/8-capable configuration. I've gotten as
far as an -8/L w/12kW (via a BM08) and PC8L high-speed papertape,
but no posibus mass storage. My negibus -8/i has 64kW of DF32,
but only 4kW of core (another long-term project is to get that up
to 8kW internally).
There are some bus converter projects there too, if those would be
of any help.
On Wed, Aug 06, 2008 at 10:26:08PM -0700, Vincent
Slyngstad wrote:
No particular reason, but the board's pretty crowded trying to get
it all to fit in 8"x11". I'd probably use SIL resistor packs and
shift down the six DIPs below the switches. (I'm not sure there'd
be enough void left to route the thing with 10mil design rules.)
There's some other "fast and loose" stuff in there too -- S240 as
bus drivers, multiple LS loads on the bus (still should be less
than a 74xx load), and (importantly) no provision to continue the
bus, so that it must be placed last. There's also no equivalent
for the G717 terminator card, so it may need to be a fairly short
bus!
Vince
N.B.: The 74LS151's in the write-protect
section need to be real
TTL, and float their inputs high without pull-ups. Of course, if
you don't care about separate write-protect switches for each 32K,
you can just hard wire the inputs to the right values :-).
I don't remember seeing too many designs with floating TTL inputs.
Why no pullups? What's the advantage?
5:51 p.m.
On Thu, Aug 07, 2008 at 02:27:19PM -0700, Vincent Slyngstad wrote:
From: "Ethan Dicks"
Sure, if it comes to that. The -8/i is out of the picture until I
get enough memory on it anyway. I have enough M-series logic to
stuff the backplane with the memory extension logic (somewhere, I
even have an M720), and I think I have enough spare G-series
logic from an -8/L that came to me as a crate of parts (broken
chassis, no front-panel, a few broken backplane pins, no core...),
or nearly enough to implement 8kW, but 32kW sure would be nice for
doing any "real" work (like playing Adventure ;-)
My main -8/L is in closer shape than anything, though if I fix
whatever is wrong with my -8/e, I do have a posibus adapter for
that, and it might be a more reliable testbed for new devices,
especially since I can easily throw an RX8E in there (borrowed
Ever since I got that first PDP-8/L when I was in
high-school, I've
been trying to get a OS/8-capable configuration. I've gotten as
far as an -8/L w/12kW (via a BM08) and PC8L high-speed papertape,
but no posibus mass storage. My negibus -8/i has 64kW of DF32,
but only 4kW of core (another long-term project is to get that up
to 8kW internally).
There are some bus converter projects there too, if those would be
of any help. from a PDP-8/a-based WPS box in the same room). I
suppose I
could just use the -8/a and the posibus adapter, but that, too,
has some odd problem (with the KM8A, IIRC).
Sigh... all of this stuff worked once, but it's been three
moves and a lot of years, and I haven't been able to spend
enough time in front of the machines to give them the attention
they deserve.
No particular reason, but the board's pretty
crowded trying to get
it all to fit in 8"x11". I'd probably use SIL resistor packs and
shift down the six DIPs below the switches. (I'm not sure there'd
be enough void left to route the thing with 10mil design rules.)
Sounds crowded.
There's some other "fast and loose"
stuff in there too -- S240 as
bus drivers, multiple LS loads on the bus (still should be less
than a 74xx load), and (importantly) no provision to continue the
bus, so that it must be placed last. There's also no equivalent
for the G717 terminator card, so it may need to be a fairly short
bus!
Well... while that's not optimal, it's workable. Given the dearth
of OEM posibus equipment these days (I have none and have never
seen any except on the web), it's not like many people are going to
be sticking this on a heavily loaded bus. The RF08 is large enough
to do interesting things with, leaving only removable media to get
things on and off the system - paper tape is certainly one way,
but floppies are more convenient... in my particular case, I suppose
the largest configuration I'd attempt would be a modern solid-state
RF08 and that RX8 you/we started and haven't gotten back to. It's
been so long that I can't remember if the RX8 can be in the middle
of the posibus or not.
-ethan
--
Ethan Dicks, A-333-S Current South Pole Weather at 7-Aug-2008 at 21:30 Z
South Pole Station
PSC 468 Box 400 Temp -47.4 F (-44.1 C) Windchill -86.9 F (-66.1 C)
APO AP 96598 Wind 21.0 kts Grid 11 Barometer 682.0 mb (10553 ft)
Ethan.Dicks at usap.gov http://penguincentral.com/penguincentral.html
6:29 p.m.
From: "Ethan Dicks"
The RX8 will go in the middle of a bus, and even has a provision for a
G717 equivalent bus terminator, in the form of an optional SIL resistor
pack. Real drivers, too. Of course, being known not to work as yet,
I doubt anyone will be rushing out to build it :-(.
Vince
On Thu, Aug 07, 2008 at 02:27:19PM -0700, Vincent
Slyngstad wrote:
It is a little. The 8"x11" restriction is really just to keep the
prototype cost low for a particular board shop. If that or the
10mil restriction were relaxed, things would get easier.
Of course it would also be fairly easy to redo the thing with
GALs instead of TTL, which should also make it easier to fit
things in, but would require me to program parts.
No particular reason, but the board's pretty
crowded trying to get
it all to fit in 8"x11". I'd probably use SIL resistor packs and
shift down the six DIPs below the switches. (I'm not sure there'd
be enough void left to route the thing with 10mil design rules.)
Sounds crowded. There's
some other "fast and loose" stuff in there too -- S240 as
bus drivers, multiple LS loads on the bus (still should be less
than a 74xx load), and (importantly) no provision to continue the
bus, so that it must be placed last. There's also no equivalent
for the G717 terminator card, so it may need to be a fairly short
bus!
Well... while that's not optimal, it's workable. Given the dearth
of OEM posibus equipment these days (I have none and have never
seen any except on the web), it's not like many people are going to
be sticking this on a heavily loaded bus. The RF08 is large enough
to do interesting things with, leaving only removable media to get
things on and off the system - paper tape is certainly one way,
but floppies are more convenient... in my particular case, I suppose
the largest configuration I'd attempt would be a modern solid-state
RF08 and that RX8 you/we started and haven't gotten back to. It's
been so long that I can't remember if the RX8 can be in the middle
of the posibus or not.
8 Aug
8 Aug
12:25 a.m.
On Thu, Aug 07, 2008 at 03:29:01PM -0700, Vincent Slyngstad wrote:
It is a little. The 8"x11" restriction is
really just to keep the
prototype cost low for a particular board shop. If that or the
10mil restriction were relaxed, things would get easier.
Gotcha.
Of course it would also be fairly easy to redo the
thing with
GALs instead of TTL, which should also make it easier to fit
things in, but would require me to program parts.
Sure. I myself have a couple of burners and more than a handful
of 16V8s and 22V10s, so _I_ wouldn't mind a GAL-inclusive design,
but there is something neat about an all-TTL design replacing
and upgrading a FLIP-CHIP disk - you can get a probe in there and
see *exactly* what's going on.
I forget... does a real RF08 have a KM11 port? I happen to
have a couple of as-yet-unassembled modern replicas, but I
got them to plug into my RK-11C and PDP-11/20 (and probably
an RX01 just to watch it blink), so they'll get use, but I
just don't remember about the RF08.
The RX8 will go in the middle of a bus, and even has a provision for a
G717 equivalent bus terminator, in the form of an optional SIL resistor
pack.
Thanks for the reminder - sounds like a very workable combination.
Oh... the RF08 is a databreak device - I think that would mean that
I'd need one more module for OMNIBUS testing, but I do have the
required M-series logic to fit that into the -8/L.
Real drivers, too. Of course, being known not to
work as yet,
I doubt anyone will be rushing out to build it :-(.
Right... I _am_ still interested in getting it working, but to be quite
honest, it'll be 6 months until I could even start to look at it - gotta
get home first.
-ethan
--
Ethan Dicks, A-333-S Current South Pole Weather at 8-Aug-2008 at 04:20 Z
South Pole Station
PSC 468 Box 400 Temp -50.4 F (-45.8 C) Windchill -88.3 F (-66.8 C)
APO AP 96598 Wind 17.5 kts Grid 13 Barometer 682.5 mb (10534 ft)
Ethan.Dicks at usap.gov http://penguincentral.com/penguincentral.html
5776
days inactive
5778
days old
8 comments
3 participants
participants (3)
-
ethan.dicks＠usap.gov -
RichA＠vulcan.com -
vrs＠msn.com