Q-bus or Unibus bus transactions in FPGA?

[Please do not mail me a copy of your followup]

Has anyone implemented a Q-bus or Unibus bus interface logic in an
FPGA that is freely available or documented?  I am interested in
making a board for an LSI-11 computer that looks to the LSI-11 like a
disk drive, but is really a communication circuit back to a PC that
serves the drive image from a file on the PC.

Any ideas on implementation strategies would be appreciated!
-- 
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
          <http://www.xmission.com/~legalize/book/>
	    Pilgrimage: Utah's annual demoparty
               <http://pilgrimage.scene.org>

Richard wrote:

>[Please do not mail me a copy of your followup]
>
>Has anyone implemented a Q-bus or Unibus bus interface logic in an
>FPGA that is freely available or documented?  I am interested in
>making a board for an LSI-11 computer that looks to the LSI-11 like a
>disk drive, but is really a communication circuit back to a PC that
>serves the drive image from a file on the PC.
>  
>
Not on an FPGA (PDP-11 was before their time) but the bus protocol
is quite simple.  Unibus has separate data and address, Q-bus has
multiplexed A/D.  The master sends out an address and a transfer
request, and when the target is done it sends an acknowledge.  It
is all async, there is no bus clock.

Jon

[Please do not mail me a copy of your followup]

Jon Elson <jmelson@artsci.wustl.edu> spake the secret code
<438CE854.8010104@artsci.wustl.edu> thusly:

>Richard wrote:
>>
>>Has anyone implemented a Q-bus or Unibus bus interface logic in an
>>FPGA that is freely available or documented? [...]
>>
>Not on an FPGA (PDP-11 was before their time) but the bus protocol
>is quite simple. [...]

Well, I didn't mean to imply that the FPGA was concurrent with the
PDP-11 :-), I was more hoping that another retro computing hobbyist
would have made something I could bootleg! 

I suppose I'll have to do my own bus handshake implementation from the
Q-bus docs (I think my processor or peripheral handbook that came with
the 11/03 has one in there somewhere).

I wonder how hard it would be to get a PCB fabbed with the bus edge
connector?
-- 
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
          <http://www.xmission.com/~legalize/book/>
	    Pilgrimage: Utah's annual demoparty
               <http://pilgrimage.scene.org>

I did one in a Xilinx XC3064... many, many years ago.
Unfortunately, I ditched all the documentation a few years ago... and have 
done a complete brain purge.  Might still have the DEC documentation, 
though.  I'll check on Wednesday.

It wasn't too hard of a bus to interface to and is quite slow.

Good luck,
Ed



"Richard" <legalize+jeeves@mail.xmission.com> wrote in message 
news:dmiiik$vqn$2@news.xmission.com...
> [Please do not mail me a copy of your followup]
>
> Has anyone implemented a Q-bus or Unibus bus interface logic in an
> FPGA that is freely available or documented?  I am interested in
> making a board for an LSI-11 computer that looks to the LSI-11 like a
> disk drive, but is really a communication circuit back to a PC that
> serves the drive image from a file on the PC.
>
> Any ideas on implementation strategies would be appreciated!
> -- 
> "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
>          <http://www.xmission.com/~legalize/book/>
>     Pilgrimage: Utah's annual demoparty
>               <http://pilgrimage.scene.org> 

>I suppose I'll have to do my own bus handshake implementation from the
>Q-bus docs (I think my processor or peripheral handbook that came with
>the 11/03 has one in there somewhere).

Does it have the specs for the bus transcievers?  I remember using
some special DEC chip.


>I wonder how hard it would be to get a PCB fabbed with the bus edge
>connector?

Needs gold fingers.  They are big relative to modern connectors.
I don't know if thats good or bad - easier to make but takes more gold.

Anybody know where to get boards with gold fingers at hobbyist prices?
I keep scheming about making a few PCI cards. 

-- 
The suespammers.org mail server is located in California.  So are all my
other mailboxes.  Please do not send unsolicited bulk e-mail or unsolicited
commercial e-mail to my suespammers.org address or any of my other addresses.
These are my opinions, not necessarily my employer's.  I hate spam.

"Hal Murray" <hmurray@suespammers.org> wrote in message 
news:vaadnSJVHo9ezBDeRVn-hQ@megapath.net...
>
>>I suppose I'll have to do my own bus handshake implementation from the
>>Q-bus docs (I think my processor or peripheral handbook that came with
>>the 11/03 has one in there somewhere).
>
> Does it have the specs for the bus transcievers?  I remember using
> some special DEC chip.

DEC did make some interface IC's for Qbus ... but *IF* I remember right... 
these were easily replaced with standard TTL logic.

>
>
>>I wonder how hard it would be to get a PCB fabbed with the bus edge
>>connector?
>
> Needs gold fingers.  They are big relative to modern connectors.
> I don't know if thats good or bad - easier to make but takes more gold.
>
> Anybody know where to get boards with gold fingers at hobbyist prices?
> I keep scheming about making a few PCI cards.
>
> -- 
> The suespammers.org mail server is located in California.  So are all my
> other mailboxes.  Please do not send unsolicited bulk e-mail or 
> unsolicited
> commercial e-mail to my suespammers.org address or any of my other 
> addresses.
> These are my opinions, not necessarily my employer's.  I hate spam.
> 

GPE wrote:
> I did one in a Xilinx XC3064... many, many years ago.
> Unfortunately, I ditched all the documentation a few years ago... and have
> done a complete brain purge.  Might still have the DEC documentation,
> though.  I'll check on Wednesday.
>
> It wasn't too hard of a bus to interface to and is quite slow.
>
> Good luck,
> Ed
>

I did interfaces many years ago for both buses using PAL's for logic
and
74F-series parts for bus drive.  Probably 5V 74FCT parts would also
work.
If you're serious about Unibus, you'll need to know that the connector
pin-out
in the DEC documentation is for the bus extender cable and not where a
board plugs in.  A and B connectors are not used for plug-in cards,
only
C, D, E and F.  I got a copy of the magic document from someone who
got it from someone who did PC board work for DEC.

I don't have any of my documentation now except for the Unibus pinout.
You'll
need a copy of the board mechanicals.  For Qbus there used to be
prototyping
breadboards available with holes on 0.1 x 0.1" centers.  If you check
with some
hobby places like Jameco there may still be such things available if
you want
to breadboard your design or just lay out a "daughtercard" to wire down
to the
breadboard.

Good Luck,
Gabor

Richard wrote:
> [Please do not mail me a copy of your followup]
>
> Jon Elson <jmelson@artsci.wustl.edu> spake the secret code
> <438CE854.8010104@artsci.wustl.edu> thusly:
>
> >Richard wrote:
> >>
> >>Has anyone implemented a Q-bus or Unibus bus interface logic in an
> >>FPGA that is freely available or documented? [...]
> >>
> >Not on an FPGA (PDP-11 was before their time) but the bus protocol
> >is quite simple. [...]
>
> Well, I didn't mean to imply that the FPGA was concurrent with the
> PDP-11 :-), I was more hoping that another retro computing hobbyist
> would have made something I could bootleg!
>
> I suppose I'll have to do my own bus handshake implementation from the
> Q-bus docs (I think my processor or peripheral handbook that came with
> the 11/03 has one in there somewhere).
>
> I wonder how hard it would be to get a PCB fabbed with the bus edge
> connector?

There was a discussion of this on S.E.D. quite recently with one of
it's denizens successfully getting  a board with gold fingers
fabricated at a very reasonable price.

X-Posted to s.e.d. for comments

Cheers

PeteS


> --
> "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
>           <http://www.xmission.com/~legalize/book/>
> 	    Pilgrimage: Utah's annual demoparty
>                <http://pilgrimage.scene.org>

[Please do not mail me a copy of your followup]

hmurray@suespammers.org (Hal Murray) spake the secret code
<vaadnSJVHo9ezBDeRVn-hQ@megapath.net> thusly:

>>I suppose I'll have to do my own bus handshake implementation from the
>>Q-bus docs (I think my processor or peripheral handbook that came with
>>the 11/03 has one in there somewhere).
>
>Does it have the specs for the bus transcievers?  I remember using
>some special DEC chip.

I do recall seeing the specs for the DEC bus trainceiver chip in the
handbook, yes.
-- 
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
          <http://www.xmission.com/~legalize/book/>
	    Pilgrimage: Utah's annual demoparty
               <http://pilgrimage.scene.org>

PeteS posted:

>Newsgroups: comp.arch.fpga, sci.electronics.design
>From: "PeteS" <p...@fleetwoodmobile.com> - Find messages by this >author
>Date: 30 Nov 2005 07:18:19 -0800
>Local: Wed, Nov 30 2005 8:18 am
>Subject: Re: Q-bus or Unibus bus transactions in FPGA?
>Reply | Reply to Author | Forward | Print | Individual Message | Show >original | Report Abuse

Richard wrote:
> [Please do not mail me a copy of your followup]

> Jon Elson <jmel...@artsci.wustl.edu> spake the secret code
> <438CE854.8010...@artsci.wustl.edu> thusly:


> >Richard wrote:

> >>Has anyone implemented a Q-bus or Unibus bus interface logic in an
> >>FPGA that is freely available or documented? [...]

> >Not on an FPGA (PDP-11 was before their time) but the bus protocol
> >is quite simple. [...]

> Well, I didn't mean to imply that the FPGA was concurrent with the
> PDP-11 :-), I was more hoping that another retro computing hobbyist
> would have made something I could bootleg!

> I suppose I'll have to do my own bus handshake implementation from the
> Q-bus docs (I think my processor or peripheral handbook that came with
> the 11/03 has one in there somewhere).

> I wonder how hard it would be to get a PCB fabbed with the bus edge
> connector?

PeteS wrote:

>There was a discussion of this on S.E.D. quite recently with one of
>it's denizens successfully getting  a board with gold fingers
>fabricated at a very reasonable price.

>X-Posted to s.e.d. for comments

>Cheers

>PeteS


I designed a Q-bus interface card about 16 years ago using a CPLD for
decoding logic and another CPLD for board functions -- both written in
ABEL.  I used DEC's ChipKit for logic buffering to the bus itself.  The
physical dimensions of the board fingers came from one of DEC's
manuals.

Alas, it was 16 years ago and 2 companies ago, so I don't have much to
offer in the way of code.  But, it has been done before, so it can be
done again!  As for the logic interface, there are so many logic
families that modern FPGA's I/O can be programmed for, I'm sure
something will match the characteristics of the Q-bus (Unibus), then
you don't need the ChipKit.

Cheers
Tom