Re: [Simh] VAX PSI

[Ethan Dicks]

On Fri, Mar 20, 2020 at 8:05 AM Peter Allan  wrote:
> Mention of VAX-2780-3780PE piques my interest. I have a long term desire to 
> be able to connect a simulated VAX (or even better, a PDP-11) to a simulated 
> IBM mainframe (using Hercules) and use the VAX/PDP11 as a remote job entry 
> system for the IBM. Has anyone tried to do this?

I used to do it for real with real hardware (MC68000-based serial
board with sync interface to IBM side and DMA interface to Unibus,
Qbus, or VAXBI machine on host side plus application software).  I
have not tried to do anything similar in the world of emulation.

We used to use the COM5025 or the Zilog Z8530 in sync mode for serial
comms.  This is definitely an interesting aspect of the process to
emulate.

-ethan
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] VAX PSI

[Ethan Dicks]

On Thu, Mar 19, 2020 at 8:16 PM Paul Koning  wrote:
> I thought 2780 uses BISYNC, which is a different protocol.  X.25 is HDLC.

Yes.  You are correct.

I used to make and sell COMBOARD 2780/3780, HASP and SNA communication
boards for Qbus, Unibus and VAXBI.

2780/3780 is Bisync not HDLC.

-ethan
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] VAX PSI

[Robert G. Schaffrath]

At my firm, we used VAX 2780-3780 to perform dialup 2780 BISYNC 
emulation to a remote data provider (1989-1992 time-frame - 
pre-internet). I wrote a program that would send the dialing 
instructions to the modem over the X.25 port and then wait for the 
carrier signal to be detected. It would then switch to BISYNC and the 
data would start downloading. One thing that was very annoying about the 
2780-3780 software was a bug I hit in either the VAXBI X.25 port driver 
(name of the BI card escapes me) or the VAX 2780-3780 software. We were 
running on a VAX 6330 and anytime I shut the system down, as soon as the 
VAX 2780-3780 daemon process was terminated, the system would BUGCHECK 
on a SPINLOCK issue. I solved the issue by doing some kernel magic and 
marking the process NODELETE early in the shutdown phase. As such, the 
process stayed alive until VMS itself terminated and the BUGCHECK's went 
away.


We started migrating to HP/UX in May 1992 and it was never a priority to 
have DEC fix it. The workaround worked and that was fine with me.


___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] VAX PSI

[dave porter]

Paul Koning wrote:

X.25 is indeed asymmetrical.  Whether that can be gotten around is not 
entirely clear to me.


Well, we did :-)

For X25-11M (Datapac, Transpac, as I recall) it involved using console 
switches to jam certain status bits into a certain lowcore word in order to 
convince the software it had initialized  :-)   For its successor, RSX-11M 
PSI (or whatever its name was) there was some slightly cleaner way to do it.


This might well have been suitable for testing only, in that you promised 
not to originate calls that would collide. Or maybe the magic was to change 
address allocation for the fake-DCE end.  Them brain cells have long since 
vanished, sorry.



___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] VAX PSI

[dave porter]

I have been involved in a project using VAX-2780-3780PE and have been able 
to connect two VAX’s together using a Black Box Synchronous Modem 
Eliminator ( SME ).


Sure, but the point I was trying to make was that if your goal is to 
specifically use VAX PSI then you may be able to operate back-to-back even 
though that is not the focus of the VAX PSI product.


___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] Mouse capture on VAXstation II/GPX (KA630) simulator V4.0-0 Current

[paulhardy2]

I successfully run the QDSS Vaxstation GPX colour graphics, using the SimH 
MicroVAX 3100M38  emulator, that Matt Burke contributed last year. My ini file 
has:

set va capture

set cpu model=VAXstationGPX

 

and mouse tracking and clicks work fine.

 

My only outstanding problem is corruption of the text fonts after pushing and 
popping  windows, which I logged some months ago.

 

I would like to see the SPX graphics working, but that still seems to be a work 
in progress.

 

Regards,

 

-- 

Paul  Hardy

Email:   web:   www.paulhardy.net

 

From: Simh <  
simh-boun...@trailing-edge.com> On Behalf Of Hittner, David T [US] (MS)
Sent: 19 March 2020 21:24
To: 'Supratim Sanyal' <  supra...@riseup.net>; 
'simh@trailing-edge.com' <  
simh@trailing-edge.com>
Subject: Re: [Simh] Mouse capture on VAXstation II/GPX (KA630) simulator V4.0-0 
Current

 

I see now; I just noticed that you were running the MicroVAX2 executable, not 
the VAX executable, which doesn’t currently have a QDSS device (it could). The 
set cpu model command implicitly enables/disables the correct device(s). Try 
doing a “show device”  before and after the set cpu model command to see how 
the enable/disable works.

 

The correct configuration is:

 

   ; for multicolor vaxstation

   set qdss capture; qdss is the color device

   set cpu model=vaxstationgpx  ;vaxstationgpx is multicolor variant

 

OR

 

  ; for monochrome vaxstation

  set qvss capture ;qvss is the monocolor device

  set cpu model=vaxstation  ;vaxstation is monocolor variant

 

David

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] VAX PSI

[Peter Allan]

Mention of VAX-2780-3780PE piques my interest. I have a long term desire to
be able to connect a simulated VAX (or even better, a PDP-11) to a
simulated IBM mainframe (using Hercules) and use the VAX/PDP11 as a remote
job entry system for the IBM. Has anyone tried to do this?

Cheers

Peter Allan

On Thu, 19 Mar 2020 at 23:51, John Welsh  wrote:

> Hi Dave,
> I have been involved in a project using VAX-2780-3780PE and have been able
> to connect two VAX’s together using a Black Box Synchronous Modem
> Eliminator ( SME ).
> Product code: ME204A-F.
> No VAX PSI involved.
> Needed Decnet Plus and Wan drivers.
>
> Was able to test using a utility that comes with the Protocol Emulator -
> XJTEST which can use either VAX as a loopback.
> In my case it was a DSV11 to a DSW42 using the SME and the V.24 cables
> that come with the adapters.
> There is also another utility called REMOTE that can transfer simple files
> from VAX to VAX.
> I don’t think there is a driver for SIMH that can emulate a sync device. (
> yet ).
>
> Regards,
> John Welsh.
>
> Sent from my  iPhone
>
> > On 20 Mar 2020, at 03:38, dave porter 
> wrote:
> >
> > I know almost nothing about VAX PSI, but I have a vague idea
> > that even though X.25 is asymmetric and intended for DTE to DCE
> communication (i.e., your computer to the phone company's premises), it may
> be possible to coerce it
> > into talking from one VAX to another (via null modem
> > cable connecting two devices such as DUP-11s).
> >
> > I worked on RSX-11M PSI and that's how we tested in
> > the lab.  I can no longer remember how that was done.
> > Undocumented option somewhere, no doubt.
> >
> >
> > ___
> > Simh mailing list
> > Simh@trailing-edge.com
> > http://mailman.trailing-edge.com/mailman/listinfo/simh
>
> ___
> Simh mailing list
> Simh@trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] EXT :Re: Mouse capture on VAXstation II/GPX (KA630) simulator V4.0-0 Current

[Matt Burke]

It looks like mouse capture is still not enabled as I'm sure the release
key should be shown in the window title bar. You can toggle the messages
region with CTRL+F2. I'll give the latest code a try to see if I can
reproduce the issues.

The QDSS does have some known issues which cause screen corruption over
time. I hope to fix these one day but debugging the QDSS can be quite
difficult.

Matt

On 19/03/2020 17:38, Supratim Sanyal wrote:
>
> Very interesting. With "qvss capture" below "cpu model", mouse now
> works (jumpy, but better than no mouse). Also, I am getting console
> replies at the top of the graphical desktop, and have lost color.
> Here's a screenshot.
>
> On 3/19/20 1:25 PM, Supratim Sanyal wrote:
>> On 3/19/20 11:09 AM, Hittner, David T [US] (MS) wrote:
>>>
>>> et cpu model=vaxstationgpx
>>>
>>> set qvss enable    ; if required – I think ‘set
>>> cpu model = vaxstationgpx’ does this implicitly
>>>
>>> set qvss capture
>>>
>> Tried this sequence:
>>
>> set cpu model=vaxstationgpx
>> set qvss enable
>> set qvss capture
>>
>> Trying to set capture after the CPU model produces this:
>>
>> "Capture Mode Can't be changed with device enabled"
>>
>> BTW, an explicit "set qvss enable" produces this twice, so I think
>> your thoughts on implicit enable is correct ...
>>
>> "QDSS Display Created.  Uncaptured Input Mode
>> QVSS Display Created.  Uncaptured Input Mode"
>>
>> -- 
>> Supratim Sanyal, W1XMT
>> 39.19151 N, 77.23432 W
>> QCOCAL::SANYAL via HECnet
> -- 
> Supratim Sanyal, W1XMT
> 39.19151 N, 77.23432 W
> QCOCAL::SANYAL via HECnet
>
> ___
> Simh mailing list
> Simh@trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh