[Simh] KS10 IMP documentation

Mon Feb 29 22:46:07 EST 2016

As Clem notes, the 1822 report defines three types of Host to IMP interfaces, Local Host, Distant Host, and Very Distant Host.

The Local host uses TTL levels to signal, with per-bit handshaking.  The Distant Host
uses, IIRC, differential signalling aomewhat like RS422 electrically, but uses the per-bit handshaking.
The VDH, on the other hand, is more like the IMP to IMP protocol which is actually BISYNC at the wire level.
This can be done with any synchronous serial line adapter on the host (no start and stop bits!). So if there are any SIMH synchronous serial controllers, they would work for VDH.  The DU-11 maybe ?

At SU-ISL we ran VDH from an 11 to the IMP at the Stanford Medical School (SUMEX-AIM).

The software for VDH is messier than for LH or DH though, because there is a BISYNC layer for framing and a link ARQ scheme for error detection and retransmission of packet fragments across the (not assumed error free) VDH link.

So there were 11’s with VDH.  I don’t know if any 10’s used them, so that the VDH software would exist.

-Larry

> On 2016, Feb 29, at 3:48 PM, Clem Cole <clemc at ccc.com <mailto:clemc at ccc.com>> wrote:
> 
> Tim I'm going to guess that the AN22 implements the PDP-10 side of the BBN 1822 IMP interface. Is that a correct belief?
> 
> Bit Savers has the BBN 1822 document from 1976 (the original I think was '72 - I had a xerox of that one).  I think I saw it about a week ago at home, so I'm nearly 100% sure I have very clean copy of the actual published in it's Yellow Cover from BBN, the later edition - which would have been the one when the IP transition was taking place or at least planned (i.e. post Arpanet - when IP was being created).  If so, I'll look into getting it scanned and send an update to Al.   I'm pretty sure there was changes when BBN implemented their own mini for the IMP (the C30 - which later got re-microcoded to be  general purpose computer - the C70 "C Machine").
> 
>  
> 
> I do not remember how CMU interfaced the 10's and C.mmp to their (Honeywell) IMP.  I suspect it was either something like the AN22 or be a Jim "Tetter Toy" that he cooked up (I never knew).   However, IIRC @ UCB Ingres 11/70s a DR-11B with a little logic  (??Bob Kriddle hack maybe?? - at one time I had my hand in it).  That was the UCB Arpanet interface for many, many years --- until CSRG finally got a C30 IMP in Evans in the early 1980s.   Ing70 had a "very long host" interface from a Honeywell IMP that was "up the hill" at LBL.    Again, IIRC the C30 could do Ethernet to the Vaxen.   Or maybe it was connected to the Ethernet via a C70 which was connected the C30 (I've forgotten). But until the C30 showed up, the UCB Arpanet/Internet connection was fairly shallow; unlike CMU, MIT or Stanford.
> 
> Clem
> 
> On Mon, Feb 29, 2016 at 12:33 PM, Timothe Litt <litt at ieee.org <mailto:litt at ieee.org>> wrote:
> On 29-Feb-16 12:34, b4 at gewt.net <mailto:b4 at gewt.net> wrote:
>> Hello,
>> 
>> Can anyone point me at IMP documentation for the KS10? I'm only seeing AN10 maintenance printsets up on Bitsavers...I'd love documentation that isn't KLH10's KS10-ITS implementation!
>> 
>> Alternatively, were any of the experimental Ethernet interfaces ever supported in TOPS-10/20 on the KS10? (I don't have access to my SRI-NIC packs due to the VM being down at present)
>> 
>> Alternatively too...the userland ANF-10 bridge could prove useful for this...
>> 
> Not clear what you want to do.
> 
> The KS does have an ARPA interface, the AN22.  I have not tracked down a spec for it.  It is KMC-11 based, not simulated by SimH at present.  The TOPS-20 monitor sources for 4.1 are a resource.  One day I'll get access to my stuff at CHM and will look for a spec.  It's on my list to see if it's worth emulating.
> 
> The DEUNA is simulated in SimH.  Neither the released TOPS-10 nor TOPS-20 support it.  There is some TCP code floating around that does.
> 
> If you're trying to move files in and out:
>   DECnet does work (phase II on TOPS-20, Phase III or IV on TOPS-20), and will talk to VAX/RSX.  You can move files by stopping on or PMR thru) one of them.
> 
>   ANF-10 on the KS does work, KS-KS.  It should be possible to build the -11 nodes on the KS & boot them on a SimH -11.  I've been meaning to get around to that, but haven't.  With ANF, besides the peripherals and terminals, you get DCP.
> 
>    Networking on the KS is supported on both OSs by the KDP emulator.  The DMR should work, though I think there's a monitor bug in 7.04 where my colleagues changed AC definitions and didn't update the DMR driver.  Also on my list.
> 
>     You can also move files in and out with Kermit (serial lines), cards, magtape, printer.
> 
> I don't know what you mean by 'anf-10 userland bridge'.  Of course, FAL on the -10 listens to both DECnet and ANF.
> 
> 
> _______________________________________________
> Simh mailing list
> Simh at trailing-edge.com <mailto:Simh at trailing-edge.com>
> http://mailman.trailing-edge.com/mailman/listinfo/simh <http://mailman.trailing-edge.com/mailman/listinfo/simh>
> 
> _______________________________________________
> Simh mailing list
> Simh at trailing-edge.com <mailto:Simh at trailing-edge.com>
> http://mailman.trailing-edge.com/mailman/listinfo/simh

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20160229/17bcd63f/attachment.html>