Kari, that is the correct answer
------Origineel bericht------
Van: Kari Uusim ki
Afzender: owner-hecnet at Update.UU.SE
Aan: hecnet at Update.UU.SE
Beantwoorden: hecnet at Update.UU.SE
Onderwerp: Re: [HECnet] Auto-booting a MicroVAX 3400
Verzonden: 13 december 2012 22:09
Set boot defines the boot device, but set halt <value> is the parameter
for restart control.
Regards,
Kari
On 13.12.2012 22:54, hvlems at zonnet.nl wrote:
Set boot 3
Set boot defines the boot device, but set halt <value> is the parameter for restart control.
Regards,
Kari
On 13.12.2012 22:54, hvlems at zonnet.nl wrote:
Set boot 3
Why? DEC-10 is big endian... :-)
paul
On Dec 13, 2012, at 3:48 PM, John Wilson wrote:
From: "Rob Jarratt" <robert.jarratt at ntlworld.com>
Big endian, I use htons() to prepare the length for network transmission.
Big-endian, for a DEC-only protocol?! Heresy!
John Wilson
D Bit
On 13 Dec 2012, at 15:48, sampsa at mac.com wrote:
I've just been donated (by Saku Setala) a pair of MicroVAX 3400s.
One of them is set up as a satellite node and boots fine off ESA0 - however, I can't figure out what I need to do to make it automatically boot when powered on - anyone know what switch / console command I need to give to make this happen?
At the moment it starts up to the >>> prompt and if I type boot, it boots happily. I'm just trying to get it to autoboot instead.
Would it be dep bdr 0 as it is in simh for auto boot on a real VAX?
(if i'm wrong, let me know)
sampsa
I've just been donated (by Saku Setala) a pair of MicroVAX 3400s.
One of them is set up as a satellite node and boots fine off ESA0 - however, I can't figure out what I need to do to make it automatically boot when powered on - anyone know what switch / console command I need to give to make this happen?
At the moment it starts up to the >>> prompt and if I type boot, it boots happily. I'm just trying to get it to autoboot instead.
sampsa
From: "Rob Jarratt" <robert.jarratt at ntlworld.com>
Big endian, I use htons() to prepare the length for network transmission.
Big-endian, for a DEC-only protocol?! Heresy!
John Wilson
D Bit
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE]
On Behalf Of John Wilson
Sent: 13 December 2012 19:30
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] VCF-East on HECnet?
From: "Rob Jarratt" <robert.jarratt at ntlworld.com>
No. It uses a TCP connection. Each buffer received from the device
driver is sent along on the TCP connection with a 2-byte length prefix,
and that is it. So no DDCMP underneath.
OK cool. Little-endian?
Big endian, I use htons() to prepare the length for network transmission.
[Paul Koning]
If you want to really make your head spin, imagine DMP/DMV emulation
(DDCMP multipoint).
I've never really understood this. Is it a broadcast net, or what?
I'm pretty sure I have a pair of DMVs in this mess somewhere...
John Wilson
D Bit
It is being merged into the official SIMH builds by Mark Pizzolato, I am
hoping it will be in the next release. I just checked with Mark and he is
not ready to share the code quite yet, but there should be some almost
official code to share soon.
Regards
Rob
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE]
On Behalf Of Paul_Koning at Dell.com
Sent: 13 December 2012 15:20
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] VCF-East on HECnet?
On Dec 12, 2012, at 5:32 PM, Rob Jarratt wrote:
...
If you want to peer with me using DMC11 let me know. I will need to
know your IP/DNS address and a port you want to listen on. You can
download a Windows executable of the VAX780 emulation with DMC11
support here:
http://sdrv.ms/Lmg33s I am hoping that the DMC11 will be in the next
release of SIMH, and perhaps in the PDP11 emulation too.
I'd be interested in trying that. The Unibus emulation should just carry
over.
Is the source available?
paul
On Dec 13, 2012, at 2:29 PM, John Wilson wrote:
From: "Rob Jarratt" <robert.jarratt at ntlworld.com>
No. It uses a TCP connection. Each buffer received from the device driver is
sent along on the TCP connection with a 2-byte length prefix, and that is
it. So no DDCMP underneath.
OK cool. Little-endian?
[Paul Koning]
If you want to really make your head spin, imagine DMP/DMV emulation
(DDCMP multipoint).
I've never really understood this. Is it a broadcast net, or what?
I'm pretty sure I have a pair of DMVs in this mess somewhere...
For the gory details, read the DDCMP spec. It's gory, all right.
Physically is's a broadcast net. Logically there is a single master and N-1 slaves. The master polls each of the slaves. The master can send at any time; the slave can send only in response to a poll. I think in theory it can be full duplex, I'm not sure if in practice that made any real difference.
Performance sucks, but if you needed lots of connected stations without using up lots of hardware on the PDP11 side, multipoint is an answer. Ethernet, of course, is vastly superior, but that came later.
It could be worse. I once spent some weeks debugging a communication problem between a PDP11 and some Harris 2200 terminals. Those used multipoint also, but Bisync, not DDCMP. And the interface was a DL11-E. So: half duplex bisync multipoint over a 9600 baud async line. I think it needed the -E variant because the modem control signals were used for poll control. Yikes.
paul
