No, that wouldn t work, because what Rob is saying is that he sends the length plus payload, but not a regular DDCMP header or either of the CRCs. So if you send that emulated DMC data stream to a real DDCMP implementation it would not recognize it because the headers are omitted. It wouldn t be a large change, though. Compared to the job of emulating a DMC (the microcontroller to bus interface logic), DDCMP itself is a simple matter. paul From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf Of Kevin Reynolds Sent: Wednesday, January 04, 2012 5:43 PM To: hecnet at update.uu.se Subject: RE: [HECnet] Hecnet and DDCMP There is a possibility to use a real system however, if a client was written that would read input from the socket and throw it out of the serial port on physical hardware? As long as the original data integrity is maintained this should work as long as timing isn't an issue, correct? You basically created a virtual driver that behaved as a serial adapter? Was this done over TCP/IP or DecNET, or LAT or something else? For synchronous comms (assuming DMC-11 RS232-C physical) you would need to use a 25 pin connector to wire it as it requires pins 15, 17 and 24 for the clock signal in addition to the standard async pins. It would just be a matter of putting in a microcontroller such as the launchpad or arduino to handle the synchronous communications, and then pass it back out async. Yes this would probably be very messy and perhaps the wrong way to go. Regardless, I would love to see what you have done to get this working SIMH->SIMH. Thanks heaps, Kevin