G'day Erik - Congratulations on your Nova 4 "save"! Having worked with DG equipment for about 31+ years here are some "Quick answers": The Nova 4 and and Nova 3 CPU cards are not compatible, but most of the I/O cards are. There are two main I/O daisey chain signals on a standard Nova/Eclipse backplane - interrupt priority in/out and data channel request in/out - and they usually should be jumpered across unused slots if other I/O cards exist higher in the chassis. Those don't sound like standard DG I/O card numbers: I believe that the DG board part numbers usually start with 005-xxxx-xx. I would have to research that. DG put its 15" x 15" card numbers along the "front" reinforcement bar, that part of the card just opposite the finger contacts and directly between the two extraction handles. There were usually a series of three numbers on the bar. The Nova 4 came in two standard physical configurations and sold/marketed as three subprocessor "types" (Nova 4/C, 4/S and 4/X) - each different than the others only in four (4) PROM chips. There was quite an aftermarket for PROM burner as these four chips cost a minimum of $1,500 from DG (when $1,500 was 'real' money). There is a picture of the front of the Nova 4 at http://www.SimuLogics.com/nostalgia.htm. BTW, the DG Eclipse S/140 and subsequent S/280 were virtually identical pieces of hardware since they used the same microprogramming core, only PROM was different. They also sold ERCC memory with the Eclipse as a market positioning strategy (i.e. maximum client $ extraction mechanism). BTW, Medtronics still uses Nova derivitives called "Strobe Hawk" boards which are Nova-like co-processors that fit into a PC's ISA slot. The Nova on the board does all the processing and then requestion all I/O through a TSR on the PC (egad!). We are currently working with Medtronic to replace their hardware with our commercial software product (ReNOVAte) as described at www.SimuLogics.com. A free version of our Nova emulator that is written in Java is also available at that site. We are also trying to preserve all DG 16-bit software, hardware and documentation and are actively working with EMC, DG, TLC and other companies in this respect. And yes, Carl Friend is an exellent [re]source of additional information Bruce ----- Original Message ----- From: "ip500" <ip500(a)home.com> To: <classiccmp(a)classiccmp.org> Sent: Wednesday, November 08, 2000 6:31 AM Subject: Re: DG Nova 4

ClassicCmp wrote: Depending on the _type_ of Nova bus backplane there are memory-only, memory-I/O and I/O-only slots. An I/O controller can be stuck in either an I/O or memory-I/O slot. But... Many DG controller carry interface signals to the device by making use of backplane pins which are unused on the slot in question (for example, in the case of an I/O slot, pins which carry memory signals) by wirewrapping the device connector to the backplane (essentially marrying the device to that and only that slot). When facing the wirewrap side of the backplane, it's the two pins, two in from the right-hand side of the left hand connector (confused yet?) So something like: ........xx.. .................. ........xx.. .................. These signals are essentially grant signals. There's a single wire-or interrupt request line and a similar line for data channel requests; when the CPU decides to honor the request it drives the INTA or DCHA line (which are the two signals being jumpered). The signals "enter" at the bottom of the slot and hit logic on the board which passes the signal out the top of the slot if the board isn't requesting service, or block the signal and drive the device's six-bit device code onto the bus if it is -- hence giving the device physically closest to the CPU higher priority. Yes, from the top of the last occupied slot to the bottom of the next occupied slot. Depending on the artwork, you may be able to see the traces which carry these signals between ajacent slots. Oh, and don't forget to remove any existing jumpers that bypass the boads that you're installing ;-) -- Chris Kennedy chris(a)mainecoon.com http://www.mainecoon.com PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97

G'day Mike - I'm not familiar with those serial number formats - DG has quite a few different main computer lines divided into the following major groups: 16-bit Nova 16-bit Eclipse 32-bit MV [also called MV/EClipse] Motorola 88000-based AViiON systems now Intel x86-based AViiON systems. DG was also recently purchased by EMC mainly for its RAID storage line, the CLARiiON. We are trying to collect, preserve and archive all DG 16-bit related hard/soft/doc-ware (www.SimuLogics.com ) and am moving to 32-bit MV preservation faster than initially intended. Carl Friend and Chris Kennedy are good resource due to enthusiasm and experience. I would need to know more to answer about the model numbers or physical description to give you better answer... Bruce ----- Original Message ----- From: "Mike Kenzie" <KenzieM(a)sympatico.ca> To: <classiccmp(a)classiccmp.org> Sent: Monday, November 13, 2000 10:54 AM Subject: Data General/MASK was Re: DG Nova 4

Okay, it's one of "DG's" Personal Computers (that they bought from several different vendors, including Intel directly). Sorry, I (and others around here) don't consider those DG machines since they are simply OEM versions of systems available through other channels. Ooooops. Bruce ----- Original Message ----- From: "Mike Kenzie" <KenzieM(a)sympatico.ca> To: <classiccmp(a)classiccmp.org> Sent: Tuesday, November 14, 2000 6:46 PM Subject: Re: Data General/MASK was Re: DG Nova 4