Guy Sotomayor wrote: Hmph; just got done configuring the boards for the MMU :). Is there a good reference for how the jumpers should be configured for just the basic CPU set? I have the engineering drawing sets (in TIFFs... anyone have these in PDF? Tiff is annoying :)), and it indicates the jumpers, and _some_ of them seem to indicate a "default" position, but not all, and I don't know whether these are really defaults or not. I know the MMU and the Stack Limit options require jumper changes, did any other options require them? Just assembled one of the KM11s last night, now I just need to order a couple of unibus extenders so I can use it :). The M9302 does not appear to have jumpers, and it doesn't appear to be of the same vintage as the 11/40 (it has a "caution" label that mentions an 11/34 or 11/04...) so maybe it is incorrect for use in an 11/40. Thanks, Josh

Guy Sotomayor wrote: Hey, alright -- checked out my boards against the drawings again, and the EIS jumper was removed (W1 on M7233) so I reinstalled it. Now examine/deposit increments properly -- 2 at a time, every time :). Now to solve the termination problem and maybe then I can get a program to run... and then... THE WORLD! Anyone have a spare M9312? Thanks again, Josh

Tony Duell wrote: I fixed an incorrect jumper (my boardset was configured for the EIS option, which I do not have) and that solved the incrementing problems. Now that that's fixed, programs I've toggled in seem to be running correctly, too. Yay! I'm fairly sure I have the grant continuity cards in the right way (they're not exactly clearly labeled -- these are the tiny ones that lack handles). They're in the D connector of the empty slots, and oriented such that the label "Grant Continuity" is the right way up (or put another way, such that the conductors on the card edge are facing toward the back of the machine). I'll check the NPG jumpers -- here's a stupid question -- how do I find where they're supposed to be? I've looked all over for a simple diagram showing which pins are CA1 and CB1, but I'm not finding them... Thanks, Josh

On Thu, May 21, 2009 at 9:06 AM, Tom Uban <uban at ubanproductions.com> wrote: Having worked with these machines a lot over the years, I can personally recommend having the right handbooks on hand when configuring things or diagnosing problems - it will save a lot of time and grief. Back in the day, we always had them and we used them frequently. Another way to see is to look at a dual-height Unibus grant card and note which pair of pins is shorted together on the "C" finger (the one that only has one pair of pins shorted, not the one that has 3 pairs of pins shorted). If you don't _have_ a dual-height grant card, you are going to have to get very familiar with where the NPR pins are because you are going to be wrapping and unwrapping that jumper on the backplane every time you insert or remove cards. At Software Results, we used to make dual-height grant cards so we could stick one in every COMBOARD box for our customers' convenience (the last batch of 100 we ever bought ran about $20 each, 15 years ago). Since we had them lying around, we always removed *every* NPR wire on every backplane and never used G727 single-height grant cards. It just took too much time to reconfigure machines otherwise. I've also seen shops that tended not to want to buy fistfuls of dual-height grant cards, so they just minimized how often they fiddled with the backplane - once they got it right, they tried not to mess with it. I still have the films (manual paste-up) for the Grantosaurus Rex cards, c. 1981., but our last grant card was a plain rectangular affair with a handle cutout. Not as interesting, but much cheaper to make. -ethan

On Thu, May 21, 2009 at 1:54 PM, Dave McGuire <mcguire at neurotica.com> wrote: Yep... there's some discussion of it on the list archives. I think they are great, but I understand why we went with cheaper ones (no silk-screen, for example). I still wish we'd added some jumper pads to allow us to flip them from Unibus to Qbus grant cards, but when that last order was being placed, the production guys just ordered more of what we'd been ordering rather than incur the expense of a change. -ethan

IIRC the FPF11 (floating point board for the 11/23 and 11/24) has that feature. Some years ago I was given a box of mostly home-made grant cards. The home-made ones turned out toe be dual-height prototyping boards with links soldered between the pins. You can guess what I did with those :-) -tony

On Thu, May 21, 2009 at 2:35 PM, Tony Duell <ard at p850ug1.demon.co.uk> wrote: I think you are right. I don't have one, but I do remember seeing that documented in a handbook. That's probably where I got the idea. -ethan

Correct. That's the 4th one from the top of the machine -- that is there should be 3 empty conenctors above the grant continuity card and 2 below it Correct. OK. The 'C' means connector C (the one above the connector where you've puy the grant card) The 'A' and 'B' mean the top 2 pins on a particular side (the 18 pins on a particular side are lettered A B C D E F H J K L M N P R S T U V) The '1' means the front (component) side of the connector. There should eb a wire-wrapped link between these 2 pins on every SPC slot in the machine. It would be fitted at the factory, but cut if an NPR (DMA, basically) device was used in that slot. In fact it _must_ be cut if you put such a device in the slot. -tony

Unless you have a DMA device - which is typically going to be something like a disk or tape controller - the NPG grant should be daisy-chained without break. I disagree slightly with Ethan's assertion that this is something you'll play with these jumpers a lot: I've only done it once, for a tape controller. Maybe I've just been lucky. :-) I think if I ever remove that tape controller and put in a non-NPR board, I'm going to find some sort of slide-on jumper for CA1-CB1, just in case.... ISTR that the 11/40 uses a M930 terminator. I think I have one I could loan you for testing, if you'd like. -- Ian

That's right. A memory board doesn't use the NPG signal, so the jumper should be in place. What I would do next is grab a logic probe and the Unibus pinout (for the terminator slot) and see which grant(s) is being asserted at the terminator, if any. IIRC, the grants are active _high_, unlike most Unibus signals. Check the NPG, BG4, BG5, BG6 and BG7 signals at the M9302 terminator -tony

Rememebr a floating signal will appear as high. Also, if there's a device that uses BG6 and the BG6 is open between the CPU and that devices, thenBG6 In on said device will appear high, so the device logic wil faithfully pass on the grant and set BG6 Out high. First check it's not stuck high at the processor end (this could be a fault with the rbitration logic or similar). Then I guess you do a half-split. Check it at a slot midway along the backplane, and move in the apropriate direction until you find what's causing it. -tony