Qbus Backplanes

On Jun 5, 22:41, Peter Turnbull wrote: QQCD OK, so I've just thought of one that might be. An early 11/03 quad-height M7264 was meant for a Q-Q H9270 backplane and presumably that's what Sridhar is thinking of. However, I don't think it's what he calls a QQQQ-only processor. I'd have to look at the print set or a board to be sure, but IIRC the M7264 actually has Q-Bus on its A&B slots, so it would be OK in a Q-CD backplane, so long as the module underneath it didn't use the upper contacts on its own C&D slots (C and D aren't bussed; the lower contacts on C&D are only connected to the upper contacts of the slot below). However, just before I hit "send", Sridhar told me what he *was* thinking of: The short answer is "I don't know". FWIW, here are my thoughts. I would be amazed if it put any signal(s) on the relevant pin(s) on *both* the A&B and C&D slots (because that would lead to multiple signal paths). I'd be rather surprised if it put some on A&B and some on C&D, but if it did, or if it put QBus signals on C&D, it would definitely have to go in the first Q-Q row. Or maybe the second, something I saw about it suggests an I/O module is supposed to go in front of it, and then the argument applies to the I/O module, really. However, the logical way to build such a board would be with the QBus signals on A&B, like I believe the M7264 does -- unless the bus length is critical, and it needs the shortest path between the first and second rows, which is the C&D slots. But would it physically fit any other backplane than its own? Don't those new-fangled things have different handles to anything that fits a Q-CD backplane? Chuck? -- Pete Peter Turnbull Network Manager University of York