Several comments on crashed drive in PDP-11/23

It's a DEC tape. The tapes are square boxy things, about 105x105x25mm, and the drives can be cantankerous. (In particular, the mechanism that picks up the tape and gets it started in the drive tends to stop working; it's a plastic leader that hooks into a hole in the tape, and the drive firmware is too stupid to recover from failure to hook in any way but "try again". If the tape hole isn't in quite the right place, the pickup plastic flaps around again and again and wrecks itself if you don't catch it fairly soon.) It is. You don't. There are no "upper bits" to the PC (nor any of the other registers); the PDP-11 is not a segmented architecture in the sense the 8086 is. With the MMU off, you can access only 64K (I think - I/D splitting happens only with the MMU on, as I recall, even on machines that have split I and D space). On most machines, that's 56K of RAM and 8K of I/O space. With the MMU on, the 17 bits of address coming from the CPU (16 bits that are usually thought of as address plus the I/D space bit) are mapped to however many your hardware supports; you can access anything you have, but you're restricted to 64K of instruction space and 64K of data space at any given moment. (Some implementations may not actually have all these lines present in hardware; this is a conceptual description. In particular, modesl without split I and D space don't implement the I/D space bit. And there may be other bits I've forgotten about.) /~\ The ASCII der Mouse \ / Ribbon Campaign X Against HTML mouse at rodents.montreal.qc.ca / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B