11/45 interrupts

True enough. I'd forgotten you were running XXDP+. OK, it would appear that the data buffers (and for that matter most instructions) must be working correcrtly. Let me explain ; We know that one particular device (the Line Time Clock) is nut interrupting correctly -- in particular the vector is being treated as 0 (it's halting with 4 in the PC, IIRC). You've (ugh!) swapped out the DL11-W board, so it would appear it's not a fault with that particular device, but what I want to see is if another interrupt has its vector treated as 0 too. The easiest device to try (and one I know you have) is the receiver interrupt from your console port). This will generate an interrupt if a character has been received from, the console. So, what I am suggesting is : 1) Set all the vectors so that they contain .+2, HALT. That is, so when the 11/45 fetches a particular vector, the PC gets loaded with the address of the second word of that vector, which is actually a halt instruction. The machine should therefore halt with the PC containing the address of the first word of the next vector (as it did for the Line time clock test -- it halted at 4, which is the address of the first word of the next vector after 0, if you see what I mean) 2) Clear any received chracters from the console port. That way it's not going to generate an interrupt. yet 3) Enable interrupts on on the console port receiver 4) Loop endlessely 5) You press a key on the console terminal .That sends a character to the console port, which then generates a receiver interrupt. The machine _should_ then halt, since all vectors effectively point to halt instructions (see (1)). What interests me is _where_ it halts. I will guess the PC will contain 4 (that is, it's totally ignored the vector again), but I don't want to assume anything. -tony