PDP11 core power

I never had any of the 11 data commited to memory, but I did (at one time) know quite a bit about the Nova-1200, which was somewhat contemporaneous.

The standby power was on the order of 50 watts per board, regardless of size. Since you can only read-write one word at a time, that power was not a factor of the number of planes, but the total current in the drivers. On the 8K nova stacks, it was 1 ampere nominal between the +15V and -15V rails -- per bit. That's 480W for a few hundred nS for a write, and even more than that for a read. Yowza, eh? The higher density stacks had slightly lower write currents IIRC.

I seem to recall that in the tight memory-heating test, it could do a read every 2805 nS. Figuring two 300 nS pulses per read, that's a duty cycle of 21% times 480W = about 100W. So, for a busily running system with 32K of core, you've got 4 x 50W quiescent current plus 100W of read-write activity, so, about 300W in memory.

My moderately full system drew just about 12A at 120V for the CPU-memory cage, and another 15A for the two Diablo 30 disk drives.

Oh, How I miss DDOS and LHS MUB.

-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Computer software consists of only two components: ones and zeros, in roughly equal proportions. All that is required is to sort them into the correct order.