On Fri, 12 Oct 2007 11:21:23 -0700, Brent Hilpert wrote:
Lovely photography but many of the blurbs leave a lot
to be desired as far
as historical and technical accuracy goes. Can't have everything I guess.
e.g. this bizarrely phrased attribution:
> #10: Core Memory
> Because it holds its state, core memory is able to retain information
> even when the electricity is turned off, thereby paving the way to modern
> computing.
Howze that again?
Many years ago, when removing magnetic core from process control systems I was required to
erase well
more like scramble the magnetic core before it could leave the plant. So there was no
chance process
variables would leave the plant. I would take off the top cover and pass a magnet over the
core array and
told them to put it back into the system, if they did not believe me, but I had no idea
what state the
production line would come back up in, if at all. Never had anyone put one back in and
test it.
One of the sales points of using magnetic core memory on early process control systems
even as late as
the early 8080 days, was core's ability to retain state on loss of power. One of the
leading process control
systems of the day did not boot as we know it today on power up It just woke back up, the
program was
always stored in 16k of magnetic core. The floppy drive was an FE tool for IMPL of core on
cold boot.
Power on was a true warm boot because the current program was still in memory in a running
state. It just
needed to be restarted or continued depending on the first interupts it gets back from
it's sensors.
Not to defend the person who wrote #10 above, but is not instant on, fast recover to
operation, computing
devices based on rewriteable static memory, the very definition of modern computing even
if they only had
a clock speed of 476khz.
These words being typed by someone who just flashed an ATT/Cingiular 8525 2gigs of storage
and 3G
connectivity.
From core to flash via eeprom, where too next ?
The Other Bob
Still a few Bobs short of a choir :)