31 Jul
2013
31 Jul
'13
12:46 p.m.
On 07/31/2013 11:41 AM, Fred Cisin wrote:
IBM wanted LOTS of money for each row of 16K RAM, but
it took the same
ones as TRS80, where there was an intense market competition.
This was also the time of the US "Let's shoot ourselves in the foot"
anti-dumping trade sanctions against Japanese semiconductor makers. So
any DRAM was in short supply and expensive. For my 5150, I used some
Intel 2109s that happened to pass diagnostics on the PC for 16Kb.
NOTE (for those not familiar): PARITY does NOT reduce
errors, in fact,
9chips instead of 8 means 12.5% more errors. What PARITY does is LOCK UP
the machine when there IS an error. IBM considered the inconvenience of
having to reboot and rerun the software to be less important than the
possibility of printing $96.00 instead of $64.00 on a check.
...and parity does not cause a machine to "lock up". What it does do is
trigger a non-maskable interrupt. The default action for that vector on
the 5150 is to display a message and halt--but that does not preclude
the user from supplying his own handler that returns control to the
interrupted code.
I'd rather know about a memory error than discover that my program was
producing garbage.
Seymour Cray is often credited with saying "Parity is for farmers" when
the 6000-series machines lacked parity checking on memory. It's also
worth noting that his later machines not only had parity but also SECDED.
--Chuck