[HN Gopher] IBM's SSEC, the first computer that could modify a s...
___________________________________________________________________
IBM's SSEC, the first computer that could modify a stored program
Author : luuuzeta
Score : 19 points
Date : 2023-02-13 13:08 UTC (1 days ago)
(HTM) web link (www.historyofinformation.com)
(TXT) w3m dump (www.historyofinformation.com)
| chx wrote:
| Persnickety but also curious comment: wasn't the System/360 the
| first mainframe?
| rst wrote:
| By the time the S/360 was announced, IBM had already been
| producing top-of-the-line systems with the main processor
| sharing memory with several I/O "channels" for several years --
| which would be "mainframes" by any reasonable definition I can
| think of other than "S/360 and derivatives" (which, by the way,
| excludes competing mainframe systems from Univac and Burroughs,
| among others).
| einpoklum wrote:
| The machine was "designed, built, and placed in operation in only
| two years"
|
| ... impressive!
| jmclnx wrote:
| I remember the time when modifying running programs were a big
| rage. Now seems many OSs are going in the opposite direction with
| OpenBSD leading the race.
| baabaloo wrote:
| [dead]
| addaon wrote:
| Is OpenBSD moving away from program loading and dynamic
| linking? Even many embedded systems allow software updates
| these days...
| Aardwolf wrote:
| So it used both vacuum tubes and electromechanical relays.
|
| Any idea which of those two was more reliable for a computer?
| Vacuum tubes definitely have a limited life and had to be
| replaced constantly from what I read about early computers, but
| afaik electromechanical relays can only do a few 100K cycles
| before they break, or a few millions at lower voltage - which
| seems easy to reach if you do 100s of additions/multiplications
| per second
___________________________________________________________________
(page generated 2023-02-14 23:01 UTC)