NORC: Multiplying By Each Digit
ICL 1901 A
I have noticed that in TP4999-1, there is one page in the table of contents which...
When NORC was built there weren't many other computers in existence. But the idea of being the fastest of what was out there does make sense for the definition of "supercomputer" and NORC definitely meets that defintion.
Definitely not true, there were many before that.
Not mentioned was the IBM 7030 or "Stretch" which was a supercomputer.
Actually, Stretch does qualify because it did contain various techniques to maximize processing time not normally used. Stretch ran into trouble in that it was hard to predict how much time would be saved by such techniques. Stretch was clearly the fastest, but not as fast as expected since typical programs didn't make as good use of its speed techniques as had been hoped for.
I don't think the specific techniques used to speed time make the definition; rather it is that special techniques that aren't found in production machines are used. Many of the techniques used in 1950s "supercomputers" are used in all computers today routinely, but at the time, they were novel and expensive.
The supercomputers push the state of the art, and the benefits-- once proven to be effective--trickle down to regular production machines.