steve@jplgodo.UUCP (04/18/86)
In article <817@umcp-cs.UUCP>, chris@umcp-cs.UUCP writes: > In article <257@pyramid.UUCP> csg@pyramid.UUCP (Carl S. Gutekunst) writes: > >[The Ridge 32C, predecessor of the 32/300, turned in 1776 Dhrystones. > > I don't believe it; I think the 32C was a lot faster than that.] > > I would believe it. The Electrical Engineering department here got > an early Ridge. The thing was usable only as long as there was one > process running on it. Start two tasks---one background job would > do---and response time became awful. It felt as though it were a > swapping system with floppies as the only backing store.... When Ridge first starting selling machines, they gave a very large discount to universities and government sites on the purchase of the first machine at a site. Unfortunately, the machines sold under this discount only had two megabytes of memory. Also, that version of ROS (Unix port) gave the same memory priority to background and foreground jobs (the system was envisioned as a number cruncher running essentially a single job at high speed). After using our 32C for about a year, we added two more megabytes of memory and upgraded to ROS 3.3 (which gives foreground priority over background). The performance increase when running multiple jobs was tremendous. BTW: I was the person who sent in the 1776 Dhrystone number and I am reasonably sure I did it right. I am in no way connected with Ridge--just a satisfied customer. -- ...smeagol\ Steve Schlaifer ......wlbr->!jplgodo!steve Advance Projects Group, Jet Propulsion Labs ....group3/ 4800 Oak Grove Drive, M/S 156/204 Pasadena, California, 91109 +1 818 354 3171