Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!hellgate.utah.edu!csn!datran2!smb From: smb@csn.org!datran2 Newsgroups: comp.arch Subject: Re: ~8-job "knee" in response curves on Suns (was Re: IBM RS6000) Summary: Tobin's benchmark on NeXT 68040 Message-ID: <1991Jan16.231017.2530@csn.org!datran2> Date: 16 Jan 91 23:10:17 GMT References: <5257@auspex.auspex.com> <3956@skye.ed.ac.uk> Sender: Steven M. Boker Organization: Data Transforms, Inc. Lines: 36 I just tried Richard Tobin's little context switch benchmark on my NeXT 68040 which was relatively quiessant with 46 active pids consuming 2% of cpu as a base. The NeXT seems to be pretty stable as far as context switches go. 3 : 1460 switches / second 4 : 1460 switches / second 5 : 1448 switches / second 6 : 1393 switches / second 7 : 1404 switches / second 8 : 1393 switches / second 9 : 1362 switches / second 10 : 1341 switches / second 11 : 1333 switches / second 12 : 1327 switches / second 13 : 1335 switches / second 14 : 1325 switches / second 15 : 1307 switches / second 16 : 1321 switches / second 17 : 1313 switches / second 18 : 1310 switches / second 19 : 1309 switches / second 20 : 1317 switches / second 24 : 1276 switches / second 28 : 1289 switches / second 32 : 1280 switches / second 36 : 1280 switches / second Steve. -- #====#====#====#====#====#====#====#====#====#====#====#====#====#====#====# # Steve Boker # "Badgers, we don't have no stinking badgers" # # datran2!smb@csn.org # -from Treasure of the Sierra Madre Zoo # #====#====#====#====#====#====#====#====#====#====#====#====#====#====#====#