Thread View: comp.lang.oberon
2 messages
2 total messages
Started by Doug Danforth
Wed, 17 Apr 1996 19:16
Oberon Cycles Per Second
Author: Doug Danforth
Date: Wed, 17 Apr 1996 19:16
Date: Wed, 17 Apr 1996 19:16
55 lines
2672 bytes
2672 bytes
OBERON CYCLES PER SECOND Below is a rank ordering of the number of Oberon Cycles that can be executed per second under various operation systems and Oberon versions. The term cycles is used rather than tasks since it is assumed that many tasks can be executed in a single cycle of the Oberon loop and not substantially change the number of cycles per second. More explicity it is assumed that if the user has installed three tasks,abc, then the following happens H0: OabcOabcOabc... where "O" is an implicit task that frees time to the operating system so that other non Oberon task can run. If, however, we have H1: OaObOcOaObOc ... then, indeed, the time would grow as the number of tasks increases (I have been assuming that Time(O) >>> Time(x) for x=a,b,c... This seems not to be the case for Emi Zeller's Multi system). In any case the reported number of cycles per second are: O/S Oberon CPS Reported by ------ ------ ----------- ------------------ Win95 Multi 124,000 Emi Zeller Win3.x S3 13,800 (1) Emi Zeller Win95 S3 5,800 Doug Danforth DOS S3 1,400 (2) Frank Hrebabetsky Linux V4 167 Wojtek Skulski Win95 V4 20 Doug Danforth (1) uncertain whether this is Win3.x or Win95. (2) best value reported (of all mouse locations on screen). We seem to see a factor of 60,000 difference between the best and worst. Even if we exclude the Multi system there is still a factor of almost 700 between best and worst. Unless this analysis is all wet, it says to me that one should avoid spitting procedures into tasks on Win95 V4. Comments? -Doug -- ----------------------------------------------------------------------------- Douglas G. Danforth, Ph.D. danforth@csli.stanford.edu Visiting Scholar (415)723-2487 x332 vox Applied Speech Technology Lab (415)725-2166 fax Center for the Study of Language and Information Ventura Hall Stanford University, Stanford CA. 94305-4115 -----------------------------------------------------------------------------
Re: Oberon Cycles/s (Linux System-3)
Author: skulski
Date: Thu, 18 Apr 1996 00:46
Date: Thu, 18 Apr 1996 00:46
31 lines
1671 bytes
1671 bytes
> In any case the reported number of cycles per second are: > > O/S Oberon CPS Reported by > ------ ------ ----------- ------------------ > Win95 Multi 124,000 Emi Zeller > Win3.x S3 13,800 (1) Emi Zeller > Win95 S3 5,800 Doug Danforth > DOS S3 1,400 (2) Frank Hrebabetsky > Linux V4 167 Wojtek Skulski > Win95 V4 20 Doug Danforth > > (1) uncertain whether this is Win3.x or Win95. > (2) best value reported (of all mouse locations on screen). > > We seem to see a factor of 60,000 difference between the best and worst. > Even if we exclude the Multi system there is still a factor of almost > 700 between best and worst. > > Unless this analysis is all wet, it says to me that one should avoid spitting > procedures into tasks on Win95 V4. One more data point: Linux System-3 v1.5, timing module by Emil Zeller, Pentium-100 MHz, X-windows, Mach64 card/Xserver, fvwm window manager. Measured: 255 to 294 cycles per second (Itried a few times). CPU load split about half/half between System-3 and X. Hope it helps, Wojtek
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads