Path: utzoo!censor!geac!torsqnt!lethe!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!cs.utexas.edu!samsung!ernie.viewlogic.com!peter From: peter@viewlogic.COM (Peter Colby) Newsgroups: comp.sys.mac.system Subject: Re: True Multitasking Message-ID: <1991Jan16.213340.12807@viewlogic.com> Date: 16 Jan 91 21:33:40 GMT References: <19019@shlump.nac.dec.com> <42588@ut-emx.uucp> Sender: news@viewlogic.com (News Administrator) Reply-To: peter@viewlogic.COM (Peter Colby) Organization: Viewlogic Systems, Inc., Marlboro, MA Lines: 28 Originator: peter@rodeo In article , n67786@lehtori.tut.fi (Nieminen Tero) writes: |> Isn't it exactly same wether the OS or the programmer gets the decisision |> on things like this. BTW, how is the programmer to know he's program is |> more important than some other program, let alone make the decision. Let |> the user choose and give us preemptive multitasking and controll over |> cpu time usage. It's important to remember that a preemptive scheduler has a definite idea on the importance of particular program getting the CPU - and that this prioritization has actually been determined by the implementor of the scheduler. No scheduler is the best of all things to all programs and most implementations can only deal efficiently with certain types of utilization profiles. The whole idea behind MultiFinder is in fact that the USER determine what program gets to run the must - by bringing that program to the forground. And the whole concept of being MultiFinder friendly is that your program check frequently for user input. Even if only one program is running is still has to respond to user generated events at reasonable intervals. Forground<=>Background switching is merely a side effect of the event-loop process. Peter C -- (O)(O)(O)(O)(O)(O)(O)(O)(O) (O)(O)(O)(O)(O)(O)(O)(O)(O) (O) !the doctor is out! (O) (0) peter@viewlogic.com (0) (O)(O)(O)(O)(O)(O)(O)(O)(O) (O)(O)(O)(O)(O)(O)(O)(O)(O)