Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!cbmvax!daveh From: daveh@cbmvax.commodore.com (Dave Haynie) Newsgroups: comp.sys.amiga Subject: Re: Multitasking at home is great!! (Was Reality check: ....) Message-ID: <17210@cbmvax.commodore.com> Date: 8 Jan 91 03:55:35 GMT References: <1990Dec13.155848.8152@maytag.waterloo.edu> <1990Dec22.082240.2443@news.iastate.edu> <26060@uflorida.cis.ufl.EDU> <17193@cbmvax.commodore.com> Reply-To: daveh@cbmvax.commodore.com (Dave Haynie) Organization: Commodore, West Chester, PA Lines: 51 In article mwm@raven.relay.pa.dec.com (Mike (My Watch Has Windows) Meyer) writes: >In article <17193@cbmvax.commodore.com> daveh@cbmvax.commodore.com (Dave Haynie) writes: > It's much more far reaching than that. For example, Personal Workstation > magazine equates "true" with "preemptive" in much of their writings on > multitasking. >So what's your point? That folks far outside the Amiga community don't consider non-preemptive task switching to be true multitasking. After all, it is simply a matter of how you define multitasking. It is simple to define the properties of "true multitasking" to either include or exclude non-preemptive systems. >So how does any of this make non-preemptive multitasking not "true" For example, _my_ definition of "multitasking" may have as one of its properties: A completely CPU-bound task at priority N may not completely block tasks at priority N+d, where d>=1. Preemptive systems will pass this test, non-preemptive systems will fail it. Since there is no single, accepted definition of "multitasking", what will or won't constitute "true" multitasking depends on who's definition you apply. A very loose definition might include user-initiated multitasking systems, such as the Macintosh "Switcher" type programs. Or possibly even things like Mac's desk assessories or MS-DOS "TSRs". Both of those are, in fact, forms of multitasking, albeit more limited than the user-initiated switcher, which is more limited than the function call initiated switcher, which is more limited than the hardware-interrupt initiated switcher. Do you draw a line as to which is "real" and which isn't? If you don't, then you're at least being consistent. If you are, then we're (for the sake of argument) simply at odds over where to draw the line, not whether a line should be drawn. Which must fall back on our personal definition of the properties of "true" vs. "false", "simulated", "pseudo", whatever-you-call-it, multitasking. Again, my original point was that a much larger segment of the industry counts non-preemptive multitasking as not quite "true" multitasking. It wasn't the Amiga folks who made it up, though you are correct in that some of them do tend to yell too loudly about it simply to make themselves feel superior to Machine X, often without really understanding the real differences. >