[comp.sys.mac.system] Turning off time slices in S7

triantos@acsu.buffalo.edu (Nick B Triantos) (05/30/91)

Dear World,

I was thinking...Since S7 is always MultiFinder, aren't apps such as Solarian,
and other games, and many other things which are heavily dependent on timing,
going to suffer?  On my SE, I can't play them anyway, but it seems like there
will be problems.  Even when my screen saver goes on, I notice that things are
*REALLY* creeping.  Any ideas?

Thanks.
Nick Triantos
--
  __      |    _____              |           University at Buffalo
 /  \ o _ |/  /  |  \ _ o __  __ -+- __  ___  triantos@acsu.buffalo.edu
|    ||/  |\     |   / \|/  |/  \ | /  \(___  
\    /|\_ | \    |   |  |\_/||  | \_\__/ ___)   "Bus Error: Passengers Dumped."

tonyrich@titanic.cs.wisc.edu (Anthony Rich) (05/31/91)

Nick B. Triantos writes:

>I was thinking...Since S7 is always MultiFinder, aren't apps such as Solarian,
>and other games, and many other things which are heavily dependent on timing
>going to suffer?

Yeah, I think so.  Color Tetris under system 7.0 got jerky on my plain
old Mac II after I installed SuperClock 3.9 -- so I dumped SuperClock. :(
Even with without SuperClock installed, Tetris is slightly slower and less
responsive under 7.0 than it was under 6.0.5 and MultiFinder.

I think the added features of System 7 are starting to push the limits of what
the lower-end Mac hardware (anything less than a 68030?) can support without
noticeable delays.  Small delays are only annoying in many situations, but
they can spoil real-time programs like arcade games.  (Hmmm...so we have to
buy a IIfx to play games, is that it?  What a marketing strategy...! ;^)

So what do System 7 users do to get "Finder-only" performance for apps that
need it?  We can still reboot with Finder and 6.0.x, of course, but I don't
think Apple intends that we keep doing that.  (I can just hear the snickers
from the DOS/Windows camp:  "How many reboots does it take to run a game on
a Mac?", etc., etc.)  What is needed is a "Single App Mode" toggle that would
temporarily turn off much of the System 7 overhead and allow only one app to
run while still booted under System 7.  I'll bet it wouldn't be an easy thing
to implement, though.

lsr@Apple.COM (Larry Rosenstein) (06/01/91)

In article <1991May31.030742.10637@spool.cs.wisc.edu> tonyrich@titanic.cs.wisc.edu (Anthony Rich) writes:

>a Mac?", etc., etc.)  What is needed is a "Single App Mode" toggle that would
>temporarily turn off much of the System 7 overhead and allow only one app to
>run while still booted under System 7.  I'll bet it wouldn't be an easy thing
>to implement, though.

It is straightforward to write a program that sends a quit Apple event to
all running apps (including the Finder) and then launches another app such
as Tetris (killing itself in the process).  When Tetris quits the Finder
would restart.


-- 
Larry Rosenstein, Apple Computer, Inc.

lsr@apple.com
(or AppleLink: Rosenstein1)

glenn@gla-aux.uucp (Glenn Austin) (06/01/91)

In article <TRIANTOS.91May29214004@lictor.acsu.buffalo.edu>, triantos@acsu.buffalo.edu (Nick B Triantos) writes:
> Dear World,
> 
> I was thinking...Since S7 is always MultiFinder, aren't apps such as Solarian,
> and other games, and many other things which are heavily dependent on timing,
> going to suffer?  On my SE, I can't play them anyway, but it seems like there
> will be problems.  Even when my screen saver goes on, I notice that things are
> *REALLY* creeping.  Any ideas?

Well, considering the foreground application has control over when the application
is switched out, this shouldn't be a problem.  This IS a problem on time-slicing
machines, like the PC, and means another level of technical information
that the user has to know.  Isn't it nice to work on the Mac?

===============================================================================
| Glenn L. Austin                | "Turn too soon, run out of room.           |
| Macintosh Wizard and           |    Turn too late, much better fate."       |
| Auto Racing Driver             |   -- Jim Russell Racing School Instructors |
|-----------------------------------------------------------------------------|
| Don't take me too seriously -- I never do!  :-)                             |
|-----------------------------------------------------------------------------|
| Usenet:  glenn@gla-aux.uucp or glenn%gla-aux.uucp@skinner.cs.uoregon.edu    |
===============================================================================

tonyrich@titanic.cs.wisc.edu (Anthony Rich) (06/01/91)

Larry Rosenstein writes:

> It is straightforward to write a program that sends a quit Apple event to
> all running apps (including the Finder) and then launches [a time-critical]
> app such as Tetris [...].

OK, but what about killing live background extensions like SuperClock that
aren't "launched apps" in the usual sense?  Maybe they shouldn't be "killed",
just "put on hold" until the time-critical foreground app is finished.
INITs/CDEVs shouldn't be killed, then require manual re-launching later.

Glenn Austin writes:

> Well, considering the foreground application has control over when
> the application is switched out, [switchouts] shouldn't be a problem.

As I understand it, an app has control over *when* it's switched out, but not
for *how long*--so I don't think cooperative multitasking solves the problem.
It's considered "Multifinder friendly" for an app to give up the CPU once
in a while so that other tasks like modem downloads don't timeout.  The hard
part for a time-critical app is getting the CPU *back* quickly enough.

An interesting solution might be to have separate foreground and background
processors.  That way, the foreground app wouldn't need to be interrupted
to handle the background stuff.  If a user interface should give the user
the illusion that the computer is always paying attention, why not dedicate
a processor to maintaining that illusion?

lsr@Apple.COM (Larry Rosenstein) (06/04/91)

In article <1991Jun1.070626.2853@spool.cs.wisc.edu> tonyrich@titanic.cs.wisc.edu (Anthony Rich) writes:
>
>OK, but what about killing live background extensions like SuperClock that
>aren't "launched apps" in the usual sense?  Maybe they shouldn't be "killed",

If you've killed all other apps besides Tetris (for example), then this is
no different than System 6 with MultiFinder turned off.  There's no general
way to put these things on hold.



-- 
Larry Rosenstein, Apple Computer, Inc.

lsr@apple.com
(or AppleLink: Rosenstein1)

ashwin@gatech.edu (Ashwin Ram) (06/04/91)

In article <13842@goofy.Apple.COM> lsr@Apple.COM (Larry Rosenstein) writes:
>   In article <1991Jun1.070626.2853@spool.cs.wisc.edu> tonyrich@titanic.cs.wisc.edu (Anthony Rich) writes:
>   >
>   >OK, but what about killing live background extensions like SuperClock that
>   >aren't "launched apps" in the usual sense?  Maybe they shouldn't be "killed",
>   
>   If you've killed all other apps besides Tetris (for example), then this is
>   no different than System 6 with MultiFinder turned off.  There's no general
>   way to put these things on hold.

How about giving (settable) priorities to different tasks, like in other
multiprocessing systems?

-- Ashwin.

umh@vax5.cit.cornell.edu (06/04/91)

In article <1991Jun3.142717.103@magnus.acs.ohio-state.edu>,
scasterg@magnus.acs.ohio-state.edu (Stuart M Castergine) writes: 
> In article <1991Jun3.015035.5222@vax5.cit.cornell.edu> umh@vax5.cit.cornell.edu writes:
> 
>>And this brings us into the brave new world of multithreaded programs and real
>>operating systems.
> 
> Sigh. Another "what a terrible operating systeem is MacOS" message. :-)

This dow not really deserve a reply except that I keep seeing this kind of junk
and it irritates me.
THE USER-INTERFACE IS NOT THE OPERATING SYSTEM !!!!!! Repeat 1000 times.

The operating system is the guts of the machine never seen by the users. It
is the software that handles how system resources (CPU time, disk space,
network access, memory, screen space etc) will be allocated among different
contenders, ie among different processes, DAs, INITs etc (in the Mac context).
The Mac OS is considered superior to DOS as an OS because it does a superior
job of such resource management- it has a more flexible file scheme, better
handling of the screen, better memory management etc.

The Mac Toolbox is software that gives the Mac its look and feel. This is *not*
whate we are arguing about. Neither are we arguing about whether or not
MultiFinder is "real" multitasking.

What I am pointing out is that, by the standards of modern operating systems,
the Mac OS is severly deficient. It does not offer protected memory- if one
program crashe, it can crash the entire machine. It does not offer threads-
allowing you to do other thing *in the Finder* while disk copying is going on,
for example. It does not offer support for multiple users on one machine- this
is useful *even if* you are the only user of that machine, to ensure that you
don't do something rash by mistake. It does not offer support for parallel
processing etc etc.

Now a whole bunch of people are going to say that we, common users, don't need
that. This is a load of bull. It doesn't matter what you dow ith your computer-
everyone benefits form a robust environment where the crash of one program does
not affect anything else. Everyone benfits from an environment where time
intensive jobs are shuffled to the background in a separate thread while the ap
you're using remains interactive. Everyone benefits from the extra CPU
performance parallel processing can bring. Have you ever tried to manipulate 2
page 24bit graphics? Wouldn't it be nice if that took half a second instead of
30 seconds. Have you ever tried editing sound? Printing out to a quickdraw
printer?

Even things the MacOS does like multi-tasking could be speeded up a great deal
if the OS were changed in some way. Context switches on a Mac are *slow* by the
stanadards of many other OSs.

Now, what of the competition. Everyone shrinks back at the thought of UNIX.
Dammit- we are interested in UNIX the OS here, not UNIX the shell. The fact is
that as an operating system, UNIX is fundamentally more powerful than the
MacOS. Mach is a kernel based on networking/parallel processing concepts that
is rather more powerful than UNIX, can be customized rather more easily for a
certain task, and can be made to look like UNIX to a programmer- and like UNIX,
to look like anything you want including a Mac, to the user.

Look at AIX. When AIX runs, it takes over the machine- the MacOS is killed.
That doesn't stop you running Mac programs that behave just like always- just
they're using UNIX as their OS, not the native Mac OS, but they still use the
Mac Toolbox.

There are a few innovative features in the MacOS. Things like resources and
parts of Quickdraw. But most of it is not nearly as good as what is available,
and it makes no sense to claim either that it is, or that we do not need the
improvements.

Maynard Handley

philip@pescadero.Stanford.EDU (Philip Machanick) (06/05/91)

In article <1991Jun3.232340.5237@vax5.cit.cornell.edu>, umh@vax5.cit.cornell.edu writes:
[Lots of stuff about limits of the Mac OS, as distinct from the user
interface]
|> Now, what of the competition. Everyone shrinks back at the thought of UNIX.
|> Dammit- we are interested in UNIX the OS here, not UNIX the shell. The fact is
|> that as an operating system, UNIX is fundamentally more powerful than the
|> MacOS. Mach is a kernel based on networking/parallel processing concepts that
|> is rather more powerful than UNIX, can be customized rather more easily for a
|> certain task, and can be made to look like UNIX to a programmer- and like UNIX,
|> to look like anything you want including a Mac, to the user.

I agree with everything you said up to this paragraph. Certainly, unix
does everything you have pointed out as missing from the Mac OS, but does
it really do these things _well_? unix was designed for small minis, and
has grown out of proportion. I think Apple ought to be able to do better
starting from scratch. This doesn't mean they shouldn't also support
unix compatibility, in the same way as A/UX supports Mac compatibility.
My ideal future OS would be capable of running Mac, A/UX and its new native
mode applications.

|> There are a few innovative features in the MacOS. Things like resources and
|> parts of Quickdraw. But most of it is not nearly as good as what is available,
|> and it makes no sense to claim either that it is, or that we do not need the
|> improvements.

Again, I agree that it's silly to pretend the Mac OS can't be
improved (or better still, completely rewritten). However, we
should also remember that there are some parts of unix that would
be a big step backwards from the Mac.
-- 
Philip Machanick
philip@pescadero.stanford.edu

neeri@iis.ethz.ch (Matthias Ulrich Neeracher) (06/05/91)

In article <1991Jun3.232340.5237@vax5.cit.cornell.edu> umh@vax5.cit.cornell.edu writes:
>What I am pointing out is that, by the standards of modern operating systems,
                                                     ^^^^^^^^^^^^^^^^^^^^^^^^
>the Mac OS is severly deficient.

The Mac OS is more "modern" than UNIX :-)

>It does not offer protected memory- if one
>program crashes, it can crash the entire machine.

This might be introduced anytime on the machines with an MMU.

>It does not offer threads-
>allowing you to do other thing *in the Finder* while disk copying is going on,
>for example.

While I haven't received my copy of Inside Macintosh VI yet, I *think* this
is no longer entirely impossible (The finder could launch an "invisible
application" to do the disk copying. I hazily remember having seen a
Process Manager chapter in the IM VI stack).

>It does not offer support for multiple users on one machine- this
>is useful *even if* you are the only user of that machine, to ensure that you
>don't do something rash by mistake.

I was wondering whether it wouldn't be possible, now with the FileShare feature
of System 7, to configure a Mac such that the User has to mount the *local* hard
disk as an AppleShare volume. This would provide the protection for files.

>It does not offer support for parallel
>processing

Yes it does. That's what A/ROSE does :-) Seriously, I have seen Macs with
transputer cards. Under System 7, access to the file system for other processors
should be possible to arrange (FileShare again).

>Now a whole bunch of people are going to say that we, common users, don't need
>that. This is a load of bull. It doesn't matter what you dow ith your computer-
>everyone benefits form a robust environment where the crash of one program does
>not affect anything else. Everyone benfits from an environment where time
>intensive jobs are shuffled to the background in a separate thread while the ap
>you're using remains interactive.

I agree fully.

>Everyone benefits from the extra CPU
>performance parallel processing can bring. Have you ever tried to manipulate 2
>page 24bit graphics? Wouldn't it be nice if that took half a second instead of
>30 seconds. Have you ever tried editing sound? Printing out to a quickdraw
>printer?

Some of these tasks sound like the job for a dedicated graphics coprocessor.
This has been done. Others sound like the job for multiple processors (not
necessarily 680X0'es) with the Mac acting as a frontend. This has also been
done. 

>Even things the MacOS does like multi-tasking could be speeded up a great deal
>if the OS were changed in some way. Context switches on a Mac are *slow* by the
>stanadards of many other OSs.

Sure, and graphics is *fast* on a Mac by the standards of many other OSs.

>Now, what of the competition. Everyone shrinks back at the thought of UNIX.
>Dammit- we are interested in UNIX the OS here, not UNIX the shell.

Which UNIX ?

>The fact is
>that as an operating system, UNIX is fundamentally more powerful than the
>MacOS.

... and fundamentally more weird. System VR4 supports 14 different file
systems. On many UNIXes, to install a new device driver, or change some
configuration parameters, you have to recompile the kernel.  While I like some
of the UNIX ideas, I think today's UNIX kernels are much too complex, partially
because they have to provide too much backwards compatibility.

>Mach is a kernel based on networking/parallel processing concepts that
>is rather more powerful than UNIX, can be customized rather more easily for a
>certain task, and can be made to look like UNIX to a programmer- and like UNIX,
>to look like anything you want including a Mac, to the user.

I like Mach a lot better than UNIX.

>Look at AIX. When AIX runs, it takes over the machine- the MacOS is killed.
>That doesn't stop you running Mac programs that behave just like always- just
>they're using UNIX as their OS, not the native Mac OS, but they still use the
>Mac Toolbox.

Finally some good news for IBM ! Mac programs run on an IBM RS/6000, you say ?
Oh, you meant A/UX, not AIX ? Never mind. (FYI, I *like* A/UX)

>There are a few innovative features in the MacOS. Things like resources and
>parts of Quickdraw. But most of it is not nearly as good as what is available,

As far as I heard, the Hierarchical File System and AppleTalk are also good.
The International Utilities/Script Manager are *very* good (How many western
computers can be booted up with Hebrew, Arab, Korean and Chinese systems ?).
AppleEvents are good, Aliases are good, the Sound Manager is pretty good. So
what is "not nearly as good as what is available" ? Multiprocessing, OK, and
MS-DOS compatibility.

>and it makes no sense to claim either that it is,

Most of the Toolbox *is* state of the art. Repent, heathen !

>or that we do not need the
>improvements.

Certainly, we need the improvements. That's why Apple is constantly improving
its System Software. What you seem to underestimate in your post is that "the
MacOS" is a moving target. To me, "the MacOS" is everything currently running
on a Mac that will let me run my Applications. Therefore, I find it ridiculous
to pick on "the MacOS", pointing to A/UX. A/UX *is* a MacOS, and probably shows
some of the further directions the "mainstream" System Software is taking.

>Maynard Handley

Matthias

-----
Matthias Neeracher                                      neeri@iis.ethz.ch
   "These days, though, you have to be pretty technical before you can 
    even aspire to crudeness." -- William Gibson, _Johnny Mnemonic_

Irv.Kanode@p3.f53.n114.z1.fidonet.org (Irv Kanode) (06/05/91)

> So what do System 7 users do to get "Finder-only" performance for apps that
> need it?  We can still reboot with Finder and 6.0.x, of course, but I don't
> think Apple intends that we keep doing that.  (I can just hear the snickers
> from the DOS/Windows camp:  "How many reboots does it take to run a game on
> a Mac?", etc., etc.)  What is needed is a "Single App Mode" toggle that 
would
> temporarily turn off much of the System 7 overhead and allow only one app
to
> run while still booted under System 7.  I'll bet it wouldn't be an easy 
thing
> to implement, though.
> 

I don't mind the System 7 overhead but I need a way to prevent kids in the 
lower
elementary grades from getting to the desktop by using "multifinder" menu.

Anyone have any ideas?

 

--  
Uucp: ...{gatech,ames,rutgers}!ncar!asuvax!stjhmc!53.3!Irv.Kanode
Internet: Irv.Kanode@p3.f53.n114.z1.fidonet.org