Why aren't computer clocks as accurate as cheap quartz watches?

[BillW50]

Date: Fri, 28 Oct 2005 03:15:07 GMT

How about offering some insight rather than just a big buzzer?

Depends on the version really, Win 3.1 and earlier didn't offer pre-
emptive multitasking,

All DOS applications ran under Windows 3.1 preemptively.

when an application was minimized it generally ground to a halt.

If the application doesn't want CPU time, it doesn't get it. This is
what makes cooperative tasking really great! I love cooperative tasking
when it is done right.

Win 9x was a big improvement over this but still mediocre. Win
NT/2K/XP is better still, and are generally quite good OS's,

That is your belief and my opinions are mixed. Take this 2595XDVD
running Windows 2000 with 192MB of RAM (its maxed out). And it can't
handle streaming audio/video anything faster than 100k. Yet the other
laptop, same thing except it runs Windows 98SE has no problems streaming
coming in at 800k or higher. So in this case, Windows 98 is better at
multitasking than Windows 2000/XP are.

but the multitasking is still rather poor compared to several other
OS's on the market. Of course any OS is a compromise, what you gain
in one area you often lose in another.

No in my humble opinion and experience.

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

 

[Mxsmanic]

Actually it was the other way around. As IBM black mailed into writing
OS/2. And IBM's master plan was to get everyone off of MS-DOS and on to
OS/2. Then IBM would have OS/2 changed to run on only true IBM PCs. Thus
killing off the clone market and MS as well. This was all documented and
shown on PBS.

IBM sounds a lot like Apple.

 

[BillW50]

Date: Fri, 28 Oct 2005 10:29:38 -0600

No, windows NT does not pre-emptively multitask.

Actually Windows 3.1 did preemptive multitasking for DOS applications.
Which was like a few weeks difference than OS/2 claimed to do so.

This is because it only multitasks, but it is not pre-emptive
multitasking. The kernel does not have complete control of each
application.

It depends on the Windows application. All DOS applications use
preemptive and 32-bit Windows uses preemptive. But 16-bit Windows
applications uses cooperative tasking (which in my experience is often
better than preemptive tasking anyway). This is true for Windows 3.1,
and Windows 9x. I'm not sure what happens under NT/2K/XP with 16-bit
Windows applications. As who runs 16-bit Windows applications anymore?

It is very good, but it is not pre-emptive. OS/2, for one, uses
pre-emptive and it is so far ahead and superior to the way windows
works, folks would not believe it. The difference between the two is
beyond night and day.

OS/2 sucked BIG TIME for preemptive tasking Windows 3.1 applications!
Some Windows applications crashed and burned under OS/2 when the same
ran stable as a rock under the real Windows. OS/2 often multitasked
Windows applications far slower than the real Windows OS. And that is
why preemptive tasking sucks! It often gives too much CPU time to
something that doesn't need it and not enough time for one that does
need it.

To fix the flaw with preemptive tasking, OS often includes an
application priority level that one could adjust so it behaves better
with other multitasking functions. Cooperative tasking has no need for
any of this tweaking nonsense. Plus everything in the multitasking sense
often runs faster because the stupid preemptive tasking OS isn't
screwing everything up with its added CPU overhead.

The difference will prove to be in your definition. The original
definition has been absconded with by microsoft in order to make it
appear that their inferior implementation actually meets the
requirements, so if it is really important that you 'win' that's okay
with me.

Mark

You have never mentioned cooperative tasking in anything you have
posted. Me thinks you really don't know about the different methods of
multitasking and the pros and cons of each.

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

 

[BillW50]

Date: Thu, 27 Oct 2005 12:58:06 -0500

... look for a program called D4. It is a free download and will
keep your clock synced to universal time. Also, Widows XP can sync
to the same time servers that D4 uses. Both work great!

Nobody I've seen yet thanked you for recommending this fine program.
Well I for one am very grateful! Although I usually set my computers
clocks about 5 to 10 times per year because they were off about a
minute. But now this is one task I don't have to worry about anymore.
<grin>

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

 

[Mxsmanic]

It depends on the Windows application. All DOS applications use
preemptive and 32-bit Windows uses preemptive. But 16-bit Windows
applications uses cooperative tasking (which in my experience is often
better than preemptive tasking anyway). This is true for Windows 3.1,
and Windows 9x. I'm not sure what happens under NT/2K/XP with 16-bit
Windows applications.

Sixteen-bit applications cooperatively multitask within a single NTVDM
(virtual DOS machine). The NTVDM is preemptively multitasked with
other processes in the system. This is done because 16-bit
applications often cannot tolerate preemptive multitasking; they
expect to run in systems that enforce only cooperative multitasking.
It's possible to preemptively multitask 16-bit applications by running
each of them in a separate NTVDM, though.

To fix the flaw with preemptive tasking, OS often includes an
application priority level that one could adjust so it behaves better
with other multitasking functions. Cooperative tasking has no need for
any of this tweaking nonsense. Plus everything in the multitasking sense
often runs faster because the stupid preemptive tasking OS isn't
screwing everything up with its added CPU overhead.

For what it's worth, I once wrote a communications program that
achieved unheard of line speeds on very slow PCs by using cooperative
multitasking instead of preemptive multitasking. The latter is indeed
much slower, although it's more consistent and it does compensate for
poorly written applications to some extent.

 

[Bob Masta]

Am I the only guy that was working with this crap back then? IBM
contracted with M$ to write OS/2 for them in like 1987. M$ drug their
feet on the release, while spending IBM's money, so that they could get
Win 3.0 out before OS/2, by saying that OS/2 just wasn't stable enough
for release yet. Yeah, no conflict of interest their. Finally IBM got
fed up and took the project away from M$. There are very many
suspicious similarities in "bugs" within the graphics system calls of
Win 3.0 and OS/2.

Interesting interview with Bill Gates on the whole OS/2 debacle in
PC Magazine, Nov 8, 2005 page 122-123.

Best regards.




Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Home of DaqGen, the FREEWARE signal generator

 

[Bob Masta]

All DOS applications ran under Windows 3.1 preemptively.

I hadn't heard of this before. Can you explain how it worked?
I had the impression that the DOS application took over and
Windows apps didn't get any time at all. If there were time
slices for Windows apps, do you recall how they did this?

Thanks!




Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Home of DaqGen, the FREEWARE signal generator

 

[BillW50]

Date: Sun, 30 Oct 2005 13:35:58 GMT

I hadn't heard of this before. Can you explain how it worked?

I had the impression that the DOS application took over and
Windows apps didn't get any time at all. If there were time
slices for Windows apps, do you recall how they did this?

Well the OS itself requires CPU time for itself as well. But we will
skip that part and just focus on the applications. Now Windows 3.1 (as
well as other Windows 16-bit) applications (which are cooperative
tasking) under W31/NT/W9x/W2K/XP always throws them into one single VDM
(virtual DOS machine). And if this is all of the applications running,
multitasking is generally just fine and runs very well. Although while I
have found it to be very rare, one bad cooperative application can ruin
it for the other cooperative applications.

Now add one DOS application which gets its own VDM. And this one DOS
application supposedly gets 50% of the CPU time while the total number
of the other cooperative applications shares the other 50%. Which can be
bad right? Yes it can be. But not always. As Windows 3.1, OS/2, etc.
tries to guess about these preemptive sessions when these DOS and 32-bit
applications are not really doing anything useful. Like running a
keyboard scan in a loop until it gets something. For example, WordStar
for DOS runs very well under Windows and OS/2. And it isn't using up 50%
of the CPU time in this example.

Now add a second DOS application. Thus the two DOS applications
theoretically gets 33.3% of the CPU each, and the cooperative
applications all have to share 33.3%. See how this does in those
cooperative applications?

All 16-bit applications (whether DOS or Windows) gets a VDM. Although
every DOS application gets its own VDM. 32-bit applications don't get a
VDM at all, but gets environment 32-bit subsystems instead. Which in a
weak way, might be thought like being VDMs as far as multitasking goes.

To really understand this stuff in detail, see like:

http://www.microsoft.com/technet/prodtechnol/Windows2000Pro/reskit/part8/proch36.mspx

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

 

[Mxsmanic]

Interesting interview with Bill Gates on the whole OS/2 debacle in
PC Magazine, Nov 8, 2005 page 122-123.

What I find most interesting is that November 8 is still over a week
in the future.

 

[Anthony Fremont]

I hadn't heard of this before. Can you explain how it worked?
I had the impression that the DOS application took over and
Windows apps didn't get any time at all. If there were time
slices for Windows apps, do you recall how they did this?

It was basically a form of cooperative multitasking. When the DOS app
called INT21 functions or made BIOS calls, Windos could then regain
control of the machine. Hardly what I'd call preemptive multitasking.

It is my opinion that even XP doesn't qualify as a proper OS. Any OS
that allows an errant application to hang things up is not right.

 

[BillW50]

Date: Sun, 30 Oct 2005 11:41:44 GMT

Am I the only guy that was working with this crap back then?
Nope!

IBM contracted with M$ to write OS/2 for them in like 1987.

It might have been in '86 actually. And MS had been working on Windows
since about '84. Although MS couldn't give the development time it
deserved because those MS programmers were mostly working on OS/2. MS
lost 3 years in Windows development because of OS/2.

M$ drug their feet on the release, while spending IBM's money, so
that they could get Win 3.0 out before OS/2, by saying that OS/2
just wasn't stable enough for release yet. Yeah, no conflict of
interest their.

IBM only paid MS for the lines of code MS produced. IBM didn't care if
MS spent more time to make the code lean, mean and faster. As IBM would
pay you less if you did so. IBM was cutting their own throats. IBM is
full of a much of morons. Impossible to work with and to get paid fairly
for. Hell I would work slowly and drag my feet as well for those morons.

Finally IBM got fed up and took the project away from M$.

Yeah, IBM got fed up alright! As Microsoft didn't want to be a slave to
IBM (who always makes slaves or crushes anybody that gets in their way
up to this point in time). And IBM wanted MS to create OS/2 which would
be made to run on only true IBM PCs after they have the world hooked on
OS/2.

Yeah that is a great plan for us, NOT! Bill Gates had taken the biggest
risk in his career. As nobody ever bucked IBM and had survived. Although
he did it! And thank goodness he did! As we all would be using real IBM
machines and OS/2 by now.

Sure IBM was ticked that Bill Gates wasn't going to play along. So they
parted ways. And IBM wouldn't sell any IBM computer with Windows
installed for a short time. Until IBM realized that they couldn't sell
IBM computers with either crappy PC-DOS or OS/2 on them. As
people wanted Windows instead, plain and simple.

There are very many suspicious similarities in "bugs" within the
graphics system calls of Win 3.0 and OS/2.

The same MS programmers wrote both OS/2 and Windows 3.0. So why should
this be a surprise?

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

 

[Mxsmanic]

It is my opinion that even XP doesn't qualify as a proper OS. Any OS
that allows an errant application to hang things up is not right.

XP does not allow applications to do that, unless they have the proper
privileges.

Unfortunately, many Windows applications today won't run without
elaborate privileges, and if they contain bugs, they can hang the
system. That's not the fault of the OS; if you tell it to run an
application as the administrator, it will, and all bets are off.

Even so, modern Windows applications are generally extremely stable,
and XP is even more stable still. I can't remember the last time I
saw an XP system crash. If the hardware fails, it may crash. A bad
driver can still crash it in certain situations. But that's it. Even
the Windows Explorer, a bastion of fragile instability when it was
first transplanted from Windows 95 into Windows NT 4.0, now rarely if
ever causes any problems.

Come to think of it, not only can I not remember the last time I saw
an XP system crash, I can't remember the last time I saw it lock up.

 

[Anthony Fremont]

Date: Sun, 30 Oct 2005 11:41:44 GMT


It might have been in '86 actually. And MS had been working on Windows
since about '84. Although MS couldn't give the development time it
deserved because those MS programmers were mostly working on OS/2. MS
lost 3 years in Windows development because of OS/2.

I suppose that's one way to look at the time that M$ spent sucking money
from IBM and using it for their own gains.

IBM only paid MS for the lines of code MS produced. IBM didn't care if
MS spent more time to make the code lean, mean and faster. As IBM

would

I think IBM had visions of stability that M$ will never attain, ever.

pay you less if you did so. IBM was cutting their own throats. IBM is
full of a much of morons. Impossible to work with and to get paid fairly
for. Hell I would work slowly and drag my feet as well for those

morons.

Yeah, morons. They only own the mainframe market even though Honeywell
made better hardware. IBM's only moronic move was to allow M$ to screw
them for a second time. The first time being with MSDOS/IBMDOS games.

Yeah, IBM got fed up alright! As Microsoft didn't want to be a slave to
IBM (who always makes slaves or crushes anybody that gets in their way

Too bad that isn't true since they would have done the world a great
favor by crushing M$.

up to this point in time). And IBM wanted MS to create OS/2 which would
be made to run on only true IBM PCs after they have the world hooked on
OS/2.

Yeah that is a great plan for us, NOT! Bill Gates had taken the biggest
risk in his career. As nobody ever bucked IBM and had survived. Although
he did it! And thank goodness he did! As we all would be using real IBM
machines and OS/2 by now.

Actually, if Gates wasn't so good at being greedy, we'd all be using
something that actually worked. OS/2 was crap too. Too bad Xerox
didn't have sense enough to stay in the game, they had the best product
for the office in 1980. Apple didn't have anything that could come
close for around 10 years. It took M$ almost another 5 years on top of
that to catch up.

Sure IBM was ticked that Bill Gates wasn't going to play along. So they
parted ways. And IBM wouldn't sell any IBM computer with Windows
installed for a short time. Until IBM realized that they couldn't sell
IBM computers with either crappy PC-DOS or OS/2 on them. As
people wanted Windows instead, plain and simple.

The only reason being that M$ delayed OS/2 was so that Win 3.0 could get
the jump on it. If OS/2 would have shipped on time, it would have
possibly eliminated windows.

The same MS programmers wrote both OS/2 and Windows 3.0. So why should
this be a surprise?

It's not a surprise to me. I think it just goes to show that M$ had no
qualms about directly lifting the code that they originally wrote for
IBM using IBM's money and, AFAICT, IBM's design goals. I'm not saying
that was illegal back then, but it certainly wouldn't happen in today's
IP obsessed world without bringing about major court battles.

Here was a true visionary: http://www.cadigital.com/kildall.htm

You obviously really like M$ so there probably isn't much point in
continuing this until it becomes a real pissing contest. I run windos
on some machines because I basically have to. When I need something
that really works, I use Linux.

 

[BillW50]

Date: Sun, 30 Oct 2005 15:02:52 GMT

It was basically a form of cooperative multitasking.

A form of cooperative tasking my eye! Each VDM session uses the Intel
v86 (Virtual-8086) mode. Windows 3.1 and later as well as OS/2 uses v86
mode to preemptive task DOS and other VDM sessions.

When the DOS app called INT21 functions or made BIOS calls, Windos
could then regain control of the machine. Hardly what I'd call
preemptive multitasking.

Doesn't sound right to me. As Windows uses (since W3.1) the CPUs v86
mode (Task State Segments) to support multitasking to preemptive task
all VDM sessions through Windows Virtual Machine Manager (VMM). Every
manual I have ever read (and I just searched the Internet as well) calls
this preemptive tasking. Looks like you're alone to me.

It is my opinion that even XP doesn't qualify as a proper OS. Any
OS that allows an errant application to hang things up is not
right.

Under any x86 machine, any buggy ring 0 code can take out any OS, bar
none! It's not just a Windows limitation, but effects *all* OS. Yes, any
clever programmer who wants to take out an x86 machine running any OS
can indeed do so (with administrator privileges of course).

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

 

[Anthony Fremont]

XP does not allow applications to do that, unless they have the proper
privileges.

That's what they say, but.......

Unfortunately, many Windows applications today won't run without
elaborate privileges, and if they contain bugs, they can hang the
system. That's not the fault of the OS; if you tell it to run an
application as the administrator, it will, and all bets are off.

Right, you don't really have much choice but to use the machine as an
admin. I log into Linux all the time as root though, and I run plenty
of bad code as root and it promptly segfaults and that's basically it.
You'd have to go to pretty good lengths to write code that would hang
Linux just because you ran it as root. Hanging the kernel is primarily
accomplished by device drivers, which are running in kernel space, so
all bets are really off there. My point is that hanging windows is
allot easier. On Linux it's fairly tricky just getting into position to
be able to start slapping the kernel around unless you're a device
driver of course.

Even so, modern Windows applications are generally extremely stable,

I'm not sure I really agree with that. It's probably a point of view
kinda thing. My background is in the mainframe world originally doing
online TP, so my definition of stability tends to be different from many
people. The same goes for security. Even Linux upsets me greatly at
times, especially MythTV and the ivtv driver. But that tends to be the
fault of the third party programmers and not the Linux kernel.

and XP is even more stable still. I can't remember the last time I
saw an XP system crash. If the hardware fails, it may crash. A bad

I can't fault the OS if hardware dies but, depending upon the particular
hardware, the driver might be graceful about it.

driver can still crash it in certain situations. But that's it. Even
the Windows Explorer, a bastion of fragile instability when it was
first transplanted from Windows 95 into Windows NT 4.0, now rarely if
ever causes any problems.

Come to think of it, not only can I not remember the last time I saw
an XP system crash, I can't remember the last time I saw it lock up.

Unfortunately, I can. Granted XP is better than their previous
offerings, but that's like saying it's better than a poke in the eye.
;-)

 

[Anthony Fremont]

Date: Sun, 30 Oct 2005 15:02:52 GMT

A form of cooperative tasking my eye! Each VDM session uses the Intel
v86 (Virtual-8086) mode. Windows 3.1 and later as well as OS/2 uses v86
mode to preemptive task DOS and other VDM sessions.


Doesn't sound right to me. As Windows uses (since W3.1) the CPUs v86
mode (Task State Segments) to support multitasking to preemptive task

Well sure it has a TSS, otherwise you couldn't switch tasks very easily.
The TSS holds all the context information (registers, pc, ldt etc...)
required to put the task back into execution without it freaking out.
It's just a mechanism to make it easy, but it doesn't magically
interrupt a running task.

Windows could use the timer tick ints to accomplish task switching or it
could even set up another spare timer to generate interrupts for task
switching. Those would then safestore the TSS values for the running
task when the interrupt occurs and then transfer control thru the
interrupt vector to the dispatcher (or whatever M$ calls it). Using a
timer of some sort could make it preemptive since they could then
conceivably interrupt between any two instructions. AFAIK though, they
just depended upon the system calls to resume control.

all VDM sessions through Windows Virtual Machine Manager (VMM). Every
manual I have ever read (and I just searched the Internet as well) calls
this preemptive tasking. Looks like you're alone to me.


Under any x86 machine, any buggy ring 0 code can take out any OS, bar
none! It's not just a Windows limitation, but effects *all* OS. Yes, any
clever programmer who wants to take out an x86 machine running any OS
can indeed do so (with administrator privileges of course).

That's why there is 4 ring levels supported in hardware. Too bad M$
doesn't utilize it properly, Linux wins hands down here and only uses 2
of the 4 levels.

 

[Mxsmanic]

That's what they say, but.......

But it's true.

Right, you don't really have much choice but to use the machine as an
admin.

That's not the fault of the OS. There are some applications that will
run without special privileges.

You'd have to go to pretty good lengths to write code that would hang
Linux just because you ran it as root.

No more so than for XP.

Hanging the kernel is primarily
accomplished by device drivers, which are running in kernel space, so
all bets are really off there.

The same is true for XP.

My point is that hanging windows is allot easier.

Except that it's not.

On Linux it's fairly tricky just getting into position to
be able to start slapping the kernel around unless you're a device
driver of course.

If you're running a GUI, it's easy.

I'm not sure I really agree with that. It's probably a point of view
kinda thing. My background is in the mainframe world originally doing
online TP, so my definition of stability tends to be different from many
people.

I have the same background. XP is stable.

The same goes for security. Even Linux upsets me greatly at
times, especially MythTV and the ivtv driver. But that tends to be the
fault of the third party programmers and not the Linux kernel.

Linux and UNIX are quite insecure, compared to NT.

I can't fault the OS if hardware dies but, depending upon the particular
hardware, the driver might be graceful about it.

The driver is usually written by the hardware vendor. Many drivers
are very poorly written.

 

[Jamie]

Date: 28 Oct 2005 13:35:00 -0700




You mean hardly useful! And IBM dropped support a few months before they
were saying they would never drop OS/2 support. IBM has never done
anything except lie to me over and over again.




I did a search through OS/2 files for the Microsoft copyright in Warp a
few years ago. And Warp was littered everywhere with Microsoft's code
throughout OS/2.

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

excuse me, if memory serves IBM had MS write the first OS which i think
was for the 286 and thought that it would be alive for a long time.
then when 386 hit they tried to get MS to rewrite it for them and thus
ms quoted such a high price just to get them to go away thus leading
the way for MS to where they are now. mean while IBM then took over
the development to carry it on with their own programmers.
that is the way i remember it.

 

[Mxsmanic]

That's why there is 4 ring levels supported in hardware. Too bad M$
doesn't utilize it properly, Linux wins hands down here and only uses 2
of the 4 levels.

NT-based versions of Windows are much more secure than Linux or UNIX.
You only need hardware support for two levels or privilege.

 

[Jamie]

Date: Fri, 28 Oct 2005 10:29:38 -0600





You have never mentioned cooperative tasking in anything you have
posted. Me thinks you really don't know about the different methods of
multitasking and the pros and cons of each.

______________________________________________
Bill (using a Toshiba 2595XDVD & Windows 2000)
-- written and edited within Word 2000

Ha., i would be careful making statements like that.
i cut out most of the original poster since you
were trying very hard to out smart him (publicly).
there are a couple of things you made an error on
which i won't get into.