Windows NT vs OS/2 2.1

Windows NT vs OS/2 2.1

Post by Wayne J. Hy » Mon, 16 Aug 1993 20:22:42



I asked someone to post my review since I was having
problems with my news program.  Now my problem is
solved.  I've changed a few things in my review and added
others.  Well, here it is:

***********************************************************
Ok!  Here is the review of NT & OS2 that I promised.  Keep
in mind that these results were taken from *MY* testing
of NT on *MY* machine.  For those who need to know again,
my machine configuration is as follows:

        Gateway 2000 486/66, 256k cache, 16 meg SIMM.
        ATI VLB GUP w/ 2 megs
        2, WD Caviar 340 meg IDE drives.
             -- NT resides on a 160 meg NTFS partition
     ** Logitech Bus Mouse
     ** Future Domain 860M SCSI card
        TeXEL 3024 CDROM
        Jumbo 250

** For some reason, NT assumes that the FD8xx controllers
use IRQ 5.  I have mine set to IRQ 11 or 12 with my bus
mouse on IRQ 5, so NT didn't recognize my mouse or SCSI
card at first.  I had to manually change the IRQ setting
under the registry to get both to work.  

NOTE:  I *REALLY* tried to give NT a fair shot at
winning over my system.  If NT turned out to be
better than OS/2, I would have purchased NT for myself.

[I was using a friend's upgrade from the beta since he
didn't have a system to run NT on and I happily
volunteered to let him test it on mine]

During my tests, I tried to run the same programs on
both operating systems to keep everything consistent.

Example:
When I used NT's Paintbrush to load the large (2 meg) BMP
that hung the system for a while, I ran the Paintbrush from
win3.1 seamlessly under OS/2.

To simulate heavy loads, I ran QRT to hog CPU time.

I don't think benchmarking multitasking OS's by running
a single application and timing them is an accurate
measure of an OS's capabilities.  Magazines that
benchmark WIN-OS/2 vs Win 3.1 by timing how long it
takes to run a macro really irk me.  I am sure Windows 3.1
will outperform WIN-OS/2 in running a single task--OS/2 is
doing other things invisible to the user.  Run a few
apps at the same time to see how things go.    

************
Winbench 3.1
(I closed all applications and minimized the ProgMan)

OS/2 2.1 using 8514/A drivers from 2.1 1024x768x256

Mode             Score         Elapsed Time
----------       ---------     ------------
FullScreen   :   6,799,416     63 seconds
Seamless     :   4,578,050     65 seconds

While running a single QRT raytrace:
FullScreen   :   3,872,493     66 seconds
Seamless     :   2,649,938     70 seconds

NT using ATI 1024x768x256 drivers.

"Seamless"   :   4,110,144     66 seconds

While running a QRT trace:

"Seamless"   :   2,953,394     75 seconds

You can make any judgement from this data that you wish.
HW_TIMER was on for OS/2.

Seamless windows under OS/2 do seem _slightly_ slower than
"normal" speed on my system.

NT ran win16 apps just as well as OS/2 from what I could tell.
Some win16 apps ran faster under OS/2 than NT.  Others ran at
the same speed.
************

DOS Compatibility and speed
---------------------------

Wolf3D
------
Wolf3D runs BL2.  (Faster than DOS).
Wolf3D also ran very well under NT, but slightly slower (~15%).

With 1 QRT running, Wolf3D still ran well under OS/2.  ~5% reduction
in speed.  NT, on the other hand, was very jerky and slow. ~40% slower.

Dune2
-----
Just about the same results as Wolf3d.

Qmodem
------
During a high speed (16.8k) transfer, OS2 and NT performed well.
2000+ cps under both OS's.

Tape Backup software (TAPE.EXE)
----
Tape ran well under OS/2.  I could restore backups while multitasking
without any problems.  While using my modem I did lose characters,
but I still have a 16450 UART (getting a 16550AFN on monday).
Tape did not run under NT.  NT doesn't allow it.

MOD/MIDI players for my GUS
-----
Run fine under OS/2.  Music pauses during heavy multitasking. I don't
have any native GUS/MIDI players, so this should go away once they are
available.
Don't work under NT.   (NT [homey] don't play that!)  ;)

**************************
Stress testing:                      

I loaded up many applications at once and compared
how responsive the two OS's stayed.  I used the applets
that came with the respective OS.  I also ran a full
screen WIN-OS/2 session under OS/2.

Under NT, I could load a few applications without any
slowdowns.  I played with File Manager, the System
Administration Tools, and other programs.  NT _DID_
start to swap immediately, but this wasn't a problem -- yet.
After I started loading more apps, swapping became a problem.
When NT isn't swapping, switching between processes is pretty
fast.  When NT starts swapping, switching windows takes
a while longer.  CPU intensive apps also caused NT's task
switching to slow down in a _BIG_ way.  The foreground app
would run OK, but if you tried to switch to another window,
you had better go get a drink.  

[I had my girlfriend use OS/2 and NT... see below]

Under OS/2, I loaded up a a few of the "Productivity" apps.  
No swapping occurred until I loaded MANY apps.  OS/2 really
does run great in 16 megs.  Anyways, swapping does slow
down things a little, but it wasn't as bad as under NT.
I am sure that Brian will explain the differences between
OS/2's and NT's swapping algorithm, but an end user doesn't
care; they just want their OS to run and run WELL.

A note about OS/2's swapfile.  I used to have the swapfile set
initially at 2048k.  My swapfile would grow up to 10 megs if
I loaded a fullscreen WIN session and a few other programs.
I recently changed the minimum size to 8192k, and my swapfile
has very RARELY grown any larger.  It seems that OS/2
over compensates when it has to expand the swapfile.  Even if
I load 3 fullscreen WIN-OS/2 sessions and a few OS/2 programs,
my swapfile does not grow.  The same procedure would have
made the swapfile grow to about 15 megs or more.   Hmmmmm.

I made the following batch file and ran it under both OS's:

:START
unzip -t temp.zip
goto START

After I ran it, system performance was slower, but NT was still
very usable.
I could notice a slight degradation of performance under OS/2, but
not as much as under NT.

Raytracing test:

(I ran a test with one raytrace, but the results were about
the same as the UNZIP test from above)                        

I ran 4 raytraces in windowed sessions under NT and OS/2 for 5 minutes.
At the end of 5 minutes I checked how many lines had been traced.
During this time I putzed around and played with system settings
and loaded up a few (small) apps like the clipboard viewer and
a readme file.  I also played with File Manager (under both OS's).

Under NT, performance became unbearable.  The raytraces were bringing
the system to almost unusable levels.  Window redraws were SLOW,
SLOW SLOW!  I tried to switch to a different window and it took 30
seconds to redraw the window.  Pretty bad performance.  Again,
Program Manager did not slow down, but I've been corrected that
this is actually a "good" feature.  I'll accept that point.

Under OS/2, performance was slower, but VERY *VERY* usable.  The
system responded very well to anything I did.  Windows were redrawn
quickly, applications loaded fast; it was nice.  This is the
way I want my OS to handle.

After 5 minutes, NT's traces had completed about 100 lines each.
OS/2's had completed about 105 each in the same time.  Even though
OS/2 and NT completed roughly the same number of lines, NT's performance
was very poor compared to OS/2's.

Oh, BTW, I'm running 11 QRT raytraces under OS/2 right now
while I type this post.  OS/2 is performing VERY well
under stress.  In fact, a while ago (under the beta), I
ran 20 raytraces at once under OS/2.  Granted system
performace was much slower than normal, but the system
was still responsive unlike NT.

---
Ok, I just had my girlfriend use my system and judge its performance.
She switched between apps and ran a few programs under OS/2.
She has a 386/25 with 4 megs and runs Windows 3.1.  She said that
it looks kinda like Windows, and "it was running fine." !!  I guess
she is just used to Windows on her "slow" machine.  :)  I'd
definitely say that OS/2 is running slow under the circumstances,
but it is still usable IMHO (and Heather's).  :)

[I'll reboot in a sec and have her judge NT while running only 4
traces instead of 11.  I don't want to wait around forever
trying to close the windows.]            

OK, I rebooted, had her play with NT, and here is what she had to say:
(this is her typing)                                                  
It's a lot slower than the other one, especially opening things up.  
Once he loaded the four programs [QRT's] things  slowed down even more.  
At that point the difference between the two systems was unmistakable.
Can I go home now?
(end of Heather's editorial)

hehe.  :)  She doesn't like my computer--it is the "other" woman.
Anyways, when I first loaded NT for her, notice that I did not run
anything before she used it.  She thought it was slower than OS/2
when OS/2 was running the 11 (ELEVEN!) raytraces.  NT's tasking
was set on "foreground gets BEST priority".  
Oh, don't flame ME, these were HER opinions.  :)
---                                                  

Oh, I wish to elaborate more on the priority boosts in OS/2 and
NT when an application is in the foreground.

OS/2 does give apps a boost when they have the focus, but it is a
very small boost.  When I ran the 4 traces, I gave one the focus.
The foreground trace finished about 30 lines while the others
finished about 15 lines in a given time interval.

When NT is set on the "give foreground app best priority", it really
means best!   During the same test, the foreground trace finished
about 60 lines while the others finished ONE or TWO!  WOW!
With the tasking set to the middle level, the boost was less drastic.
Multitasking was even when set to EQUAL.

I know, I know.  Raytraces are very CPU intensive and don't
represent the every day ...

read more »

 
 
 

Windows NT vs OS/2 2.1

Post by Wayne J. Hy » Mon, 16 Aug 1993 22:05:54


Someone has suggested to me that my using a DOS application may not
have been the ideal choice to test NT's handling of large loads.  I
am in the process of putting together a 100% native "test".  I'll let
ya know what happens.  

--
Wayne Hyde


 
 
 

Windows NT vs OS/2 2.1

Post by Brian Ho » Mon, 16 Aug 1993 22:57:08



Quote:>   Someone has suggested to me that my using a DOS application may not
>   have been the ideal choice to test NT's handling of large loads.  I
>   am in the process of putting together a 100% native "test".  I'll let
>   ya know what happens.  

Well, too bad. :-)  The fact is, NT is going to be running mostly Win16 and
DOS apps for a long time.  How about a comparison of the following:

        OS/2 running 32-bit OS/2 apps   vs.  NT running NT apps
        OS/2 running Win16+DOS apps     vs.  NT running Win16+DOS apps

Then you could compare their ability to do things "natively" and to do
things while running other types of software.

Brian
--
//----------------------------------------------------------------------------
// Brian W.K. Hook                 "Stop!  Stop in the name of all that

// ( 72144,3662 CI$ )
//
// "Oooooooo-la-tec!  Ooooooo-la-tec!"
//

//----------------------------------------------------------------------------

 
 
 

Windows NT vs OS/2 2.1

Post by Craig Kule » Tue, 17 Aug 1993 09:22:44


First, let me start this off by stating that I have run and used both
Windows NT (March Beta -- waiting for my CD to arrive) and OS/2 2.1 GA
on a 386/25 (8 MB, et4000 video) and my home Gateway 486/66 (single 16 MB
SIMM, ATI Ultra XLR video).

Ironically, this appears to be nearly the same system Wayne Hyde is using.
Even on the March Beta, though -- I come to rather different results
than Mr. Hyde.

Soooo...let's get started!

======================================================================

In my tests on both systems, I have come up with a rough guidline which
seems to hold rather well in my experience:

Windows NT (3/93) in 8 MB = OS/2 2.1 in 5-6 MB = OS/2 2.0 in 6 MB
Windows NT (3/93) in 12 MB = OS/2 2.1 in 8 MB
Windows NT (3/93) in 16 MB = OS/2 2.1 in 12 MB

The feel and swapping slowdown is roughly comparable using this rule
of thumb. Networking services in NT were not auto-started (set using
Services and Network applets in NT). The paging file in NT was set to
20 MB for all configurations and worked well. OS/2's swap file varied,
depending upon load and memory configuration (naturally).

As a stand-alone system, I have found NT to be very useable in 16 MB,
decent in 12 MB, and just barely useable in 8 MB. As a networked
machine, I would suggest 12 MB as tolerable, 16 MB+ as useable.
(I would suggest 12 MB+ as useable as a networked OS/2 2.x machine).

NT does implement a lot of lazy reads and writes, so the HD blips
frequently for no real obvious reason (as compared to DOS/Windows 3.1),
but I have found that the UI remains very responsive. Loading SDK
help files, Word for Windows, MathCAD 4, and Borland C++ (DOS)
simultaneously while watching the Performance Monitor shows that
minimal swapping is going on, and bringing up the mimimized File
Manager while all of this is going on is snappy. No delays whatsoever.
Doing the same under OS/2 (Substituting MathCAD 3 for MathCAD 4 since the
latter doesn't run in OS/2, and the OS/2 help system instead of the
Win32 SDK help system) takes somewhat longer and freezes the WPS the
majority of the time.

Time to do the above in NT: 16 seconds
Amount of swapping while switching between apps: barely perceptable

Time for OS/2 to do the above: 23 seconds
Amount of swapping while switching between apps: barely perceptable

The time "benchmark" is really pointless -- but the fact that I could
not use the WPS and still use NT is significant -- coupled with the fact
that NT did not swap noticeably, even with File Manager and the Performance
Monitor running.

Side Note: Since NT re-allocates memory blocks to applications depending
upon relative useage, if any of the above apps in NT were to be ignored
for a long time, there would be a small swapping period (1 sec) in which
NT brings up memory pages from the paging file. This is not a major
slowdown.

Since Wayne seemed to have the largest complaints against the responsiveness
of NT during CPU-intensive tasks, I tried some tests here. The program(s)
I used are DOS and Windows 3.1 versions of a galactic collision simulation
model I had previously written (and will port to Win32 sometime...and OS/2
if someone has a compiler handy...).

OS/2 handles the programs nicely. All of the apps (2 DOS, 1 Windows) run
at approximately the same priority -- the DOS program finishes in 60
seconds (both, actually), the Windows one in 70 seconds.

NT (3/93) is given the same task. (This is all being done on the 486, BTW).
With tasking set to 'all apps equal', the DOS programs finish in the
same time -- 65 seconds. The Windows app takes 95 seconds (due to a FP
bug in the Win16 subsystem -- fixed in the 475 release). The system
remains responsive -- the next time I loaded the Control Panel; it
took 10 seconds instead of the usual 5 or so.
The task list came up immediately -- as did File Manager and (of course)
Program Manager.

Setting the tasking options to the middle setting ('give application with
focus some priority boost'), I re-ran the simulations. Leaving the focus
upon one of the DOS apps allowed it to complete in 30 seconds, but the
others slowed appreciably -- 80 seconds for the DOS apps (much slower),
and 100 seconds for the Windows app (just a pinch slower). Pulling up
the task list gave it the priority boost -- but since it didn't do
anything, the other apps ran happily along with times similar to
that described by the 'all apps have same priority' setting. Made sense.
Loading the Control Panel gave it the higher priority, and it loaded in
6 seconds -- but it only affected the simulations by about 1 second --
and visually, not really at all.

Increasing the load to 5 simulations (3 DOS, 2 Windows) didn't change
anything but the times. Responsiveness was still snappy.

I ran the same simulations on the 386 box, and although the times were
longer, and the HD swapped quite a bit, the system was still
quite responsive. Swapping seemed to slow the machine down by about
50% or so -- but File Manager, Program Manager, and the task list were
all readily available within 1 second or less about 95% of the time.

So, I would say that, in my experience, Windows NT (3/93) is VERY
useable on 16 MB systems (as is OS/2). If anything, I have found that
the WPS tends to be less responsive than the NT desktop in situations
of app loading etc.

So what is happening here?
First, I would suggest that Wayne try some other CPU-intensive app
and see if he still has responsiveness problems. My guess is that NT
is having some problems with this app. I have found, for instance,
that the DOS Borland C++ compiler (286 protected mode) runs well, but
if a low priority app (screen saver) takes focus, BC++ hogs it's
CPU cycles. The system is still extremely reponsive, but the screen
saver crawls as somehow NT believes BC++ is trying to do something.
(It, of course, is sitting idle).

My guess is that your DOS app is playing some funny games (compiled
with DOS extender, a 32-bit app, protected mode program, etc etc.)
that NT's VDM just wasn't programmed to interpret.

Try another DOS program -- or a Windows program -- or even an NT app.
I think you will find the desktop responsive.

If you don't need the networking options, you might go into
Control Panel --> services and Control Panel --> Network and remove
any boot-up services you don't need. You'll find the system to
be more spry afterwards.

If it's not, could you go into the Performance Monitor, open up
a graph on 'committed bytes' and tell me what you see?

Until then, I'm waiting for my NT CD-ROM to come in. And I understand
that SDK users will get a pre-release of VC++ for NT soon?

Take care everyone!

Craig kulesa (AA8CM)
Graduate student
Department of Astronomy
University of Arizona

======================================================================
This message was written using recycled phosphors.

 
 
 

Windows NT vs OS/2 2.1

Post by Wayne J. Hy » Tue, 17 Aug 1993 06:21:02


Quote:Craig writes:
>Ironically, this appears to be nearly the same system Wayne Hyde is using.
>Even on the March Beta, though -- I come to rather different results
>than Mr. Hyde.

Nice system, eh?  :)  

I have not run the March Beta, so I can't compare its performance to the GA.
I do know that someone posted in c.o.m.m that they experienced the same
problems I had while the March Beta ran very fast on their system.  Let
me know how the release runs on your computer.  

Quote:>Since Wayne seemed to have the largest complaints against the responsiveness
>of NT during CPU-intensive tasks, I tried some tests here. The program(s)
>I used are DOS and Windows 3.1 versions of a galactic collision simulation
>model I had previously written (and will port to Win32 sometime...and OS/2
>if someone has a compiler handy...).

[Grrrr, the damn OS/2 "bug" occured again.  I marked, copied, and pasted
the first quote just fine.  The second had all the first characters of
each line replaced with a space... I had to retype the damn letters.]

I am looking for other CPU intensive apps to run under NT so I can more
thouroughly test out the system.  If anyone has any native apps for NT
that suck CPU time for breakfast, please mail them to me.  

Quote:>So, I would say that, in my experience, Windows NT (3/93) is VERY
>useable on 16 MB systems (as is OS/2). If anything, I have found that
>the WPS tends to be less responsive than the NT desktop in situations
>of app loading etc.

I found NT usable if I was running native apps and didn't do much
multitasking.  Once I loaded up a few apps and the system started
swapping, all hell broke loose.  WPS gives me the Hourglass sometimes,
but usually this occurs only when I launch an application.  I'm trying
out a program called PC/2 (very nice, time to send my money!) which
lets you launch an app as a background task.  I've pretty much done
away with using the folders to launch programs; my system looks more
like a UNIX box now.  

Quote:>My guess is that your DOS app is playing some funny games (compiled
>with DOS extender, a 32-bit app, protected mode program, etc etc.)
>that NT's VDM just wasn't programmed to interpret.

Actually, QRT is a pretty old program ported from the Amiga.  It doesn't
use any DOS extender or protected mode.  It wouldn't have worked under
OS/2 if it had.  

Quote:>If it's not, could you go into the Performance Monitor, open up
>a graph on 'committed bytes' and tell me what you see?

I'll try this.  I have looked at the pages/sec while I was using my
system, and the graph was very active.  

--
Wayne Hyde

 
 
 

Windows NT vs OS/2 2.1

Post by a.. » Tue, 17 Aug 1993 12:44:15



>First, let me start this off by stating that I have run and used both
>Windows NT (March Beta -- waiting for my CD to arrive) and OS/2 2.1 GA
>on a 386/25 (8 MB, et4000 video) and my home Gateway 486/66 (single 16 MB

             ^^^^^                                                  ^^^^^^
So you have one machine with 8 MB and the other with single 16 MB SIMM. Will
get back to this a little later.  

Quote:>SIMM, ATI Ultra XLR video).

>Ironically, this appears to be nearly the same system Wayne Hyde is using.
>Even on the March Beta, though -- I come to rather different results
>than Mr. Hyde.

>Soooo...let's get started!

>======================================================================

>In my tests on both systems, I have come up with a rough guidline which
>seems to hold rather well in my experience:

>Windows NT (3/93) in 8 MB = OS/2 2.1 in 5-6 MB = OS/2 2.0 in 6 MB
>Windows NT (3/93) in 12 MB = OS/2 2.1 in 8 MB
>Windows NT (3/93) in 16 MB = OS/2 2.1 in 12 MB

Now this seems strange (curious; no flame intended). How did you reach the
above conclusions? I understand you ran NT on the 8 MB machine you have
(right?); which 5 MB or 6 MB machine you ran OS/2 on? Also, did you use
different 12 MB machine (since you have a single 16 MB SIMM on the other
machine) for running NT and OS/2 or are these guesses?

I have seen several posts in the last few days about running OS/2 on 4 MB and
one strange 5 MB machines and they all have been running it for quite some
time. But even Brian Sturgill (and Mike Timbol) suggested that NT is virtually
unusable in 8 MB. Strange.

- Show quoted text -

Quote:

>The feel and swapping slowdown is roughly comparable using this rule
>of thumb. Networking services in NT were not auto-started (set using
>Services and Network applets in NT). The paging file in NT was set to
>20 MB for all configurations and worked well. OS/2's swap file varied,
>depending upon load and memory configuration (naturally).

>As a stand-alone system, I have found NT to be very useable in 16 MB,
>decent in 12 MB, and just barely useable in 8 MB. As a networked
>machine, I would suggest 12 MB as tolerable, 16 MB+ as useable.
>(I would suggest 12 MB+ as useable as a networked OS/2 2.x machine).

>NT does implement a lot of lazy reads and writes, so the HD blips
>frequently for no real obvious reason (as compared to DOS/Windows 3.1),
>but I have found that the UI remains very responsive. Loading SDK
>help files, Word for Windows, MathCAD 4, and Borland C++ (DOS)
>simultaneously while watching the Performance Monitor shows that
>minimal swapping is going on, and bringing up the mimimized File
>Manager while all of this is going on is snappy. No delays whatsoever.
>Doing the same under OS/2 (Substituting MathCAD 3 for MathCAD 4 since the
>latter doesn't run in OS/2, and the OS/2 help system instead of the
>Win32 SDK help system) takes somewhat longer and freezes the WPS the
>majority of the time.

That's understadable. Single message queue for the WPS.

Quote:

>Time to do the above in NT: 16 seconds
>Amount of swapping while switching between apps: barely perceptable

>Time for OS/2 to do the above: 23 seconds
>Amount of swapping while switching between apps: barely perceptable

Since I don't have NT (can't afford $300 price; happy with OS/2 etc. :)), I
can't  comment on the tests you performed.

        (deleted)

Quote:

>Take care everyone!

>Craig kulesa (AA8CM)
>Graduate student
>Department of Astronomy
>University of Arizona

>======================================================================
>This message was written using recycled phosphors.

Max.

 
 
 

Windows NT vs OS/2 2.1

Post by Brian Sturgi » Tue, 17 Aug 1993 07:26:35


...

Quote:>>In my tests on both systems, I have come up with a rough guidline which
>>seems to hold rather well in my experience:

>>Windows NT (3/93) in 8 MB = OS/2 2.1 in 5-6 MB = OS/2 2.0 in 6 MB
>>Windows NT (3/93) in 12 MB = OS/2 2.1 in 8 MB
>>Windows NT (3/93) in 16 MB = OS/2 2.1 in 12 MB

>Now this seems strange (curious; no flame intended). How did you reach the
>above conclusions? I understand you ran NT on the 8 MB machine you have
>(right?); which 5 MB or 6 MB machine you ran OS/2 on? Also, did you use
>different 12 MB machine (since you have a single 16 MB SIMM on the other
>machine) for running NT and OS/2 or are these guesses?

>I have seen several posts in the last few days about running OS/2 on 4 MB and
>one strange 5 MB machines and they all have been running it for quite some
>time. But even Brian Sturgill (and Mike Timbol) suggested that NT is virtually
>unusable in 8 MB. Strange.

To the best of my knowledge I've never said anything about NT being
unusable in 8 megs.  I've never seen NT in 8 megs.

Brian
--
C. Brian Sturgill           Want good, low cost, POSIX (Unix-like) tools and

Center for Software

 
 
 

Windows NT vs OS/2 2.1

Post by Craig Kule » Tue, 17 Aug 1993 14:22:48


I (Craig Kulesa) wrote,

>>In my tests on both systems, I have come up with a rough guidline which
>>seems to hold rather well in my experience:

>>Windows NT (3/93) in 8 MB = OS/2 2.1 in 5-6 MB = OS/2 2.0 in 6 MB
>>Windows NT (3/93) in 12 MB = OS/2 2.1 in 8 MB
>>Windows NT (3/93) in 16 MB = OS/2 2.1 in 12 MB

>Now this seems strange (curious; no flame intended). How did you reach the
>above conclusions? I understand you ran NT on the 8 MB machine you have
>(right?); which 5 MB or 6 MB machine you ran OS/2 on? Also, did you use
>different 12 MB machine (since you have a single 16 MB SIMM on the other
>machine) for running NT and OS/2 or are these guesses?

>I have seen several posts in the last few days about running OS/2 on 4 MB and
>one strange 5 MB machines and they all have been running it for quite some
>time. But even Brian Sturgill (and Mike Timbol) suggested that NT is virtually
>unusable in 8 MB. Strange.

As a networked PC (which is how NT is configured as default), 8 MB is virtually
unusable. But disabling net services completely helps a lot -- so it becomes
*just* about usable.

As for your other question: I achieved 12 MB by borrowing 4 1x9 SIMMs from
another lab computer (just for a day, honest!). One of the other 386/25's
(a Dell) uses 72-pin SIMMS and had a 4 MB SIMM and two 1 MB SIMMS. It seemed
just a hair faster in use (OS/2 2.1) than did the 8 MB NT box -- so I wrote
5-6 MB in the post.

Thanks for helping me clear that up.

Take care,

Craig Kulesa (AA8CM)
Graduate student
Department of Astronomy
University of Arizona
===========================================================================

 
 
 

Windows NT vs OS/2 2.1

Post by Pira » Wed, 18 Aug 1993 05:30:58



Quote:>>I have seen several posts in the last few days about running OS/2 on 4 MB
>>and one strange 5 MB machines and they all have been running it for quite
>>some time. But even Brian Sturgill (and Mike Timbol) suggested that NT is
>>virtually unusable in 8 MB. Strange.

>To the best of my knowledge I've never said anything about NT being
>unusable in 8 megs.  I've never seen NT in 8 megs.

>Brian
>--

I doubt you will.  Even after Microsoft's "silent upgrades" NT will still
require 12M to run.

Strange-- When I went to the first Beta demo of NT about a year ago, I was
told the kernel was only 50K.  Tiny little kernel.  HUGE OS.  Go figure.

 
 
 

Windows NT vs OS/2 2.1

Post by Dan Weism » Wed, 18 Aug 1993 10:12:33



Quote:> I asked someone to post my review since I was having
> problems with my news program.  Now my problem is
> solved.  I've changed a few things in my review and added
> others.  Well, here it is:

> ***********************************************************
> Ok!  Here is the review of NT & OS2 that I promised.  Keep
> in mind that these results were taken from *MY* testing
> of NT on *MY* machine.  For those who need to know again,

[well thought out review deleted]

One request:

can you all keep in mind that NT is for all intents and purposes, version
1.0...And OS/2 is what? 2.1? Why not compare OS/2 1.0 with NT and see how it
goes. I know it's easy to compare OS/2 to NT right now and say how much better
it is than NT, but let's face it, do you think NT is never gonna change? Do you
realize that this was started from the ground up, brand new, no previous
version to build upon? NT is a little kid and OS/2 has had time to mature.
I have been using NT (yes Release NT) for a week or so now and I have had no
problems whatsoever with it. I don't find it slow at all. Yeah sure I haven't
run 20 ray traces in the background, but who the hell EVER does that? Man, if
I'm gonna run 20 raytraces I'm gonna buy another machine or something. I think
that NT is a remarkable achievement for a fledgling OS and that's all there is
to it. I hope you didn't expect the be-all end-all OS right out of the box. I
say give NT as much time as OS/2 has had to mature and then compare the two.

I am not allowed to discuss the next version of NT, but let's just say that it
will be different and better than NT is right now. Can we compare THAT to OS/2
2.1? Why not? Would that not be fair? 'Nuff said

Dan
____
Dan Weisman
Software Test Engineer
Microsoft Corporation
Redmond, WA

 
 
 

Windows NT vs OS/2 2.1

Post by a.. » Wed, 18 Aug 1993 16:44:47




>can you all keep in mind that NT is for all intents and purposes, version
>1.0...And OS/2 is what? 2.1? Why not compare OS/2 1.0 with NT and see how it
>goes. I know it's easy to compare OS/2 to NT right now and say how much better
>it is than NT, but let's face it, do you think NT is never gonna change? Do you

Sure it'll change. Do you think OS/2 won't?

Quote:>realize that this was started from the ground up, brand new, no previous
>version to build upon? NT is a little kid and OS/2 has had time to mature.

Really? Some people think this is what was supposed to be OS/2 3.0 before MS
and IBM decided to go seperate ways.

Quote:>I have been using NT (yes Release NT) for a week or so now and I have had no
>problems whatsoever with it. I don't find it slow at all. Yeah sure I haven't

Compared to what?

Quote:>run 20 ray traces in the background, but who the hell EVER does that? Man, if
>I'm gonna run 20 raytraces I'm gonna buy another machine or something. I think
>that NT is a remarkable achievement for a fledgling OS and that's all there is
>to it. I hope you didn't expect the be-all end-all OS right out of the box. I
>say give NT as much time as OS/2 has had to mature and then compare the two.

>I am not allowed to discuss the next version of NT, but let's just say that it
>will be different and better than NT is right now. Can we compare THAT to OS/2
>2.1? Why not? Would that not be fair? 'Nuff said

Now why would I wait for something when I can buy a similar product today? Did
you forget to put smileys or was this really a serious article?
Quote:

>Dan
>____
>Dan Weisman
>Software Test Engineer
>Microsoft Corporation
>Redmond, WA

 
 
 

Windows NT vs OS/2 2.1

Post by David Boy » Wed, 18 Aug 1993 19:42:38




>>[well thought out review deleted]

>can you all keep in mind that NT is for all intents and purposes, version
>1.0...And OS/2 is what? 2.1? Why not compare OS/2 1.0 with NT and see how it
>goes.

Because the whole point of the review (which was very good, and
quite fair) was to compare the choices that people have *now*.
If you want to compare OS/2 v1 against anything, then by rights it should
be the version of NT that was available at the time. (Pretty boring
review though ;)

Quote:>I know it's easy to compare OS/2 to NT right now and say how much better
>it is than NT, but let's face it, do you think NT is never gonna change?

No, but I'd be surprised if OS/2 never changes either. But look:
Microsoft has been telling everyone that NT is the greatest thing
since sliced bread - certainly that it's better than OS/2. All Wayne
did was judge NT by the standards that Microsoft had set for it,
and he found it lacking.

Quote:>Do you
>realize that this was started from the ground up, brand new, no previous
>version to build upon? NT is a little kid and OS/2 has had time to mature.

NT has been in development for over 5 years, and been at tens of thousands
of beta sites for over a year, so it's not as if it's *that* new.
Nor *brand* new either, since Microsoft has a long period of
DOS, Windows and (yes) OS/2 experience to build on, plus there's
the Cutler team.

Quote:>Yeah sure I haven't
>run 20 ray traces in the background, but who the hell EVER does that?

NT is being marketed as a server OS, with lots of
competing demands on resources. Its process scheduling has to be
a lot better than it apparently is at the moment for it to be
able to do that effectively. (I think the original test only
used 2 or 3 ray-traces, BTW, but I may be wrong...)

Quote:>I hope you didn't expect the be-all end-all OS right out of the box.

That's a shame: we did, because that's what Microsoft has been
telling us for the past two years. That's actually the problem.
Everyone's expectations have been raised too much over the last couple
of years, and the reality just hasn't lived up to it. (BTW. I guess
I should point out that I don't run either of NT or OS/2, but Unix.
I'm just an interested observer, reading this newsgroup to glean
what little information is available. What I've noticed is that apart
from a very few obviously pro-Microsoft zealots (better than calling them
bigots, I guess), people with experience of both systems invariably
say that NT has a real performance problem, and that OS/2 beats it.
It says that Microsoft has really blown it, since the least that
should have been expected of NT is that it be better than OS/2,
and that doesn't appear to have happened. It doesn't bode well
for NT's long-term survival.)

Quote:>I am not allowed to discuss the next version of NT, but let's just say that it
>will be different and better than NT is right now. Can we compare THAT to OS/2
>2.1?

Hmmm, I realise you aren't a Microsoft spokesman but that sounds very
similar to the usual line of marketing doublethink: OS/2 may
be better now, but we've got something *much* better waiting in
the wings! It's pretty much the line I've heard when people talk
about using NT: the next beta will be better, the next beta will be
better, production will be better, the *next* production will be
better. It doesn't inspire much confidence in what's actually
available now,

        - Dave

 
 
 

Windows NT vs OS/2 2.1

Post by Jay Mayna » Wed, 18 Aug 1993 21:32:20



>can you all keep in mind that NT is for all intents and purposes, version
>1.0...

Not according to Microsoft: it's 3.1!

Quote:>NT is a little kid and OS/2 has had time to mature.

...no thanks to Microsoft.

Quote:>I hope you didn't expect the be-all end-all OS right out of the box.

Why not? That's exactly how MS has promoted it, and how MS sycophants like
Brian Sturgill have been advocating it.

Quote:>I say give NT as much time as OS/2 has had to mature and then compare the two.

There's no historical reason to believe that MS will allow it to live that
long.

Quote:>I am not allowed to discuss the next version of NT, but let's just say that it
>will be different and better than NT is right now. Can we compare THAT to OS/2
>2.1? Why not? Would that not be fair? 'Nuff said

No, we can't. The next version of NT is vaporware, and, given the incredible
delays NT itself showed, I expect it to remain so for some time to come.

OS/2 2.1 is here now. So is NT 3.1/1.0. Comparing the two is fair.
--
Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can

                     "iHaTeX." -- Andrew Burt

 
 
 

Windows NT vs OS/2 2.1

Post by Doug Botim » Wed, 18 Aug 1993 22:08:02


Dan Weisman writes

Quote:

>can you all keep in mind that NT is for all intents and purposes, version
>1.0...And OS/2 is what? 2.1? Why not compare OS/2 1.0 with NT and see how it
>goes.

deletia

Quote:

>I am not allowed to discuss the next version of NT, but let's just say that it
>will be different and better than NT is right now. Can we compare THAT to OS/2
>2.1? Why not? Would that not be fair? 'Nuff said

>Dan
>____
>Dan Weisman
>Software Test Engineer
>Microsoft Corporation
>Redmond, WA

What is this Dan?  NT is called Windows NT 3.1 to convince us all that it is  
not Microsofts first stab at creating an operating system.  Now Microsoft, you,  
are crying fowl when it gets compared to the competition.  And more of  
Microsoft's "just wait for our next trick, I promise it will be better".

What about all the effort Microsoft spent telling us how much better NT would  
be than OS/2?  It was fine for Microsoft to compare promises to IBM's shipping  
product, but now it's unfair to compare Microsoft's shipping product to IBM's?

No you can't compare THAT (promised new version of NT) to OS/2 2.1.  You can  
compare it, when it ships, to whatever version of OS/2, or any other vendor's  
operating system, is also shipping at the same time.

By the way, lots of people are saying that, of course, NT is not intended for  
the desktop.  It needs a powerful computer.  My response to that is you can  
make a rock fly with a big enough engine.

Doug

These opinions are my own!

 
 
 

Windows NT vs OS/2 2.1

Post by ASI.. » Wed, 18 Aug 1993 22:51:42



(Dan Weisman) says:

Quote:

>One request:

>can you all keep in mind that NT is for all intents and purposes, version
>1.0...And OS/2 is what? 2.1? Why not compare OS/2 1.0 with NT and see how it
>goes. I know it's easy to compare OS/2 to NT right now and say how much better

OK. And we compare the DEC Alpha chip to the Intel 8080 not to the Pentium.
And if I would start manufactoring cars I would expect you compare them to
the Ford Model T. And when we are at it, we should compare OS/2 1.0 to what?
To Windows 1.0 of course. FYI Windows NT 3.1 is a 3.1 version. No matter what
YOU think. If MS says it it 3.1 they better make it as good as a 3.1 should be.

Quote:>version to build upon? NT is a little kid and OS/2 has had time to mature.

                        ^^^^^^^^^^^^^^^^^^
But it wants to play with the big boys.

Quote:>I am not allowed to discuss the next version of NT, but let's just say that it
>will be different and better than NT is right now. Can we compare THAT to OS/2
>2.1? Why not? Would that not be fair? 'Nuff said

Sure, if 2.1 is the current version of OS/2 by then. But more likely it will
be compared to OS/2 2.2 or 2.3 or 3.0 or Workplace OS.

Michael Bode