Hi,
It appears that I can force my machine to behave like it has lesser mem
that it does with the /maxmem switch in boot.ini. Is there something that
can be done to make the processor
run slower either actually or simulate such? Also, is it possible to ensure
that all floating point calculations are done in software rather than in
hardware? Can somebody please throw
some light on this.
The reason this intrigues me being that I have an existing code base that
works under Windows but would like to determine how it performs under PDA
like conditions. The
reason i cant run it on a PDA being having to port the code to Win CE which
currently is a bottleneck.
thanks
kutty
Well, I suppose there’s the obvious “put a load on the machine”
approach. If you start up a higher-priority thread somewhere that eats
up cycles on some periodic basis, it will look much like a slower
machine. However, it’s not a very good simulation. Among other things,
it won’t affect any device drivers’ interrupt handlers unless you go to
extraordinary effort.
I don’t think there’s any way to make FP calcs happen in software.
They’re part of the instruction set. Of course, you could write your own
FP functions and call those instead of the built-in ones. This is one of
those rare instances where operator overloading in C++ can actually be a
good idea.
Kutty Banerjee wrote:
Hi,
It appears that I can force my machine to behave like it has lesser mem
that it does with the /maxmem switch in boot.ini. Is there something that
can be done to make the processor
run slower either actually or simulate such? Also, is it possible to ensure
that all floating point calculations are done in software rather than in
hardware? Can somebody please throw
some light on this.
The reason this intrigues me being that I have an existing code base that
works under Windows but would like to determine how it performs under PDA
like conditions. The
reason i cant run it on a PDA being having to port the code to Win CE which
currently is a bottleneck.
thanks
kutty
–
…/ray..
Please remove “.spamblock” from my email address if you need to contact
me outside the newsgroup.
Thanks for the response but from my own studies i figured it is easier to
actually port the code to PocketPC than do funny simulation stuff:-))
kutty
“Ray Trent” wrote in message
news:xxxxx@ntdev…
> Well, I suppose there’s the obvious “put a load on the machine” approach.
> If you start up a higher-priority thread somewhere that eats up cycles on
> some periodic basis, it will look much like a slower machine. However,
> it’s not a very good simulation. Among other things, it won’t affect any
> device drivers’ interrupt handlers unless you go to extraordinary effort.
>
> I don’t think there’s any way to make FP calcs happen in software. They’re
> part of the instruction set. Of course, you could write your own FP
> functions and call those instead of the built-in ones. This is one of
> those rare instances where operator overloading in C++ can actually be a
> good idea.
>
> Kutty Banerjee wrote:
>> Hi,
>> It appears that I can force my machine to behave like it has lesser mem
>> that it does with the /maxmem switch in boot.ini. Is there something that
>> can be done to make the processor
>> run slower either actually or simulate such? Also, is it possible to
>> ensure that all floating point calculations are done in software rather
>> than in hardware? Can somebody please throw
>> some light on this.
>> The reason this intrigues me being that I have an existing code base
>> that works under Windows but would like to determine how it performs
>> under PDA like conditions. The
>> reason i cant run it on a PDA being having to port the code to Win CE
>> which currently is a bottleneck.
>>
>> thanks
>> kutty
>
> –
> …/ray..
>
> Please remove “.spamblock” from my email address if you need to contact me
> outside the newsgroup.
>
Why simulate when you could just buy a slow machine. Just find the slowest machine that supports the lowest level OS that your driver will run on and buy that. Good news is old == cheap (as long as it works).
I don’t actually know old x86 CPU would have comparable “grunt” to the sort of CPUs that you’d find in a modern PDA. However seeing as you can still buy 386 based laptops on ebay, I’m guessing that something appropriate is available. Taking the differences in overhead from Windows CE to Windows ??? might also be difficult though.
Just a thought in case it never occurred to you.
Richard McNally
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Kutty Banerjee
Sent: Tuesday, March 01, 2005 9:02 AM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] simulate a slower machine
Thanks for the response but from my own studies i figured it is easier to
actually port the code to PocketPC than do funny simulation stuff:-))
kutty
“Ray Trent” wrote in message
news:xxxxx@ntdev…
> Well, I suppose there’s the obvious “put a load on the machine” approach.
> If you start up a higher-priority thread somewhere that eats up cycles on
> some periodic basis, it will look much like a slower machine. However,
> it’s not a very good simulation. Among other things, it won’t affect any
> device drivers’ interrupt handlers unless you go to extraordinary effort.
>
> I don’t think there’s any way to make FP calcs happen in software. They’re
> part of the instruction set. Of course, you could write your own FP
> functions and call those instead of the built-in ones. This is one of
> those rare instances where operator overloading in C++ can actually be a
> good idea.
>
> Kutty Banerjee wrote:
>> Hi,
>> It appears that I can force my machine to behave like it has lesser mem
>> that it does with the /maxmem switch in boot.ini. Is there something that
>> can be done to make the processor
>> run slower either actually or simulate such? Also, is it possible to
>> ensure that all floating point calculations are done in software rather
>> than in hardware? Can somebody please throw
>> some light on this.
>> The reason this intrigues me being that I have an existing code base
>> that works under Windows but would like to determine how it performs
>> under PDA like conditions. The
>> reason i cant run it on a PDA being having to port the code to Win CE
>> which currently is a bottleneck.
>>
>> thanks
>> kutty
>
> –
> …/ray..
>
> Please remove “.spamblock” from my email address if you need to contact me
> outside the newsgroup.
>
—
Questions? First check the Kernel Driver FAQ at http://www.osronline.com/article.cfm?id=256
You are currently subscribed to ntdev as: xxxxx@dsto.defence.gov.au
To unsubscribe send a blank email to xxxxx@lists.osr.com
Why buy an older machine when you can just run the wince simulator that
comes with the .net framework? That thing is plenty slow 
=====================
Mark Roddy
Windows .NET/XP/2000 Consulting
Hollis Technology Solutions 603-321-1032
www.hollistech.com
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
McNally, Richard
Sent: Monday, February 28, 2005 10:29 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] simulate a slower machine
Why simulate when you could just buy a slow machine. Just
find the slowest machine that supports the lowest level OS
that your driver will run on and buy that. Good news is old
== cheap (as long as it works).
I don’t actually know old x86 CPU would have comparable
“grunt” to the sort of CPUs that you’d find in a modern PDA.
However seeing as you can still buy 386 based laptops on
ebay, I’m guessing that something appropriate is available.
Taking the differences in overhead from Windows CE to Windows
??? might also be difficult though.
Just a thought in case it never occurred to you.
Richard McNally
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Kutty Banerjee
Sent: Tuesday, March 01, 2005 9:02 AM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] simulate a slower machine
Thanks for the response but from my own studies i figured it
is easier to actually port the code to PocketPC than do funny
simulation stuff:-))
kutty
“Ray Trent” wrote in message
> news:xxxxx@ntdev…
> > Well, I suppose there’s the obvious “put a load on the
> machine” approach.
> > If you start up a higher-priority thread somewhere that
> eats up cycles
> > on some periodic basis, it will look much like a slower machine.
> > However, it’s not a very good simulation. Among other
> things, it won’t
> > affect any device drivers’ interrupt handlers unless you go
> to extraordinary effort.
> >
> > I don’t think there’s any way to make FP calcs happen in software.
> > They’re part of the instruction set. Of course, you could
> write your
> > own FP functions and call those instead of the built-in
> ones. This is
> > one of those rare instances where operator overloading in C++ can
> > actually be a good idea.
> >
> > Kutty Banerjee wrote:
> >> Hi,
> >> It appears that I can force my machine to behave like it
> has lesser
> >> mem that it does with the /maxmem switch in boot.ini. Is there
> >> something that can be done to make the processor run slower either
> >> actually or simulate such? Also, is it possible to ensure that all
> >> floating point calculations are done in software rather than in
> >> hardware? Can somebody please throw some light on this.
> >> The reason this intrigues me being that I have an
> existing code base
> >> that works under Windows but would like to determine how
> it performs
> >> under PDA like conditions. The reason i cant run it on a PDA being
> >> having to port the code to Win CE which currently is a bottleneck.
> >>
> >> thanks
> >> kutty
> >
> > –
> > …/ray..
> >
> > Please remove “.spamblock” from my email address if you need to
> > contact me outside the newsgroup.
> >
>
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
> xxxxx@dsto.defence.gov.au
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
> xxxxx@hollistech.com To unsubscribe send a blank email to
> xxxxx@lists.osr.com
>
Oh and if you really want an older machine and you happen to live in a town
that still has a dump (or a ‘transfer station’ as they are now called,) just
go and hang out with the rest of the teenage-linux nerds and wait for the
junk pcs to show up. Then fight them for first dibs.
=====================
Mark Roddy
Windows .NET/XP/2000 Consulting
Hollis Technology Solutions 603-321-1032
www.hollistech.com
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of
McNally, Richard
Sent: Monday, February 28, 2005 10:29 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] simulate a slower machine
Why simulate when you could just buy a slow machine. Just
find the slowest machine that supports the lowest level OS
that your driver will run on and buy that. Good news is old
== cheap (as long as it works).
I don’t actually know old x86 CPU would have comparable
“grunt” to the sort of CPUs that you’d find in a modern PDA.
However seeing as you can still buy 386 based laptops on
ebay, I’m guessing that something appropriate is available.
Taking the differences in overhead from Windows CE to Windows
??? might also be difficult though.
Just a thought in case it never occurred to you.
Richard McNally
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Kutty Banerjee
Sent: Tuesday, March 01, 2005 9:02 AM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] simulate a slower machine
Thanks for the response but from my own studies i figured it
is easier to actually port the code to PocketPC than do funny
simulation stuff:-))
kutty
“Ray Trent” wrote in message
> news:xxxxx@ntdev…
> > Well, I suppose there’s the obvious “put a load on the
> machine” approach.
> > If you start up a higher-priority thread somewhere that
> eats up cycles
> > on some periodic basis, it will look much like a slower machine.
> > However, it’s not a very good simulation. Among other
> things, it won’t
> > affect any device drivers’ interrupt handlers unless you go
> to extraordinary effort.
> >
> > I don’t think there’s any way to make FP calcs happen in software.
> > They’re part of the instruction set. Of course, you could
> write your
> > own FP functions and call those instead of the built-in
> ones. This is
> > one of those rare instances where operator overloading in C++ can
> > actually be a good idea.
> >
> > Kutty Banerjee wrote:
> >> Hi,
> >> It appears that I can force my machine to behave like it
> has lesser
> >> mem that it does with the /maxmem switch in boot.ini. Is there
> >> something that can be done to make the processor run slower either
> >> actually or simulate such? Also, is it possible to ensure that all
> >> floating point calculations are done in software rather than in
> >> hardware? Can somebody please throw some light on this.
> >> The reason this intrigues me being that I have an
> existing code base
> >> that works under Windows but would like to determine how
> it performs
> >> under PDA like conditions. The reason i cant run it on a PDA being
> >> having to port the code to Win CE which currently is a bottleneck.
> >>
> >> thanks
> >> kutty
> >
> > –
> > …/ray..
> >
> > Please remove “.spamblock” from my email address if you need to
> > contact me outside the newsgroup.
> >
>
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
> xxxxx@dsto.defence.gov.au
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
> xxxxx@hollistech.com To unsubscribe send a blank email to
> xxxxx@lists.osr.com
>
Mark:
My question is really out of context in this group but would you know
:“Does the WinCE emulator do software floating point operations?”
thanks
kutty
“Mark Roddy” wrote in message news:xxxxx@ntdev…
> Why buy an older machine when you can just run the wince simulator that
> comes with the .net framework? That thing is plenty slow
>
>
> =====================
> Mark Roddy
> Windows .NET/XP/2000 Consulting
> Hollis Technology Solutions 603-321-1032
> www.hollistech.com
>
>> -----Original Message-----
>> From: xxxxx@lists.osr.com
>> [mailto:xxxxx@lists.osr.com] On Behalf Of
>> McNally, Richard
>> Sent: Monday, February 28, 2005 10:29 PM
>> To: Windows System Software Devs Interest List
>> Subject: RE: [ntdev] simulate a slower machine
>>
>> Why simulate when you could just buy a slow machine. Just
>> find the slowest machine that supports the lowest level OS
>> that your driver will run on and buy that. Good news is old
>> == cheap (as long as it works).
>>
>> I don’t actually know old x86 CPU would have comparable
>> “grunt” to the sort of CPUs that you’d find in a modern PDA.
>> However seeing as you can still buy 386 based laptops on
>> ebay, I’m guessing that something appropriate is available.
>> Taking the differences in overhead from Windows CE to Windows
>> ??? might also be difficult though.
>>
>> Just a thought in case it never occurred to you.
>>
>> Richard McNally
>>
>>
>> -----Original Message-----
>> From: xxxxx@lists.osr.com
>> [mailto:xxxxx@lists.osr.com]On Behalf Of Kutty Banerjee
>> Sent: Tuesday, March 01, 2005 9:02 AM
>> To: Windows System Software Devs Interest List
>> Subject: Re:[ntdev] simulate a slower machine
>>
>>
>> Thanks for the response but from my own studies i figured it
>> is easier to actually port the code to PocketPC than do funny
>> simulation stuff:-))
>>
>> kutty
>>
>> “Ray Trent” wrote in message
>> news:xxxxx@ntdev…
>> > Well, I suppose there’s the obvious “put a load on the
>> machine” approach.
>> > If you start up a higher-priority thread somewhere that
>> eats up cycles
>> > on some periodic basis, it will look much like a slower machine.
>> > However, it’s not a very good simulation. Among other
>> things, it won’t
>> > affect any device drivers’ interrupt handlers unless you go
>> to extraordinary effort.
>> >
>> > I don’t think there’s any way to make FP calcs happen in software.
>> > They’re part of the instruction set. Of course, you could
>> write your
>> > own FP functions and call those instead of the built-in
>> ones. This is
>> > one of those rare instances where operator overloading in C++ can
>> > actually be a good idea.
>> >
>> > Kutty Banerjee wrote:
>> >> Hi,
>> >> It appears that I can force my machine to behave like it
>> has lesser
>> >> mem that it does with the /maxmem switch in boot.ini. Is there
>> >> something that can be done to make the processor run slower either
>> >> actually or simulate such? Also, is it possible to ensure that all
>> >> floating point calculations are done in software rather than in
>> >> hardware? Can somebody please throw some light on this.
>> >> The reason this intrigues me being that I have an
>> existing code base
>> >> that works under Windows but would like to determine how
>> it performs
>> >> under PDA like conditions. The reason i cant run it on a PDA being
>> >> having to port the code to Win CE which currently is a bottleneck.
>> >>
>> >> thanks
>> >> kutty
>> >
>> > –
>> > …/ray..
>> >
>> > Please remove “.spamblock” from my email address if you need to
>> > contact me outside the newsgroup.
>> >
>>
>>
>>
>> —
>> Questions? First check the Kernel Driver FAQ at
>> http://www.osronline.com/article.cfm?id=256
>>
>> You are currently subscribed to ntdev as:
>> xxxxx@dsto.defence.gov.au
>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>>
>> —
>> Questions? First check the Kernel Driver FAQ at
>> http://www.osronline.com/article.cfm?id=256
>>
>> You are currently subscribed to ntdev as:
>> xxxxx@hollistech.com To unsubscribe send a blank email to
>> xxxxx@lists.osr.com
>>
>
>
>
>
I have no idea. Install it and find out. Its free with VS.NET.
=====================
Mark Roddy
Windows .NET/XP/2000 Consulting
Hollis Technology Solutions 603-321-1032
www.hollistech.com
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Kutty Banerjee
Sent: Tuesday, March 01, 2005 1:09 PM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] simulate a slower machine
Mark:
My question is really out of context in this group but would
you know :“Does the WinCE emulator do software floating point
operations?”
thanks
kutty
“Mark Roddy” wrote in message
> news:xxxxx@ntdev…
> > Why buy an older machine when you can just run the wince simulator
> > that comes with the .net framework? That thing is plenty slow
> >
> >
> > =====================
> > Mark Roddy
> > Windows .NET/XP/2000 Consulting
> > Hollis Technology Solutions 603-321-1032 www.hollistech.com
> >
> >> -----Original Message-----
> >> From: xxxxx@lists.osr.com
> >> [mailto:xxxxx@lists.osr.com] On Behalf Of McNally,
> >> Richard
> >> Sent: Monday, February 28, 2005 10:29 PM
> >> To: Windows System Software Devs Interest List
> >> Subject: RE: [ntdev] simulate a slower machine
> >>
> >> Why simulate when you could just buy a slow machine. Just find the
> >> slowest machine that supports the lowest level OS that your driver
> >> will run on and buy that. Good news is old == cheap (as long as it
> >> works).
> >>
> >> I don’t actually know old x86 CPU would have comparable “grunt” to
> >> the sort of CPUs that you’d find in a modern PDA.
> >> However seeing as you can still buy 386 based laptops on ebay, I’m
> >> guessing that something appropriate is available.
> >> Taking the differences in overhead from Windows CE to Windows ???
> >> might also be difficult though.
> >>
> >> Just a thought in case it never occurred to you.
> >>
> >> Richard McNally
> >>
> >>
> >> -----Original Message-----
> >> From: xxxxx@lists.osr.com
> >> [mailto:xxxxx@lists.osr.com]On Behalf Of
> Kutty Banerjee
> >> Sent: Tuesday, March 01, 2005 9:02 AM
> >> To: Windows System Software Devs Interest List
> >> Subject: Re:[ntdev] simulate a slower machine
> >>
> >>
> >> Thanks for the response but from my own studies i figured it is
> >> easier to actually port the code to PocketPC than do funny
> simulation
> >> stuff:-))
> >>
> >> kutty
> >>
> >> “Ray Trent” wrote in message
> >> news:xxxxx@ntdev…
> >> > Well, I suppose there’s the obvious “put a load on the
> >> machine” approach.
> >> > If you start up a higher-priority thread somewhere that
> >> eats up cycles
> >> > on some periodic basis, it will look much like a slower machine.
> >> > However, it’s not a very good simulation. Among other
> >> things, it won’t
> >> > affect any device drivers’ interrupt handlers unless you go
> >> to extraordinary effort.
> >> >
> >> > I don’t think there’s any way to make FP calcs happen in
> software.
> >> > They’re part of the instruction set. Of course, you could
> >> write your
> >> > own FP functions and call those instead of the built-in
> >> ones. This is
> >> > one of those rare instances where operator overloading
> in C++ can
> >> > actually be a good idea.
> >> >
> >> > Kutty Banerjee wrote:
> >> >> Hi,
> >> >> It appears that I can force my machine to behave like it
> >> has lesser
> >> >> mem that it does with the /maxmem switch in boot.ini. Is there
> >> >> something that can be done to make the processor run
> slower either
> >> >> actually or simulate such? Also, is it possible to
> ensure that all
> >> >> floating point calculations are done in software rather than in
> >> >> hardware? Can somebody please throw some light on this.
> >> >> The reason this intrigues me being that I have an
> >> existing code base
> >> >> that works under Windows but would like to determine how
> >> it performs
> >> >> under PDA like conditions. The reason i cant run it on
> a PDA being
> >> >> having to port the code to Win CE which currently is a
> bottleneck.
> >> >>
> >> >> thanks
> >> >> kutty
> >> >
> >> > –
> >> > …/ray..
> >> >
> >> > Please remove “.spamblock” from my email address if you need to
> >> > contact me outside the newsgroup.
> >> >
> >>
> >>
> >>
> >> —
> >> Questions? First check the Kernel Driver FAQ at
> >> http://www.osronline.com/article.cfm?id=256
> >>
> >> You are currently subscribed to ntdev as:
> >> xxxxx@dsto.defence.gov.au
> >> To unsubscribe send a blank email to
> xxxxx@lists.osr.com
> >>
> >> —
> >> Questions? First check the Kernel Driver FAQ at
> >> http://www.osronline.com/article.cfm?id=256
> >>
> >> You are currently subscribed to ntdev as:
> >> xxxxx@hollistech.com To unsubscribe send a blank email to
> >> xxxxx@lists.osr.com
> >>
> >
> >
> >
> >
>
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
> xxxxx@hollistech.com To unsubscribe send a blank email to
> xxxxx@lists.osr.com
>
> I don’t actually know old x86 CPU would have comparable “grunt” to the
sort of CPUs > that you’d find in a modern PDA.
Just a note here: I still have about 5 old AMD and Cyrix 233/266 machines
that I used for software and driver testing on Win95 OSR2 up through XP SP1.
Performance of these machines is very good on NT 4 and pretty accepttable on
W2K, as long as you don’t bring up Media Player past version 7. Media
Player 8 on W2K on a 266MHz machine can no longer play a wave file that
played fine on older version of MP or the OS. It just sits there and
stutters a lot.
Performance on XP or XP SP1 just plain sucks, but if you are patient they
will boot, and if you turn off enough GUI “pretties” you can run an
application with considerable patience. (Except MP.)
I’ve had XP SP2 trying to boot for a couple of days on one of these
machines, and while it made it to the logon prompt yesterday, it still
hasn’t finished painting the basic user screen background and task bar since
then. It has as usual displayed the screen saver, but it doesn’t have
enough horsepower to actually do anything after the initial draw, so it is a
static image on part of the screen.
I’d recommend not even trying anything slower than a 500MHz machine as an
“old” machine if you want to target a current OS. My 800MHz AMD is quite
usable even on XP SP1, and the 2.4GHz machine can actually run XP SP1 and
host XP SP2 inside VMware.
Loren
> “old” machine if you want to target a current OS. My 800MHz AMD is quite
usable even on XP SP1, and the 2.4GHz machine can actually run XP SP1 and
800MHz machines run XP SP2 fine.
Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com
Loren Wrote:
I’d recommend not even trying anything slower than a 500MHz machine as
an “old” machine if you want to target a current OS. My 800MHz AMD is
quite usable even on XP SP1, and the 2.4GHz machine can actually run XP
SP1 and host XP SP2 inside VMware.
FYI until not so long ago (that’s months rather than years), I was doing
some Win32 development on an old machine at home - a P90.
My suggestion for simulating slower machines: 2nd hand shops, Ebay,
reject shops etc.
Another observation: It’s amazing how many people will sling out
perfectly decent kit just because they’ve misconfigured it / trashed a
drive etc, and all it needs is a reinstall…
MH.
This email and any attachments is confidential, may be legally privileged and is intended for the use of the addressee only. If you are not the intended recipient, please note that any use, disclosure, printing or copying of this email is strictly prohibited and may be unlawful. If received in error, please delete this email and any attachments and confirm this to the sender.