Page file

"CBFalconer" wrote in message
...
GT wrote:

... snip ...

If a swapfile exists, windows will use it in an attempt to keep as
much physical RAM available at any one time. In other words, if you
have 2GB of RAM and are using only 400MB, then windows will still
insist on swapping pages out to the pagefile, to keep the available
RAM as high as possible. If something that has been swapped out is
then required it is reloaded from the swapfile, introducing a
performance degredation which we have been trying to tell you about.
If you choose to run without a swapfile, then there would be no
performance drop - simple as that.

Are you saying that MS is so stupid that they swap pages out, and
then swap them back in even if they haven't been disturbed? I can
well believe that, considering their other myriad stupidities.

Interesting! I don't know what windows does about a paged out page if the
RAM hasn't been disturbed. I presume in most cases the RAM will have been
used by something else, but I expect that the algorithm marks the pages out
RAM area as blank, so it won't even know where the page was originally
anyway!

[Rod] AND I have proven that its just plain pig ignorant drivel with
the last ram upgrade. The system uses the swap file MUCH
less it did with the smaller amount of physical ram, as it should.

[GT] So you have finally taken on board the fact that windows does use the
swapfile when there is still sufficient free RAM. Good.




[GT] If something that has been swapped out is
then required it is reloaded from the swapfile,

[Rod] And that is just plain wrong too, and its completely trivial to
prove that that stupid pig ignorant claim is just plain wrong too.

[GT] Given that you have told us your PC still pages out to swapfile, please
explain to us all, in an adult sentence, how your special modified version
of windows gets those paged-out pages back into RAM without hard disk access
and a corresponding delay?

GT wrote:

[Rod] AND I have proven that its just plain pig ignorant drivel with the last ram upgrade. The
system uses the swap file MUCH
less it did with the smaller amount of physical ram, as it should.

[GT] So you have finally taken on board the fact that windows does
use the swapfile when there is still sufficient free RAM. Good.

[GT] If something that has been swapped out is
then required it is reloaded from the swapfile,

[Rod] And that is just plain wrong too, and its completely trivial to
prove that that stupid pig ignorant claim is just plain wrong too.

[GT] Given that you have told us your PC still pages out to swapfile,

I never ever said anything even remotely resembling anything
like that in the sense that swapfile use doesnt change.

please explain to us all, in an adult sentence, how your special
modified version of windows gets those paged-out pages back into RAM without hard disk access and
a corresponding delay?

You aint even established that it does get anything back
from the swap file when you have enough physical ram
to be able to do without a swap file and have one anyway.

And its completely trivial to prove that it doesnt in fact do that with the test I spelt out.

GT wrote
CBFalconer wrote
GT wrote

If a swapfile exists, windows will use it in an attempt to keep as
much physical RAM available at any one time. In other words, if you
have 2GB of RAM and are using only 400MB, then windows will still
insist on swapping pages out to the pagefile, to keep the available
RAM as high as possible. If something that has been swapped out is
then required it is reloaded from the swapfile, introducing a
performance degredation which we have been trying to tell you about.
If you choose to run without a swapfile, then there would be no
performance drop - simple as that.

Are you saying that MS is so stupid that they swap pages out, and then swap them back in even if
they haven't been disturbed?

Nope, he's actually stupid enough to claim that, even tho
its completely trivial to prove that that doesnt happen.

Interesting! I don't know what windows does about a paged out page if the RAM hasn't been
disturbed.

It doesnt bother to load it back from the swap file, stupid.

I presume in most cases the RAM will have been used by something else,

Nope, NOT WHEN YOU HAVE ENOUGH PHYSICAL RAM
TO NOT NEED A SWAP FILE AND HAVE ONE ANYWAY.

but I expect that the algorithm marks the pages out RAM area as blank,

Your pig ignorant expectations are your problem, as always.

so it won't even know where the page was originally anyway!

Thanks for that completely superfluous proof that you have never
ever had a clue about how the swapfile is used when you have
enough physical ram to not need a swapfile and have one anyway.

And you are actually so stupid that you cant even manage to do the test
I spelt out and prove that its nothing like what you pig ignorantly claim.

On Wed, 24 Jan 2007 07:23:27 +1100, "Rod Speed"
wrote:


Nope, NOT WHEN YOU HAVE ENOUGH PHYSICAL RAM
TO NOT NEED A SWAP FILE AND HAVE ONE ANYWAY.


If you have one anyway, either windows doesn't write to it
or it does. Do you claim it doesn't at all? If it does
even a tiny bit, to that extent it was slower.

If it does only to a tiny extent, we could similarly claim
it won't make much difference having it enabled or disabled
except that if you are so quick to assume you couldn't
possibly estimate the total memory allocation, you probably
can't and would end up not bothering to determine actual
physical memory needs and should then - through in ability,
leave it enabled for that reason alone, OR instead of you
don't have enough physical memory to handle all situations
of use. Remember that not everyone uses a PC as a
one-system-does-everything box, some people have a box to be
their main pc but there are many many uses for a computer
where the needs are quite a bit more fixed, finite.

kony wrote:
Rod Speed wrote
GT wrote
CBFalconer wrote
GT wrote

If a swapfile exists, windows will use it in an attempt to keep as
much physical RAM available at any one time. In other words, if you
have 2GB of RAM and are using only 400MB, then windows will still
insist on swapping pages out to the pagefile, to keep the available
RAM as high as possible. If something that has been swapped out is
then required it is reloaded from the swapfile, introducing a
performance degredation which we have been trying to tell you about.
If you choose to run without a swapfile, then there would be no
performance drop - simple as that.

Are you saying that MS is so stupid that they swap pages out,
and then swap them back in even if they haven't been disturbed?

Nope, he's actually stupid enough to claim that, even tho
its completely trivial to prove that that doesnt happen.

Interesting! I don't know what windows does about
a paged out page if the RAM hasn't been disturbed.

It doesnt bother to load it back from the swap file, stupid.

I presume in most cases the RAM will have been used by something else,

Nope, NOT WHEN YOU HAVE ENOUGH PHYSICAL RAM
TO NOT NEED A SWAP FILE AND HAVE ONE ANYWAY.

If you have one anyway, either windows doesn't write to it or it does.

Irrelevant. What matters is WHETHER IT READS WHAT IS IN
THE PAGE FILE BACK INTO PHYSICAL RAM, performance wise.

Do you claim it doesn't at all?

Nope, didnt say that about WRITES.

If it does even a tiny bit, to that extent it was slower.

Wrong again IF IT DOES THAT IN THE BACKGROUND.

If it does only to a tiny extent, we could similarly claim
it won't make much difference having it enabled or disabled
except that if you are so quick to assume you couldn't
possibly estimate the total memory allocation, you probably
can't and would end up not bothering to determine actual
physical memory needs and should then - through in ability,
leave it enabled for that reason alone, OR instead of you don't
have enough physical memory to handle all situations of use.

Or you actually have enough of a clue to have noticed that
some apps make stupid requests for ram they will never ever
use and that its usually impossible to be sure that you wont
ever run one of those, even if only when having a quick try of it.

Remember that not everyone uses a PC as a
one-system-does-everything box, some people have a box
to be their main pc but there are many many uses for a
computer where the needs are quite a bit more fixed, finite.

Irrelevant when still having a page file when you have enough physical
ram to be able to do without one has no effect on performance.

The short story is that Win does indeed write some stuff to the
page file when you have lots of physical ram, basically bits of
what is in physical ram that are less likely to be used soon,
and it does that in the background, where that has no impact
on performance. It does that because its impossible to predict
when it will be necessary to free up physical ram, and its faster
if that stuff is already in the page file, no need to pause while
its written to the page file to free up physical ram.

When you have enough physical ram to be able to do without
a pagefile, there is never any need to load that stuff out of the
pagefile again, because it was never removed from physical
ram because more free physical ram was never needed.

Bet you STILL wont be able to comprehend something as
basic as that, or even try the simple test I listed that proves
thats how Win operates when you have enough physical
ram to be able to do without a page file and have one anyway.

Bet you keep digging that hole you have dug for yourself.

Give my regards to the current top banana
in china when you end up coming out in there.

On Wed, 24 Jan 2007 12:27:11 +1100, "Rod Speed"
wrote:


Irrelevant. What matters is WHETHER IT READS WHAT IS IN
THE PAGE FILE BACK INTO PHYSICAL RAM, performance wise.


You'll be right the moment there is a system that uses zero
time and resources to write to a pagefile. Let us know when
you discover such a magical beast.

kony wrote
Rod Speed wrote

Irrelevant. What matters is WHETHER IT READS WHAT IS IN
THE PAGE FILE BACK INTO PHYSICAL RAM, performance wise.

You'll be right the moment there is a system that
uses zero time and resources to write to a pagefile.

Wrong, as always. Even someone as stupid as you should
be able to grasp that as long as those writes are done
IN THE BACKGROUND WITH SURPLUS RESOUCES,
there will be no impact what so ever on PERFORMANCE.

Let us know when you discover such a magical beast.

Never ever could bull**** its way out of a wet paper bag.

Or manage to grasp the Rule of Holes either.

"kony" wrote in message
...
On Wed, 24 Jan 2007 12:27:11 +1100, "Rod Speed"
wrote:


Irrelevant. What matters is WHETHER IT READS WHAT IS IN
THE PAGE FILE BACK INTO PHYSICAL RAM, performance wise.

You'll be right the moment there is a system that uses zero
time and resources to write to a pagefile. Let us know when
you discover such a magical beast.

I quote here from the postings of Rod Speed:

We have learned that Rod's system, "...uses the swap file MUCH less it did
with the smaller amount of physical ram". Much less is not none, so while
his PC still swaps out to swapfile, he assures us that it never reads in
from that swapfile, "But not returned from the swap file to physical ram, so
no effect on performance". So his version of windows can predict which pages
it will not require, so swaps them out of RAM into virtual memory, safe in
the knowledge that they will never be required. All this happens without
using the hard disk as he said there is no performance hit - Fantastic! I
challenged him on this and he replied with a series of insults, which
clearly means he is right.

We told him that with sufficient RAM, he doesn need a swapfile, he said he
does. Microsoft says he doesn't!
We told him about the recommendation that it should be 1.5x the amount of
RAM, he said that was a childish claim. It is Microsoft's recommendation!
We told him that the swapfile is used regardless of the amount of RAM in a
PC, he said both that it is and it isn't (in the same post) - so he can't be
wrong! Microsoft says it is used!

A few quotes from a page on Microsoft's website concerning virtual memory
(http://support.microsoft.com/kb/555223):

"the normal recommendation of 1.5 times the amount of RAM in the computer is
a good place to start."

"The operating system uses the paging system for purposes other than
swapping pages due to memory over commitment."

There is also text in there about the situation whereby a system has
sufficient RAM to run without a pagefile. If you have a pagefile anyway,
then it is still used.

Clearly microsoft don't know how their own OS works, because Rod Speed must
be right - how can you argue with supporting statements such as "wrong" and
"Nope". So compelling!

I have left spaces between these quotes so that Rod can disagree directly
with microsoft.

I've just had a thought - perhaps his hard disk light is broken, so he
doesn't know when the pagefile is being used! Please feel free to continue
the battle Kony, but I've given up on him - he just doesn't understand!

GT wrote
kony wrote
Rod Speed wrote

Irrelevant. What matters is WHETHER IT READS WHAT IS IN
THE PAGE FILE BACK INTO PHYSICAL RAM, performance wise.

You'll be right the moment there is a system that uses zero
time and resources to write to a pagefile. Let us know when
you discover such a magical beast.

I quote here from the postings of Rod Speed:

You lie about what I actually said.

We have learned that Rod's system, "...uses the swap file MUCH less it did with the smaller amount
of physical ram". Much less is not none, so while his PC still swaps out to swapfile, he assures
us that it never reads in from that swapfile, "But not returned from the swap file to physical
ram, so no effect on performance".

I didnt say that about mine, I said that about a system that has enough
physical ram to do without a swap file and you have one anyway.

So his version of windows

Nothing to do with my version of windows.

can predict which pages it will not require, so swaps them out of RAM into virtual memory, safe in
the knowledge that they will never be required.

I never ever said anything even remotely resembling anything like that.

I ACTUALLY said that win does write some stuff to the swap file that
it decides is less likely to be used any time soon, and it does that
while there is still plenty of free physical ram, because it can do that
in the background and have no effect on performance, and its better
to do that while there is still plenty of free physical ram, because that
can be written to the swap file without affecting performance when
its done at that time, rather than waiting till there isnt enough free
physical ram to do that and having to wait while that happens then.

Obviously if you have enough physical ram to not need a swap file
and have one anyway, what was written to the swap file in case
there is ever a shortage of physical ram wont in fact occur, so
there wont be any need to read it into physical ram again later,
its just in the swap file AND in physical ram until its rebooted.

All this happens without using the hard disk as he said there is no performance hit - Fantastic!

Even someone as stupid as you should be able to grasp that
if the stuff that is written to the swapfile is done in the background
when nothing much is going on, it wont affect the performance.

We told him that with sufficient RAM, he doesn need a swapfile, he said he does.

I never ever said that either.

Microsoft says he doesn't!

Lying again.

We told him about the recommendation that it should be 1.5x the amount of RAM, he said that was a
childish claim. It is Microsoft's recommendation!

No it isnt. MS doesnt say it anything like that absolutely.

AND that is nothing like you claim just above that claim that
MS says you dont need a swap file either. Cant even manage
a consistent line of lies from one para to the next one.

We told him that the swapfile is used regardless of the amount of RAM in a PC, he said both that
it is and it isn't

No I didnt.

(in the same post) - so he can't be wrong!

Never ever could lie its way out of a wet paper bag.

Microsoft says it is used!

A few quotes from a page on Microsoft's website concerning virtual
memory (http://support.microsoft.com/kb/555223):

"the normal recommendation of 1.5 times the amount of RAM in the computer is a good place to
start."

Nothing like your previous claim. Concentrate on A GOOD PLACE TO START.

"The operating system uses the paging system for purposes other than swapping pages due to memory
over commitment."

But some of those are irrelevant to most, most obviously with crash dumps.

There is also text in there about the situation whereby a system has sufficient RAM to run without
a pagefile. If you have a pagefile anyway, then it is still used.

But it does NOT say that it used the pagefile identically
to how its used when you dont have enough physical ram.

That was YOUR pig ignorant drivel.

Clearly microsoft don't know how their own OS works, because Rod Speed must be right

I told you how to test your stupid pig ignorant claim that the pagefile use
doesnt change when you have enough physical ram to be able to do without it.

- how can you argue with supporting statements
such as "wrong" and "Nope". So compelling!

Lying, again.

I have left spaces between these quotes so that Rod can disagree directly with microsoft.

I havent ever done that, liar.

I've just had a thought

Obvious lie, ear to ear dog **** isnt capable of thought.

- perhaps his hard disk light is broken,

Nope.

so he doesn't know when the pagefile is being used! Please feel free to continue the battle Kony,
but I've given up on him - he just doesn't understand!

Everyone's noticed that its you that doesnt understand. They'd be putting
the boot in if they agreed with you, and they aint doing that child.

Have you the remotest concept of how many are laughing at your predicament ?