How to CORRECTLY optimize your SSD for windows 7

[SimMike2]

I have to say that I had previously used a tool "SSD Tweaker." This turned off much of the prefetch and superfetch stuff you describe. After reading this post, I decided to launch this program and set things back to Windows 7 defaults.

My computer seems noticeably faster since I reverted to standard Windows 7 configuration. Windows 7 automatically does the best optimizations anyway.

The only thing I did was change my system restore to the bare minimum, which is 1% of my SSD drive. I too have found this feature useful, so I am not that tight for space where I want to switch it off.

In fact, many of these tweaks seem to be because the smaller size of SSD drives in general. I have 40% of my 80GB Intel SSD drive open, so I am not tight for space.

BTW, I do use hibernation mainly because I like how fast my system comes back. I refuse to baby this SSD. If it wants to be a hard drive, I'm going to let it be a hard drive. Who cares if I shorten the lifespan a little bit. I'll have another drive for my boot OS drive long before that ever happens.

 

[flamenko]

Taltamir?

Ummm You asked me to refer to your signiature. The link seems dead. You obviously think that because of the number of posts, you are taken at your word here. Perhaps you might want to tackle each issue at a time so we can have a healthy debate about whether or not the advise in the thread is valid.

I believe it probably is as the site is hitting 200-300 hits daily and I get probably 10-12 e-mails daily thanking me for the site. If you read the entire thread, you will understand that, quite frankly, their is little you can do to speed a ssd even faster than its given speeds and the rest will get you some space back.

With respect to things such as Superfetch, if you are going to enlist in a bit of a discussion, do your homework and then explain what Superfetch and prefetch do followed by what purpose they would serve as the sheer access speeds of the ssd render both moot. Its the same with Indexing and Search I am afraid to say and if you are going to claim a performance advantage, please post the results at which time I will gladly concede your knowledge and thank you for the education.

Yes I am new here. I came to help a few people and thank people for their efforts rather than grab a hold of something that will get me a moment of glory and then watch all my wannabes stand by my side...

Have you got anything you have done to assist others in SSDs or operating systems by the way? I'de like to learn a bit about the person I am hoping to enter into a great conversation with respecting any of the issues he has raised. Lookin forward to the challenge.

Hmmm what issue do we start with...Superfetch/Prefetch...Defrag... Oh...yes.... how about the driver issue...Lets start with chipset drivers.

Ok...simple really. We always recommend a new user do a fresh install of Win7 on a SSD. There are just too many examples of cloning and moving a system over that has had bad results which are mainly sub par performance. Anyway, the biggest thing people get caught with in their new installation is the dreaded yellow triangle and exclamation mark in system devices. Yup....pretty much always. 90% of the time do you know what solves this? You got it my friend....the manufacturers download of chipset drivers. I could be wrong but I got the impression that you never agreed with this but never understood why. You see, the AHCI driver is simply either that of Win7 or that of Intel if you elect to install it. There just may be a copy on the driverset but guess what, it pretty much just works out that TRIM is working...or rather...the commands are being sent from the software to the hardware.

Your ball.

 

[StinkyPinky]

I don't think you need 1.5x your system ram for a pagefile anymore. sure that was sound advice when your pc had 512mb of ram, but most of us have between 4 to 12 gigs these days.

Nothing has ever used my full 8 gigs of ram, and any programs that require a pagefile to run don't really need 12 gigs of it! I set my pagefile to 4000MB (min and max)

 

[GoSharks]

With respect to things such as Superfetch, if you are going to enlist in a bit of a discussion, do your homework and then explain what Superfetch and prefetch do followed by what purpose they would serve as the sheer access speeds of the ssd render both moot.

From Microsoft:

First, SuperFetch is an enhancement of the Prefetcher that you have probably seen mentioned in previous versions of Windows. The Prefetcher is in charge of storing program information so that often-used programs and processes can run faster. In Vista, the Prefetcher has been tweaked and changed to be much more aggressive and intelligent in its caching to give even greater performance. This new functionality is called SuperFetch.

SuperFetch keeps track of which applications you use most and loads this information in RAM so that programs load faster than they would if the hard disk had to be accessed every time. Windows SuperFetch prioritizes the programs you're currently using over background tasks and adapts to the way you work by tracking the programs you use most often and pre-loading these into memory. With SuperFetch, background tasks still run when the computer is idle. However, when the background task is finished, SuperFetch repopulates system memory with the data you were working with before the background task ran. Now, when you return to your desk, your programs will continue to run as efficiently as they did before you left. It is even smart enough to know what day it is in the event you use different applications more often on certain days.

http://blogs.technet.com/askperf/archive/2007/03/29/windows-vista-superfetch-readyboost.aspx

RAM is faster than a SSD.

 

[flamenko]

Go Sharks...

From Your article...

SuperFetch keeps track of which applications you use most and loads this information in RAM so that programs load faster than they would if the hard disk had to be accessed every time.

Now...lets examine why this came into play and leave the quality of Vista out of it. It comes into play because the bottle neck of your system is the hard drive and not the ssd.

Can you list for me a single program that will start faster with Pre/Superfetch enabled with a SSD vice one that is not. You need to understand that Win7 and other applications are created to account for the fact that most people are using systems with hdd where Pre/Superfetch has a purpose.

From MSDN...I think they would qualify as understanding Superfetch much better than you or I.

Will Superfetch be disabled on SSDs?

Yes, for most systems with SSDs.

If the system disk is an SSD, and the SSD performs adequately on random reads and doesn’t have glaring performance issues with random writes or flushes, then Superfetch, boot prefetching, application launch prefetching, ReadyBoost and ReadDrive will all be disabled.

Initially, we had configured all of these features to be off on all SSDs, but we encountered sizable performance regressions on some systems. In root causing those regressions, we found that some first generation SSDs had severe enough random write and flush problems that ultimately lead to disk reads being blocked for long periods of time.

 

[taltamir]

Ummm You asked me to refer to your signiature. The link seems dead.

Anandtech has changed their forum software and apparently that link format changed.
However you do not need to follow the link. I bolded the relevant info in my signature.

[quoteI do not have a superman complex; for I am God, not superman!

No matter how bad things are, the government can always make it worse!

main: Q9400, 8GB DDR2, eVGA GTX260 SC, gigabyte EP35-DS3R. X25-M 80GB G2.
fileserver: AMD X2 EE @ 2.4ghz, 4GB DDR2, 5x750GB WD CaviarGP drives in raidz2 (ZFS raid6).
http://www.anandtech.com/myanandtech.aspx?member=141398[/quote]

You obviously think that because of the number of posts, you are taken at your word here. Perhaps you might want to tackle each issue at a time so we can have a healthy debate about whether or not the advise in the thread is valid.

I have made a point to answer EACH and every one of the points you raised, not only the ones I could easily answer, but each and every one of the 7 of them.
I assure you I do not believe that my post count lends me credibility.

I believe it probably is as the site is hitting 200-300 hits daily and I get probably 10-12 e-mails daily thanking me for the site. If you read the entire thread, you will understand that, quite frankly, their is little you can do to speed a ssd even faster than its given speeds and the rest will get you some space back.

"there is little you can do" not "their is little you can do"
Their = That which belongs to them, their house, their car, their pain.
There = a location; over there; there, by that house.
They're = short for They are

And as for the site getting 200 hits a day... so?
last I check <insert political figure X> and <insert political figure Y whose views contradict X> are greatly popular and yet one must be wrong.

With respect to things such as Superfetch, if you are going to enlist in a bit of a discussion, do your homework and then explain what Superfetch and prefetch do followed by what purpose they would serve as the sheer access speeds of the ssd render both moot. Its the same with Indexing and Search I am afraid to say and if you are going to claim a performance advantage, please post the results at which time I will gladly concede your knowledge and thank you for the education.

http://en.wikipedia.org/wiki/Windows_Vista_I/O_technologies#SuperFetch
http://www.anandtech.com/show/2163/4

How about you (or the site you extol) do some benchmarking and show why it doesn't work? every benchmark I have seen has shown it to improve performance, or at worst not affect it. The above link is just a quick google search, I am drawing on my recollection of many other reviews that are harder to find.

Yes I am new here. I came to help a few people and thank people for their efforts rather than grab a hold of something that will get me a moment of glory and then watch all my wannabes stand by my side...

You accused me of lording my post count, I have never mentioned it.
You get defensive about your post count, I have never mentioned it.
You act defensive as if you have been somehow criticized, yet all I did was respond to your arguments with counter arguments which were completely neutral and never once attacked you or your character.
Heck, while you link the SSD guide in your signature, it is seperate then an entry you call "My Computer and Win7 Optimization Guide" so I must guess that you are not even the author.

Have you got anything you have done to assist others in SSDs or operating systems by the way? I'de like to learn a bit about the person I am hoping to enter into a great conversation with respecting any of the issues he has raised. Lookin forward to the challenge.

My advice to the layman was "wait for windows 7", for the more adventurous I could point at a dozen or so guides...
I am so very tempted to actually point at things and say "look, I created THIS!" but I don't find such e-peen waving to be necessary or conductive. I have laid out my points, if you seek to counter any I shall either counter those, or concede my error. Learning from others is the ultimate goal of any argument I enter.

 

[taltamir]

Can you list for me a single program that will start faster with Pre/Superfetch enabled with a SSD vice one that is not. You need to understand that Win7 and other applications are created to account for the fact that most people are using systems with hdd where Pre/Superfetch has a purpose.

An SSD boasts access times about 100 times faster then a spindle drive, and throughput about 2 to 2.5x.
Ram boasts access times MILLIONS of times faster than a spindle drive, and a throughput ~100x greater.
I leave it up to you to deduce how much faster ram is than an SSD.

 

[flamenko]

I don't think you need 1.5x your system ram for a pagefile anymore. sure that was sound advice when your pc had 512mb of ram, but most of us have between 4 to 12 gigs these days.

Nothing has ever used my full 8 gigs of ram, and any programs that require a pagefile to run don't really need 12 gigs of it! I set my pagefile to 4000MB (min and max)

Agreed to an extent... Now Pagefile. The ONLY purpose of pagefile is to make up for physical RAM that your system does not have. If your ram has enough physical RAM, pagefile is useless except, it is still in use whether you like it or not. Shutting pagefile down, as countless people have done with adequate ram, has resulted in NO PERFORMANCE GAIN, but rather, just under 4Gb of valuable SSD property being returned for other things.

Quite frankly, why would we want services being allowed to run which serve no purpose? And further....before someone jumps on the 'but applications need it'...applications are created to grab your physical ram before pagefile and, in some cases, grab pagefile on the belief that physical ram is not there. If you shut down your pagefile and force your system to use that valuable ram you paid for, it would only make sense.

My experience with pagefile.... I have been fighting to get 64 bit moved forward since early 2007. I have done ALOT of work with Dell in order to get it out in their systems and, well, look where we are now right? Since Vista was released and to this post I have never used Pagefile. I am somewhat of a power user trust me and you can google the net to see my background. My system has not crashed or stated it had low memory even once. Not a problem...

Next consider that Win7 uses physical storage in a totally different way consider to earlier OSs? Did you know that, well, lets imagine that you had only 4Gb ram and Pagefile turned off as an example... Would you believe you could run over 50 programs sude by side without any problem whatsoever??? Hmmm pretty amazing huh??? Its how Win7 negotiates its storage between the SSD and physical ram....without Pagefile.

 

[GoSharks]

Now...lets examine why this came into play and leave the quality of Vista out of it. It comes into play because the bottle neck of your system is the hard drive and not the ssd.

I wish there was an explaination for why they turn Superfetch off, because RAM is still faster than a SSD.

 

[SimMike2]

An SSD boasts access times about 100 times faster then a spindle drive, and throughput about 2 to 2.5x.
Ram boasts access times MILLIONS of times faster than a spindle drive, and a throughput ~100x greater.
I leave it up to you to deduce how much faster ram is than an SSD.

You are right, no matter fast an SSD is compared to hard drives, RAM is faster. So the prefetch and superfetch stuff serve a valuable purpose.

 

[flamenko]

An SSD boasts access times about 100 times faster then a spindle drive, and throughput about 2 to 2.5x.
Ram boasts access times MILLIONS of times faster than a spindle drive, and a throughput ~100x greater.
I leave it up to you to deduce how much faster ram is than an SSD.

Thats convenient and I have to say that in the early days I used a very similar analogy for pagefile...

By shutting down pagefile, performance HAS to be improved because your physical ram is much faster than a hard drive swap file which also has to consider access times. Unfortunately, as hard as I tried, I couldnt document any performance change so I had to concede the fact that Pagefile ONLY gains you property back.

By logic however, its a natural assumption that No Pagefile would be faster right? In fact, MS also has a radio button to shut it off? Ever wonder why?

So having said that and when entering into our little debate, you cant just say, "It has to be" if you cannot backup your words, similar to what I did above with respect to MSDNs words pertaining to Superfetch.

 

[flamenko]

You are right, no matter fast an SSD is compared to hard drives, RAM is faster. So the prefetch and superfetch stuff serve a valuable purpose.

No it doesnt and even Microsoft engineers concede such...see my quote. So we are left with a simple problem then...and its only fair right?

If we are going to make a point, we need to be somewhat credible and back it up.

Love this kind of thing by the way...


AND IF I MAY.....How can we make an argument that RAM is faster than a SSD for Superfetch and then even remotely try and say Pagefile serves a purpose? Do we know the mechanics of how Pagefile works? Its horrible and if anyone was around for the orig release of XP, they would remember the feeling of Pagefile slowing the system down immensely and the grinding of the drive.

 

[taltamir]

Quite frankly, why would we want services being allowed to run which serve no purpose? And further....before someone jumps on the 'but applications need it'...applications are created to grab your physical ram before pagefile and, in some cases, grab pagefile on the belief that physical ram is not there. If you shut down your pagefile and force your system to use that valuable ram you paid for, it would only make sense.

The game spellforce the order of dawn would, when experiencing large battles with many units, attempt to grab page file space. if your page file was below 2GB this could result in a blue screen as it attempted to handle non existing memory addresses.
I have personally tracked down that problem and reported it to the developer, but they ignored me and I do not know if it was ever fixed. Many people have told me that my solution has solved the crashing for them. And it has eliminated it for me as well. (I deduced it originally from actually reading what the blue screen said )
It is possible that this particular case has been rendered obsolete on newer machines with more ram. But it is possible that there are issues with other programs which are similar. Even if newer versions of windows would prevent such a thing from causing a bluescreen, it would still result in the program in question crashing.

I hear many cases of programs crashing if you completely disable your pagefile, and supposedly a pagefile of mere 256mb can solve most of it.

Of particular interest is that with modern version of windows, such as windows vista and windows 7, a copy of your pagefile is stored in the ram as long as you have enough space for it. So you do not actually lose performance from having it enabled, you only lose out the hard-drive space it claims.

EDIT:

By shutting down pagefile, performance HAS to be improved because your physical ram is much faster than a hard drive swap file which also has to consider access times. Unfortunately, as hard as I tried, I couldnt document any performance change so I had to concede the fact that Pagefile ONLY gains you property back.

Only in very old versions of windows, for newer ones see what I wrote about about handling of pagefile i newer MS OS

Quite frankly, why would we want services being allowed to run which serve no purpose? And further....before someone jumps on the 'but applications need it'...applications are created to grab your physical ram before pagefile and, in some cases, grab pagefile on the belief that physical ram is not there. If you shut down your pagefile and force your system to use that valuable ram you paid for, it would only make sense.

If you disable the pagefile those applications would cause windows to blue screen, or at least crash themselves. They will not be forced to use your ram instead. And again, I point you at what I said about pagefile being stored in ram in newer MS OS.

My experience with pagefile.... I have been fighting to get 64 bit moved forward since early 2007. I have done ALOT of work with Dell in order to get it out in their systems and, well, look where we are now right? Since Vista was released and to this post I have never used Pagefile. I am somewhat of a power user trust me and you can google the net to see my background. My system has not crashed or stated it had low memory even once. Not a problem...

I am very glad you are fighting the good fight for 64bit adoption.

AND IF I MAY.....How can we make an argument that RAM is faster than a SSD for Superfetch and then even remotely try and say Pagefile serves a purpose? Do we know the mechanics of how Pagefile works? Its horrible and if anyone was around for the orig release of XP, they would remember the feeling of Pagefile slowing the system down immensely and the grinding of the drive.

Please, no strawmen. None of us said that pagefile results in faster performance, we said it is needed to prevent programs from crashing themselves or windows. Or even get a benign "out of memory" error from windows when trying to lunch one (I have seen that on a windows vista machine with too little ram and pagefile... actually I have seen all 3 things happens)

And performance is not degraded as pagefile is mirrored into ram in modern windows as long as there is enough free ram.

BTW... what do you think of readyboost.

 

[flamenko]

The game spellforce the order of dawn would, when experiencing large battles with many units, attempt to grab page file space. if your page file was below 2GB this could result in a blue screen as it attempted to handle non existing memory addresses.
I have personally tracked down that problem and reported it to the developer, but they ignored me and I do not know if it was ever fixed. Many people have told me that my solution has solved the crashing for them. And it has eliminated it for me as well. (I deduced it originally from actually reading what the blue screen said )
It is possible that this particular case has been rendered obsolete on newer machines with more ram. But it is possible that there are issues with other programs which are similar. Even if newer versions of windows would prevent such a thing from causing a bluescreen, it would still result in the program in question crashing.

I hear many cases of programs crashing if you completely disable your pagefile, and supposedly a pagefile of mere 256mb can solve most of it.

Of particular interest is that with modern version of windows, such as windows vista and windows 7, a copy of your pagefile is stored in the ram as long as you have enough space for it. So you do not actually lose performance from having it enabled, you only lose out the hard-drive space it claims.

We are starting to see eye to eye I think. You have to have sufficient RAM...absolutely and quite frankly 2Gb is not it for gaming. Would that of happened on an 8Gb system running Win7..

And with respect to your Pagefile being stored in RAM, can you show me somewhere where this is stated? Anywhere? I have never heard this and, well, Pagefile has always been and will reside on your C: drive as pagefil.sys. You can find it if you unhide system files.

It would be kind of fruitless than to store it in Ram since the whole purpose of it is to account for lack of Ram. Thats like putting a Porshe in side a VW so it would drive faster.

If you disable the pagefile those applications would cause windows to blue screen, or at least crash themselves. They will not be forced to use your ram instead. And again, I point you at what I said about pagefile being stored in ram in newer MS OS.

This is incorrect because all applications are created to search for RAM first and foremost. If you have the ram it will not bluescreen....Bluescreen? I thought that went out with IBM' OS eheheh.

BSOD = Blue Screen Of Death

I know and speak to many many gamers who DO NOT use Pagefile....no prob whatsoever as long as you have the RAM....oh...and are using a somewhat frequent OS.

 

[flamenko]

Ok so...Superfetch is discounted by Microsoft itself. Pagefile is not only discounted by you in pointing out the speed of physical ram compared to that of the Pagefile but also, well, we cant find anywhere where it shows any performance benefit or purpose can we? The simple fact is that an ssds sheer speed takes care of this. Can you find anything that will start and run better with Pagefile vice without? Even the mechanics of the issue bolis down to a simple logic here.

Found this as well....

If you have more than 4GB RAM you can disable it. If you have less, move it to the HDD. As its hardly used; only in emergency situations. And if its used; your performance will be extremely bad; like Pentium 1 systems. Toms hardware has an article about disabling swap, if you have enough memory and are not running ancient operating systems, it should be no problem. Putting it on your SSD only wastes valuable space.

We can probably dig up the Toms article I am sure...

___________________________________________________________________

Prefetch...off
Pagefile (with enough RAM) off to gain back almost 4Gb SSD space

Next issue? I will let you bring this one out.


EDIT: Readyboost does not work at all with SSDs...I can dig the article up if you like...might take a bit as my folder has so many different references...

Here is is...again from MSDN Engineers...

If the system disk is an SSD, and the SSD performs adequately on random reads and doesn&#8217;t have glaring performance issues with random writes or flushes, then Superfetch, boot prefetching, application launch prefetching, ReadyBoost and ReadDrive will all be disabled.

 

[taltamir]

We are starting to see eye to eye I think. You have to have sufficient RAM...absolutely and quite frankly 2Gb is not it for gaming. Would that of happened on an 8Gb system running Win7..

I have not yet changed my position on anything. There might be some communication issues involved rather then disagreements.

And with respect to your Pagefile being stored in RAM, can you show me somewhere where this is stated? Anywhere? I have never heard this and, well, Pagefile has always been and will reside on your C: drive as pagefil.sys. You can find it if you unhide system files.

a copy of it is stored in ram, if there is enough space. I don't remember the source on that one but I could search. Do I really need to find links for every single claim I utter, because I have a finite amount of time. My first final is friday.

It would be kind of fruitless than to store it in Ram since the whole purpose of it is to account for lack of Ram. Thats like putting a Porshe in side a VW so it would drive faster.

Actually it will make a lot of sense to store it in otherwise FREE ram... unused RAM is wasted ram. Some developers wrongly put stuff in pagefile that should go into ram. Pagefile is meant to prevent crashing due to lack of ram, but placing excess data on the HDD. Since you say that disabling pagefile increases performance by forcing its contents into ram instead, then allowing them to be placed into "pagefile" and placing that in ram provides an equal increase to ram... the point is that when you have a LEGITIMATE lack of space you can discard the pagefile contents from ram (since they are also mirrored to the HDD) and place unique content there.

This is incorrect because all applications are created to search for RAM first and foremost. If you have the ram it will not bluescreen....Bluescreen? I thought that went out with IBM' OS eheheh.

BSOD = Blue Screen Of Death

On my third day of C++ programming I accidentally wrote a bad pointed app that blue screened any OS it was run on... I REALLY wish I had saved the code for that one.
I assure you not all applications are written correctly. They SHOULD be, but they are not.

I know and speak to many many gamers who DO NOT use Pagefile....no prob whatsoever as long as you have the RAM....oh...and are using a somewhat frequent OS.

Which has exactly nothing to do with warning about specific programs causing crashes without pagefile.

 

[Voo]

This is incorrect because all applications are created to search for RAM first and foremost. If you have the ram it will not bluescreen....Bluescreen?

That's how it should be (and I've got really no idea why anyone ever did something else, but hey other programs created their own page files also), BUT you can still do it the other way round, so he's right in that point..

But your claims about SSDs not being slower than RAM and wanting to see proofs.. you're kidding right? You're not really claiming that RAM (access time: <20ns) isn't faster than a SSD (access time: ~100.000ns - we shouldn't even start talking about throughput.. that'll get ugly). It's not us who have to find proofs, but you since you're making that rather hilarious claim. I mean just look at the technical specs xX

@taltamir: Nah, storing the pagefile in RAM is completely, utterly crazy and useless. The whole point of a pagefile is to swap out pages when you get a pagefault and have no free memory. If you need the pagefile the only reason is (well should be.. you can abuse everything) that you don't have enough memory. If you can hold a page in memory you can just assign it to it's correct process, no point in doing that otherwise.
I think you mixed that up with page tables (and actually they're using multi level page table so they don't even have to hold all page tables in memory)

 

[taltamir]

Ok so...Prefetch is discounted by Microsoft itself.

Where?

Pagefile is not only discounted by you in pointing out the speed of physical ram compared to that of the Pagefile but also, well, we cant find anywhere where it shows any performance benefit or purpose can we? The simple fact is that an ssds sheer speed takes care of this. Can you find anything that will start and run better with Pagefile vice without? Even the mechanics of the issue bolis down to a simple logic here.

Why are you insisting on persisting with this strawman? we said countless times now that it has nothing to do with performance.
The argument is:
1. Pagefile neither harms not improves performance.
2. Leaving pagefile on prevents crashes or other issues.

Yet you insist on misrepresenting our argument as "pagefile results in greater performance" or now as "so you admit disabling pagefile improves performance"...

That's how it should be (and I've got really no idea why anyone ever did something else, but hey other programs created their own page files also), BUT you can still do it the other way round, so he's right in that point..

Oh yea... like photoshop, it actually creates its OWN pagefile. And unlike windows, this one is 100&#37; HDD based.

 

[flamenko]

I also commented on the Readyboost issue above...

All I can ask is that you provide some sort of backing as I try. I have never heard of Pagefile being loaded into RAM if RAM is available and even you have to admit, it is kind of foolish since the premise of Pagefile is to emulated RAM with hd/ssd space to account for the lack of true physical RAM.

I don't write code at all so if you go there I cannot respond, although I have a great network of people that can assist. You have 8Gb now and I will bet you never when you had your last BSOD... Hmmmmm??? eheheheh. Win7 with enough RAM and an SSD is bulletproof.

I know because Im a PC and I created Windows 7!!!

eheheh little humour.

 

[flamenko]

That's how it should be (and I've got really no idea why anyone ever did something else, but hey other programs created their own page files also), BUT you can still do it the other way round, so he's right in that point..

But your claims about SSDs not being slower than RAM and wanting to see proofs.. you're kidding right? You're not really claiming that RAM (access time: <20ns) isn't faster than a SSD (access time: ~100.000ns - we shouldn't even start talking about throughput.. that'll get ugly). It's not us who have to find proofs, but you since you're making that rather hilarious claim. I mean just look at the technical specs xX

No I am not making that claim whatsoever. I am only stating that there is no proven performance gain period. MS has stated that Superfetch is moot with a SSD and, well, to date anyway, nobody has shown any proof of any performance gain in keeping on Superfetch to use RAM period.

EDIT: Im really trying to keep up here but you guys are hitting 4 or 5 different things simultaneously. I posted exactly what is in print with respect to the MSDN posting...word for word...

With respect to Pagefile, your view is that it prevents crashing??? My view is that, with sufficient RAM it won't crash because Pagefile emulates the lack of physical RAM and nothing more. I only brought forward the performance thing to show how we can't claim performance without backing up our claim.

You can shout from the rooftops that Superfetch speeds things up but if you cannot show something that proves it, it doesnt especially since MSDN itself discounts the use of Superfetch and readyboost with an SSD.

 

[taltamir]

And with respect to your Pagefile being stored in RAM, can you show me somewhere where this is stated? Anywhere? I have never heard this and, well, Pagefile has always been and will reside on your C: drive as pagefil.sys. You can find it if you unhide system files.

a copy of it is stored in ram, if there is enough space. I don't remember the source on that one but I could search. Do I really need to find links for every single claim I utter, because I have a finite amount of time. My first final is friday.

Now you have me second guessing this... I don't exactly recall the source on this one and it could have been a lie I heard on some forum and accidentally internalized. I am going to have to do some research to verify whether this is true or false.

All I can ask is that you provide some sort of backing as I try. I have never heard of Pagefile being loaded into RAM if RAM is available and even you have to admit, it is kind of foolish since the premise of Pagefile is to emulated RAM with hd/ssd space to account for the lack of true physical RAM.

Backing to what, specifically? I have linked things on occasion. I have not seen you back things up yet with anything but strawman (aka, misrepresent our claim and then disprove the misrepresentations).

All I can ask is that you provide some sort of backing as I try. I have never heard of Pagefile being loaded into RAM if RAM is available and even you have to admit, it is kind of foolish since the premise of Pagefile is to emulated RAM with hd/ssd space to account for the lack of true physical RAM.

It isn't foolish, it is the only sensible way to handle it.

I don't write code at all so if you go there I cannot respond, although I have a great network of people that can assist. You have 8Gb now and I will bet you never when you had your last BSOD... Hmmmmm???

I don't remember when I had my last BSOD, but I have had my pagefile enabled the whole time so it is utterly irrelevant to the discussion at hand.

 

[taltamir]

No I am not making that claim whatsoever. I am only stating that there is no proven performance gain period. MS has stated that Superfetch is moot with a SSD and, well, to date anyway, nobody has shown any proof of any performance gain in keeping on Superfetch to use RAM period.

No they didn't.
1. I don't see the text you quoted in the link you attribute it to.
2. The text you quote does not in any way shape or form discredit superfetch... it says MS has chosen to disable it on SSDs, but discovered that on some SSDs it caused terrible performance degredation to disable it, so they leave it on for those.
Their reasoning for disabling/enabling it on others is not delved into and its performance results with those are unmentioned. All they say is that some SSDs have unacceptably bad performance with superfetch disabled and must have it enabled. And that their ORIGINAL plan was to disable it for ALL SSDs before they found out, empirically, that it degrades performance on SSDs.

As for proof... show us proof it doesn't. It is known that ram is faster, you do not dispute it... what more do you want?

 

[Voo]

No I am not making that claim whatsoever. I am only stating that there is no proven performance gain period. MS has stated that Superfetch is moot with a SSD and, well, to date anyway, nobody has shown any proof of any performance gain in keeping on Superfetch to use RAM period.

Well I see only quotes but not the original source so there could be a couple of points. First of all clearly a SSD is faster than a HDD so superfetch is less needed than on a HDD, second I just checked and SuperFetch is enabled on my desktop with a 160GB Intel G2, so even if the MSDN engineer said that, it seems MS changed their mind.

And I think of a way to test that for sure, maybe I'll write some small scripts to start some programs several time with/without superfetch, but there are a couple of rather obvious problems so we've to be careful - but if SuperFetch works correctly it is obvious that it should be a lot faster. Throughput, access times, RAM is faster in every possible metric and that by several orders of magnitude.

Other than that: You're right on the pagefile (it's not stored in memory, that'd be stupid - I think taltamir thinks about page tables) and it really shouldn't be used by any program (and I haven't seen one in a long time to use it). But it IS possible..

 

[flamenko]

Lets cut out the digs ok? You started, I shot back, you came back and on and on and on....

I am not a strawman and have presented quotes to back up my views on 2 occasions at least. I speak of education, experience and one hell of a support network that has helped me through these battles for years now.

I know you are a smart guy...its obvious. I am not perfect and without error I am sure. In the end though, it is wrong to discredit a blog and article as you have done on a number of posts here without having the support to back it.

By the way...God is a simple creation with no existence of proof whatsoever whereas the theory of evolution is hard as a rock. We, as a society, have to have this as a framework in order to survive. Everything throughout history backs on to religion as a form of framework for their society...except for Hitler of course.

It was a brilliant plan early on....it still works but the fact lies...their is no proof whatsoever...

Just going back to your sig for a bit. I can even go into the genetic code and evolution if you want eheheh.

 

[taltamir]

Other than that: You're right on the pagefile (it's not stored in memory, that'd be stupid - I think taltamir thinks about page tables) and it really shouldn't be used by any program (and I haven't seen one in a long time to use it). But it IS possible..

That is a probably explanation.

I am not a strawman and have presented quotes to back up my views on 2 occasions at least. I speak of education, experience and one hell of a support network that has helped me through these battles for years now.

Of course you aren't a strawman, you are a person... you USED a strawman, which is a technique where you misrepresent an argument then disprove the argument which nobody else ever claimed.
Also, presenting evidence has nothing to do with strawman status. neither does education or experience.
All a strawman means is that the things you are disproving are distortions of what the other guy actually claims, that you are not disproving the claims of the opposition, but claims you invented and attribute to the opposition.

I am not a strawman and have presented quotes to back up my views on 2 occasions at least. I speak of education, experience and one hell of a support network that has helped me through these battles for years now.

You presenting a quote which you attributed to a link, the link does NOT contain that quote, that quote says:
1. We wanted to disable it on all SSDs, but we saw it lowered performance a lot on jmicrons so we leave it on for those and disable it for high end SSDs like intel.
2. I and Voo own an intel G2 and both of us has it enabled, so the quote is obviously incorrect.

By the way...God is a simple creation with no existence of proof whatsoever whereas the theory of evolution is hard as a rock. We, as a society, have to have this as a framework in order to survive. Everything throughout history backs on to religion as a form of framework for their society...except for Hitler of course.

huh?

Just going back to your sig for a bit. I can even go into the genetic code and evolution if you want eheheh.

Oh my signature... yea, a little joke on my part... someone once told me that I think I am all that, that I have a superman complex... and I retorted with "nah, I don't think I am superman... I know who I am, I am God!"
Please lets not discuss genetics, evolution, god, etc here... I distinctly avoid those as samples for what a strawman is for a reason. Lets keep things tech related.