Windows Vista forum

General discussion

Need opions informatin on Superfetch service

by Dango517 / November 25, 2008 1:25 AM PST

My OS is Windows Vista Business.

The purpose of superfetch is to improve a PC's performance over time, after some time is it a good idea to restart this service? This might be especially true if a PC has developed problems.

I think so because it will reach a point where it's limit and ability to do further optimizations will be accomplished. At this point restarting it might further improve a PC.

Any comment about resetting "Microsoft .Net Framework NGen" to automatic from it's default setting, manual?

What do you think?

Note of warning ......... When tinkering with any Windows OS services write down the default settings, also a restore point is a good idea. This applies to "features" as well.

Discussion is locked
You are posting a reply to: Need opions informatin on Superfetch service
The posting of advertisements, profanity, or personal attacks is prohibited. Please refer to our CNET Forums policies for details. All submitted content is subject to our Terms of Use.
Track this discussion and email me when there are updates

If you're asking for technical help, please be sure to include all your system info, including operating system, model number, and any other specifics related to the problem. Also please exercise your best judgment when posting in the forums--revealing personal information such as your e-mail address, telephone number, and address is not recommended.

You are reporting the following post: Need opions informatin on Superfetch service
This post has been flagged and will be reviewed by our staff. Thank you for helping us maintain CNET's great community.
Sorry, there was a problem flagging this post. Please try again now or at a later time.
If you believe this post is offensive or violates the CNET Forums' Usage policies, you can report it below (this will not automatically remove the post). Once reported, our moderators will be notified and the post will be reviewed.
Collapse -
Just one.
by R. Proffitt Forum moderator / November 25, 2008 2:02 AM PST

When I need to see what effects these have I go get a guinea pig machine and then run a benchmark with the stock settings. Then I make ONE change and get new benchmarks.

This is from a method called The Scientific Method which helps to remove opinion and inject facts and solid(?) reasoning to support your theory.

Why would you do anything else?
Bob

Collapse -
That makes no sense
by Jimmy Greystone / November 25, 2008 2:34 AM PST

That makes no sense what you said. This service spends let's say an hour maximizing the performance of your system via effective memory caching. Why then, would you think that shutting it down, and flushing that cache in the process, would then improve performance?

If it reaches some point of maximum efficiency, that's a Good Thing(tm). That's exactly what you want to happen. This is the desired outcome of the service, so why you would then want to intentionally throw that all away and start over is beyond me; especially when you do it in the name of improved performance.

I don't mean to be overly rude here, but did you even think about this for a second? Because it makes no sense. A car is running at it's maximum speed, so I'll hit the breaks, come to a dead stop, and then start again because maybe this time I can go faster. Does that make any sense? It makes about as much sense as what your proposing.

For the most part, Windows is already tuned about as much as is possible when you install it. All these supposed performance tweaks and what not are just myths, they have absolutely no basis in reality. Half of them come from people misinterpreting or not fully comprehending what they read. A few might have once had some small effect, but have long since been rendered pointless (like defragmenting). An even smaller number might actually have some positive effect, but it's so small that you would have to have a stopwatch to really tell that it's there. The rest are just pure urban legend with no basis in reality whatsoever.

If you're a tweaker, and enjoy spending your free time squeezing every last bit of efficiency out of a computer as possible, you have no business using Windows. You should be using something like Linux. Specifically, the Gentoo distribution where you can tinker with compiler optimizations, and even dig into the code for some program yourself, start improving the efficiency. There's very little you can do to improve Windows short of buying new hardware, but Linux is a land of nearly endless opportunity to tweak anything and everything about the operating system.

Collapse -
SuperFetch information from Microsoft
by Dango517 / November 25, 2008 1:31 PM PST
http://www.microsoft.com/windows/windows-vista/features/superfetch.aspx

1) You down load a program you'd like to try you use it for three days as you review it carefully, you change your mind and now you want this program off your PC. Superfetch has put it in "priority" memory in your RAM drives.

2) You down load a big 300 MB software program to do a task. You use this software for three days straight. Superfetch does it's thing. You remove the program to save drive space.

3) Your PCs screwed up your using all kinds of programs you normally would not use to fix it, Superfetch puts these in "priority" memory.

None of these a far fetched scenarios. All of these I've done at one time or another.

The logical question would be how often does Superfetch renew itself, if it does? If it doesn't then restarting it would be the logical thing to do. This would insure that it has a current set of programs your using.

Another question might be how long does it take to do this? If weeks then restarting it might be a bad idea, if days, hours, minutes then "go for it".

Is it logical to expect average home PC users to use benchmarking software?

Which one would you suggest?

http://www.netlib.org/cgi-bin/search.pl
Collapse -
Look up.
by R. Proffitt Forum moderator / November 25, 2008 9:59 PM PST

You have 2 replies that at first may seem harsh but lead you to the answer.

"Is it logical to expect average home PC users to use benchmarking software? "

You're not average.
Bob

Collapse -
You're currently fighting Superfetch anyway...
by John.Wilkinson / November 25, 2008 11:49 PM PST

If you read about Superfetch and RAM optimizers/cleaners in depth you'll find you are currently contradicting your own commands. Specifically, you use Superfetch, which loads data into memory when it's not being used at the exact moment, but you also use a RAM optimizer/cleaner, which removes from RAM what is not being used at the moment (what was just loaded). Thus, in your case it is meaningless to discuss as a potential performance boost as you're preventing it from working and, in many aspects, slowing down your computer, something I brought up before.

To answer your questions, though:
1.) Superfetch is dynamically adjusted in the background. (More or less real-time.)
2.) Restarting your computer does not affect Superfetch.
3.) Uninstalling a program/deletign its files eliminates all but a few bytes of the space Superfetch would have consumed regarding that application.
4.) The longer you run/do not run a program the further it moves up/down the listing until it reaches #1 or is dropped from the list entirely.
5.) If memory is needed, any Superfetch setting is overridden then and there.
6.) No, most home users have no need for benchmarking software. That's primarily a gamer's tool.

John

Collapse -
Good point
by Dango517 / November 26, 2008 2:02 AM PST

I do use a RAM optimizer, sorry, still disagree and have no intention of stopping it's use. SuperFetch has no dependencies upon it, maybe I'll just shut it down. Still have questions thou.

We're not discussing whether stopping the computer has an effect on Superfetch. I was wondering if it was a good idea to restart Superfetch in services from time too time to give it a fresh start, much like a RAM optimizer works.

Please explain five (5) more fully. Seems to be a contradiction to me.

A few last questions for John. One for Bob.

Is SuperFetch the reason why the manual "set priority" feature in Process Explorer has changed? It no longer works with Vista. This feature could be adjusted in XP. To get to it right click a process then choose "set priority.

BTW, John, just yesterday I loaded several "real time" programs into a fast USB. I'm running them out of the USB instead of the hard drive. Immediately noticed an up tick in PCs performance and an increase in fan usage by the system. Do you think I could move Vista services to work from this USB drive (configured as NTFS)? Can the Services Manager be moved? Why not? "Real time" programs, services etcetera are few .........Vista virtual memory, Vista services, Security software ( <<<< this one may not move; these are picky), some software optimizers and few elective programs like McAfee site advisor. All of these together can fit on a 4 GB USB drive, easily.

Bob you listening? Is it possible to configure a USB as a dynamic disk? 4-6 port USB card + USBs configured as dynamic disks(?) = software RAID = a cheap way to tinker with RAID drives. Not just RAID drives ...... SSD RAID drives. Follow the logic? Happy

Collapse -
Let's make this simple
by Jimmy Greystone / November 26, 2008 2:35 AM PST
In reply to: Good point

Let's make this simple, shall we?

RAM "optimizers" are not necessary, and in fact will tend to HINDER performance. Free RAM is wasted RAM, and pretty much all these programs do is flush the cache Windows builds up to improve performance. The memory manager for XP and Vista is actually pretty good. It's not great, and could be improved, but given that for RAM "optimizer" programs to really be effective they would need to completely replace the Windows memory manager, and that's not possible... Well, the quandary should be clear.

As was explained, the Superfetch system is dynamic. Every time something on your system changes, this service evaluates that change and adjusts things as necessary. What you're proposing, using an "optimizer" along with restarting the service, is like putting a speed governor on a car that limits it to 60mph, then hoping that speeding up to 60mph, with the occasional bout of slamming on the brakes bringing the car to a complete stop, then starting again... Hoping that somehow this will allow you to break free of the speed governor that YOU YOURSELF put on the car.

Or if you want a non-car analogy... It's like putting a humidifier and dehumidifier in the same room, where the condensed water from the dehumidifier is put into the humidifier's tank. It's completely pointless, and any intelligent person would immediately ask you the very simple question of, "Why?" Probably with a bit of an incredulous inflection that hints at them questioning your intelligence and/or sanity.

Collapse -
One more time
by Dango517 / November 26, 2008 4:31 AM PST
In reply to: Let's make this simple

Processors leak data. What this means is that the people who manufacture CPU chips have done such an incredible job at shrinking the die in your CPU that the distance between the dies in your Processor and the external world are only 5 Atoms deep because of this wholes develope (more then likely due to impurities in manufacture.) and electrons literally fall out of your CPU. Those electrons are your data, your programs, your OS. (In fact Intel processors beginning with the phenom processor have an extra layer of copper in them to capture these wondering electrons.) Over time this leads to two problem, serious problems both due to data corruption. First your OS become corrupt and over time fails. The data degrades and degrades till it reaches a point when a critical component fails within the OS and the OS must be reinstalled. Second the RAM memory is effected, it stores this corrupt data from the CPU and needs to be refreashed. (reinstalled with good data from the hard drive.) If it isn't the corrupt data will slow down the PCs operation and accelerate OS failure. I would suggest you refreash "your" memory with this information and install a RAM optimizer. Processors leak data. What this means is that the people who manufacture CPU chips have done such an incredible job at shrinking the die in your CPU that the distance between the dies in your Processor and the external world are only 5 Atoms deep because of this wholes develope (more then likely due to impurities in manufacture.) and electrons literally fall out of your CPU. Those electrons are your data, your programs, your OS. (In fact Intel processors beginning with the phenom processor have an extra layer of copper in them to capture these wondering electrons.) Over time this leads to two problem, serious problems both due to data corruption. First your OS become corrupt and over time fails. The data degrades and degrades till it reaches a point when a critical component fails within the OS and the OS must be reinstalled. Second the RAM memory is effected, it stores this corrupt data from the CPU and needs to be refreashed. (reinstalled with good data from the hard drive.) If it isn't the corrupt data will slow down the PCs operation and accelerate OS failure. I would suggest you refreash "your" memory with this information and install a RAM optimizer.

Collapse -
First, how did that happen?
by Dango517 / November 26, 2008 4:52 AM PST
In reply to: One more time

Second, think of a RAM optimizer as the oil filter in your car. Does yours have a filter?

Collapse -
I've found a lug wrench more useful than
by R. Proffitt Forum moderator / November 26, 2008 4:59 AM PST

Any ram optimizer.

-> Why not do your benchmarks and put up web sites proving your points?

Sadly all I've read from you in this discussion is the stuff I overhear from the usual sales droids all over. It's painful to read.

At least with the lug wrench you can achieve something useful in no time at all.
Bob

Collapse -
Sorry, no
by Jimmy Greystone / November 26, 2008 5:17 AM PST
In reply to: One more time

Sorry, no, you're confusing a whole bunch of things which I don't have the time or inclination to get into. Let's just say that if Intel and AMD (who has the Phenom CPU not Intel) released a CPU where data could become corrupted due to electrons going off every which way, they wouldn't be able to sell it. If they did, they'd be facing lawsuits like you wouldn't believe, shareholders would likely revolt, and the entire top brass of the company would be looking for a new job. Look back to the release of the original Pentium and how much flak Intel took for a fairly minor flaw in the math coprocessor. This was back when computers were still something of an expensive toy that was more of a status symbol than anything else.

And let's just assume for a moment that your hypothesis is correct, and we take your analogy of RAM optimizers as an oil filter. Well, are these RAM optimizers not programs themselves? So shouldn't they be subject to the same gradual corruption effect as everything else? If these programs are made somehow immune to the process, why couldn't that then be applied to every other program? Your analogy falls apart, even if we look at it from the flawed perspective of your hypothesis.

What you are positing is simply not true. There is no way AMD or Intel would sell a CPU that had this kind of a problem. Same with RAM manufacturers. A few bad chips due to manufacturing issues is one thing, but you're talking a systemic failure in the product's very design. There's just no way a company would try to sell something like that, knowing of the problem beforehand, because they would be sued into oblivion.

Finally, it's either memory OR RAM, not "RAM memory". RAM memory is redundant, since they both refer to the same thing.

In any case, if anyone needs a refresher it would appear to be you. I don't have the time or inclination to get into the finer details of electrical engineering theory with you to explain exactly how you are wrong, but if you happen to be majoring in Electrical Engineering or Computer Engineering at a university... Might be time to consider a different major. In any case, feel free to pick up a few texts on Electrical Engineering and it shouldn't take too long to figure out that whatever your source of information for this, you didn't understand it as well as you thought you did.

You are of course free to do what you want with your own system. However, when you start posting this kind of erroneous information as some kind of divine revelation, people are going to call you on it. So maybe when several knowledgeable people tell you it's a load of crap, you should listen to them.

Collapse -
The bell has already rung...
by John.Wilkinson / November 26, 2008 6:34 AM PST
In reply to: One more time

The Phenom is an AMD processor, NOT Intel. You have (previously) claimed to have received much of your information from Intel and attempt to speak about their processors on a technical level, but do not even know what processors they make? You lose ALL credibility with that statement.

That aside, we've already had this conversation, and you stopped responding after I:
1.) Contacted the companies that develop the three RAM optimizers you recommended and relayed the message from two of them that your claims were factually inaccurate. In fact, the president of IOBit wanted to know where the misinformation was coming from.
2.) Cited sources which discredited your statements regarding a performance enhancement on Vista.
3.) Cited the very Intel documentation you were using, pointing out that your claim of a need to manually refresh RAM to prevent electron loss was factually inaccurate.

That said, the information you are spreading is factually inaccurate, as your own sources attest to. There is no reason to perpetuate the dissemination of these claims, nor rehash the argument of fact vs fiction which has already taken place. Therefore, I am locking this thread before this debate goes any further, getting more heated and attracting more attention from those who may purchase RAM optimizers based on a feeling of need generated from your false claims and general lack of understanding of the subject matter.

John

Collapse -
Several things...
by John.Wilkinson / November 26, 2008 3:20 AM PST
In reply to: Good point

1.) Bob, Jimmy, and myself, along with people who designed two of the RAM optimizers you have recommended, have explicitly disagreed and provided factual foundations for why your previous claims are invalid. I thought you might believe the developers of the software you use, but evidently you will not. Doesn't matter to them, really, since it means more money in their pockets, but your computer would rather you listen to them. After all, they might have some minuscule amount of knowledge about how the software works.

2.) Think about it: Shutting down terminates all processes and services, including Superfetch. Terminating Superfetch terminates Superfetch. Therefore, if terminating Superfetch by shutting down does not have an effect, neither does terminating the service alone. The settings are persistent.

3.) Elaborating on #5, if you need RAM for an active task it will make room by removing data for inactive tasks, including that loaded into RAM by Superfetch.

4.) The "set priority" feature is not broken. Some process priorities are locked, but otherwise you can change it provided you have not corrupted the feature through system modifications.

5.) No, you cannot 'move' services.

6.) USB is a standard, not a device.

John

Popular Forums
icon
Computer Newbies 10,686 discussions
icon
Computer Help 54,365 discussions
icon
Laptops 21,181 discussions
icon
Networking & Wireless 16,313 discussions
icon
Phones 17,137 discussions
icon
Security 31,287 discussions
icon
TVs & Home Theaters 22,101 discussions
icon
Windows 7 8,164 discussions
icon
Windows 10 2,657 discussions

CNET FORUMS TOP DISCUSSION

Help, my PC with Windows 10 won't shut down properly

Since upgrading to Windows 10 my computer won't shut down properly. I use the menu button shutdown and the screen goes blank, but the system does not fully shut down. The only way to get it to shut down is to hold the physical power button down till it shuts down. Any suggestions?