Tips For Speeding Up Windows XP, Without Utilizing 'Defrag'
If you're still relying on 'Defrag' to improve system performance, you are behind the times. Defragmenting is the process of reorganizing all files on a hard drive so that each file is arranged into...
So pointing out you do not understand how prefetching works is derogatory?
no.
But continuing to question my intelligence and knowledge is, especially when you do not know me, or how my system is running/behaving.
Quote:
Whoever wrote this simply copied and pasted these from other misinformed people.
quite possible, but i cannot prove anything.
Your sources are much more creditable than any source which will encourage the above guide, but i cannot agree with this:
Quote:
The facts and reality of how things work with software is not a debatable issue, there is only one way it works.
There is only one way it is supposed to work, yes i will give you that, but to the resourceful many changes are easily at hand.
I acknowledged many of your points as being credable, and most of the rest not to be tried by the common user, but you are going to have to trust me that there is more than 1 type of user, and there is such thing as a system which functions better without prefetch.
There is only one way it is supposed to work, yes i will give you that, but to the resourceful many changes are easily at hand.
Are you a programmer? Unless the software has a bug (there are none for XP Prefetching) then it only works one way. There is only one way Windows XP Prefetching works. I have personally spoken to engineers on the Windows Client Performance Team and have exhaustively tested it on a myriad of systems. I have provided you with exhgaustive documentation on it. Mark Russinovich for instance was the one who discovered the Sony Rootkit, do you think he knows what he is talking about, let alone the whole Windows Client Peformance Team?
Quote:
I acknowledged many of your points as being credable, and most of the rest not to be tried by the common user, but you are going to have to trust me that there is more than 1 type of user, and there is such thing as a system which functions better without prefetch.
No there is not a system that behaves "better without prefetching". What does that even mean? If you have prefetching disabled or broken in some way Windows XP and all your applications take longer to load period. This is not debateable. There is a reason there are no documented reproduceable tests on a clean, non overclocked, non nlite install of XP proving that any of these prefetching "tweaks" work. Because as soon as someone understands it and properly tests it they realize they had no idea what they were talking about. I can prove it irrefutably. The fact that you think otherwise simply means you still do not understand how it works nor even how to properly test it. You have stated multiple comments about how it works that are blatantly wrong, which only proves you do not understand how it works. The last person I would trust is someone who does not even understand how something works.
un les configured otherwise u wil have thes 3 colections of files
the first runing the basics of the computer (os drivers ect)
the second holding your aplications (oface browser ect)
the third holding your documents setings desktop favrates
(Pictures Music)
all stored on one partition ! 1 file alication table ,mtf bmp ect
on 1 phiscal (rotating) hdd
#phisical limitations practical implications
2hdds can read and write more data in les time than 1hdd
eg hdd1 = c:\ containg %SYSTEMROOT% = C:\Windows
hdd2 = ?:\ containg %programfiles% & %HOMEPATH%
\Program Files \Documents and Settings\
the setup above will give faster file aces time
than the all in 1 default windows layout
this is clearly the best option performance wize
offering the largest maximum potental
further stability and security
could be gained by separating the types of files most comanly acessed
eg pictures or music and storing them on seprate dedicated partitions created on one hdd .
*this method could be used with seprate partitions for sysytem, program files, and documents all tho helping provent defragmentation between these groupes the performance gaines of phisical isolation between disks are grater ! than
hopfully iwill have outlined the performance disadvantages of m$ os deployment
and given posibilties of improvement =>
bonnng
havent a clue about Prefetching
desabling ntfs time staping will speed up system
eg [write les often to disk (time stap) ]
page file 1/2 size of phisical ram (seprat partition) wil also help
for performance imprivment try
ramdisk!!
=hyperdrive
-iram
page file 1/2 size of phisical ram (seprat partition) wil also help
for performance imprivment try
ramdisk!!
Both of these are Myths. Setting the Paging File smaller does nothing but conserve disk space. By default the paging file is set to 1.5x RAM this is the system managed setting and it is not recommended to set it smaller as you may REDUCE performance by not having enough paging space. Moving the paging file to a separate partition on the same HD will further reduce performance since the HD head now has to travel farther to access it. RAMDisks are just a complete waste of time.
hmm depends on application for photo editing(ps5) audio(working temp_dir) and also archiving. or other
processes generating large temp file content
a 200+ mb ramdisk can boost performance considerably [however stability is a DEFiNATE factor]
when dealing with large amounts of data 2>gb
{ddr pc3200 or better ,hyper threading and dual core wil also help}
not to mention large hdd cache ,dedicated controller ect.. Gigabyte's i-RAM! Replace Existing Hard Drives
TheHyperDrive4 has an access time of 1100 nanoseconds read and 250 nanoseconds write (as measured by Data Transit's Bus Doctor), as opposed to 8 milliseconds access time (seek time + spindle latency) for the latest 10,000 rpm SATA Hard Disks and 5.5ms for the latest 15,000 rpm SAS Hard Disks.
Average spindle latency is 3 ms for a 10,000 rpm drive and 2ms for a 15,000 rpm drive.
This means that DDR drives, which we can ship you today for £599/£699, are faster than the next generation optical storage drives which are costing hundreds of millions of dollars in development and may become commercially available in 2011.
Setting the Paging File smaller does nothing but conserve disk space. By default the paging file is set to 1.5x RAM this is the system managed setting and it is not recommended to set it smaller as you may REDUCE performance by not having enough paging space
newer systems running xp with 512+ memory and dedicated gfxram
have no real need for page file:
setting the page to "0" is the best option
how ever systems can vary dramatically
in hw as well as software configuration , use ,ect
how your system will be affected
depends on multitudes of factors
them selves with multitude upon multitude of variables
the best way is to test ! use full considerations:
total ram in system :
/
memory size used by: file\application
One HyperDrive4 is 500% faster than one of the very latest Western Digital Raptors at running a basket of popular applications (H2BenchW). And it is at least 200% faster than any number of the world's fastest Hard Disk Drives in any configuration you like. It is the revolution that the IT industry overlooked. If you are 'waiting for your screen to come up' you are using gramophone technology. You need to try Silicon.
You have absolutely no idea what you are talking about:
Quote:
Putting a Paging File in a RAM drive is a ridiculous idea in theory, and almost always a performance hit when tested under real-world workloads. You can't do this unless you have plenty of RAM and if you have plenty of RAM, you aren't hitting your paging file very often in the first place! Conversely, if you don't have plenty of RAM, dedicating some of it to a RAM drive will only increase your page fault rate. Now you might say "yeah, but those additional page faults will go faster than they otherwise would because they're satisfied in RAM." True, but it is still better to not incur them in the first place. And, you will also be increasing the page faults that have to be resolved to exe's and dll's, and the paging file in RAM won't do diddly to speed those up. But thanks to the paging file in RAM, you'll have more of them. Also: the system is ALREADY caching pages in memory. Pages lost from working sets are not written out to disk immediately (or at all if they weren't modified), and even after being written out to disk, are not assigned to another process immediately. They're kept on the modified and standby page lists, respectively. The memory access behavior of most apps being what it is, you tend to access the same sets of pages over time... so if you access a page you lost from your working set recently, odds are its contents are still in memory, on one of those lists. So you don't have to go to disk for it. Committing RAM to a RAMdisk and putting a paging file on it makes fewer pages available for those lists, making that mechanism much less effective. And even for those page faults resolved to the RAMdisk paging file, you are still having to go through the disk drivers. You don't have to for page faults resolved on the standby or modified lists. Putting a paging file on a RAMdisk is a self-evidently absurd idea in theory, and actual measurement proves it to be a terrible idea in practice. Forget about it.
Flash memory on the HD ("i-RAM) or Flash memory drives have nothing to do with a RAMdisk. Taking away system RAM for a RAMdisk is couter productive and will reduce performance. Windows Vista takes advantage of Flash memory devices to further improve performance via ReadyBoost but there is no RAMdisk in system RAM and the paging file is NOT disabled.
Telling someone to disable the paging file is even more absurd. I suggest you read up on how the paging file works. Please STOP spreading completely inaccurate misinformation.
i love this debate about pagefile.
and really when it comes to audio apps.
i run a system with audio apps, and for your average user who does some dabling in making mp3`s all the arguments are legit.
but lets face it, when you get out to 56 tracks at 24bit 48khz what you are arguing becomes a joke at best. to take the load off of the track drive you need a large page file. loading 16 gigs into a 2 gig ram drive? all righty then. not going to work, first off the files have to be streamed, weather it be off the hdd or out of a second item to be loaded by the track hdd, you still run into bottlenecks.
best way i have found to acomplish this mass migration of info is thus.
C:\ apps and os - pagefile 1.5 x ram. 10k rpm
D:\track drive - no pagefile. 10k rpm
E:\pagefile 1.5 x ram 7200 rpm
letting windows manage my pagefile = D:\ running at 60 - 70 percent capacity. not good when running serious audio apps, playback stops.
page file layed out as mentioned drops disk usage to under 20 percent, tracks are now spread across 3 drives which results in faster access to streaming data as no one drive is trying to carry the full load. if the project fits in alloted pagefile, usage drops further, thus enabling one to get more work done.
getting rid of all the useless crap running in the background is of equal importance as well. running real time fx is the holy grail, getting sounds is not a majical process, different sounds need to be tried to avoid frequency stacks ( mud in the mix ) this cannot be accomplished with all the useless crap microsnot "feels" is important on all computers. granted not even 10 percent of the population will ever push their system as hard as i do, not even gamers. fx have to be run real time before you commit to sounds. once rendered ( destructive editing - yes as bad as it sounds, once done and out of the application, its done and cant be undone ) rendering takes place when i am happy with the mix. so i may end up with 60+ tracks with at least 20 to 30 running real time fx. my computer is not a piece of crap, and i found a few of the hints in the article at the top of this thread usefull, so to the author i say, thankyou, you helped me squeaz a little more headroom out of my rig.
just so you know.
p4 duo @ 2.4 ghz, 1066 fsb
4 gig ram, 800 mghz
C:\ 10000 rpm 40 gig sata raptor
D:\ 10000 rpm 150 gig sata raptor
E:\ 2x 160 gig 7200 rpm sata barracuda on raid 1
F:\ 2x 250 gig 7200 rpm sata barracuda on raid 1
@ Mastertech, yes, there are many misconceptions about the way pagefiles are implemented in Windows. You are completely correct these are myths, good catch
You seem to have a great knowledge in this area, but i do ask that you try to be more "approachable" on the subject, rather than athoritative. There is always room for debate, especially around the windows OS, there is much information that has not been released, and likely never will be from MS.
@above unregistered:
I too love these debates, because everyone thinks they are right, when although some users are more right, it is not always a clear cut right and wrong.
For the most part if you are modifying your pagefile and it happens to increase your speed, etc... there are many other things you should be looking at first. If you are uncomfortable stripping your OS of windows services, you likely should not be playing w/ the pagefile either.
~my 2 cents.
best way i have found to acomplish this mass migration of info is thus.
C:\ apps and os - pagefile 1.5 x ram. 10k rpm
D:\track drive - no pagefile. 10k rpm
E:\pagefile 1.5 x ram 7200 rpm
letting windows manage my pagefile = D:\ running at 60 - 70 percent capacity. not good when running serious audio apps, playback stops.
page file layed out as mentioned drops disk usage to under 20 percent, tracks are now spread across 3 drives which results in faster access to streaming data as no one drive is trying to carry the full load. if the project fits in alloted pagefile, usage drops further, thus enabling one to get more work done.
A Windows Managed paging file on any drive is by default 1.5 x system RAM. You can enable the paging file per drive thus "Windows managed" really only means that once you set a paging file to enabled on a drive Windows is able to resize it as needed, which is recommended. Your concern for disk usage if a paging file was enabled on your D drive is not warranted since Windows will use the paging file on the less frequently used partition over the paging file on the heavily used partition and uses an internal algorithm to determine which paging file to use for virtual memory management. You cannot disable this feature if more than one paging file is present. Thus if you are streaming off D: and you have a paging file on another drive Windows will use that paging file for paging. There is absolutely no need to set the paging file manually in Windows XP and no reason to disable it on any drive (outside of disk space concerns).
Quote:
Originally Posted by Unregistered
getting rid of all the useless crap running in the background is of equal importance as well. running real time fx is the holy grail, getting sounds is not a majical process, different sounds need to be tried to avoid frequency stacks ( mud in the mix ) this cannot be accomplished with all the useless crap microsnot "feels" is important on all computers. granted not even 10 percent of the population will ever push their system as hard as i do, not even gamers. fx have to be run real time before you commit to sounds. once rendered ( destructive editing - yes as bad as it sounds, once done and out of the application, its done and cant be undone ) rendering takes place when i am happy with the mix. so i may end up with 60+ tracks with at least 20 to 30 running real time fx. my computer is not a piece of crap, and i found a few of the hints in the article at the top of this thread usefull, so to the author i say, thankyou, you helped me squeaz a little more headroom out of my rig.
Then you are running the wrong OS because Windows XP is not a Real-time OS, Windows XP Embedded can be run in real-time with third party utilities. But I suspect you are misusing the term and don't understand it. Many people do not understand how Windows works and make assumptions off these misunderstandings. What I have stated is backed up with sources and facts from Microsoft. How these features work is not up to interpretation, this is software and it only works one way. If you want to stick your fingers in your ears and not take the time to read what I have linked then by all means it is your computer and your choice.
The reason I sound authoritative is because I have already researched this and all the sources are provided which mainly go to Microsoft. I am simply trying to prevent people from wasting their time and even worse slowing down their system.
Last edited by Mastertech; 09-19-2007 at 04:04 PM..
Quite comical. Mastertech sounds like he/she's working for the marketing department at Microsoft. Being a programmer, I can't believe that someone states that an application (or an OS for that matter) can only go one way! No matter how hard we try someone will ALWAYS find a path that we didn't realize was possible to take through our code. So does that mean that the originally envisioned path is the best path? NO.
As far as someone being concerned that others will slow/break their machines, I find myself chuckling. I say let them do it. The reasoning is that they will be inclined to learn more about their machine, OS, and any other software installed on it. Will some of them come to the same conclusions? Yes. Others may walk away and find that they weren't cut out for it. Either way it encourages thought.
Tips For Speeding Up Windows XP, Without Utilizing 'Defrag'
No that is not how it works. 
hmm depends on application for photo editing(ps5) audio(working temp_dir) and also archiving. or other 
[however stability is a DEFiNATE factor]
