Poll! Built in defragger vs 3-rd party. Your opinions of using it

I have noticed this too. Don't understand how that is possible.
This is the reason why I switched to O&O.

 

this si because these are active used files in the OS no matter what you use it'll happen. Also a few thousand isn't an issue unless it's all in a single file. The problem is when you get tens-hundreds of thousands of fragments....like my drive lol

 

Page file. Current running programs. System files being used by the OS.

 

just don't bother. not all forms of fragmented files are performance problems. the in-use-system files are designed to be accessed that way, they are designed to be fragmented and are NOT slower because of it. it all depends on the individual fragment sizes, and the individual access sizes.

that's why microsofts defragmentation tool doesn't show the individual fragments anymore: because it's missleading. there are tons of files on your system that can be fragmented and you don't feel ANY difference, there are tons of WAYS a file can be fragmented and you don't feel a difference.

and that's why one shouldn't use any other defragmentation tool on the SYSTEM drive than the one from microsoft. they know best what kind of performance-issues they have to fight. and they only fight those.

defragmenting too much leads to problems, too. very often, the files are then not in an optimal order to allow for editing them without creating much more fragments.

so, everyone, just get over it. it's just a bad habit from old tech days. let the os do it for you and enjoy life instead. we're in 2011, it's time to not have to care about such stuff anymore.

 

I disagree. I don't see how they would already be in an "optimal" state.

The OS definately doesn't know whats best always. Especially when it comes to on OS stuff. The OS defrager is garbage and a solid 3rd party with optimization tools is best...if you know what ur doing. My charts shows the difference in latency is exactly why optimization is so important. Also why this coming 72/96 holiday i am reinstalling windows to a 40gig partition on the outer platter. That will make a big difference for me. My OS is installed retardedly Also some reason i got 10 gig on my outer platter blocked. It's in a unformatiable removeable section. No clue how ASUS did that but I am installing to a new drive and completely formatting that drive to fix it.

 

I use SSD. - But back in days Auslogics defrag was my choise

 

well, you think that some random small developer knows better than microsoft itself, who have written ntfs and know exactly where what takes how long to do?

well, dream on.

a simple example: if your mom cleans your room, and puts everything in order, you will have a much harder time to find everything again. the result: much higher seek times, much slower speed when "quickly having to find something". the end result: much quicker refragmentation, and most likely, much worse refragmentation than before.

a defragmentation program that cleans up everything perfectly has to take into account that those files will change again, will have to have space to move in, grow, shrink, get rewritten. and there's only one who knows exactly, what the purpose of all those tiny files is: the ones that wrote the os. this is why microsofts defragmentation software does BY DESIGN not defragment everything. only there, where there is actual gain. everything else would be a) nonsense or b) actual damage.

but you don't have to believe me. believe the marketing of the defragmentation tools on how they boost your performance magically.

 

you make me smile

EDIT: I'll run a few tests before i optimize my drive and we shall see the difference. I'll have this done by the end of the month. Need to wait for a 72 or 96

 

i noticed back in it's days that using perfectdisk enhanced my disks performance. but after a week, performance was LOWER than it was before defragmenting. the result: having to run perfect disk at least once a week to stay at high performance.

after a reinstallation, i just have not used anything except windows defragmentation, automatically running. performance stayed the same.

defragmentation tools are like coffee (or like cocaine): they push your system to higher performance, but afterwards it'll get more quickly more slow again.

reason, as stated, is that the defragmentation does not focus on long-term effects on the way it arranges it's files. only the quick boost is what matters.


other tools resulted in similar results btw.

 

Would those charts be the proof you keep claiming to have and yet never share? And would the same charts show some artificial testing difference or a demonstrable difference in real world use? Sure you might be able to reduce latency from 8.67 ms to 8.65 ms but does it MATTER? Would anyone really notice a difference? No one has ever said that defraging might not reduce latency, what has been said is "so what". If the difference can't be measured in HUMAN terms (i.e. with a stopwatch) then it is really not too important for a desktop or a laptop. Now if we are talking about a server farm with a bazillion transactions being thrown at it, we have a whole different set of requirements. But this is NBR and that first letter stands for Notebook.

Really? You can't figure out why or how they did that, and yet you expect us to believe your expertise on defragmentation and that you know what "ur doing"? Yikes.

We anxiously await your charts and don't bother to post some low level hardware stats that have no meaning in real world use. Post some tests done with real applications, measured in a way perceptible to a real human on the other side of the keyboard.

Gary

 

Looks like you like to argue just for arguing and I here write why.

You are saying that in real world you will not feel the difference between latency 5 ms and 25 ms or in transfer speeds from 139.75MBps to 67MBps?

Well then you can see here what difference is in changing Access Time from 0.06 ms to 0.1 ms or changing speed on 10 MB per sec AND WHAT IT GIVES IN REAL WORLD How To: Improve Low SSD Performance in Intel Series 5 Chipset Environments | StorageReview.com

Now compare that tiny changes on SSD to this big one on HDD

 

Those are still mostly 'benchmarks', the real world difference(to me anyway) is 'noticeable changes in day to day usage'.

I think no one is saying there is no difference, it is whether the gain warrant to effort which as always depends on individuals.

 

In the middle of that you can find information about real using. Perhaps you missed it.
There are described changes in installing time of MS Office 2010 and other soft, also copying files and scanning files

 

realworld difference is the moment where people will say "damn, this pc is slow!". sometimes, defragmentation can help there (really old xp installations that never got it, and are overloaded with data).

on win7, there's no such moment where after defragmentation, stuff feels faster (if you do the defrag without notifying the user). on the other hand, if you INFORM the user, but don't do anything, then it'll feel faster.

perception is a massive liar.

just get over it, james. you fixate on something that actually will gain you less time than you wasted writing all your stuff.

 

How many times have you installed MS Office 2010 or Visual Studio 2010 ?

Edit:
I disabled the builtin scheduled defrag and did use perfect disk 11 when I do the HDD->SSD upgrade, mainly because I need to shrink the system drive which needs to pack things to the front of the partition.

 

I never said any such thing, so don't try to put words in my mouth.

But you have not shown any correlation between the numbers you show above and fragmentation or the lack of it.

Gary

 

Well how many different software installation you would prefer in 1 article? Or perhaps you want a full report on 168 hours a week using a computer before and after speeding up and quicker responsivity with described time intervals on everything including opening a music or movie folder or even calculating this with different item viewing (large icons, small icons, list, table etc.)

Come on.

Besides O&O Defrag can put files in such order as how often they are using by a system and put them in the beginning of HDD making a free area for them and making areas for and between 3 types of zones. That was about how 3-rd party programs can know something about usage systems files.

I never put that words in your mouth, I asked it ("?"). So don't try to put words in my mouth

IN CONCLUSION.

The idea is that 3-rd party software really may enhance productivity of computer. We may not feel this or see this. We may remember our bad experience with some exact program and believe or not believe in this. Our optimization level can be different thanks to different expectations, hardware, usage and even our mood!

At the end I would like to tell you something from my history:
I had HP nc6000 laptop with 32 MB Video Memory, 1.6 Ghz Penrium M CPU speed and 1 GB RAM. I installed Halo 2 on XP (YES i did really did this ON XP not on Win 7 but however) and when I was playing it the game took 3 minutes AT LEAST to load each map. I could drink a coffee meanwhile. I thought it was because of my slow machine. But in a week or more I figured out to make defragmentation and after that It took 10-15 SECONDS to load a map. After that I understood that how defrag program can speed the system in real Life no matter what make your system slow (or not the whole system but just a player or winRAR or else): bad game installator, download programs or usual usage of a computer.

Another history which still continuous:
When I am watching HD movies some of them may freeze on half a second every 5-10 minutes or more often and I better should not use another program at the same time. It happens because of huge fragmentation of ISO images or Rips which were made from my legally bought DVD or BD disks of course.

 

As i said i will use boot timer and test my drive boot times before and after optimization. I would test the different AAM settings but i can't make those permanet. Crystal mark info has to be open to work so i can't test boot time deferency from that but i can test the boot time difference when my drive is installed like it is and after i optimize it. Also after work i'll post screenshots of my drive in hardrive management or whatever it is called in windows administative tools. I'll show you how they got 10gig of space just sitting on the outer platter and that i can't make the 10 gig merge into the C partiton. For me to merge those two spaces I think i need to not have the OS on and reformat it than. AKA put the drive in another pc and format it from there.


EDIT: i have clearly shown the difference in AAM and moving the whole OS to the outer rim will make a big difference. I'll also make sure i don't have windows readyboost on since that shortens the boot time alot. Ready boost is amazing if you got a slow 5400 drive(i got a 7200rpm thouhh). I use my 16 gig SD card it has 15-25MBps transfers and 15MBps read and a .3ms ^^ its better than my class 10 SD card lol. The class 10 has a slow .7ms and only slightly faster rights....but we all know with this its all about latency so the slightly faster sequencial transfer is useless.

 

Disktrix UltimateDefrag 3.0
It has the most customizability, you can choose specific files and folders that you want placed after the rest of the files in the partition. You have a high performance section where seelcted files and folders are placed.
It takes advantage of the layout.ini files that Windows Prefetch produces.
Has a defrag simulator.
Can sort all files and folders by name, recency, volatility.

Has a visual of the drive as a drive platter (circular) instead of a rectangular box like all other defraggers show you.

This program will give you the best performance your hard drive can provide, although ultimately a hard drive is still a hard drive. SLOW

Get an SSD and forget about it. Problem solved.

 

a good hardrive is plunty fast. Yes spending 6 times more on an SSD is faster but only good for an OS and a single program......where do i put my 705gigs of steam games?

 

...Tape Storage?

Who needs 705GB of games anyway.

-> if you need to store data get a nice storage setup, Home Server v1, a NAS etc.

 

GREAT ADVICE!!!!........thats even slower......

 

NAS over GBit Lan tops out at about 120MB/s, nothing slow there.
Home Server should be able to do similar speeds on good hardware.

 

how likely that an icon image is fragmented ? Don't just throw out all sorts of problem which are not related to file fragmentation.

 

My drive is 138MBps=caped....then add on top of that high latency.....an extra 100feet of copper and a switch or two and 2 different network cards....that makes it a ton worse.

eSATA or internal=best idea.
Network drive is only good for data storage like movies.

 

-> One thing to remember via the network, or a multiple drive eSATA enclosure you can run several drives and hence get better speed

Oh, and on that note, I wrote MB/s not Mbit/s

 

yes i know and my drive is 138MBps=capped....my god read

EDIT:Again you still add alot of latency using network drives...eSATA no.....network yes

eSATA 1-3foot/feet cable

network drive
Computer insides->network card->wires->router->wires->netowork card->computer. That will add alot of ms to those times making that network drive automaticly slower no matter if its even an SSD.


EDIT: ping your computer to your network computer. That is how much more latency to add to hardrive latency.....not cool

 

actually as a hdd has a latency from 8 to 15ms, and the network latency is <1ms, network doesn't matter.

btw, 3.5" hdds have lower latency than 2.5"s mostly => having them in a server and accessed over network can be faster than using the internal disk

 

Sorry i was referring to a 3.5 internal my bad....was not thinking of laptop. Also eSATA external is better. second the latency is >1ms. If you ping your computer in the network your not going to get a 1ms ping. it'll bee like 5ms or so. Atleast that is what i had in my home network back at home. I don't currently.........scratch that i got a friend connect to the network give me a min and i'll find their ip and test it....maybe i am wrong...lets hope so.


crap...i don't know my login and password anymore...crap!

 

That's g wifi at my home... not a very good router either...


Uploaded with ImageShack.us

 

well i tried typing 192.168.1.1-15 and i got from 0-15ms I even did 192.168.1.1(my router) and i got 0ms 2ms 2ms and 8ms. The other tests were 0-15ms and those are through wireless and network cables. One of those not sure which was my sling box(wired) and the others were my roommates 2 laptops on my wireless.

I don't know how you get 0ms on wireless...i think thats garbage but ok. Lucky you.

Also i got Cat6 cables to my laptop and slingbox also i got a netgear 3700.

 

Rubbish router or rubbish hardware

I just redid the pings to other other devices - the first is the same you see above, then another laptop in the network, a print server, and again twice the home server.



Uploaded with ImageShack.us

 

Gentlemen, Start Your Buying: NETGEAR WNDR3700 RangeMax Dual Band Wireless-N Gigabit Router Reviewed - SmallNetBuilder

not a rubbish router. Definitely not bad cat6 cables.

internal network card....no clue

http://img833.imageshack.us/img833/9237/pingtestj.png

those are my tests. Again wired i could see you getting those pings.....wireless no way


EDIT: You can see something is messing ip. I get 1ms across thee board except for one out of 4 tests. so something and i am assume laptop side is causing it to choke. But still with my high end router and cat6 cables i get 1ms....than from there it has to go to another computer and back...thats an easy 1-3ms i would say. <1ms no way. I think your tests are bs. You can't get a 0 ping on wifi just not posssible

 

Well, can't be a good one

And yes, that's g wireless Belkin too... but it works...

 

read my edit. your tests are bull. no way

 

I'm just typing ping and the IP address, nothing hidden there.


Uploaded with ImageShack.us

My laptop via Wi-FI to the router, to the Home Server (via the cheapo supplied cable).

In my previous one i pinged my mother's laptop (also on Wi.Fi) and the Wi-Fi print server.

Did I mention my network is WPA2

On that note, is is rather ludicrous that a at least "Not good" router is getting better results than yours

Edit:
Because you enjoy them so much, here is another one

Uploaded with ImageShack.us

 

http://forum.notebookreview.com/networking-wireless/546174-0ms-g-wifi.html#post7034649

made a new thread. I want download to read this. I am going to pm him.






EDIT: Back to other part of topic


Uploaded with ImageShack.us

you can see here that there is 14.65gig of free space on the outer part of the platter i can't use. No extend option. I have tryed shrinking, formatting, and expanding and nothing will get that to be part of the C drive. It can even be made in its own little drive...can't get it undone. Asus some how messed it up. So i will be taking the 2 320gig drives out putting my D in C slots and making my D drive as the C drive with a 40 gig and a 270ish gig size drive. This will force all data to the outer platter for the OS and primary programs. Rest garbage will go on the second partition. Then i will format the original C drive so that i get that 14.65gig back.

 

i let them be on steampowered.com and only install those games i actually play..

you're completely wrong. networking is <1ms even in bigger company networks, and definitely in home land. and you're on a notebook forum here => by default, we always compare to notebook setups.

 

yea yea i admit i was wrong...gladly...i thought netowrks has a couple ms latency....didn't know it was that fast lol. But still i use an external for all my games and it will be capped on a gigabit network.


EDIT: also with a NAS i will have to spend a crap ton of extra money to build it....not cool. Though down the road when i get the money i will make a 4 2TB NAS

 

Considering that your max. speeds are only for sequential reads, I wouldn't be so sure that it matters.

Try a SSD array in a NAS it'll beat your HDD not for several GB files, but for small file transfers


On that note... do you ever reach that theoretical read speed outside of testing a freshly formatted drive?

 

no crap ton of extra money per se. and the advantage to have all your stuff in one place (so all ordered, and no duplications). you have a pc and a laptop, not?

and well, you lose maybe 10% speed over the MAX a hdd can deliver (but doesn't over most of it's range).

so yeah, as detlev said. ssds in all your systems, and big cheap storage for the big things == best solution right now.

 

Just to add on the cost side

A Home Server will cost you about the same, if not slightly less than a medium/upper medium specced laptop.

My HP came to about 750€ (50€ cashback from HP) - 450€ server, drives - that's one 2,5GHz Pentium Processor in the Home Server, 1 1TB System drive and 3 2TB drives of my choice, expandable via eSata or USB

-> I guess you could have bought that for the money you spent on computer games... -> and simple low end NAS systems with say 2 2TB drives cost less obviously

 

yes.....benchmarks aren't theoretical and show that my drive can easily real 130-140MBps on the outer 25% of the drive. If i had a drive that was as fast as this i would show you. Can a ramdisk do this? If so i'll run a ramdisk and transfer 2gigs from my external to it.

Also the point is that a optimized hardrive is plenty fast enough. A much better option than 3000 dollars worth of SSDs

At that point i rather buy a bunch of ram and put it on a battery back up and make that my hardrive for games. Be better to just dump a 2-30gig game in ram at that point and run it from there.....which isn't a bad idea. Can you set up a ramdisk through a eSATA port? Would be an interesting project

 

Benchmarks are theoretical

They can tell you what sort of rough category your system falls into, or if you have a serious problem, but else they do nothing for you.

So your drive can reach 130-140MB/s on the outside of the platter - so? The only ever case where this speed can be reached is in sequential read/write operations, something that is only needed for storage.
Computers for the most part do small random reads and writes.

My SSD can reach 250MB/s read speed - does it matter - no. What matters are the random reads and writes.

I suppose if you store some ISO files at several GB a piece on the outer platter you could benefit from such speeds, but else, no.

 

most my reads are fairly sequential...movies music games. My drive is nearly always clean and without fragments. Also once i get a program that optimizes i will place all my most used games on the outer edge....so yea those 130+MBps is very practical. Again my point is a good hardrive is more than good enough....much better than spending 3k in SSDs.....i love how you just ignore my point

EDIT:i only spent a little over 500 for all my games over my whole life. That is about 200 games total.


http://www.steamcalculator.com/id/DCMAKER
Than i got a bunch of games on cds and DVDs Those are mostly all iso files now on my external.

 

Music and videos will not need 130MB/s
-> 320kBit/s MP3s play fine at 45KB/s (I know that, I streamed them from home over the web, although my upload has more than doubled a little while ago)
Let's assume it's wave - 1174kBit/s -that's about 4 times as much data (a bit less), so it'll work fine at 200KB/s

-> on that note, divide by 8 and you get your Bytes more accurately

 

I think the only activity that needs 130MB/s(may not be enough) are real time video capture or that sort of thing, 1920x1080x30x3

 

my god i have my games on there too...their loads take alot of speed. My games load so much faster on my external than on my laptop hardrives....god you are ignorant

 

Without knowing what you have for your internal and external, nothing we can gauge here.

Calling people ignorant is a bit rude.

 

i already stated many times that i have games and the reason for my external is for my games. a 50-100MBps hardrive is 17ms is far slower than my 70-138MBps 14.3ms hardrive.