From: Ben Myers on
On 2/15/2010 6:00 PM, David Harper wrote:
>
> "Ben Myers" <ben_myers(a)charter.net> wrote in message
> news:hlcil1$1de$1(a)news.eternal-september.org...
>
>> The performance gain from using a defragmenter varies widely with each
>> system and each user. I have to disagree that a defragger does not
>> improve performance.
>
> I think the important question is: Does a defragger improve performance
> over the one that comes with Windows? I would guess not. Anyone have any
> evidence that proves otherwise?
>
> - David Harper

Does Defraggler or any defragmenter improve performance over Windows'
built-in defrag? Probably not, and if it does, measuring the difference
is counting angels dancing on a pin. And it does not matter, because
after-defrag system performance is only one metric of the value of a
product. Other important metrics: speed of defrag process and useful
info provided.

Here is where Defraggler shines. If a hard disk does not have much free
space, 15% or less, Windows' defrag either does not run at all, or it
runs in a very much degraded mode. Defraggler manages to run and get a
lot of a drive defragged when the drive is crowded with data.

Here is where else Defraggler shines. It does its analysis very
quickly. It tells you exactly which files are fragmented, and
fragmented they are. This is useful info. For example, if I defragment
my second hard drive, the one with all kinds of neat stuff on it, I tell
it NOT to defrag the 4GB Linux ISO files. Why bother? And Defraggler
defragments very quickly. It also runs just fine on 64-bit Windows 7,
and I'll assume that it also runs OK on the 64-bit versions of Vista
(gag reflex suppressed) and XP Pro.

I installed and ran UltraDefrag 4.0. It appears to run OK, but I did
not want wait for it to finish the job, because it runs slower than
Defraggler. And it is far less polished and professional looking than
Defraggler. People need to be careful which version they choose to
download and install.

Defragmenting software is not the answer to all the world's problems and
it has not yet cured cancer, but it can prove very useful... Ben Myers
From: RnR on
On Mon, 15 Feb 2010 18:24:30 -0500, Daddy <daddy(a)invalid.invalid>
wrote:

>Monica wrote:
>> I've always used Window's defrag program. This is adequate or can I/Should
>> I be using something better?
>> A newsletter I got today mentioned a program called UltraDefrag. It's free
>> so maybe it's worth what it costs <g>
>> What would you guys suggest?
>> Thanks,
>> Monica
>>
>>
>Unless you're running a server, Windows built-in defragmenter is all you
>need.
>
>"Don�t become a defrag junkie"
>http://winhlp.com/node/82
>
>Daddy


That sorta backs up what I was telling Ben (at least to a degree). I'm
running a defragger now and with 46% fragged, I'm anxious to see what
I will find hopefully by morning. In the past when I ran one (perhaps
10 to 20% fragmentation) I didn't notice any difference to my eye.
That said, I did not run any benchmarks which probably would say it
was faster of course. I read a lot and it just seems now that a lot
seem to say it's not necessary. Of course there will be exceptions to
anything. And I can't say that Ben is wrong just because I read
otherwise so I guess ultimately it's up to the owner to decide. If
you think it runs faster, that's all that counts <g>. I can say that
normally now I don't run it but years ago, I ran it on an automatic
basis (diskeeper I believe). I guess I agree with the "don't be a
defrag junkie". Of course it can't hurt to run it tho. Obviously if
I ran it more often now, I wouldn't have to wait over nite to get the
C drive done :(
From: Ben Myers on
On 2/15/2010 5:48 PM, RnR wrote:
> On Mon, 15 Feb 2010 13:17:01 -0800 (PST), "William R. Walsh"
> <wm_walsh(a)hotmail.com> wrote:
>
>> Hi!
>>
>>> I just tried MyDefrag and it not only wouldn't complete the process
>>> but messed up a program or two. I uninstalled it.
>>
>> You very probably have a deeper, more sinister problem lurking --
>> either you have high level filesystem corruption in the areas where
>> those programs are living, or you have a disk with bad spots that
>> haven't been mapped out/can't be mapped out because they can't be
>> fully read.
>>
>> Make a backup before you take corrective action, if that's what you
>> plan to do. Perhaps make a backup and test your hard drive anyway.
>>
>>> Actually many say now that Defrags aren't necessary due to
>>> the inherit speed of the drives today vs. years ago.
>>
>> Speed of the drives and data density have both improved, and to a
>> certain extent, that has improved the amount of time it takes for the
>> drive to find and start retrieving data. If you've got a file that's
>> broken up in many pieces, however, and the drive has to stop
>> transferring data to seek over to the next piece of the file, you lose
>> some time there. (Not much time--maybe nanoseconds at most--although
>> it can add up.)
>>
>> Buffering techniques have also improved to the point where if a drive
>> has to stop writing, something else can pick up the slack for a while.
>> (This is important if you're capturing data in real time and cannot
>> stop...as you might be if you were capturing video or multi-track
>> audio. People who do those things defragment their drives mainly to
>> have gobs of free space stacked up together so the drive won't have to
>> pause for seeking purposes once writing starts.)
>>
>> That used to be a big deal, it was possible to buy so-called "A/V"
>> drives that wouldn't do thermal recalibration in the middle of a long
>> running write operation, amongst other things.
>>
>> Microsoft said back in the days of NT4 that NTFS would not need
>> defragmentation. So far as I know, Executive Software provided some
>> interesting statistics that suggested NTFS actually fragmented quite
>> badly. How biased these were I don't know. Executive Software (today
>> known as Diskeeper Corporation) has always made a lot of noise that
>> suggests fragmentation will end up robbing your house while you sleep.
>>
>> The developer of JK/MyDefrag says that Microsoft provided a
>> defragmentation API (in other words, a set of functions and tools for
>> use by defragmentation software) in Windows NT, 2000 and XP. I'm not
>> sure the bit about NT is correct; Microsoft was very adamant at the
>> time that NTFS did not suffer problems from fragmentation.
>>
>> I don't know if it's actually a Microsoft designed API or not. When a
>> disk defragmenter first showed up in Windows NT family products, it
>> was a licensed, stripped down version of Executive's Diskeeper.
>>
>> It's always possible that the defragmentation APIs are just "move this
>> file here, move that file there" commands and the actual strategy is
>> left to whatever defragmentation utility you use. I haven't
>> investigated it that closely.
>>
>> For most people, defragmentation probably isn't a big deal. The world
>> won't end if you don't, but given the nature of personal computers to
>> do a wide variety of things, it wouldn't hurt to do it periodically so
>> your computer is performing at its best.
>>
>> Windows has allegedly shipped with some sort of optiimzation scheme
>> for program launch acceleration since the days of Windows 98 and its
>> "WinAlign" tool. Office 2000 even got in on the game, with a post-
>> installation "file optimizer" that claimed it would run any time
>> Office files needed to be optimized. I've never seen it happen more
>> than the one time after installation, and some of the O2K
>> installations I have are quite old.
>>
>> If you want to defragment your hard disk, set up a utility to do so on
>> a schedule, at a time when you won't be using your computer for
>> anything.
>>
>> William
>
>
> I admit hd could be going bad but my gut feeling is due to all the
> programs and redundancy I have for viri checking. Of course I can
> check the drive fast and determine the outcome. After that I'll give
> defraggler a try and see how that goes.
>
> Truthfully I doubt I'll notice much difference in defragging my drive
> but I'll report back what I find and how my eye observes the
> difference. Let me check the drive first.
>
> thanks.

The other important questions related to hard drive performance are:

1. Hard drive going south? You've got that covered with appropriate
diagnostics, like HDAT2 and the mfr's diagnostics.
2. Hard drive nearly full? This KILLS performance. A lady brought her
old Dell C610 over, complaining that it had poor performance. It had
only 700MB or so of free disk space out of a capacity of 30 or 40GB. I
looked at the contents, found the biggest offenders with WINDIRSTAT
(another fine and free analysis tool), deleted some of them, ran
CCLEANER (of course!), a defragged the drive. The C610 now performs
respectably.
3. Incorrect (too low) virtual memory setting? Usually after a memory
upgrade.

.... Ben Myers
From: RnR on
On Mon, 15 Feb 2010 23:35:43 -0500, Ben Myers <ben_myers(a)charter.net>
wrote:

>On 2/15/2010 5:48 PM, RnR wrote:
>> On Mon, 15 Feb 2010 13:17:01 -0800 (PST), "William R. Walsh"
>> <wm_walsh(a)hotmail.com> wrote:
>>
>>> Hi!
>>>
>>>> I just tried MyDefrag and it not only wouldn't complete the process
>>>> but messed up a program or two. I uninstalled it.
>>>
>>> You very probably have a deeper, more sinister problem lurking --
>>> either you have high level filesystem corruption in the areas where
>>> those programs are living, or you have a disk with bad spots that
>>> haven't been mapped out/can't be mapped out because they can't be
>>> fully read.
>>>
>>> Make a backup before you take corrective action, if that's what you
>>> plan to do. Perhaps make a backup and test your hard drive anyway.
>>>
>>>> Actually many say now that Defrags aren't necessary due to
>>>> the inherit speed of the drives today vs. years ago.
>>>
>>> Speed of the drives and data density have both improved, and to a
>>> certain extent, that has improved the amount of time it takes for the
>>> drive to find and start retrieving data. If you've got a file that's
>>> broken up in many pieces, however, and the drive has to stop
>>> transferring data to seek over to the next piece of the file, you lose
>>> some time there. (Not much time--maybe nanoseconds at most--although
>>> it can add up.)
>>>
>>> Buffering techniques have also improved to the point where if a drive
>>> has to stop writing, something else can pick up the slack for a while.
>>> (This is important if you're capturing data in real time and cannot
>>> stop...as you might be if you were capturing video or multi-track
>>> audio. People who do those things defragment their drives mainly to
>>> have gobs of free space stacked up together so the drive won't have to
>>> pause for seeking purposes once writing starts.)
>>>
>>> That used to be a big deal, it was possible to buy so-called "A/V"
>>> drives that wouldn't do thermal recalibration in the middle of a long
>>> running write operation, amongst other things.
>>>
>>> Microsoft said back in the days of NT4 that NTFS would not need
>>> defragmentation. So far as I know, Executive Software provided some
>>> interesting statistics that suggested NTFS actually fragmented quite
>>> badly. How biased these were I don't know. Executive Software (today
>>> known as Diskeeper Corporation) has always made a lot of noise that
>>> suggests fragmentation will end up robbing your house while you sleep.
>>>
>>> The developer of JK/MyDefrag says that Microsoft provided a
>>> defragmentation API (in other words, a set of functions and tools for
>>> use by defragmentation software) in Windows NT, 2000 and XP. I'm not
>>> sure the bit about NT is correct; Microsoft was very adamant at the
>>> time that NTFS did not suffer problems from fragmentation.
>>>
>>> I don't know if it's actually a Microsoft designed API or not. When a
>>> disk defragmenter first showed up in Windows NT family products, it
>>> was a licensed, stripped down version of Executive's Diskeeper.
>>>
>>> It's always possible that the defragmentation APIs are just "move this
>>> file here, move that file there" commands and the actual strategy is
>>> left to whatever defragmentation utility you use. I haven't
>>> investigated it that closely.
>>>
>>> For most people, defragmentation probably isn't a big deal. The world
>>> won't end if you don't, but given the nature of personal computers to
>>> do a wide variety of things, it wouldn't hurt to do it periodically so
>>> your computer is performing at its best.
>>>
>>> Windows has allegedly shipped with some sort of optiimzation scheme
>>> for program launch acceleration since the days of Windows 98 and its
>>> "WinAlign" tool. Office 2000 even got in on the game, with a post-
>>> installation "file optimizer" that claimed it would run any time
>>> Office files needed to be optimized. I've never seen it happen more
>>> than the one time after installation, and some of the O2K
>>> installations I have are quite old.
>>>
>>> If you want to defragment your hard disk, set up a utility to do so on
>>> a schedule, at a time when you won't be using your computer for
>>> anything.
>>>
>>> William
>>
>>
>> I admit hd could be going bad but my gut feeling is due to all the
>> programs and redundancy I have for viri checking. Of course I can
>> check the drive fast and determine the outcome. After that I'll give
>> defraggler a try and see how that goes.
>>
>> Truthfully I doubt I'll notice much difference in defragging my drive
>> but I'll report back what I find and how my eye observes the
>> difference. Let me check the drive first.
>>
>> thanks.
>
>The other important questions related to hard drive performance are:
>
>1. Hard drive going south? You've got that covered with appropriate
>diagnostics, like HDAT2 and the mfr's diagnostics.
>2. Hard drive nearly full? This KILLS performance. A lady brought her
>old Dell C610 over, complaining that it had poor performance. It had
>only 700MB or so of free disk space out of a capacity of 30 or 40GB. I
>looked at the contents, found the biggest offenders with WINDIRSTAT
>(another fine and free analysis tool), deleted some of them, ran
>CCLEANER (of course!), a defragged the drive. The C610 now performs
>respectably.
>3. Incorrect (too low) virtual memory setting? Usually after a memory
>upgrade.
>
>... Ben Myers


All excellent tips. I've got it all covered tho after years of
tweaking <grin> but appreciate the help regardless. Thank you.
From: Hank Arnold on
On 2/15/2010 1:21 PM, Monica wrote:
> I've always used Window's defrag program. This is adequate or can I/Should
> I be using something better?
> A newsletter I got today mentioned a program called UltraDefrag. It's free
> so maybe it's worth what it costs<g>
> What would you guys suggest?
> Thanks,
> Monica
>
>

I've been using UltraDefrag for a while. It's good and works fast. A lot
faster than the built in XP defrag....

--

Regards,
Hank Arnold
Microsoft MVP
Windows Server - Directory Services
http://it.toolbox.com/blogs/personal-pc-assistant/