From: Dustin Cook on 2 Jan 2010 21:39
Cronos <cronos(a)sphere.invalid> wrote in news:hhoual$sab$4(a)news.eternal-
> PajaP wrote:
>> I agree with you.
>> See I just agreed with another troll. Though this time he is also a
>> I am putting you in the killfile too as you are the one who started all
>> the hostilities in the thread. Proving you are a fuckwit troll!!
> Not a "fuckwit" troll, just a very good troll. I consider trolling an
> art form and I am one of the best at it.
Imho, I consider it a waste of time and have a picture of a person who
enjoys doing it in my head of a fat slob who never leaves the room.
.... Those are my thoughts anyways...
From: Cronos on 3 Jan 2010 11:53
Dustin Cook wrote:
> That's just not true, yes; windows makes so many temp files and turns the
> place into a mess; but the idea that you shouldn't bother defragging is
> just.. bad. NTFS is much better at preventing it from occuring, but it
> still does happen.
> I'm not interested in a flame war; I'm only stating a fact that's been
> known to most (I thought) computer persons since the days of fat... And
> likely anyone who's been in the ripping scene. Contigous space is your
> friend, especially when doing audio/video things.
Well, they are storage experts and you are security expert so maybe they
know more about this topic than you. You don't need to go start a
flame war with them because I trolled them on this subject just recently
and below are some of the responses from various people. I already knew
what their response would be but Roddy is fun to troll now and again.
And for most people's workloads would there be enough of a difference
to actually make a difference? If I can't tell there's a difference it
doesn't matter for any practical purposes.
I can imagine nasty combinations of workloads and platforms that would
make defragging helpful, but they must be really rare now.
In a former life (~15 years ago), doing backup from HD=>tape, it was
obvious that defragging before starting a backup kept the tape mostly
streaming, while skipping the defrag step led to a lot of shoe-shining.
That system was a 486/33 with 4MB of RAM, running Win 3.1, with a
3600 RPM non-DMA FAT16 HD, and a QIC (definitely not quick) tape
connected over a shared parallel port; and, the (Colorado) backup
software was very primitive.
In that era, I used to say that any mag.tape had only two speeds:
"It streams or it sucks".
Over the past dozen or so years, I've never been able to notice any
performance gain due to defragging, which is why I always recommend
using a defragger which is free: either none, or whatever is bundled
with the OS.
Oil changes /do/ make user-detectable changes, albeit over a long time.
Defragging doesn't, no matter how long you leave it - except perhaps
in that the increased wear and tear on the disk due to unnecessary
defragging may lower its lifetime.
Judging by this post and your signature, you would be one of these
"professional recovery services" I was talking about. So yes, I can see
how your job might be a easier with defragmented disks in cases when
there is a trashed FAT/MFT/etc. and files that can be recognized and put
However, making your job slightly easier is not a good reason to defrag.
Disaster recovery would also be easier if people never compressed or
encoded files, or used binary formats for word processor files, or used
jpeg instead of bmp for images, or any one of a thousand other ways to
make hard-to-identify files and file fragments.
The huge majority of people will never need or use your services -
either they have good backups, or they never have a hard disk disaster,
or they don't know about such services, or can't (or won't) afford them.
Therefore, it is not a reason for people to defrag their disks.
Nope, the fools that decide the defaults cant grasp the basics, that
unless the user
can detect the difference that defragging makes, there isnt any point in
Respect and credibility is something a person or company must work hard
to earn, and can quickly loose. MS has worked long and hard to ensure
they have as little credibility with technically knowledgeable people as
they possibly can.
I am /not/ saying that they are always wrong. But you must be very
na�ve to assume that what they say is right, without looking for
independent confirmation or proof.
It is generally true that defragging will be faster if the last time you
ran it was a week ago rather than two months ago. But the total time
wasted on weekly defrags over those two months is much more than the
time wasted for a single defrag once every two months. But whether you
do it once a week or every second month, it is still wasted time.
You wanted to know the reason newer Windows defaults to auto defragging
once a week? It's because lots of people, such as yourself, assume that
this is a "new feature" - another "reason" for "upgrading" to Vista /
Windows 7. Companies like DiskKeeper have done a great false
advertising job persuading people that they need scheduled defragmenters
- MS is simply cashing in on their marketing.
Yes, but then they have done that in the past and ended up doing
it my way too, most obviously with sleeping hard drives by default.
What matters is that there are very few situations where extra seeks
due to fragmentation are even detectable by the user in a proper
double blind trial. The vast bulk of linear access to large files is with
media files where the speed of access to the file is entirely dependant
on the media play speed and so extra seeks arent even visible at all.
The bulk of other access to large files is not linear, most obviously
with databases, so there arent even any extra seeks at all with those.
One situation where you do see extra seeks matter is copying very
large files, most obviously video files, but it makes a lot more sense
to avoid copying those than to furiously defrag to improve the speed
of copying those files.
Similarly with backups, it makes a lot more sense to do the
backups in the background or when the system is not being
used than to furiously defrag or even defrag monthly.
> I ask because I have seen people claim their load times for games has
decreased after a defrag. What say you to that?
Bet they wouldnt be able to pick it in a double blind trial.
>> Nope, not with most modern systems anymore.
> > I've found one scenario where defragging has proven critical:
No you havent.
> > I use a Windows Explorer 'replacement' called Directory Opus.
> > One of the features I like about it is its ability to show, on the fly,
> > the total space consumed by a directory and all its subdirectories.
> > If there are enough subdirectories involved, that will end up with
> > my machine grinding for a *long* time.
Just because that app is completely fucked, proves
nothing except that that app is completely fucked.
> > The fix for this, assuming I want this functionality, is
> > to 'defragment' the directories, as well as the files -
> > rewriting them to the innermost tracks on the disk.
Directorys dont get fragged and fragmented files have no effect
whatever on repeated mindless rereading of directorys anyway.
> > This is something that most defraggers just won't do. I have only
> > found *one* product that will relocate directories - Ultimate Defrag.
> > Once I allow that tool to reorg my volume in this fashion,
> > the heads don't need to move around the whole bloody
> > disk surface to read the directories, and as a result, this
> > little bit of functionality in Directory Opus is quick, and quiet.
Obvious lie on the speed.
And caching the directorys would have a much bigger effect anyway.
> > So I would agree, that under most circumstances
> > it's doubtful whether regular defragging is essential.
Its not doubtful, its absolutely certain that it is not essential.
> > But it absolutely has its place, depending on your usage of the
Like hell it does.
> > and shouldn't be discounted altogether.
Wrong, as always.
>> Run Linux, an OS that does not cause disks to become fragmented in
the first place.
> > All file systems have some fragmentation,
Nope, there are a few that never fragment at all. They just refuse to
write a file when there is no block big enough to write the file to and
only allow a file to be extended into the free space that follows it.
RT-11 is an example of that.
> > but FAT is probably the worst as it is often implemented by low
That has nothing to do with the file system, everything to do with how
> > exFat is a better file system as it maintains a bitmap of clusters
> > This bit map makes it easier to allocate a contingous range of
> Rod Speed <rod.speed....(a)gmail.com> wrote
>> >> Meat Eaters are Sinners - Repent! wrote
>> >> No more than those who rip the sexual organs off plants and munch
>>> >>> Rod Speed wrote
>>>> >>>> There are plenty who cant grasp that defragging most modern
systems is pointless.
>>> >>> Do you play computer games at all?
>> >> Yes.
>>> >>> I ask because I have seen people claim their load times for
>>> >>> games has decreased after a defrag. What say you to that?
>> >> Bet they wouldnt be able to pick it in a double blind trial.
> > Defragging has it's place.
Nope, not with most modern systems anymore.
> > I tend to do it when ever I tidy my office, ie every few months
I'm not stupid enough to tidy the office either.
> > and not every week.
I'm not stupid enough to do it every few months either.
From: Leythos on 3 Jan 2010 16:42
In article <hhqi29$e95$4(a)news.eternal-september.org>,
> Well, they are storage experts and you are security expert so maybe they
> know more about this topic than you.
And then there are people who do more than write for magazines, have
decades of real-world experience, who work with small systems to systems
with hundreds of terra-bytes....
My experience with Windows NTFS and file fragmentation gives me 100%
confidence that performance is increased when the file system is not
fragmented. This holds true across single and multi-drive systems.
The benefit is seen at different levels and is based on the amount of
fragments, their location on the physical disk(s) and how much an app
uses those files.
I've seen SQL queries fail because of fragmented SQL Database files, not
just fragmented indexes, but a database file with 30,000+ fragments.
Same for web site performance, where an app serves PDF files and allows
changes to those PDF files - 8 million PDF files, hundreds of millions
of fragments in less than a years time, and random app failures as more
and more users connect - a defrag completely cured the problem.
On servers, depending on how many ADD/DELETES you do, they also benefit
from unfragmented files, same with roaming profiles, same with a single
computer reading a large (4GB file), etc....
Most people could easily get by with a yearly full defrag, with
compacting space, and they would notice the difference, others may do it
monthly or every 6 months.....
If you've used your computer, heavily, for a year and don't notice any
benefit after a full defrag and compacting then you really won't notice
much else about your computer.
You can't trust your best friends, your five senses, only the little
voice inside you that most civilians don't even hear -- Listen to that.
spam999free(a)rrohio.com (remove 999 for proper email address)
From: Dustin Cook on 3 Jan 2010 21:21
ASCII <me2(a)privacy.net> wrote in news:4b4069e7.1080140(a)EDCBIC:
> Dustin Cook wrote:
>>Contigous space is your
>>friend, especially when doing audio/video things.
> ...or trying to recover (hopefully intact) files after some failure
Without a doubt.
.... Those are my thoughts anyways...
From: Dustin Cook on 3 Jan 2010 21:24
Cronos <cronos(a)sphere.invalid> wrote in
> Dustin Cook wrote:
>> That's just not true, yes; windows makes so many temp files and turns
>> the place into a mess; but the idea that you shouldn't bother
>> defragging is just.. bad. NTFS is much better at preventing it from
>> occuring, but it still does happen.
>> I'm not interested in a flame war; I'm only stating a fact that's
>> been known to most (I thought) computer persons since the days of
>> fat... And likely anyone who's been in the ripping scene. Contigous
>> space is your friend, especially when doing audio/video things.
> Well, they are storage experts and you are security expert so maybe
> know more about this topic than you. You don't need to go start a
> flame war with them because I trolled them on this subject just
> recently and below are some of the responses from various people. I
> already knew what their response would be but Roddy is fun to troll
> now and again.
hehe, I'm a technician first and foremost; having been doing this a very
very long time. :) I read the snippits, and based on personal experience
with servers and systems which create/empty alot of files on a routine
basis, I would respectfully have to disagree with many of them. You will
notice a difference, and if your into recovering data; a fragged to hell
drive isn't your friend in this case.
Zvi netiv would be another individual who I consider highly knowledgable
in the storage field; and even he would agree with me. :)
> I'm not stupid enough to do it every few months either.
.... Those are my thoughts anyways...