Anyone hear about the exFAT file system, the successor to FAT32?
Yousuf Khan
Microsoft's latest FAT file system driver, meant mainly for use with the
newest generation of bigger thumb drives. It's called exFAT (extended),
and it's the successor to FAT32.
Description of the exFAT file system driver update package
http://support.microsoft.com/kb/955704
Looks like it's got support for Windows CE, XP, and Vista. And it'll
handle theoretical file systems upto 64 ZB, and files theoretically upto
64 ZB too. The recommended sizes for both file system and files is 512
TB, however.
I guess NTFS is too complex or too proprietary for thumb drives.
Yousuf Khan
Harry
> Came across this article the other day, it's a description of
> Microsoft's latest FAT file system driver, meant mainly for use with the
> newest generation of bigger thumb drives. It's called exFAT (extended),
> and it's the successor to FAT32.
IMHO, it's another crap from M$.
According to extFAT wiki, the disadvantages are:
1. devices formatted using exFAT cannot be read by Windows
Me, Windows 2000, and Windows Server 2003 or earlier,
and most other non-Microsoft operating systems
2. 32-bit Windows XP requires an update to be applied in
order to support exFAT
3. 64-bit Windows XP does not offer exFAT support
4. devices using exFAT are unable to use Windows Vista's
ReadyBoost capability (but Windows 7 does support the
new exFAT file system with ReadyBoost and also enables
a bigger ReadyBoost cache due to the bigger space limit
with exFAT)
5. Microsoft has not released the official exFAT file
specification, and a licence from Microsoft is required
in order to make and distribute exFAT implementations[8]
6. limited or no support outside PC environment at present —
most current consumer electronics, such as televisions
and A/V receivers, can only handle previous FAT versions
(this may change with the new SDXC cards requiring exFAT).
The only feasible advantage is that, if you have files bigger than
4GB to be carried .....
Arno
> On Aug 28, 11:21?am, Yousuf Khan <bbb...@yahoo.com> wrote:
>> Came across this article the other day, it's a description of
>> Microsoft's latest FAT file system driver, meant mainly for use with the
>> newest generation of bigger thumb drives. It's called exFAT (extended),
>> and it's the successor to FAT32.
> IMHO, it's another crap from M$.
> According to extFAT wiki, the disadvantages are:
> 1. devices formatted using exFAT cannot be read by Windows
> Me, Windows 2000, and Windows Server 2003 or earlier,
> and most other non-Microsoft operating systems
And there is the clue: They are (again) trying to be
as incompatible as possible. Just what you expect from
a low-quality vendor.
> 2. 32-bit Windows XP requires an update to be applied in
> order to support exFAT
> 3. 64-bit Windows XP does not offer exFAT support
> 4. devices using exFAT are unable to use Windows Vista's
> ReadyBoost capability (but Windows 7 does support the
> new exFAT file system with ReadyBoost and also enables
> a bigger ReadyBoost cache due to the bigger space limit
> with exFAT)
I thought ReadyBoost was a dud anyways?
> 5. Microsoft has not released the official exFAT file
> specification, and a licence from Microsoft is required
> in order to make and distribute exFAT implementations[8]
See my comment at 1.
> 6. limited or no support outside PC environment at present ?
> most current consumer electronics, such as televisions
> and A/V receivers, can only handle previous FAT versions
> (this may change with the new SDXC cards requiring exFAT).
> The only feasible advantage is that, if you have files bigger than
> 4GB to be carried .....
Just another effort to lock the competition out or to have
it pay money to be in the market. The anti-market tactics
of scum.
Arno
YKhan
> 5. Microsoft has not released the official exFAT file
> specification, and a licence from Microsoft is required
> in order to make and distribute exFAT implementations[8]
>
> 6. limited or no support outside PC environment at present —
> most current consumer electronics, such as televisions
> and A/V receivers, can only handle previous FAT versions
> (this may change with the new SDXC cards requiring exFAT).
>
> The only feasible advantage is that, if you have files bigger than
> 4GB to be carried .....
If Microsoft is licensing it, then I can't see why flash disk vendors
would even bother with it? Typical of Microsoft.
Yousuf Khan
Rod Speed
> Harry <harryooopot...@hotmail.com> wrote
Because it fixes the 4GB problem.
> Typical of Microsoft.
Havent seen anyone else fix the 4GB problem tho.
YKhan
> YKhan wrote
> > If Microsoft is licensing it, then I can't see why
> > flash disk vendors would even bother with it?
>
> Because it fixes the 4GB problem.
>
> > Typical of Microsoft.
>
> Havent seen anyone else fix the 4GB problem tho.
Isn't there any number of open-source sophisticated file systems that
fix the 4GB problem? With the additional features of exFAT such as
cluster bitmaps and stuff, the overhead of the metadata might not be
much larger between them. Most vendors adopted FAT on flash drives
because of the small overhead of the metadata, and the fact that most
implementations of FAT were already public domain.
I guess we'll have to wait to see if flash vendors adopt exFAT.
Yousuf Khan
Mike Tomlinson
writes
>Came across this article the other day, it's a description of
>Microsoft's latest FAT file system driver, meant mainly for use with the
>newest generation of bigger thumb drives. It's called exFAT (extended),
>and it's the successor to FAT32.
Would have thought it's more likely to be an underhand way of
reinforcing their patent claim on the FAT filing system, to be surer of
extracting licencing fees.
--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png
Arno
> On Aug 29, 2:18?pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
>> YKhan wrote
>> > If Microsoft is licensing it, then I can't see why
>> > flash disk vendors would even bother with it?
>>
>> Because it fixes the 4GB problem.
>>
>> > Typical of Microsoft.
>>
>> Havent seen anyone else fix the 4GB problem tho.
> Isn't there any number of open-source sophisticated file systems that
> fix the 4GB problem?
Basically all modern ones. I guss the comment is about FAT with
> 4GB support. However, given that almost nothing needs files > 4GB
on mobile storage devices, I guess calling it a "problem" is a
bit of an overstatement.
> With the additional features of exFAT such as
> cluster bitmaps and stuff, the overhead of the metadata might not be
> much larger between them. Most vendors adopted FAT on flash drives
> because of the small overhead of the metadata, and the fact that most
> implementations of FAT were already public domain.
Indeed, especially the later reason.
> I guess we'll have to wait to see if flash vendors adopt exFAT.
I will not use it. But I reformat my new media anyways.
Arno
Rod Speed
>>> If Microsoft is licensing it, then I can't see why
>>> flash disk vendors would even bother with it?
>> Because it fixes the 4GB problem.
>>> Typical of Microsoft.
>> Havent seen anyone else fix the 4GB problem tho.
> Isn't there any number of open-source sophisticated file systems that fix the 4GB problem?
Yes, but thats no use to someone who wants to put the flash
disk stick into a system that only supports FAT and NTFS.
> With the additional features of exFAT such as cluster bitmaps and stuff,
> the overhead of the metadata might not be much larger between them.
The problem aint the overhead.
> Most vendors adopted FAT on flash drives because
> of the small overhead of the metadata, and the fact that
> most implementations of FAT were already public domain.
The real reason is because their drives are mostly
used on systems that can only read FAT and NTFS.
> I guess we'll have to wait to see if flash vendors adopt exFAT.
Its more likely they'll keep using FAT and just break the files up into <4GB parts with PVRs etc.
jj
> YKhan <yjk...@gmail.com> wrote
>> On Aug 29, 2:18?pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
>>> YKhan wrote
>>>> If Microsoft is licensing it, then I can't see why
>>>> flash disk vendors would even bother with it?
>>>
>>> Because it fixes the 4GB problem.
>>>
>>>> Typical of Microsoft.
>>>
>>> Havent seen anyone else fix the 4GB problem tho.
>
>> Isn't there any number of open-source sophisticated file systems that
>> fix the 4GB problem?
>
> Basically all modern ones. I guss the comment is about
> FAT with > 4GB support. However, given that almost
> nothing needs files > 4GB on mobile storage devices,
The most obvious thing that does is video recording devices and there
are a hell of a lot of those out there now. The better set top boxes for
digital TV will record to any USB device now, flash drive or hard drive.
> I guess calling it a "problem" is a bit of an overstatement.
Only in the sense that its easy to chop the files up into <4GB segments on file creation.
Dan Lenski
> 5. Microsoft has not released the official exFAT file
> specification, and a licence from Microsoft is required in order to
> make and distribute exFAT implementations[8]
Any guesses on how long until it's reverse-engineered and there's a Linux/
xBSD driver? Software patents be damned...
Dan
Dan Lenski
Apparently I should have Googled that first. There's been working read-
only support for exFAT for the Linux kernel since January: http://
groups.google.com/group/fa.linux.kernel/browse_thread/
thread/903eb182127e8f21/e2e23eeb3a016268
It seems they haven't worked on it since because... no one uses or cares
about exFAT?
Dan
Arno
Quite possible.
Arno
mscot...@aol.com
>
> - Show quoted text -
I gather exFat is part of Vista SP1
I've never seen a disk using it though.
Michael
YKhan
> In article <4a98202...@news.bnb-lp.com>, Yousuf Khan <bbb...@yahoo.com>
> writes
>
> >Came across this article the other day, it's a description of
> >Microsoft's latest FAT file system driver, meant mainly for use with the
> >newest generation of bigger thumb drives. It's called exFAT (extended),
> >and it's the successor to FAT32.
>
> Would have thought it's more likely to be an underhand way of
> reinforcing their patent claim on the FAT filing system, to be surer of
> extracting licencing fees.
It's looking that way.
Yousuf Khan
Rod Speed
> Mike Tomlinson <m...@jasper.org.uk> wrote
>> Yousuf Khan <bbb...@yahoo.com> wrote
>>> Came across this article the other day, it's a description of
>>> Microsoft's latest FAT file system driver, meant mainly for use
>>> with the newest generation of bigger thumb drives. It's called
>>> exFAT (extended), and it's the successor to FAT32.
>> Would have thought it's more likely to be an underhand
>> way of reinforcing their patent claim on the FAT filing
>> system, to be surer of extracting licencing fees.
> It's looking that way.
Only for conspiracy theorists.
The industry will just ignore exFAT, you watch.
Mike Tomlinson
s.com>, YKhan <yjk...@gmail.com> writes
>It's looking that way.
I wonder if they'll try a retrospective land grab? Claim that exFAT is
based on prior art, so they can extract licence fees for FAT too?
Wouldn't put anything past M$.
YKhan
> YKhan <yjk...@gmail.com> wrote:
> > On Aug 29, 2:18?pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
> >> Havent seen anyone else fix the 4GB problem tho.
> > Isn't there any number of open-source sophisticated file systems that
> > fix the 4GB problem?
>
> Basically all modern ones. I guss the comment is about FAT with> 4GB support. However, given that almost nothing needs files > 4GB
>
> on mobile storage devices, I guess calling it a "problem" is a
> bit of an overstatement.
I can see DVD ISO's going over 4GB's easily. Even the internal VOB
files on DVD's would be that large.
If you need to maintain Windows-world compatibility, why not just
format to NTFS in this case? And there's decent reverse-engineered
NTFS fs read/write drivers for Linux these days.
> > With the additional features of exFAT such as
> > cluster bitmaps and stuff, the overhead of the metadata might not be
> > much larger between them. Most vendors adopted FAT on flash drives
> > because of the small overhead of the metadata, and the fact that most
> > implementations of FAT were already public domain.
>
> Indeed, especially the later reason.
Also pretty much all of file systems available for Linux, apart from
the Windows ones, are open source, like IBM JFS, SGI XFS, not to
mention the default EXTxFS fs's, and ReiserFS. If the set top
designers standardized on one of these fs's, then it's simply a matter
of packaging a user-space driver on Windows which will work with all
versions of Windows, and then eventually Microsoft would have to build
it in.
Yousuf Khan
Mike Ruskai
<mi...@jasper.org.uk> dribble thusly:
>In article <95bf7c70-e94a-430e...@d34g2000vbm.googlegroup
>s.com>, YKhan <yjk...@gmail.com> writes
>
>>It's looking that way.
>
>I wonder if they'll try a retrospective land grab? Claim that exFAT is
>based on prior art, so they can extract licence fees for FAT too?
>
>Wouldn't put anything past M$.
You can't patent something based on prior art.
mscot...@aol.com
wrote:
> On or about Wed, 2 Sep 2009 02:48:20 +0100 did Mike Tomlinson
>
> >In article <95bf7c70-e94a-430e-b744-68d9e1758...@d34g2000vbm.googlegroup
> >s.com>, YKhan <yjk...@gmail.com> writes
>
> >>It's looking that way.
>
> >I wonder if they'll try a retrospective land grab? Claim that exFAT is
> >based on prior art, so they can extract licence fees for FAT too?
>
> >Wouldn't put anything past M$.
>
> You can't patent something based on prior art.
I did hear of patent case where a company was trying to patent an
idea / method that I had been sold publically, world wide for 20
years. If felt like they were trying to patent the wheel.
Yousuf Khan
> In article <95bf7c70-e94a-430e...@d34g2000vbm.googlegroup
> s.com>, YKhan <yjk...@gmail.com> writes
>
>> It's looking that way.
>
> I wonder if they'll try a retrospective land grab? Claim that exFAT is
> based on prior art, so they can extract licence fees for FAT too?
>
> Wouldn't put anything past M$.
Prior art means something that existed before the patent application.
It's the exact opposite of applying for a patent, therefore prior art is
used to undo a patent, but never to reinforce a patent.
Yousuf Khan
Yousuf Khan
> I gather exFat is part of Vista SP1
>
> I've never seen a disk using it though.
I don't think it'll catch on. I have a 500GB external disk here that
came originally fully formatted in FAT32. FAT32 can span these large
volumes, so I'm not sure why Microsoft is saying FAT32's maximum volume
size limit is only 32GB! Sure it's got the 4GB file size limit, but
that's never a major problem.
Yousuf Khan
Mike Tomlinson
<BUTth...@DONTearthlinkLIKE.netSPAM> writes
>You can't patent something based on prior art.
Even if it's your prior art?
mscot...@aol.com
32GB is the largest size that XP will format from within the operating
system. It is a completely artificial maximum. There are many tools
that allow formatting a 500GB disk as FAT32.
I have tried formatting an old FAT32 disk using Vista SP2, and the
default option is exFat, so I am trying it. Its a 250GB drive.
Rod Speed
> mscot...@aol.com wrote
>> I gather exFat is part of Vista SP1
>> I've never seen a disk using it though.
> I don't think it'll catch on. I have a 500GB external disk here that
> came originally fully formatted in FAT32. FAT32 can span these large
> volumes, so I'm not sure why Microsoft is saying FAT32's maximum
> volume size limit is only 32GB!
There is a theoretical problem with volumes bigger than 32GB.
Thats why XP etc wont format them, tho it is happy to use them.
> Sure it's got the 4GB file size limit, but that's never a major problem.
That never is just plain wrong. The most obvious
problem is with video files produced by PVRs etc.
Hordes of those are bigger than 4GB.
Rod Speed
> Mike Ruskai <BUTth...@DONTearthlinkLIKE.netSPAM> wrote
>> You can't patent something based on prior art.
> Even if it's your prior art?
Yep, it you were stupid enough to not patent the prior art, you're fucked.
You cant do it retrospectively that way.
Mike Ruskai
<mi...@jasper.org.uk> dribble thusly:
>In article <mhss959tm4ck6oi9h...@4ax.com>, Mike Ruskai
><BUTth...@DONTearthlinkLIKE.netSPAM> writes
>
>>You can't patent something based on prior art.
>
>Even if it's your prior art?
Patents apply to novel and non-obvious advances. If something is prior art,
it is ineligble for a patent.
Moreover, patents are only valid for 17 years. You can't refresh that by
making a minor change. You have to invent something new, even if it's
something that modifies an existing entity. That patent applies to the update
only, though, not the original entity.
Microsoft has no patents for FAT. All they have (four patents) are for VFAT
and FAT32, and they're quite questionable at that (no more valid than the
patent for a crustless peanut butter and jelly sandwich). But they're enough
to extort money with, due to the wretched state of the US patent system these
days.
David Brown
How about this:
<http://www.newscientist.com/article/dn965-wheel-patented-in-australia.html>
(Yes, I know it was "teething problems" with a new patent system that
was supposed to make the system faster, easier and cheaper.)
David Brown
> On or about Wed, 2 Sep 2009 18:43:43 +0100 did Mike Tomlinson
> <mi...@jasper.org.uk> dribble thusly:
>
>> In article <mhss959tm4ck6oi9h...@4ax.com>, Mike Ruskai
>> <BUTth...@DONTearthlinkLIKE.netSPAM> writes
>>
>>> You can't patent something based on prior art.
>> Even if it's your prior art?
>
> Patents apply to novel and non-obvious advances. If something is prior art,
> it is ineligble for a patent.
>
Being ineligible for a patent has never stopped people applying for
patents, and has never stopped patent offices approving them (especially
in the US, but not limited to them). As long as companies are willing
to pay license fees for unreasonable patents, rather than challenge them
and get them overturned, then "innovators" are going to keep patenting
obvious ideas with plenty of prior art. MS excels at this (though they
are not alone) - I would not be surprised if they can figure out a way
to consider exFAT patents as being relevant to older FAT versions.
> Moreover, patents are only valid for 17 years. You can't refresh that by
> making a minor change. You have to invent something new, even if it's
> something that modifies an existing entity. That patent applies to the update
> only, though, not the original entity.
>
Again, that applies only in theory, not in practice. Some companies use
such modifications as a way to extend the existing patents, contrary to
the intentions of patents, and to patent laws. But as long as other
companies would rather pay small license fees than large legal fees in a
court room, such abuses happen.
> Microsoft has no patents for FAT. All they have (four patents) are for VFAT
> and FAT32, and they're quite questionable at that (no more valid than the
> patent for a crustless peanut butter and jelly sandwich). But they're enough
> to extort money with, due to the wretched state of the US patent system these
> days.
That's the key point. Theoretically, you can't patent a crustless
peanut butter and jelly sandwich - in practice you can, and MS is one of
the biggest abusers of the system (though they haven't often sued other
companies).
MS are also experts at FUD and sowing confusion. You are entirely
correct that MS has no patents for FAT - but "fat patents" gives about a
million hits in Google because people /think/ they have patents on FAT.
And as long as people are convinced of that, MS can continue to extort
license fees (or other control) over other companies.
Rod Speed
> Rod Speed wrote
>> Mike Tomlinson wrote
>>> Mike Ruskai <BUTth...@DONTearthlinkLIKE.netSPAM> wrote
>>>> You can't patent something based on prior art.
>>> Even if it's your prior art?
>> Yep, it you were stupid enough to not patent the prior art, you're fucked.
>> You cant do it retrospectively that way.
> How about this:
> <http://www.newscientist.com/article/dn965-wheel-patented-in-australia.html>
Thats not a real enforceable patent.
> (Yes, I know it was "teething problems" with a new patent system that was supposed to make the system faster, easier
> and cheaper.)
It isnt even teething problems, its not a real enforceable patent.
Rod Speed
> Mike Ruskai wrote
>> Mike Tomlinson <mi...@jasper.org.uk> wrote
>>> Mike Ruskai <BUTth...@DONTearthlinkLIKE.netSPAM> wrote
>>>> You can't patent something based on prior art.
>>> Even if it's your prior art?
>> Patents apply to novel and non-obvious advances. If something is prior art, it is ineligble for a patent.
> Being ineligible for a patent has never stopped people applying for patents,
Yes, but they dont end up with a real enforceable patent for prior art.
> and has never stopped patent offices approving them (especially in the US, but not limited to them).
Yes, but they dont end up with a real enforceable patent for prior art.
Any attempt to enforce that patent in a court will see it tossed out.
> As long as companies are willing to pay license fees for unreasonable patents, rather than challenge them and get them
> overturned, then "innovators"
> are going to keep patenting obvious ideas with plenty of prior art.
Doesnt mean that they end up with an enforceable patent for prior art.
> MS excels at this (though they are not alone) - I would not be surprised if they can figure out a way to consider
> exFAT patents as being relevant to older FAT versions.
They can consider whatever they like. What matters is whether they
can ever enforce license fees for FAT implementations. They cant.
>> Moreover, patents are only valid for 17 years. You can't refresh
>> that by making a minor change. You have to invent something new, even if it's something that modifies an existing
>> entity. That
>> patent applies to the update only, though, not the original entity.
> Again, that applies only in theory, not in practice.
Wrong. It applys in practice too with the technology being discussed.
> Some companies use such modifications as a way to extend the existing patents, contrary to the intentions of patents,
> and to patent laws.
Bet you cant list even a single example with the technology being discussed.
> But as long as other companies would rather pay small license fees than large legal fees in a court room, such abuses
> happen.
Bet you cant list even a single example with the technology being discussed.
>> Microsoft has no patents for FAT. All they have (four patents) are for VFAT and FAT32, and they're quite
>> questionable at that (no more valid than the patent for a crustless peanut butter and jelly sandwich). But they're
>> enough to extort money with, due to the wretched state of the US patent system these days.
> That's the key point. Theoretically, you can't patent a crustless
> peanut butter and jelly sandwich - in practice you can,
Like hell you can and end up with a patent thats enforceable.
> and MS is one of the biggest abusers of the system (though they haven't often sued other companies).
Bet you cant list even a single example with the technology being discussed.
> MS are also experts at FUD and sowing confusion. You are entirely correct that MS has no patents for FAT - but "fat
> patents" gives about a million hits in Google because people /think/ they have patents on FAT.
What matters is that they dont. And most of those patents are for obesity anyway.
> And as long as people are convinced of that, MS can continue to extort license fees (or other control) over other
> companies.
Bet you cant list even a single example with the technology being discussed.
dennis
> There is a theoretical problem with volumes bigger than 32GB.
Theoretical? FAT32? No, the specs supports much bigger drives than 32GB
just fine.
Rod Speed
> Rod Speed wrote
>> There is a theoretical problem with volumes bigger than 32GB.
> Theoretical? FAT32? No,
Yep.
> the specs supports much bigger drives than 32GB just fine.
There is a theoretical problem with volumes bigger than 32GB.
dennis
> Yep.
You can download a copy of the FAT32 specs from Microsoft's website. Try it.
Rod Speed
> Rod Speed wrote
>> Yep.
Irrelevant to what MS said about the problem with FAT32 partitions bigger than 32GB.
dennis
> Irrelevant to what MS said about the problem with FAT32 partitions bigger than 32GB.
Irrelevant? Right. And XP supports FAT32 partitions much larger than
32GB just fine. The builtin formatting tool just won't format larger
than 32.
Rod Speed
> Rod Speed wrote
>> Irrelevant to what MS said about the problem with FAT32 partitions bigger than 32GB.
> Irrelevant? Right.
Fraid so.
> And XP supports FAT32 partitions much larger than 32GB just fine.
What I said LONG before you ever showed up.
> The builtin formatting tool just won't format larger than 32.
And it refuses to do that for a reason.
And it aint just XP either.
Yousuf Khan
> There is a theoretical problem with volumes bigger than 32GB.
And what's that theoretical problem?
Yousuf Khan
Rod Speed
> Rod Speed wrote
>> There is a theoretical problem with volumes bigger than 32GB.
> And what's that theoretical problem?
MS claims its because many tasks on a very large FAT32 file system become slow and inefficient.
http://technet.microsoft.com/en-au/magazine/2006.07.windowsconfidential.aspx
mscot...@aol.com
That is not a theoretical problem, it is just that handling more than
32GB gets clumsy and NTFS is better. By putting the 32GB limit on
formatting just helps save some people from complaining about
performance.
Looking at exFAT very briefly, I'm not sure it actually much better.
The example (250GB disk) I am looking appears to have clusters of
128K, unless there is something I haven't seen yet, it will not be
very effecient at storing short files.
Michael
Rod Speed
>>>> There is a theoretical problem with volumes bigger than 32GB.
>>> And what's that theoretical problem?
>> MS claims its because many tasks on a very large FAT32 file system become slow and inefficient.
>> http://technet.microsoft.com/en-au/magazine/2006.07.windowsconfidenti...
> That is not a theoretical problem,
Corse its a theoretical problem when some configs have
plenty of cpu power and so wont see that particular problem.
> it is just that handling more than 32GB gets clumsy and NTFS is better.
Its more complicated than that when the amount of cpu power available varys considerably.
> By putting the 32GB limit on formatting just helps save
> some people from complaining about performance.
And is a theoretical problem when they have plenty of cpu power.
> Looking at exFAT very briefly, I'm not sure it actually much better.
No one ever said it was.
> The example (250GB disk) I am looking appears to have
> clusters of 128K, unless there is something I haven't seen
> yet, it will not be very effecient at storing short files.
Having fun thrashing that straw man ?
Arno
I doubt it exists. And even if it does, what do we care about
a _theoretical_ problem?
Arno
YKhan
I don't think we needed a theory to figure that one out. :-)
Yousuf Khan
YKhan
Regardless, flash disks aren't really all that performance critical as
components. Seeing as FAT32 is obviously capable of being extended out
to greater than 32GB as long as you use a non-Microsoft formatter
util, I'm sure most mfgs will avoid paying the Microsoft tax and just
continue to use FAT32. Those applications which require file sizes
greater than 4GB, such as PVR's will most likely be using more
advanced file systems anyways like NTFS, or one of the *Nix file
systems.
Yousuf Khan
Arno
> On Sep 4, 7:25?am, Arno <m...@privacy.net> wrote:
>> Yousuf Khan <bbb...@yahoo.com> wrote:
>> > Rod Speed wrote:
>> >> There is a theoretical problem with volumes bigger than 32GB.
>> > And what's that theoretical problem?
>>
>> I doubt it exists. And even if it does, what do we care about
>> a _theoretical_ problem?
> Regardless, flash disks aren't really all that performance critical as
> components. Seeing as FAT32 is obviously capable of being extended out
> to greater than 32GB as long as you use a non-Microsoft formatter
> util, I'm sure most mfgs will avoid paying the Microsoft tax and just
> continue to use FAT32. Those applications which require file sizes
> greater than 4GB, such as PVR's will most likely be using more
> advanced file systems anyways like NTFS, or one of the *Nix file
> systems.
Or alternatively us a tivial split scheme, for example as DVDs do.
Incidentially, quite a few NAS boxes and I suspect at least
part of the PVRs alreasy run Linux and use for example
ext2 or ext3. Windows is not only more expensive, it requires
more expensive hardware for the same performance and is
a lot less stable and more difficult to program, not to
mention less secure. And it does not really run well on
MIPS. (AFAIK you have to revert to the brain-damaged CE
version.)
Arno
Mike Ruskai
dribble thusly:
DVR's don't use normal file systems. All of them fragment way too much for
video recording purposes. TiVo units, for example, use a proprietary file
system called MFS (Media File System).
Rod Speed
Its a real problem with large FAT32 partitions and low
horsepower system with calculating the free space alone.
Obviously not bad enough to make it unusuable, but thats a different matter entirely.
All MS is saying is that NTFS works a lot better in those partitions bigger than 32GB and they are right.
Rod Speed
> On or about Fri, 4 Sep 2009 08:52:02 -0700 (PDT) did YKhan
> <yjk...@gmail.com> dribble thusly:
>
>> On Sep 4, 7:25 am, Arno <m...@privacy.net> wrote:
>>> Yousuf Khan <bbb...@yahoo.com> wrote:
>>>> Rod Speed wrote:
>>>>> There is a theoretical problem with volumes bigger than 32GB.
>>>> And what's that theoretical problem?
>>>> Yousuf Khan
>>>
>>> I doubt it exists. And even if it does, what do we care about
>>> a _theoretical_ problem?
>>
>> Regardless, flash disks aren't really all that performance critical
>> as components. Seeing as FAT32 is obviously capable of being
>> extended out to greater than 32GB as long as you use a non-Microsoft
>> formatter util, I'm sure most mfgs will avoid paying the Microsoft
>> tax and just continue to use FAT32. Those applications which require
>> file sizes greater than 4GB, such as PVR's will most likely be using
>> more advanced file systems anyways like NTFS, or one of the *Nix file
>> systems.
> DVR's don't use normal file systems.
Quite a few of them do.
> All of them fragment way too much for video recording purposes.
Fragmentation is irrelevant with DVRs, because the speed
of access to the file is entirely determined by the play
speed of the media, so extra seeks arent even noticeable.
> TiVo units, for example, use a proprietary file system called MFS (Media File System).
And hordes of other DVRs dont, particularly those that use USB connected media.
YKhan
wrote:
> DVR's don't use normal file systems. All of them fragment way too much for
> video recording purposes. TiVo units, for example, use a proprietary file
> system called MFS (Media File System).
That would make some sense, especially since you need to be able to
read from the file while it may still be being recorded.
Yousuf Khan
YKhan
> > TiVo units, for example, use a proprietary file system called MFS (Media File System).
>
> And hordes of other DVRs dont, particularly those that use USB connected media.
Which DVR's use flash disks to store video? They may use it to copy
stuff already recorded off of them, but do any of them really use the
flash disks for direct recording?
Yousuf Khan
Rod Speed
> Rod Speed <rod.speed....@gmail.com> wrote
>>> TiVo units, for example, use a proprietary file system called MFS (Media File System).
>> And hordes of other DVRs dont, particularly those that use USB connected media.
> Which DVR's use flash disks to store video?
All the ones that dont have an internal hard drive.
> They may use it to copy stuff already recorded off of them, but
> do any of them really use the flash disks for direct recording?
Yep, one of mine does. No internal hard drive, size of a cigarett packet, its
a set top box as well, and records to flash ram sticks and SD cards too.
Completely standard FAT format sticks, it just breaks the file into 4GB segments.
Its happy to write to any USB device and uses a standard format on that.
My cellphone uses flash disks to store video too.
Rod Speed
> Mike Ruskai <BUTthann...@DONTearthlinkLIKE.netSPAM> wrote
>> DVR's don't use normal file systems. All of them fragment way too
>> much for video recording purposes. TiVo units, for example, use a
>> proprietary file system called MFS (Media File System).
> That would make some sense,
Fraid not. Fragmentation doesnt matter a damn with media
files because the speed of access is entirely determined by
the media speed so a few extra seeks dont matter a damn.
> especially since you need to be able to read
> from the file while it may still be being recorded.
Perfectly possible to do that with completely standard file systems.
My PVR which has a number of capture cards does
that fine with either FAT32 or NTFS formatted drives.
And you're mangling the file format with the file system in that case anyway.
Capella
guys, stop criticizing exfat. It is not such a bad file system
especially if you have windows 7. The advantages of exfat is mainly to
serve as an alternative to ntfs. Exfat is also used to eliminate
fragmentation on drives and for use in drives with low space
(surprised?)
Microsoft claimed that exfat does not fragment a volume until that
volume only has 15 percent free space. I have tested that on my own
system. 0 percent framentation.
In addition, exfat requires little space for the actual formating. I
heard that ntfs takes like 40 mb just for the file system. However,
exfat takes much less than that.
Consider an external hard drive. Usually, if you buy say for example a
1tb external hdd, you would have it formated in fat 32 which most people
would change immediately to ntfs (why would you get so much space and
NOT have files greater than 4 gigs?) However, Ntfs becomes inefficient
and slow over time ESPECIALLY as the drive fills up. exFat is
Microsoft's solution to the problem.
Please respect the existence of exFAt, and don't randomly criticize it.
It is already being used in limited areas of the industry as the default
format for flash drives and external hdd. Without exfat, you are forced
to use ntfs, which is slow and inefficient.
Microsoft created this for a purpose and lets respect that.
Rod Speed
> guys, stop criticizing exfat.
Take your orders and shove them where the sun dont shine.
> It is not such a bad file system especially if you have windows 7.
The most obvious problem with it is that fuck all supports it.
> The advantages of exfat is mainly to serve as an alternative to ntfs.
> Exfat is also used to eliminate fragmentation on drives
It doesnt eliminate fragmentation, and there is no point in eliminating fragmentation anyway.
> and for use in drives with low space
Where it clearly doesnt eliminate fragmentation.
> (surprised?)
> Microsoft claimed that exfat does not fragment a volume
> until that volume only has 15 percent free space. I have
> tested that on my own system. 0 percent framentation.
Thats the implementation, not the file system.
> In addition, exfat requires little space for the actual formating.
Who cares with modern TB hard drives ?
> I heard that ntfs takes like 40 mb just for the file system.
> However, exfat takes much less than that.
Who cares with modern TB hard drives ?
> Consider an external hard drive. Usually, if you buy say for example
> a 1tb external hdd, you would have it formated in fat 32 which most
> people would change immediately to ntfs (why would you get so
> much space and NOT have files greater than 4 gigs?) However,
> Ntfs becomes inefficient and slow over time
Like hell it does.
> ESPECIALLY as the drive fills up.
Wrong, as always.
> exFat is Microsoft's solution to the problem.
Pity that fuck all supports it.
> Please respect the existence of exFAt, and don't randomly criticize it.
Take your orders and shove them where the sun dont shine.
The criticism aint random.
> It is already being used in limited areas of the industry
> as the default format for flash drives and external hdd.
By fuck all.
> Without exfat, you are forced to use ntfs,
No you arent.
> which is slow and inefficient.
Bare faced pig ignorant lie.
> Microsoft created this for a purpose
Yeah, to gouge license fees.
> and lets respect that.
Take your orders and shove them where the sun dont shine.
Ed Light
Is it as damage-resistant as NTFS?
On my flash drives, for large numbers of small files NTFS is much faster
than FAT.
--
Ed Light
Better World News TV Channel:
http://realnews.com
Iraq Veterans Against the War and Related:
http://ivaw.org
http://couragetoresist.org
http://antiwar.com
Send spam to the FTC at
sp...@uce.gov
Thanks, robots.
Ed Light
> Where do you get exFAT?
>
> Is it as damage-resistant as NTFS?
>
> On my flash drives, for large numbers of small files NTFS is much faster
> than FAT.
I guess it is just for Windows Embedded CD? Maybe needs an exFAT aware OS.
David Brown
I might not have used the same language myself, but I agree with /all/
your points here, Rod.
Capella must be an astroturfer quoting some MS press release.
Rod Speed
>>> guys, stop criticizing exfat.
>>> (surprised?)
>> Like hell it does.
>> Wrong, as always.
>> The criticism aint random.
>> By fuck all.
>> No you arent.
>>> and lets respect that.
Bugger, now I will have to reverse myself. Have you no sense of common decency and decorum at all ?
> Capella must be an astroturfer quoting some MS press release.
Just a silly little pig ignorant wanker.
mscot...@aol.com
>
> - Show quoted text -
Just to upset everyone I would like to say that exFAT is not a bad
file system. For many files it does not use the FAT and so is much
faster. It seems to encourage writing in long sequences rather than
filling each hole in the FAT.
It is part of Vista and Windows 7
However, FAT32 survives because it is compatible with many systems,
from cameras, to Macs, to Linux. exFAT will never replace it in those
areas and I do not expect to see it being used anywhere, except by
mistake by Windows 7 users formating and external drive.
Michael
Rod Speed
>>>>> guys, stop criticizing exfat.
>>>>> (surprised?)
>>>> Like hell it does.
>>>> Wrong, as always.
>>>> The criticism aint random.
>>>> By fuck all.
>>>> No you arent.
>>>>> and lets respect that.
> Just to upset everyone
You didnt upset me.
> I would like to say that exFAT is not a bad file system.
Sure, its certainly got some useful improvements, particular support for files
bigger than 4G but the massive problem with it is that fuck all supports it.
> For many files it does not use the FAT and so is much faster.
> It seems to encourage writing in long sequences rather than
> filling each hole in the FAT.
Sure, but thats not necessarily any big deal with modern 1TB drives and flash drives.
> It is part of Vista and Windows 7
Vista SP1 and Win 7
> However, FAT32 survives because it is compatible
> with many systems, from cameras, to Macs, to Linux.
And requires no license fee to Microsoft.
> exFAT will never replace it in those areas and I do not expect to see it being used
> anywhere, except by mistake by Windows 7 users formating and external drive.
I expect it will eventually do better than just that last.
Ed Light
> Just to upset everyone I would like to say that exFAT is not a bad
> file system. For many files it does not use the FAT and so is much
> faster. It seems to encourage writing in long sequences rather than
> filling each hole in the FAT.
>
> It is part of Vista and Windows 7
> Michael
Michael,
I missed part of this thread.
Just one question: Is exFAT as crash damage resistant as ntfs?
Ed Light
> Rod Speed wrote:
>> Capella wrote:
>>
>>> guys, stop criticizing exfat.
>>
>> Take your orders and shove them where the sun dont shine.
> I might not have used the same language myself, but I agree with /all/
> your points here, Rod.
Lots of people have roddy filtered out. You might consider it. It's so
friendly most of the time, without seeing his filth.
mscot...@aol.com
> mscotgr...@aol.com wrote:
> > Just to upset everyone I would like to say that exFAT is not a bad
> > file system. For many files it does not use the FAT and so is much
> > faster. It seems to encourage writing in long sequences rather than
> > filling each hole in the FAT.
>
> > It is part of Vista and Windows 7
> > Michael
>
> Michael,
>
> I missed part of this thread.
>
> Just one question: Is exFAT as crash damage resistant as ntfs?
> --
> Ed Light
>
> Better World News TV Channel:http://realnews.com
>
>
> Send spam to the FTC at
> Thanks, robots.
I don't think it is as crash resistant NTFS. Also, I don't think it
supports compression
I have no plans, and see no reason to change my NTFS drives to exFAT.
I haven't check speed of exFAT vs FAT32 vs NTFS, but I might try so on
an external USB drive
Michael
David Brown
> mscot...@aol.com wrote:
>
>> Just to upset everyone I would like to say that exFAT is not a bad
>> file system. For many files it does not use the FAT and so is much
>> faster. It seems to encourage writing in long sequences rather than
>> filling each hole in the FAT.
>>
>> It is part of Vista and Windows 7
>> Michael
>
> Michael,
>
> I missed part of this thread.
>
> Just one question: Is exFAT as crash damage resistant as ntfs?
NTFS journals metadata changes, making it somewhat crash resistant (but
anyone who has ever run chkdsk on a an NTFS partition knows this
resistance is limited). exFAT should be a bit safer than FAT, but is
not journalled.
exFAT is in no way a replacement for NTFS. It is aimed to be a bit of
an improvement over FAT (higher limits and a free space bitmap to
improve allocation), while still being light enough (in terms of
processing power, and disk space overhead) for things like cameras.
If I were a cynic, I'd say the main reason for its existence is to try
to lock people into Vista / Windows 7 or force them to upgrade older
Windows systems. MS don't like that people use FAT on usb drives and
other memory devices without paying them - few manufacturers pay the
licenses even though they are low cost, and their patents are tenuous at
best and will soon expire. Worst of all, non-MS systems can work
happily with FAT. exFAT gives MS a new chance - it is poorly specified
and documented so that these evil open-source commies can't implement
it, and the license agreements are secret so that they can give cheap
deals to boost exFAT's market penetration, then charge more once
manufacturers and users are locked in. There are certainly some patents
involved, and you can be sure that more will magically appear if anyone
else implements exFAT.
David Brown
> David Brown wrote:
>> Rod Speed wrote:
>>> Capella wrote:
>>>
>>>> guys, stop criticizing exfat.
>>>
>>> Take your orders and shove them where the sun dont shine.
> ....
>> I might not have used the same language myself, but I agree with /all/
>> your points here, Rod.
>
> Lots of people have roddy filtered out. You might consider it. It's so
> friendly most of the time, without seeing his filth.
>
I've considered it, but Rod does have useful things do say - he has a
fair amount of knowledge and experience, and people do get helpful
advice from him. The problem comes if he is contradicted - as long as
his posts are correct, they are useful. So I currently have my rodbot
filter on manual.
Yousuf Khan
> I missed part of this thread.
>
> Just one question: Is exFAT as crash damage resistant as ntfs?
I originally posted this thread:
http://preview.tinyurl.com/yzdqgvj
Basically, I saw an article about the new exFAT file system on
Microsoft's site. I thought that maybe Microsoft was giving it away for
free, since they were trying to make flash memory manufacturers and
others adopt it. But later it turned out that Microsoft is charging
money for it, so people are now not so keen on it anymore, including
myself. :-)
Yousuf Khan