Microsoft Increases the FAT32 Limit From 32GB To 2TB (windows.com) 83
Longtime Slashdot reader AmiMoJo writes: Windows has limited FAT32 partitions to a maximum of 32GB for decades now. When memory cards and USB drives exceeded 32GB in size, the only options were exFAT or NTFS. Neither option was well supported on other platforms at first, although exFAT support is fairly widespread now. In their latest blog post, Microsoft announced that the limit for FAT32 partitions is being increased to 2TB. Of course, that doesn't mean that every device that supports FAT32 will work flawlessly with a 2TB partition size, but at least there is a decent chance that older devices with don't support exFAT will now be usable with memory cards over 32GB.
just in time (Score:5, Funny)
Just in time for the 21st century.
Re: (Score:3)
Re: (Score:1)
I use greek gods for my hostnames, but my challenge is to find the short names because I'm too lazy to type long names.
Re: (Score:2)
Re: (Score:3)
Re:Why (Score:4, Interesting)
Re: Why (Score:5, Interesting)
Yes, and no.
For PCs of the late 90s and early to mid 00s, yes. 'To promote NTFS.'
For Memory cards?
No.
That's much more closely tied to 'convenience of the memory card consortium'. Specifically, 'It's VERY convenient for the filesystem to have an allocation unit size that is a whole divisible factor of the erase unit size, with 1:1 being *oh so VERY convenient*!'
For devices up to about 32gb in size, this 'convenient coincidence!' Held. Larger devices however, have erase unit sizes far too large for FAT32 to be 'convenient' for.
This is, (among other reasons), my microsoft created ExFAT. It's 'FAT', with 'Absurdly large cluster sizes!'
Exactly what the memory card consortium THIRSTED for.
It also let mictosoft get a shiny new exclusive patent that *everyone* would need to get a license for, which is what *microsoft* THIRSTED for.
This marriage of convenience saw fat32 not having 'official' large volume support for ages.
I would not at all be surprised if this recent change coincides with ExFAT patent expiration.
Re: (Score:3)
ExFAT has other useful features for applications that stream data to storage (e.g. recording audio/video). It's a lot more efficient than FAT32 for these applications.
Re: (Score:2)
[ExFAT] also let mictosoft (sic) get a shiny new exclusive patent that *everyone* would need to get a license for, which is what *microsoft* THIRSTED for.
I have re-formatted several USB drives to ExFAT using Linux. Not to create partitions greater than 32 GB, but to support boot-drives with ISO files that are too large for Fat32. Did the creators of the Linux distro or exfatprogs need to send a check to Redmond for supporting ExFAT? I don't think so.
Re: Why (Score:4, Interesting)
Re: (Score:2)
It didn't really, Linux support was via a FUSE using Paragon's proprietary driver, and the latter definitely paid Microsoft. Samsung also had a driver of dubious quality and even more dubious legality. It was used in Samsung products (paid for by Samsung) but the rest of the Linux world didn't touch it. The second exFAT had it's IP released (August 2019) exFAT was submitted to the 5.4 Kernel release.
Re: (Score:3)
Did the creators of the Linux distro or exfatprogs need to send a check to Redmond for supporting ExFAT?
Yes they did. How did you miss this? This was the specific reason that the Linux kernel did *NOT* explicitly support exFAT (no capitalisation on the first letter) until 2019. Prior to 2019 if you used exFAT it was via a user space and closed source FUSE driver that was provided and paid for by Paragon. If you wanted to include exFAT support in Linux you did so via Paragon's *paid for* software. In 2019 Microsoft opened the specification and made them patent cost free and a very VERY rough exFAT driver made
Re: (Score:2)
Yeah, Microsoft has stupid partition limits on NTFS as well - even the 64-bit Windows does. They should be able to set a limit at 2^48 Bytes, but no! It's a measly 256TB-8192TB. Why not simply set it to 2^63, and not worry about it in the foreseeable future?
Re: (Score:1)
Yeah, Microsoft has stupid partition limits on NTFS as well - even the 64-bit Windows does. They should be able to set a limit at 2^48 Bytes, but no! It's a measly 256TB-8192TB. Why not simply set it to 2^63, and not worry about it in the foreseeable future?
Because you humans cannot perceive yet will demand 16K and then 32, 64, 128K resolution for your porn videos.
Re:Why (Score:5, Informative)
They did.
Earlier windows versions didn't have this limitation, they introduced it in xp as far as i remember. Other systems also had no trouble formatting fat32 volumes up to 2tb in size, and windows would happily read such volumes it just couldn't format them.
Re: (Score:2)
It could format them from the command line. Just the GUI gave you no option to do so.
Re: (Score:3)
Re: (Score:2)
It was hard coded into the GUI format utility by Dave Plummer, as hard drives were small back then and supporting larger drive sizes would make tiny files use exponentially more space on disk.
Then the limitation stuck around for decades because no-one thought to fix it. Notably the limitation didn't exist on on 3rd party OSes. Or for that matter, Windows itself, as it can mount and read/write such partitions fine.
FAT32 Gaslighting (Score:5, Interesting)
The limit of the filesystem was always 2TB, but Microsoft had this artificial 32GB limit in their built-in format tool. You could easily format a larger FAT32 partition using third-party or Linux-based tools.
I think this was an artificial limit kept in place to ensure exFAT adoption, specifically to ensure manufacturers paid their licenses.
Of course, there is still the 4GB file size limit to contend with, but I've never had a problem using a 256GB FAT32 partition on older devices, for example.
Re: FAT32 Gaslighting (Score:2)
Thank you, I thought I remembered the limit being 2TB.
Re: FAT32 Gaslighting (Score:4, Informative)
Of course, there is still the 4GB file size limit to contend with, but I've never had a problem using a 256GB FAT32 partition on older devices, for example.
The file size limit has always been more of a problem for me. One of our LG TVs refuses to read memory sticks unless they're FAT32, while an older LG model has no problems with exFAT or NTFS. A good quality rip of a film can easily exceed 10GB.
Re: FAT32 Gaslighting (Score:1)
So you cant watch pirated films on one of your TVs? Oh the humanity!
Re: (Score:3)
What an odd thing to say.
I rip my DVD and Bluray discs to a NAS and play them with Jellyfin. There's not even an optical player near a TV anymore. The last one died in about 2008.
But I have put some of those files on USB before for the kids.
Used Blurays are about $5 usually and much better quality than any streaming service rental.
Re: FAT32 Gaslighting (Score:3)
You know you can pick up a new dvd player for the price of a few hardback books, right?
Re: FAT32 Gaslighting (Score:2)
it is my hardware, i use it as i see fit. if the hardware is 2T, i don't need an artificial limit.
what i do or don't with it is nobody's concern until they can prove I've done something legally unacceptable.
Problem?
Re: (Score:2)
Who said a movie rip is pirated? Are you ignorant or a corporate bootlicker continuing to spread the lie that format conversion is 100% legal and not piracy?
Re: (Score:2)
Breaking the DRM is illegal in most jurisdictions due to the Americans insisting on it as part of trade deals. Illegally acquired rips would be considered piracy, even if no one cares. I know I don't care and neither does my government but strictly speaking, it is illegal.
Re: FAT32 Gaslighting (Score:2)
I believe this is a real limit of FAT32 though as the header only allocates 4 bytes for size. Same as .wav files. I guess nobody wanted to waste the bytes imagining files could ever exceed that size.
Re: (Score:1)
ffmpeg -i bigfile.mkv -c copy -to 1:00:00 part1.mkv
ffmpeg -i bigfile.mkv -c copy -ss 1:00:00 part2.mkv
plus mkclean if your TV is fussy like mine
Re: (Score:2)
Re: (Score:1)
If you omit menu and extra features, most DVDs could be ripped 100% quality to fit a 4.7 GB dvd rw+. What are you talking about?
Re: (Score:2)
I think this was an artificial limit kept in place to ensure exFAT adoption, specifically to ensure manufacturers paid their licenses.
Oh boy did that dredge up some fun memories from an embedded-Linux project I was working on a while back. Like this gem here:
15. What are some of the risks OEMs face when using open-source code for a Linux Kernel versus a proprietary solution from Paragon Software?
If an issue is found within the open-source code and no one is willing or able to fix or maintain it, then the code can be excluded from Linux. For example of this was the removal of EXOFS from Linux in version 5.1.
If an issue is caused by the open-source code (e.g. fails completely, causes extensive memory or CPU usage with an OEM’s hardware configuration, performance degrades with a new Linux update, etc.), then the only resource for assistance is the open source community. OEMs should not expect any assistance from commercial exFAT providers since GPL v2 requires the source code to be published.
OEMs can become a target for legal inquiries for GPL compliance concerning full disclosure of the source code related to the OEM’s product. The main problem with such demands for source code disclosure is determining to what extent an OEM is obligated to disclose the source code of its product – especially since OEM products typically use a mix of both proprietary and open source GPL code. An experienced (i.e. expensive) lawyer is required to address such GPL compliance inquiries properly and determine to what extent source code should be disclosed.
Open-source code is not free of vulnerabilities that are easy to exploit for viruses and cyber attacks. Vulnerabilities of SMB protocol have already led to major virus attacks. Proprietary code, which is not available as source code, is safer and decreases the risk of an OEM’s product being susceptible to such vulnerability-based attacks.
Re: (Score:2)
Proprietary code, which is not available as source code, is safer
[Citation Needed]
Re: (Score:2)
Wish I had modpoints. I never created a FAT32 partition in Windows, so I never ran into this limitation.
The headline is so misleading it's basically a lie.
Re: (Score:2)
Except the guy (Dave Plummer) who wrote the dialog (back in the 90s) picked 32GB as a "big enough" limit that never was updated.
In other words, back then if you had a large enough drive, you were forced to use NTFS. exFAT came well after the dialog limit.
https://www.youtube.com/watch?... [youtube.com]
Re:FAT32 Gaslighting (Score:4, Interesting)
Bingo. Until recently, vendors implementing exFAT required licensing from Microsoft. Now that those patents have expired, there is no reason to arbitrarily limit FAT32 anymore.
Re: (Score:2)
This makes no sense. There was no patent related reason FAT32 couldn't be have larger sizes, and in fact since the issue was Windows GUI related only vendors could do any number of user friendly yet completely solutions:
a) Format FAT32 in hardware without limit.
b) Provide FAT32 formatting in their software without limit.
c) Provide a simple GUI front end that calls Windows existing formatter from the command line to format without limit.
Specifically I use a lot of memory cards in a lot of different devices.
Re: (Score:2)
Re: (Score:2)
Can't MS increase that 4 GB file size limit when updating its FAT32?
Re: (Score:3)
They didn't update the format; just the Windows tool that had the limit--which was never part of the format.
Re: (Score:2)
No, the 4GiB file limit is part of the filesystem itself. Changing that would be a breaking change. That's why, when they changed it, they also changed the name to ExFAT, along with a number of other fixes and updates. The 32GiB limit was just an artificial limitation of the GUI formatting tool. IIRC, the command line tools have never had such a limitation, and the drivers can access larger filesystems just fine. The 2TiB limit is a hard limit of the filesystem spec and going higher would be a breaking chan
Re: (Score:2)
Just a GUI change (Score:5, Interesting)
This is just a GUI change. Windows already supports mounting these larger drives and already supports formatting these larger drives on the command line. Nothing about the actual spec is changing.
Just a marketing "change". (Score:2)
This is just a GUI change. Windows already supports mounting these larger drives and already supports formatting these larger drives on the command line. Nothing about the actual spec is changing.
What you casually call a GUI "change", I would more call an error/bug in need of correction, long overdue.
How long have the other methods built into Windows, functioned properly in supporting these larger drives again?
Call a spade a spade already. Otherwise, the next Spectre Ping Death Touch Meltdown will be dismissed as a Class-3 Annoyance, because stock price.
Fuck that.
Re: (Score:2)
Just as in that's the only thing changing, not that it's trivial or not overdue.
Re: (Score:2)
huh? (Score:4, Interesting)
Personally I just use exFAT if I need to format e.g. a USB stick gtr thn 32 that I need on linux / mac / windows.
Fatter FAT? (Score:2)
Unproductive improvement (Score:1)
Re: Unproductive improvement (Score:5, Insightful)
All modern OS are now using NTFS for 2 decades now.
Except, you know, everyone who doesn't use Windows.
Incidentally, the XBox 360 couldn't mount NTFS volumes, but HFS+ was fine. I always thought that was odd.
Re: Unproductive improvement (Score:2)
The article is about removable media (Score:2)
You are correct with respect to their internal storage.
However, say you want to interchange files among several computers using removable media, such as an SD card, USB flash drive, or USB hard drive. One is a Windows PC that prefers NTFS, another a Mac that prefers Apple's FS, and another a Linux PC that prefers ext4. What file system would you use on the drive?
Re: (Score:2)
And if you purchase a USB flash drive make sure you get a 3.x not a 2.0, for performance. But the naming is tricky, USB 3.2 gen 1 is just a 3.0.
Re: (Score:2)
I think the HFS+ bit probably has to do with the idea that Microsoft used PowerMac G5 machines to develop the Xbox 360 [fandom.com] because of the PowerPC processor. I'd imagine there were some development steps that might have required mounting an HFS+ drive to pull code over to the 360.
Re: Unproductive improvement (Score:2)
Re:Unproductive improvement (Score:4, Insightful)
All modern OS are now using NTFS for 2 decades now.
Hmm. I recently bought a game camera. Following the instructions, I inserted and formatted (using the camera's utility) a 32G microSD card. After use, I pulled it out and stuffed it in my Linux system. Linux says exFAT.
I suspect that a lot of smaller devices will avoid NTFS for its complexity and go with FAT, exFAT, etc. for their relative simplicity.
Re: Unproductive improvement (Score:5, Informative)
It has *much* more to do with the memory card's erase block size.
NTFS wants to use a 512 BYTE or 1kbyte allocation unit size. (Dont believe me? Right click your system volume, and choose properties. See what your allocation unit size is.)
This size was selected because it is 1:1 the sector size of original winchester style hard disk drives, which makes those sizes the most efficient to transfer to or from the disk controller.
Modern drives tend to favor 4kbyte sized sectors, but still emulate 512 BYTE ones.
FAT had cluster (allocation unit) sizes quite a bit larger than this. Usually between 4k and 16k, but 32k and 64k clusters are supported.
For early flash memory cards, 32k and 64k cluster sizes were 1:1 what the eraseblock sizes of the flash array were, meaning having the filesystem use that size gave the best possible efficiency with the device controller.
SDHC and SDXC devices though, have erase block sizes that (cough), 'greatly exceed' (cough) what FAT32 can support.
ExFAT however, happily lets you use cluster sizes in the MULTIPLE MEGABYTES size range, allowing the flash makers to still have 1:1 cluster->erase unit parity, and maximized device IO efficiency.
Your camera formats that card as ExFAT because that's what the SDCard Assn demands.
The SDCard Assn demands it, so that they can reliably claim the write speeds written on the top of the card.
NTFS will annihilate flash cards with write amplification, and have piss-poor io performance writing to them.
Re: Unproductive improvement (Score:2)
I assumed there was some technical reason for it to have been set that way so I left it, even though it's pretty annyoing if every little plain textfile note takes up 8 MB. (Don't npm install on that sucker!) I couldn't find a confident explanation back then, but this finally explains it.
Re: (Score:2)
All modern OS are now using NTFS for 2 decades now.
Really? I don't think I've ever used NTFS on any machine I've owned.
Lack of information.... (Score:2)
I've certainly never formatted removable media in NTFS. That sounds like a way to make your life more difficult than necessary.
Re: (Score:2)
I've certainly never formatted removable media in NTFS. That sounds like a way to make your life more difficult than necessary.
As does using Windows.
Re: (Score:2)
Re: Unproductive improvement (Score:2)
About time (Score:2)
Windows 98 SE could format and use 64GB FAT32 volumes out of the box and even larger after patching, so kudos to MS for finally making Windows NT series exceed the capabilities of mighty Windows 98.
Useless comment (Score:2)
A million years ago I was on a call at work, with the person on the other end on speaker. They mentioned that "So-and-so lost his FAT." So-and-so was quite obese, so of course I said, "Good! He needs to lose his fat!" So-and-so was present and heard my comment.
Fortunately So-and-so had a sense of humor.
Oy Vey! (Score:2)
2TB of un-journaled data? I recommend you just don't.
I'll still use NTFS (Score:2)
As a Linux user I primarily use EXT4, so I've gotten used to assigning 'natural language' filenames to my files. Being able to use spaces, and many punctuation characters, makes filenames more descriptive and easier to read. That's important to me for large libraries of certain file types.
With few exceptions NTFS does allow that; but FAT32 doesn't. I like some of my external drives to be accessible by Windows machines 'just in case'.
Granted, Linux's handling of NTFS drives is a bit limited. An accidental di
mkfs.vfat has so such limitation (Score:2)
I've been using 128 GB SD card in my phone for many years. Linux had no problem formatting it to fat32.
They probably didn't metion that pasky patent (Score:2)
unless you dig deeply.
Anyone remember the Microsoft FAT patents in the 90's?
gues educated (Score:2)
It would not surprise me that patents are involved.
Ancient stuff (Score:1)
Why is this outdated system full of flaws still being pushed onto us? Even patents on exFAT expire next year btw (unless this new FAT32 trick is patented, of course).
Dave Plummer's Fault (Score:2)