“Western Digital has settled a class action lawsuit concerning the deceptive labeling of their hard drives. The problem occured when consumers bought hard drives with a stated capacity. As an example, let’s say you purchased a Western Digital 250 gigabyte hard drive. After you install the drive into your computer, you will notice that the drive size has become 231 gigabytes. The reason for this is that Western Digital, along with most hard drive manufacturers, calculates the storage capacity of their hard drives using the decimal system. This in contrast to virtually all operating systems that calculates the capacity of the hard drives with the binary system. As hard drives become larger, this discrepancy grows larger.”
Sometimes I kick myself being too much of a geek … I’ve always been aware that the stated capacity is higher than the “usable” capacity.
Can I feign stupidity (not hard for me 🙂 ) and somehow get in on the settlement? 😉
Edited 2006-06-28 20:24
I’m with you. Actually I don’t think it’s worth a lawsuit, but this issue has irritated me a little for many years, so I’m glad it’s going to be cleared up.
Western Digital and the rest of the hard drive companies are correct, according to the metric system 10^9 is Giga and 2^30 is Gibi.
http://zapatopi.net/labs/kibioctets.html
Western Digital and the rest of the hard drive companies are correct, according to the metric system 10^9 is Giga and 2^30 is Gibi.
You’ve got a better chance of the US going metric than of anyone but the most anal retentive geeks using the gibi/kibi/etc prefixes.
If the U.S. had gone SI long ago then its inhabitants would use such units correctly and this might never have happened. Standard prefixes were misappropriated and continuing to misuse them simply serves to devalue the utility of such standards. I admittedly fall into the poor habit when joining existing conversations where the meaning is clear, but I consider that a bug rather than a feature.
No they wouldn’t, i live in a country where we use the metric system and SI units and i still see NOONE using the kibi, etc, units.
The problem with these SI units is that they were introduced AFTER the words “megabyte”, “gigabyte” became common words, millions of geeks wont change their language just because a commitee says they should.
Now it might also have helped if the SI names sounded less wacky, it sure will be a cold day in hell before you will hear me say the our application takes up 700 mebibytes
It will be pretty easy to fix though, the day Windows starts only displaying base 10 units the majority of people will automatically use the SI units without even knowing it, and only the geeks will know the binary version. And it is hardly uncommon that some words have a different meaning when used in a technical context compared to when used by normal people.
The decimal prefixes misappropriated for use predate the usage of terms like “megabyte.” The binary prefixes necessarily postdate such misuse because it was never the intention for such prefixes to be used thusly and they were created to address this misuse by the IEC. Those are not SI names, since the SI system serves to standardize the metric system. They aren’t, really any more absurd-sounding than SI prefixes, you simply aren’t used to them. yottabye : yobibyte, wow how incredibly silly!
The point of standardization in technical discourse is to reduce ambiguity that results in errors. The misuse of standard prefixes in jargon resulted in the misuse of prefixes for reporting information to users, which resulted in a frivolous lawsuit that was settled at the future expense to the consumer. Real efficient that.
You’ve got a better chance of the US going metric than of anyone but the most anal retentive geeks using the gibi/kibi/etc prefixes.
Well, if gibi/kibi/etc can effectively clear up the clout, there is more than enough reason to support it. And name-calling isn’t nice. Having standards are simply important. They clear up many things for us to see the real problem…
I cannot understand how such frivilous lawsuits can be tolerated. When the article started implying that they had a problem with the settlement, I thought for sure they were going to dismiss it as crap. However, they think that the settlement wasn’t tough enough on WD, and suggest that this is a conspiracy to get people to buy more hard disks!!
First of all, the notion that 1 GB = 1 billion bytes is actually the official SI standard. The binary definition is not. Hard disks are block devices, and therefore there’s no binary addressing scheme involved, unlike most RAM technologies. For the hard disk manufacturers, the binary capacity measures are an inconvenience, albeit a very small one. However, they made their decision, and especially because it is in line with international standards, they are entitled to their decision.
On the other hand, Microsoft and other OS vendors made an arbitrary decision to report capacity and usage based on the binary system even though it is widely known that hard disks use the SI system. They knowingly report these quanitities in a deceiving manner. It’s not that customers are only getting 231 of the 250 GB they paid for, it’s that their OS is under-reporting the number of GB. The hard disk manufacturers should not be liable for deceiving their customers, the OS vendors should bear that burden!!
On Linux, starting with fileutils-4.0, basic disk utilities such as du and df can report capacity and usage using either system: -h indicates the binary system and -H indicates the SI system. It is really easy for DE developers and distributors to comply with the ways of the hard disk manufacturers simply by replacing the -h flag with -H in their graphical filemanagers and other disk utilities.
I think that consumer protection is important, but this is a bit too overzealous…
>> I think that consumer protection is important, but this is a bit too overzealous…
It is not about cosumer protection. The whole thing is about some greedy lawyers trying to sqeeze $$ out of HDD manufacturer. Consumer reparations is really just a side effect. See this: http://www.theinquirer.net/default.aspx?article=32683
This lawsuit is really a big BS. There has always been serveral different representation systems used concurrently inside a computer:
RAM capacity: binary byte, 1KB = 1024byte
RAM speed: decimal byte/sec, 1KB/SEC = 1000byte/sec
HDD capacity(HDD OEM): decimal byte, 1KB = 1000byte
HDD capacity(OS): binary byte, 1KB = 1024byte
HDD speed: decimal byte/sec, 1KB/sec = 1000byte/sec
HDD speed(SATA): decimal bit/sec, 1kb/sec = 1000bit/sec
Network speed: decimal bit/sec, 1kb/sec = 1000bit/sec
etc etc etc
Choice of unit are purely cause by historical reasons/customs. No one is representation particuarly right or wrong. As WD could be found guilty in this suit, we can sue the RAM manufacturer as well; cause speed*time != stated capacity.
This kind of stupid suit clearly shows how run down the US legal system has become.
First of all, the notion that 1 GB = 1 billion bytes is actually the official SI standard. The binary definition is not. Hard disks are block devices, and therefore there’s no binary addressing scheme involved, unlike most RAM technologies. For the hard disk manufacturers, the binary capacity measures are an inconvenience, albeit a very small one.
Wrong. Take a look at the ATA standard some time. All addressing in hard disks is done via sectors, which are a fixed size of 512 bytes. The hard disk size *MUST* be a multiple of a half kilobyte. You can’t read or write 1 byte to a hard disk – you have to do mutliples of 512. If your code writes 1 byte to a file, the OS reads in the 512 byte block, modifies the 1 byte you’re changing, and writes back the entire 512 bytes.
Anyway, the point is, hard disks are inherently sized based on powers of 2, so it makes sense to rate the capacity using power of 2 units.
One could see this coming for the last 20yrs but if you know the decimal v binary difference, then is was just a growing nuisance. Since we generally deal in K,M G,T bytes that are powers of 1024, the disk industry should have caved in along time ago and followed the same idea before the math looked too bad. Worse some HD vendors used a mix of decimal and binary.
Now I would really like to see alot more honesty regarding USB2.0 advertised at 480Mb when it most certainly never is. It can only sustain peak 480Mb during packet transfers after switching to that higher rate, otherwise it must also run signalling at the old 12Mb and even 1Mb rates to see what’s there. Net avg must be far less than 480Mb.
Also when one sees USB memory sticks or external drives with the USB2 moniker, they never actually suggest the device is usually far slower than even the 480 suggests. If it actually said avg 20Mbytes sustained for typical transfers on Windows with IDE66 supports, I would be quite happy with that.
As usual the lawyer makes off with the dough and Windows users get some trinkets they probably already have.
Dang. Had I known this I would have saved weeks spent looking for my missing 7 gigs.
what is the software they’re giving away? i keep hearing about it, but no specifics other than “backup software”.
is it something written by wdc or is it a copy of “dd” that wdc have compiled for windows?
i have five qualifying drives, so is it worth the hassle of sending off for this software – if it’s a copy of symantec ghost 10 corp, or acronis true image 9.1 enterprise server, then i might be interested….
but perhaps the question is, how is backup software going to make amends for “losing” 20gb per drive? if it was compression software it would make sense to me, you could gain storage space, but backup?????
what is the software they’re giving away? i keep hearing about it, but no specifics other than “backup software”.
Its probably something a vb hack threw together over a weekend
No, a gigabyte is 1,073,741,824 bits, no matter what the prefix says.
Hard drive manufactures just want to use lazy math or maybe programmers are bad at metrics prefixes.
It’s all about which number is bigger, so it looks like you get more than you do, and they get more money.
No one I know uses kibi, mebi, gibi, or tebi.
It was introduced in the late 90s, and you just can’t uproot 40 plus years or computer history.
Actually, it’s 1,073,741,824 bytes…
As for staying with the old definition because of history, I don’t know. Should we go back to the imperial system because it was used for years? Anyway, many open-source software are starting to use the binary prefix.
You can make new computer history, where the integrity of the SI units is maintained and those newly entering the field are saddled with a less ambiguous jargon that carries on into the future, instead of us pointlessly heaping our own failings onto their laps and telling them to live with it because that’s the way it is.
1,073,741,824 just lacks the beauty of 1024 or 200 or simple numbers like that. I doubt the kind of people who expected more really had a critical use for exactly the advertised amount of storage. Labels on boxes should only be used to tell you which drive is bigger, not exactly how many bits you can place on it.
Until the end of the nineties, ALL harddisks were sized with proper binary megabytes (or gigabytes).
Go check out anything below 10GB, it’ll be binary!
Then one company started selling (i think it was 40GB) disks with gibibytes instead gigabytes…and since this makes a difference in larger capacities, everyone followed.
As you can see below, i still have an honest 80 GIGAbyte, and a fake 60 GIBIbyte…
root@barry:~# fdisk -l | grep Disk
Disk /dev/hda: 60.0 GB, 60022480896 bytes
Disk /dev/hdb: 81.9 GB, 81964302336 bytes
cheers,
Paul
One programmer to another:
– Can you lend me 500$?
– Shall we make it a round number of 512$?
They should really specify the size as
1 GB=1000000000 octets, or else they will get a new lawsuit from someone whos byte got more then 8 bits. (Yes that does exists, a byte is just the minimum memory unit a cpu can referer directly)
well, can anyone just tell me what i need to do with this matter? i hv received an email from WD saying that i need to fill out the claim form… coz english isn’t my first language, i want to know what’s the claim about..
conincidently my HDD disappeared right after i got the email from them, does it mean my HDD will no longer be working if i dun submit the claim form?
thx
esther from Hong Kong