Slashdot Mirror
Computer Makers Sued Over Hard Drive Size
FPCat writes "Finally, some one is doing something about one of my pet peeves. It seems a group of people are suing Apple, Dell, Gateway, HP, and others for misleading consumers about hard disk sizes. About time someone spoke up and said '1000 MB != 1 GB'" It's not much of a mystery to anyone who's up on industry practices, but it's similar to the way graphic displays are sized, cereal boxes are filled, and so on. Andy Rooney could have a field day with this one.
Thats just stupid. I think the lawsuit is innapropriate.
HD manufacturers always measuered their disks like that.
What next? Will people sue that their 56k modems are not 56kilobytes/second? Or that their DSL line is 1.5Mbits and not bytes?
This is just silly. They might as well complain that they lose size in formating.
In Soviet Russia, the television watches YOU!
A gigabyte IS 1000 megabytes. A megabyte is, however, NOT 1024 kilobytes or 104576 bytes. That's MiB and GiB you're thinking of. Giga and Mega are SI prefexes. Not binary compatible.
The opinions in this post are ficticious. Any similarity to actual opinions, real or imagined, is purely coincidental.
did somebody buy a 100gb drive with the intent of using EVERY LAST BYTE of it when they realized it actually works out to a touch less? if i tell somebody it's 100 degrees outside and it's actually 97, it's HOT.
people need to get a life, seriously...
well, it's nothing one behind the ear wouldn't cure
What if it's both? I have a multivolume disk array. What happens if I end up replacing one of my drives with a disk that looks like it should be big enough by the specs only to find out that its four or five megabytes too small? They're probably not going to buy my logic for why I'm returning it...
Do not look into laser with remaining eye.
Every hard drive (even on web sites that sell systems) has a disclaimer that it isn't the "actual" byte size.
I really think that people should standardise the meaning of kilo-, and giga- to their SI meanings. The is a google cache link to a web page about the proposed changes where they would change to SI definitions, and new prefixes (kibi, gibi) would come into to define the warped computer terminology defintions of kilo- and giga-. It would be less fuss for most people, and everyone could then get on without all this trivial garbage.
I would support a government mandate that tech companies have to use binary SI prefixes on labels.
Mandating the current use of gigabyte but that it means 10^9 is too trouble some, but saying gibibytes is simple, people that don't care will either read it as "giga" not realizing, or be told by sales-people that its "the same thing". and they won't be surprised when the drive is the wrong size.
We have mandates on product labeling for many other products I think its time we force the industry to be upfront. Don't think this is an accident, the drive manufacturers knew EXACTLY what they were doing when they started using standard SI meanings for the prefixes, rather than the industry accepted practice.
Car, truck, and motorcycle represent their motors rounded usually to the nearest 100. My 1100cc motorcycle is actually onlt 1085cc. Isnt this sort of behavior rampany? Are 50mg pills always 50mg? Certainly 2x4 lumber is not actually 2x4. I would think making everything absolutely accurate would simple confuse the average consumer.
This just seems silly.
Heh, I remember when Apple actually did it THE OTHER way. I was trying to look up a Maxtor 4.0 GB HD that shipped in one of my macs, and could not find any mention of such a hd. The Apple specs clearly said it was 4.0 GB. But it turned out that Maxtor classified these as 4.3GB, whereas apple used the 1024 size cound, rather than the 1000 that maxtor used. Heh.
Yes, but using an already established system of measurement and then changing what it means in the fine print isn't exactly "right" either.
Imagine how quickly Coke would get sued if they made a new 2 liter bottle, but it was really only 1.8 litres and somewhere it has in small print "1 litre means 0.9 litres".
I doubt that would last very long... so why has it lasted this long with Hard Drives?
Consider my newest hard drive. Western Digital, who manufactured it, says it's 120GB. Windows 2000, written by Microsoft, tells me it's 111GB. Wieghing in the fact that it's slightly over 120,000,000,000 bytes, it's apparent to me that Western Digital is right and Microsoft is wrong. Had Windows 2000 been prgrammed to say "GiB" instead of "GB", Microsoft would be right as well.
Those are the real problems. I just got a new HP (complain all you want, it was cheap and it had a DVD burner) with an 80 GB drive... open it up and look at the size of C: (D: was a backup) and we've got 71 GB. Luckily I moved the 80 GB hard drive from my older computer into that one. That 9 GB doesn't hurt me so much anymore.
those systems that ship with 128mb or more of memory, but in the fine print says part of it is shared with the Video Card.
:)
I remember a long time ago my IBM PCjr had 128k of memory, but 16k of it was shared with the display card, such that only 112 was available. Consequently, many PC software apps that required 128k of ram didn't work. Thank god for the sidecar memory expansion kit
So Apple went with the flow and started marketing 12" monitors as 13". And for a time it was good.
Until the industry got slapped with a deceptive advertising suit or something. But rather than market it CORRECTLY, now more ink is wasted when ads are printed with disclaimers, like "* 18.1" viewable" on 19" CRT screens.
I've always had such problems with RAIDs, etc, of finding an identical drive to plug in when one dies, that I've gotten in the habit of buying a couple at a time, and just leaving them on a shelf. The one time a customer called me on it, I pointed out that, since I was billing at 150.00 an hour, buying a drive 200 dollar drive that would save me 3 hours of work was a bargain for them.
ad logicam Claiming a proposition is false because it was presented as the conclusion of a fallacious argument.
Nonsense. Since words can have different definitions in different industries or professions (compare casual vs. legal usage of "property"), prefixes can too.
The computing industry isn't the science or engineering industry.
And electrical engineers are fine with base 2 prefixes.
So, don't be a fruit and walk around talking crazy shit about kibi- and gibibytes.
An excellent book on the subject is The Litigation Explosion: What Happened When America Unleased the Lawsuit by Walter K. Olson, in case anyone is interested.
I read it many years ago and thought it an excellent analysis on the the underlying causes of litigiouness in the American legal system.
It's no longer in print, apparently, but you can get it used or pick it up at your local library.
quiquid id est, timeo puellas et oscula dantes.
If you wish to beat me over the head with your analogy, how about making it a valid one? Adjectives and nouns are not the same thing. If I say you have a big gun, a big server, or big breasts, the word big means the same darn thing in all three contexts.
"I'd rather be a lightning rod than a seismometer." -Ken Kesey
For CD-R, DVD-R/w, the industry defines 1024 MB = 1 GB
You'd be surprised: all the writable DVDs I have claim 4.7GB but offer 4,700,000,000 (+/- a tiny amount) bytes = 4.3*2^30. (CDs, on the other hand, do use 1024: the "700MB" CDs I use are 736,966,656 (data) bytes = 703*2^20.
Good lord, this is confusing...
Look at the context. Most everything at the lowest level in a binary computer operates on the base 2 system. The ONLY reason that HD makers round like this is to make the marketing teams' job easier.
I also wouldn't get hasty about the government intervening on standards. If they really cared about universal standards, I'd be bitching about my car getting less than 9 kilometers to the liter. In fat, if they did step in, I'd be worried about them upholding the other standard.
Elvis Presley said: 'My voice alone is just an ordinary voice. What people come to see is how I use it. If I stand still while I'm singing, I'm dead!'
Close to later re-phrased:
"It's not my voice. It's what I am doing with it."
Back to the topic:
IMHO if my HDD is used just as a big CD - it's dead.
Less is more !
What about all the thousands of bytes spread about your disk of empty space? Depending on the file system in use, blocks vary in size. If your block size for a partition is 1024 bytes, and you want to write a 500 byte file, then you just wasted 524 bytes - or over 50% of the size of the file. Multiply this times the thousands of files on your system, and you are losing a good chunk (maybe 20% of the space used) of the disk anyway - particularly if you have alot of small files.
On a 20GB drive, we are talking about 3.6 Gigabytes... give or take.
Lodragan Draoidh
The more you explain it, the more I don't understand it. - Mark Twain
The computer science community has accepted, by long use, the definition of 1KB=2^10 bytes. This means that, although it is inconsistent with the SI definitions of the quantifiers, this is a de facto industry standard; one which hardware manufacturers have intentionally defied for years. That this is not the SI meaning of those quantifiers is a moot point.
It seems to me that, to be a legitimate claim, there must be an established alternate standard for computation and reporting of file and disk sizes. The question then becomes, how does Microsoft, Linux, Solaris, Apple, and HP/UX compute file sizes? Also, what does Intel, Sun, TI, etc use compute megahertz? If they are using 2^10 (1024) instead of 10^3 (1000), then the lawsuit may have a basis. If, however, the OSes utilize the same measuring stick as the drive manufacturers, then there should really be no problem. As an example, ls -al shows clarkconnect-1.3.iso on my Gentoo Linux box as 183009280 bytes. This corresponds to 183.0MB, or 174.5 MiB. An ls -alh (added h for human readable) shows the same file as 175MB. Therefore, Linux seems to use M = 2^20, as anyone who has used computers for more than 10 years would KNOW is correct. Computers are based on a binary system. 2^1 is a bit, 2^2 bits is a nibble, 2^3 bits is a byte, 2^10 bytes is a KB, or kilobyte, 2^10*2^10 bytes is a MB, or megabyte. It is unfortunate that the originators of these names used the metric prefixes, but they did, and it was understood by those dealing with it. This lawsuit is really pathetic. Users should read the box, where it plainly states the meaning of the abbreviation (Maxtor 120 GB drive says "A gigabyte (GB) means 1 billion bytes."). I could see how someone could be frustrated, though if they lost 7.37+% of their storage to an "executive decision" (1024*1024*1024)/(1000*1000*1000) = 1.0737.
Hard Drive
Disk Size: 111.79 GB (1K = 1024) 121 GB (1K = 1000)
People have it backwards. As the previous poster correctly observed, SI units are well defined and widely used. Only in this backwards, litigious society of ours would a group of people complain... nay, sue over the fact that hard drive makers were sticking to standard units and cheating the customers out of some measly number of extra bytes.
Let's see... If I buy and eighty gigabyte drive, and I'm thinking in powers of two, I would expect to have 85,899,345,920 bytes of usable space. What? You're telling me it's only 80,000,000,000? I want my extra 7%! Never mind that at current street prices drives sell for around $1/GB, which means that buy the extra space (in SI units) would probably cost less than the cost of shipping the drive.
Imagine this scenario: I buy some ram for my computer. Lessee... I got 512MB, so I should have 512,000,000 bytes. Say again? You mean I actually have 536,870,912 bytes? Cool! 36.87x10^6 bytes (mega-) for free!
Mega means "1 million". Computer scientists started using the term Mega to identify 2^20, which isn't actually 1 million....they were approximating. In this case, it's the computer scientists who gave an alternate meaning to a common numerical term...not HD manufacturers.
A modern day witchhunt.
Good example. Getting screwed at the gas station over a gallon of gas.
.
As a Canadian, where a gallon = 4.54 litres, I get screwed every time I go to the US and buy a 3.9 litre US gallon.
The last time I drove you to the US.....
I spent my $500 dollars (which you call $350)
I also drive down in a car that I bought with 0% financing (yet paid $2300 more than I would have if I paid cash).
I wore my size 42 shoes (about size 9 to you) and drove at 100 most of the way there (60 for you).
SIO standards actually define 1000 MB (MegaByte) as 1 GB (GigaByte). If you want to use the 1024 thingie, use should say 1024 MiB == 1 GiB. This stands for MebiByte and GebiByte resp. The latter are only to be used for main memory sizes afaik, not for hard drive sizes. So you should still specify the hard drive size in MB, GB or TB. However, the difference between various manufacturers l;ies in the fact that they do not allways refer to size after formatting the drive with you favorite filesystem (e.g. ReiserFS). But do you really care if your 120GB drive is actually only 115GB after formatting? I think not.
I am the Shield Anvil. And I am not yet done.
In the computer world, metric prefixes are base 10 when they refer to time or a rate and base 2 when they refer to data size.
Why?
Times/rates use base 10 because, as is suggested earlier, it's the standard for the metric system. It makes sense to have the same prefixes mean the same thing across the board. Since time is arbitrary anyway, it makes sense to use the "regular" metric prefixes.
Data size uses base 2 because of the way computers access memory. If you have a 8-bit address, you can have 256 chunks of memory. Likewise, if you have a 10-bit address, you can have 1024 chunks of memory. It takes the same number of address bits to have 1000 as it does to have 1024 and the hardware is generally simpler if you only deal with powers of 2. So since data-storage hardware is organized into groups that are powers of 2, it makes sense to use powers of 2 for the prefixes.
The argument is very much similar to the argument against the English system. Both systems are used because you want simple conversions. I have 1,000 fluid drams of something. How many fluid ounces is that? Likewise, if you always end up with 1024 units of something, why would you use 1000 as your base instead of 1024? Does it make more sense to have all of your data coming in multiples of 1.024 or in multiples of 1?
Hard drive manufacturers get away with it because hard drives aren't optimized in quite the same way as RAM. Hard drives are circular, so there's really very little benefit to making them contain 2^3n bits. You may end up dividing a hard drive in any event, and chances are your hard drive isn't exactly the maximum size your OS can handle. RAM, on the other hand, can easily be at the absolute maximum amount that a motherboard can handle. In that case, it makes sense to have 2^3n bits instead of 10^n bits. It needs the same size address and 2^3n is easier to manufacture, so ram always comes in powers of 2.
>No! They are not even close to being in the same galaxy as "more >correct". Within the context of the computer world,
>1K = 1024 or 2^10
>1M = 1048576 or 2^20
>1G = 1073741824 or 2^30
>1T = 1099511627776 or 2^40
Where have you been?
1KB = 1000 bytes
1MB = 1000000 bytes
1GB = 1000000000 bytes
1TB = 1000000000000 bytes
1KiB = 1024 bytes
1MiB = 1048576 bytes
1GiB = 1073741824 bytes
1TiB = 1099511627776 bytes
This "new" standard is from December 1998 (when it was adopted by the IEC).
Check here or here for reference.
Google for "SI binary prefix" for many more references if you care to.
//TheToon
Just because people were abusing the metric prefixes for years does NOT mean that they should keep abusing them.
Let me guess, you live in the US and don't have a fucking clue what the metric system is all about. For your information, the metric prefixes kilo, mega, and giga stand for 10^3, 10^6, and 10^9, and NOT 1024, 1024^2, 1024^3. The new units (KiB, MiB, etc) are meant to stop the abuse of the metric prefixes.
First of all, everybody chill. When you can buy 120+/- GB disk drives for $60, who cares about a 10% difference? Especially given that filesystems waste so much of the drive with unused portions of clusters, etc.
And why are they suing the computer companies rather than the disk drive companies? This is just a nuisance suit because the drive companies generally remember to put the footnote on the box, but the computer ads are already too packed to squeeze in any more caveats.
Different disk drive companies use MB to mean different quantities. Believe it or not, there are more than two choices. You may think MB should be 2^20 or 10^6 bytes, but many disk drive companies use 1000 KB (or 1000 * 1024 bytes).
And hard drives actually have more space than advertised. A significant portion of the drive holds spare sectors, and there's quite a few ECC bits on there, too.
It's the computer folks that corrupted the meanings of the SI prefixes. To distinguish the difference, many used to use capital K to mean 1024 and reserved lowercase k for 1000. And nobody really cared about a ~2.5% difference. That was fine until we got to megabytes, since M and m are both standard SI prefixes.
The odds of a cosmic ray flipping one of your bits when you had 64KB RAM was infinitessimal, and we had parity bits just in case. But now RAM sizes have grown a million fold, and we've practically eliminated parity and ECC bits. (Though the odds of a cosmic ray flipping an important bit is still tiny since most of your bits are stupid bitmaps, MP3 samples, and spyware data.) In a sense, aren't the hard drive companies more noble by using that few extra percent to protect your data than the RAM manufacturers who give you a few percent more buy no longer make an effort to ensure data integrity?
And finally, I find the claims of the plaintiffs amusing when they estimate how many digital photos you could store in the "missing" space. Isn't a vague estimate without regard to image size, resolution, color depth, file format, and file system potentially just as misleading as the footnote on your hard drive's retail box?