Seagate Offers Refunds on 6.2 Million Hard Drives

An anonymous reader writes "Seagate has agreed to settle a lawsuit that alleges that the company mislead customers by selling them hard disk drives with less capacity than the company advertised. The suit states that Seagate's use of the decimal definition of the storage capacity term "gigabyte" was misleading and inaccurate: whereby 1GB = 1 billion bytes. In actuality, 1GB = 1,073,741,824 bytes — a difference of approximately 7% from Seagate's figures. Seagate is saying it will offer a cash refund or free backup and recovery software."

wow

[Anonymous Coward]

free backup and recovery software... yea, that'll help alot.

Think this will set precedent?

[Adradis]

Wow, I'm surprised that actually went through, if only because the court systems seem so broken. Hopefully, other manufacturers will get the hint and start changing their plans. I could just see this going after other manufacturers too, who insist on using smaller sizes for their measurements to seem bigger.

Seems Silly to me

[IcarusMoth]

Seriously, the blame could just as easily be laid at the feet of the OS developers. There is a long standing history of disk manufacturers using base 10 counting numbers. It would not be so horribly difficult for the OS developers to conform to the base 10 measurement. I mean what next are the consumers going to sue because the formatting and allocation tables take up room? or perhaps because it hides space for virtual memory? seriously. come on people.

Re:Seems Silly to me

[gadzook33]

Yeah, other than the fact that computers suck at base-10 counting and are really really good at base-2 counting, you're absolutely right.

Re:Think this will set precedent?

[DiSKiLLeR]

Wow, I'm surprised that actually went through, if only because the court systems seem so broken. Hopefully, other manufacturers will get the hint and start changing their plans. I could just see this going after other manufacturers too, who insist on using smaller sizes for their measurements to seem bigger.

I bloody well hope so.

Re:Seems Silly to me

[hakr89]

You are buying the drive to store base 2 numbers, so why shouldn't the value be rated in terms of base 2?

Re:SI units

[bh_doc]

Exactly. The information technology sector is and has always been wrong to suggest that k is 2^10. It is not, and it will never be. k=10^3, M=10^6, G=10^9, etc.

Much like a RAM settlement offer in the mail

[phantomlord]

I recently got a class action settlement in the mail offering money for memory that I overpaid for back in the early 2000s. The catch is, to receive anything, I need to provide detailed information about how much memory I bought from what merchant, the brand and how much I paid. To receive the hard drive settlement, they want the same info (serial number, proof of purchase, name of retailer, price paid, etc).

I have those receipts... somewhere. Who really keeps receipts for computer parts going back a couple generations though? As an individual, I doubt the money I would receive is worth the hassle of digging up the receipts. Sure, MegaCorp may have purchased 1,000 units and have the receipt of that order and will get a hefty sum at 7% for their trouble, but most people are just going to get a couple dollars.

I'm not sure why they don't offer a token minimum amount for those who can't provide receipts (I don't see all 300 million people in the US clamoring to get a $10 check). Of course, like most class action suits, this was probably just a way for a law firm to cash in on a settlement (they get a cool $1.8 million while you get some free backup software or a couple dollars).

Re:SI units

[DRobson]

Regardless of whether the IT sector is _technically_ in the wrong it's commonly accepted that in this area we work with powers of two. The fact that people have to explicitly explain this fact shows that everyone expects it to be that way. The HDD manufacturers damn well know this and fairly blantantly use measurements which would commonly be interpreted more favourably.

Ahh, another valueless settlement.

[Harik]

yeah pretty worthless, I've bought $1000 worth of drive from them, but that's after jan 1 2006. Even if if it was before that, I would have to file 10 seperate claims for ~$5 each. Meanwhile the cocksucking trial lawyers get a cool 1.8mn in cash.

Seriously - class action lawsuits are utterly worthless. "Whoops we ripped you off by conspiring to raise memory prices tenfold. Here's a 2 dollar coupon that expires the day we get around to mailing it out and is only good at a single retailer in northern alaska. "

Seriously - How many people here paid nearly a grand for 32 meg SIMMS? Remember the "welp we had a glue factory fire so prices skyrocketed!" bullshit? Special glue just for memory ICs - and that scaled exactly with capacity? Yeah, that "glue factory fire."

"Oh yeah our batteries in our ipods are horribly defective here everyone who spent $300 on this shitty self-destructing rev of hardware and can cough up documentation gets 2 free songs on our own music store."

I'd really prefer the courts just fine the fuck out of the companies and it goes to something worthwhile - letting them use legal judgements as cheap advertising is just bullshit.

Either way.. we will pay

[Maxo-Texas]

and some lawyer is going to be flushing the money on hot cars and girls or boys.

Here is your $5.99.

By the way.. did we mention our $5.99 price increase on our drives?

Re:Seems Silly to me

[DiSKiLLeR]

Seriously, the blame could just as easily be laid at the feet of the OS developers. There is a long standing history of disk manufacturers using base 10 counting numbers. It would not be so horribly difficult for the OS developers to conform to the base 10 measurement. I mean what next are the consumers going to sue because the formatting and allocation tables take up room? or perhaps because it hides space for virtual memory? seriously. come on people.

You're moronic.

Every operating system, whether it be Windows NT, XP, or Vista, Linux, FreeBSD, or Solaris, states that 1Kb = 1024bytes, 1Mb = 1024Kb, and so on.

Every application, does too.

Why rewrite all software, and god forbid, patch all old software going back however many DECADES into the past to implement this change, when harddrive manufacturers could simply start labelling their drives correctly?

Besides, when you buy a gigabyte of ram, are you really getting 1 billion bytes? or 1073741824 bytes? You tell me :)

Last I checked, bios reported 1024Mb was a 1gb, and 4096mb was 4gb's of ram :)

I don't see why hdd manufactureres are the ONE single exception to this long standing rule, and SI units be damned.

Re:Yet Again, the Courts Drop the Ball

[SaidinUnleashed]

Yes, but no one uses the *bi- prefixes, because they sound stupid, and make one sound stupid for trying to use them. The word "gigabyte" has meant 1,073,741,824 bytes in common usage for over thirty years. So, to steal an apparantly legitimate proof of factuality, the consensus among IT professionals is that a gigabyte is 1,073,741,824 bytes. If consensus among professionals in a field can make something a fact in any one field, it can make it a fact in every field.

Re:SI units

[DiSKiLLeR]

Regardless of whether the IT sector is _technically_ in the wrong it's commonly accepted that in this area we work with powers of two. The fact that people have to explicitly explain this fact shows that everyone expects it to be that way. The HDD manufacturers damn well know this and fairly blantantly use measurements which would commonly be interpreted more favourably.

Exactly.

This says it perfectly.

RAM manufacturers do it correctly, and Application Vendors and Operating System Vendors have been doing it this way for DECADES. SI units be damned, this is the way it has always been and there is no reason for it to be changed.

Re:WTF??

[I'm Don Giovanni]

" When referring to RAM sizes and file sizes, it traditionally has a binary definition, of 1024 bytes. For every other use, it means exactly 1000 bytes. In order to address this confusion, currently all relevant standards bodies promote the use of the term "gibibyte" for the binary definition."

Seems to me that since hard drives' primary function is storing files, that hard drive capacity should use the same unit of measurment that file size does, no? Doesn't that make simple sense? So if file sizes use 1024 rather than 1000, then hard drive capacity should as well.

It's not a longstanding history

[Solandri]

Seriously, the blame could just as easily be laid at the feet of the OS developers. There is a long standing history of disk manufacturers using base 10 counting numbers.

It's not a longstanding history. It started in mid-1990s. In the early 1990s, if you bought a 300 MB drive, you got 300*1024^2 = 314,572,800 bytes.

In the mid-1990s, one marketing dweeb at a low-end hard drive manufacturer (I want to say Maxtor but don't recall for sure) convinced his company to start defining 1 MB = 1,000,000 bytes. It let them sell a smaller (and thus cheaper to manufacture) drive while labeling it as the same capacity as everyone else's drives. The others resisted for about a year, then gave in and started mis-labeling their drives. IBM was the last holdout, I think they went for 3 years selling bigger drives than everyone else labeled with the same capacity. Eventually they gave in too, shortly before selling their hard drive division to Hitachi.

Around 1998, the international standards bodies mandated that MB = 1,000,000 and GB = 1,000,000,000, while MiB = 1,048,576 and GiB = 1,073,741,824. But like metric in the U.S., these units have never really caught on in the computer industry. Personally I can see the standards bodies' point, but they're going to have to collaborate with OS, memory, hard drive, and other computer hardware manufacturers to get the change implemented. They can't just stand on a pedestal mandating that this change be made, and expect it to happen.

The whole fiasco is an example of a class of situations I haven't found a name for but which is similar to the Tragedy of the Commons [wikipedia.org]. In these situations, one member of the group does something which gives him an advantage of the others. The others then follow suit to remain competitive, and in doing so eliminate the advantage. The end result is that the situation is now identical to what it was before the change (everyone's 500 GB drives are the same size), but now everybody is worse off because of the change (1 GB on a drive does not equal 1 GB in memory). Other situations within this class include campaign spending in politics (everyone has to spend more on advertising each year just to stay even with everyone else), and net neutrality (if everyone pays the Telecos more money for priority, they have gained nothing because the total bandwidth hasn't increased, and are now losers because they're paying more for the same bandwidth).

Pointless

[ewhac]

Hard drive makers have, for some considerable time, have meant 10**9 (1,000,000,000) when referring to a gigabyte. They always so declare in their literature. I have some old IBM Deskstar drives with exactly this clarification.

However, the various SI prefixes -- kilo, mega, giga, exa, and others -- were overloaded by the computer industry to refer to powers of two ("kilo" = 2**10, "mega" = 2**20, "giga" = 2**30) which were "pretty close" to their SI counterparts.

This has actually caused some confusion as computer people speaking of "kilo" this and "mega" that have worked with scientists who have always used the traditional SI meanings. These differences in interpretation can mean your chemical process doesn't work, the patient dies, you miss Jupiter, etc.

To help redress this problem, a new set of prefixes [nist.gov] have been coined to refer to powers of two. These new prefixes have seen uneven but increasing adoption in the industry (if you have a recent Ubuntu/Debian release, run the command ifconfig -- the byte counts have the new prefixes).

So, the hard drive makers have been using the SI meanings for "giga" and, in case there was any confusion, explicitly printed in their literature, "One gigabyte is equal to 1000000000 bytes."

So, at first reaction, I think Seagate got screwed here. This makes me wonder if there aren't other layers of nuance that came out in court, but are lost in these stories.

Schwab

What a crock

[SurturZ]

What a crock. Anyone that knows enough about computers to know that GB, MB, and KB are usually base-2 should also know enough to check whether the HDD measurement is in base-2 or base-10. Non-computer people would probably assume that they are base-10... or, more likely, merely that the bigger the number, the better. In my experience non-computer people have difficulty distinguishing between hard-drive space and RAM. Saying that they are somehow miraculously able to distinguish between base-2 and base-10 measurements is ridiculous.

The Kilo-, Mega- and Giga- prefixes are always base-10 in SI. The IT industry should come up with different terms. Misusing them was a mistake in the '60s and it is a mistake now.

Re:Think this will set precedent?

[hedwards]

I bloody well hope so.

Why? What they advertised was correct, the hard disks do have the number of gigabytes that they claim to, 8,000,000,000 bits, or 1,000,000,000 bytes. The fact that the OS uses base 2 as the numbering system doesn't negate the fact that there are 8,000,000,000 bits on the hard disk per gigabyte. It isn't the responsibility of the manufacturer to know in all cases how much of that is going to be usable or how it is going to be notated.

It is just a nominal difference, anybody ignorant enough not to understand that, shouldn't be purchasing hard disks separately from a computer. Better yet, keep them from having a computer, I get enough spam and virus exposure as it is.

It doesn't really matter what the units are called, as long as they are standard amongst the industry. In this case, they have been standard since before I began using computers 20 years ago.

Are you kidding me?

[joe_n_bloe]

I understand that a "gigabyte" of RAM is 2^30 bytes, but that's just because memory addresses come in powers of two. I don't expect bytes on a hard disk to be counted in powers of two, because there is no need for them to be counted that way. But apparently there are some bargain-hunters and their lawyers who have a more self-serving style of counting.

Oh well.

Re:SI units

[IWannaBeAnAC]

While I can see the technical merit in using the Ki/Mi/Gi prefix instead of K/M/G, I object to it for the simple reason that kibibyte, mibibyte and gibibyte are stupid sounding words and I refuse to use them for that reason alone.

It might be, for a newcomer, initially confusing that a kilobyte is 1024 bytes instead of 1000 bytes, but the original scheme is a consistent exception. The powers of 2 apply to bytes and only bytes, nothing else. 1Km = 1000 meters. 1KW = 1000 Watts. 1KB = 1024 bytes. 1 KN = 1000 Newtons. Not completely uniform, but there is no ambiguity.

On the other hand, if someone came up with a set of power of 2 prefixes that didn't suck, I'd happily switch.

Re:SI units

[dgatwood]

The problem is that a 2^k organization of bytes is fundamental to the way computers operate. It can't be changed to a power-of-ten unit just because it is "more convenient to work with" as the SI folks want. You can't realistically design a RAM chip with 1000 bytes of memory. You could do it, but you'd end up building one with 1024 bytes of memory and just burning out the last 24 cells. Ditto for all other forms of electronic storage, including the caches on hard drives. Only magnetic and optical storage have the luxury of defining units in non-power-of-2 ways, and yet they generally do not, choosing to standardize on 512-byte blocks primarily because if they didn't, the VM system's paging path would be heinously inefficient.

So we have a choice: we can either standardize on one unit---the base-2 definition of a gigabyte---or we can standardize on two units---one for RAM and one for hard drives---or we can foolishly standardize on the base-10 definition and have RAM chips described as 1.074 GB. I, for one, can't imagine that last choice being too popular, and the second choice (the status quo) is sufficiently confusing to an average layman that it really doesn't work, either. Thus, the only -reasonable- choice is to standardize on base-2 definitions of these units. There's a reason the standards were bent a bit fifty years ago. The SI units just don't work. They can't work. They will never work. And the sooner we stop trying to force a base-10 unit of measurement into a base-2 world---the sooner we can dispose of this fundamentally flawed view that everything must be in base 10---the sooner we can resume actually getting things done instead of quibbling over crap like this that was set in stone before most folks on Slashdot were even born.

Put another way, it's 9 years later, and the term kibibyte is still almost universally guaranteed to get you modded "troll" in any computing forum. Maybe it's type for the SI folks to realize that perhaps the reason their standard has been near-universally rejected in computing circles for almost a decayear is that it is fundamentally brain damaged from a practical use perspective.... It makes about as much sense if the SI had standardized base-10 units of time other than the second. Kiloweeks, anyone? Decidays? The SI folks wisely realized that moving time to a base 10 unit was not practical because the natural division of days into years could never be forced into base-10 units comfortably. Instead, they acknowledge the usefulness of these non-SI units as acceptable for use in spite of their non-base-10 nature. The same is true for computing, and they would be wise to acknowledge that the same fundamental problems hold true in this area.

Re:Seems Silly to me

[nocomment]

Because computers are base2.

It might not sound like a big deal, but as HD's get bigger so does seagates 'edge' over the competition. They get to trim 73MB (or so) off every gig. This means that a 250GB drive from seagate is missing 18,435,456,000 bytes. A 500GB drive: 36,870,912,000. In the olden days, this wouldn't have mattered (much) because you weren't talking about a lot of space. People complained back then too. Now it's getting a little silly. If you need to build a 5TB array, there will be 368GB that's just missing (and that's not even counting the FS overhead).

Seagate isn't doing it to be a champion of change for a switch to base10 counting (if they were then it would make more sense), they are doing it to rip people off on a technicality.

Re:SI units

[bh_doc]

While much of what you say is insightful...

Put another way, it's 9 years later, and the term kibibyte is still almost universally guaranteed to get you modded "troll" in any computing forum.

...I must say I find the attitude exhibited by the profession against simply using a slightly different moniker to avoid any ambiguity with an already established metric disturbingly egoistic.

Go ahead. Don't use base 10 for measuring RAM sizes, use base 2, I really don't care. Just don't go calling it a Gigabyte, because it isn't. I would have no problem picking up a 2 GiB stick of RAM. If you'd prefer not to call it a Gibibyte, either, fine, call it something else. Just please stop calling it what it isn't. For an industry that regularly has to deal with and resolve ambiguities, it's surprising to me how inert it seems to have been on this one.

Re:SI units

[amorsen]

It makes about as much sense if the SI had standardized base-10 units of time other than the second.

No, the comparison is if everyone decided to call weeks dekadays, but keep their length as 7 days. It's simply wrong. If you want to use the SI units, use the SI definition. Otherwise come up with your own terms.

Re:SI units

[amorsen]

The powers of 2 apply to bytes and only bytes, nothing else

They apply to bytes, when you happen to talk about RAM. Anything else, even flash, is in powers of 10. Sometimes, but rarely, they apply to bits as well -- 2Mbps E1 is 2048kbps. ADSL can go either way. In short, "consistent" is certainly not a good description of this mess.

Concerning this Gibi nonsense

[PaladinAlpha]

Ok, there's people everywhere in here saying it's stupid to say 1GB = 1024 MB, instead use GiB, blah blah, but I have an honest question: if everyone did, for whatever reason, use GiB and MiB and whatnot, of what use would MB and GB be? None, right? No one in their right mind would ever measure units on binary hardware in powers of ten. Just like I don't measure grams of soup or whatever in powers of two. I guess what I'm saying is if the switch to GiB and MiB was made in earnest then GB and MB would be utterly, completely useless -- and I see that as an argument that we might as well just use GB and MB, since it involves no conflict, fewer meaningless terms, and a more intuitive, uh, interface, if you will.

What?

[msimm]

If a megabyte is counted as 1024 kilobytes how's your math working? Still 8 bits to a byte right? I mean I thought a byte was a byte, are you telling me a hard disk follows different conventions? Because last time I checked binary units [nist.gov] were pretty stable, not a lot of 'wiggle room' in the interpretation.

Re:RTFM

[Mycroft_VIII]

While he was a bit harsh, the 'gibi' 'mibi' 'mabi' and what not are NOT accepted units of measure irregardless of what any organization (even a 'standards' organization)says.
    The key word word is accepted.
Computers work in base 2 natively and when the field was first started and the closest prefixes for base 2 'round' numbers in that base were adopted by the vast majority.
      Now this was and is a bit of a kludge, but it's was nearly universal in use (and still dominant) and anyone who was serious about learning computers learned this fairly early on.
    This same numbering scheme continued well into the era of commodity parts including retail hard drives until some nitwit realized they could make more money by selling according to the base 10 numbering system definitions which are smaller.
      The real truth here is that what the hdd makers did was attempt to deliberately create a false impression of size by relying on the fact the for computers mega meant 2^20 and yet changing what they meant by it (after years of using the de-facto standard).

Mycroft

Re:Think this will set precedent?

[Eivind]

The difference does however grow as capacity grows.

The difference between 2^10 bytes and 10^3 bytes is 2.4%. (kilobyte)

The difference between 2^20 bytes and 10^6 bytes is 4.9% (megabytes)

The difference between 2^30 bytes and 10^9 bytes is 7.4% (gigabytes)

The difference between 2^40 bytes and 10^12 bytes is 10% (terabytes)

In other words, treating 10^(3n) as equivalent to 2^(10n) makes less and less sense as the capacities go up.

Re:Seems Silly to me

[darthflo]

Technically, you're right. Most "advanced" systems ("advanced" as in "catering a crowd that understands the difference between GB/GiB") are actually doing just that right now. Try `df --si`. Most commonly used other tools also include a similar if not the exact same switch.
On the other hand: If Joe sixpack buys a 200 Gigabyte hard drive, he'll expect Windows Explorer to show 200 Gigabytes of space. Vista will most probably eat about 50 of 'em, but that doesn't concern him, as long as there's 200 Gigabyte of total space. What also won't concern him is if that unit right behind the number says GB or GiB. He doesn't commonly use the mega or even giga prefix (if he's american he might even still be used to using medieval terms like feet and miles (no prefixes for those, of course)), 'cause commonly thousands of kilometers aren't expressed as megameters. He'll ignore the i and be as annoyed as before, cause they still took his bytes.

Re:Ahh, another valueless settlement.

[Atario]

*SIGH*

Ok, let's cover this one more time. Class action lawsuits are only sometimes intended to give a substantial settlement to all members of the class. The real point of them is not to get you rich, but to take down a wrongdoing company a few notches so that, with any luck, they'll know better next time — or at least think twice.

Re:Misleading by being correct?

[darthflo]

Actually, to a scientist, '1K' would be very fucking cold (to the point where the nerves probably wouldn't be quick enough to even transmit the coldness info to his brain). '1k' is a thousand. :)

Re:SI units

[jimicus]

For an industry that regularly has to deal with and resolve ambiguities, it's surprising to me how inert it seems to have been on this one.

There wasn't an ambiguity before hard disk manufacturers decided to invent one.

Re:SI units

[The Clockwork Troll]

In chemistry, it's very common to see heat capacity [wikipedia.org] expressed in terms of kilojoules per Kelvin (kJ / K).

I'd agree with your latter statement, but as long as I'm trolling, I'll point out that it was not technically the logical converse [wikipedia.org] of your former statement but rather a wholly different proposition (1: "I have never seen 'k' and 'K' together in a single unit." vs. 2: "I have never seen an ambiguous 'k' or 'K'").

In Soviet Russia, it fails you!

Re:wow.... are you clueless!

[kongit]

Well no, they do deserve this. The actual capacity of the hard drives is not what is specified; unlike in your examples where a kilogram of sugar is a kilogram of sugar: what Seagate (and other companies) say is 1GB is not 1GB. While I like Seagate hard drives, I am following this for the software -- which I will never use -- because I think all digital storage manufacturers should use the correct measurement. The only reason I can think of the for the current manufacturer's measurement is retail. Their current method makes hard drives appear to have more space then they really do.

If we ignore this and Seagate has to do hardly anything, all the other manufacturers will see that people don't care that they are getting cheated and will continue selling their products with false information.

Re:SI units

[bm_luethke]

You are not going to win this argument - the sides have been drawn and you bring no new information to the table. All of us have already made up our minds.

You have the two sides that care and then the rest of us that do not. It's not that I do no understand it - I've been an academic weenie for quite some time and find these things highly entertaining, yet I just don't care when it comes to my money. If I had to choose I prefer the base 2 definition as it makes the most sense, yet as long as I can compare two devices I do not care of they use "miblywinks" to rate their storage size - I just want something I know what they mean and standard across devices.

In the end what file table I decide to use has more impact as to the usable size of my hard drive than if they use base 2 or base 10. That is WAY more confusing and "dishonest" than the esoteric idea of which standard you use. I find that I spend more time explaining why their 105gb (converted to base 2) device shows much less than that (also in base 2) when they click on it to non technical people. But then, do we then standardize on NTFS, FAT16, FAT32, EXT3, REISERFS, or ExoMibFS? Again, as long as I can compare apples to apples I could care less, just lets standardize on one.

It's not a big deal to anyone who has half an idea what is being discussed, but most people do not know it. That isn't making fun of anyone - that is what those of us in IT are payed for. Lets face it, how many of us know what the building set backs for Residential One zones are - that is what Land Surveyors are for and there are going to be more of us worried if our house meets zoning laws than if our hard drives capacity is based on base 2 or base 10. Yet how many that feel that average people should know this know their local zoning laws (or land laws) to the extent we expect people to know computer technology (and there is WAY more money involved with land)? Unfortunately to many do no trust their IT professionals like they do the professionals they hire in other areas - but that is a different complaint.

Re:RTFM

[Kickasso]

"irregardless" is not standard either.

Re:SI units

[trentblase]

But the base 2 units have been used for decades. Calling that use anarchy is similar to saying that we shouldn't use "bug" to describe a software malfunction simply because it has another meaning. Context will always tell you the answer. In the computer context, kilo is 1024.

Re:wow.... are you clueless!

[pipatron]

Apparently you didn't understand what we are talking about here, and what the parent argued.

The SI-prefix G has always been 1,000,000,000. When you buy a hard drive with one gigabyte of storage capacity, it will always have a little bit more than this, due to cylinders/sectors/platters rounding. When they sell a hard drive with 500 gigabyte, it will have slightly more than 500,000,000,000 bytes. No one is trying to fool anyone here.

Now, if you go to the store and want to buy one gigagram of sugar, you expect to get no less than 1,000,000,000 grams. Anything else would be cheating. But when you go to the store and want to buy one gigabyte of storage, you suddenly expect to get a lot more?

Re:SI units

[The_Noid]

Mega is a Prefix to Byte.
This whole discussion is about the use of the prefixes Mega and Giga as a multiplier other then the 1000 they have been defined as.

Re:Seems Silly to me

[bestinshow]

The fact that the storage of data on a floppy disc (and hard drives) is segregated into blocks that are sized in powers of two (512 bytes for most floppies) suggests that the capacity of said drives should be expressed as a number that is a multiple of said block size.

E.g., Floppy Drive: 2 sides * 80 tracks * 9 sectors * 512 bytes = 720KB

Any other argument is totally pointless to be totally fair. If you want a base 10 capacity for binary data that is not stored in blocks of base 2 size you use bits, e.g., 20Mbit. For example TCP/IP packets can be any size, so using bits is common for networking capacities/bandwidths.

I appreciate that there are SI units, but they are not in common use apart from total SI/standards nerds, and their creation was misguided and based upon a core misunderstanding of the issue.

I hope this means that hard drives, flash memory, and so on, are now sold with the true capacity on the label.

1GB is really 1,000,000,000 bytes

[OrangeTide]

They were using the correct measurement. Why don't you sue Intel and 3COM because your 100Mbps ethernet card is actually 100,000,000 million bits per second.

The problem is some OS vendors, like Microsoft, incorrectly report the drive size using a strange base-1024 system. While this system might make sense for RAM which due to technical reasons must be a power of 2. (due to binary encoding for the addressing and inability to support "gaps")

Also it's not false or misleading if everyone knows what is being done. And did you overlook that * on the box that says "* 1GB is 1,000,000,000 bytes."

Re:Direct Link to claims

[tepples]

"750,000,000,000 byte hard drive for sale" - is that so terribly hard?

That or "750 GB (698 GiB) hard drive for sale".

Re:SI units

[Anonymous Coward]

Go ahead. Don't use base 10 for measuring RAM sizes, use base 2, I really don't care. Just don't go calling it a Gigabyte, because it isn't.

Except, it is a gigabyte.

Gigabyte (and megabyte and kilobyte and all the other -bytes) are not SI measurements, although they share some similarities with them. The prefixes "kilo" and "mega" (and the rest) are used by SI measurements, but they aren't unique to them; their origins date back to Greek and the computer industry has as much right to co-opt them as SI did.

The computer industry decided from its infancy that its measurements would be in powers of two. It decided to use prefixes similar to those used by SI measurements for each order of magnitude; hence kilobyte, megabyte, etc.

Arguing that a gigabyte isn't 1,073,741,824 bytes because it sounds similar to SI measurements is like arguing a foot isn't a unit of measurement to 0.3048 meters because it also happens to refer to a human appendage. While you could argue against the logic of this sort of measurement, it is the proper term; it's just not SI.

Re:Ahh, another valueless settlement.

[R2.0]

Wrong. Class action suits are largely based in common law, not criminal law. Criminal law deals with punishment, common law with equity. At least, that's how it was...

Class actions started as a matter for convenience for the courts - if there are a lot of people with the same basic claim, just litigate once instead of many times. Saves costs/more efficient.

Punitive damages were supposed to be a secondary function of the court, where particularly egregious behavior was punished.

Then came Congress, and the Trial Lawyer Association.

Congress at some point decided that, instead of making certain activities subject to criminal law, they would be subject to civil law. So if some corporation does something bad, the FBI doesn't bust ass - someone sues. And to make sure that there was punishment meted out, we got statutory punitive damages, mandatory, and set at 3x the judgement. Lawyers loved this - they get to make a shitload of money doing the Fed's dirty work.

At the same time, mega law firms developed. In order to increase revenue, they becan the rampant campaign of class action suits we see today, which are specifically designed NOT to punish, or to give their clients redress - they simply occurr to make the law firms more money. Witness the silicosis debacle, and the "tobacco settlement", where select, politically connected law firms made profits orders of magnitude greater than their cost (I'm looking at you, Angelos).

Class actions have been abused into a legal sham and a public laughing stock. as an instrument of equity, they are useless, and that's because a bunch of large law firms want it that way.

Re:1GB is really 1,000,000,000 bytes

[Spleen]

In the old days we learned:
1 Byte = 8 bits [google.com]
1 kilobyte = 1024 Bytes [google.com]
1 megabyte = 1024 kilobytes [google.com]
1 gigabyte = 1024 megabytes [google.com]
1 terabyte = 1024 gigabytes [google.com]
Which means that: 1 gigabyte = 1073741824 bytes [google.com] This isn't a strange "Microsoft" scale. If you've ever watched your memory count on boot it uses the same scale. Harddrive manufacturers for as long as I can remember have not used the correct scale, but the hardware and OS's do. They've also had a notice on the packaging that states the scale they measure by, and probably on their website too. I never liked that they didn't follow the correct scale, but its a stupid lawsuit.

Yes there was.

[camperdave]

There wasn't an ambiguity before hard disk manufacturers decided to invent one.

Yes, there was an ambiguity, and it started with marketing people. Back in the Early Eighties, in the days when music didn't suck, when the 6502 microprocessor ruled supreme in the personal computer arena, that this trouble started. Prior to this time, it was universally accepted that in the computer world K=1024. The 6502 microprocessor, found in the Apple II, the Commodore PET, the Atari 400/800, and a host of other machines, had a 16 bit address bus which means it was capable of addressing 64K of memory. In fact, that's the basis of the name Commodore 64.

Now somewhere along the line, some marketing bonehead read that 64K meant 65536 bytes, and thought "If we use K=1000, like the science nerds instead of K=1024 like the computer geeks, then we can say our machines have 65K instead of 64K. The sheeple buying these widgets will buy ours instead". And it worked. Sheeple bought the more expensive Commodore64 that had "over 65K" instead of the cheaper Commodore64 with only 64K, not knowing that they were being fleeced. Soon everyone was marketting their 64K machines as having 65K. These machines were rarely sold with hard drives. And when they were, the K from the manufacturer was 1024 bytes, but the K from the marketer was 1000.

It got worse when the IBM, Amiga, Mac, and the Atari ST lines hit the market. These machines could deal in Megabytes. The field was muddied by the folks saying a megabyte was 1000K, instead of 1024K, and further muddied by the crowd saying that 1K was 1000 instead of 1024. So, the megabyte came in three sizes, 1,000,000 bytes, 1,024,000 bytes, and it's true size, 1,048,576 bytes.

Re:wow.... are you clueless!

[tilandal]

A Byte is not an accepted SI unit of measurement therefore there is not a reasonable expectation that a gigabyte be 1 billion bytes. Since a byte is 2^3 bits to begin with its is even more unacceptable to think of as a base 10 operation. If seagate was advertising 8 Gigabits instead of 1 Gigabyte that would be more acceptable. 1 Gigabyte has always been accepted to mean 2^30 Bytes. 1 gigabit has always been 10^9 bits. They could have advertised as 1 Billion Bytes as well. You can not change the accepted notation just because it suits you. A mile is 5280 feet if you are in a car but 1852 meters if you are in a boat. Context matters.

Re:1GB is really 1,000,000,000 bytes

[tilandal]

A bit is not an SI unit. A Byte is not an SI unit. A GB and a Gb are not accepted SI notation. Each has its own accepted context. A bit is a base 10 unit a byte is a base 2 unit. If you want to report your storage space in bytes it is accepted that you are reporting in base 2. If you would rather report in base 10 thats fine. You can advertise 8Gigabits, You can advertise 1 Billion Bytes, but you can not say that a GB is anything other then 2^30 bytes.

Re:RTFM

[Mattintosh]

What's hilarious here is that most, if not all, of the supporters of SI+RetardedBi prefixes are pointing to what the US Government says about the "standard". Things to think about here include (but are not limited to):

1) The US Government is in the pocket of any company willing to pay up, including HDD manufacturers that ponied up a bit-o'-cash to get this "standardized".
2) This is made more hilarious by the fact that the US Government and its citizenry don't use SI units at all. Oh, except for 2-liter soda bottles, which are, surprisingly, about 2 fl. oz. more than a 2-quart bottle would be.

Here's my suggestion: When talking about non-byte-based values (liters, grams, rods, hogsheads, LoC's, AU's, VW Beetles, etc.), understand the prefixes to be multiples of 1000. When talking about byte-based values (bits, bytes, words, nibbles, long words, etc.), understand that the context means multiples of 1024. The human brain is a wonderful thing, and languages are too. Both are very complex and context-sensitive. Use your brain to understand context. It's not that hard.

Re:Think this will set precedent?

[Selivanow]

It might not be "new" but it wasn't always this way. I do remember drives that were labeled in the "correct" binary values. Of course it has been a while, the last drive I can absolutely remember is my WD 525MB drive (being 525.7MB or something). Manufacturers are just trying to get something (our money) for nothing (missing drive capacity) and to make themselves look better and more productive. Consumers also have something to do with this. We are never happy with the latest and greatest. We will always demand more capacity, speed, etc.

Re:Think this will set precedent?

[tiny1877]

Regardless of what FS you use, you will NOT lose 80GB just by formatting it. If I'm buying a 1TB drive (1024GB!!), it should not report in ANY OS any less than 1023GB allowing for all FS data, and that's being generous. On the flip side, food products are allowed a 20% margin in weights. Those granola bars you buy that say they're 2oz? They can range from 1.6oz to 2.4oz and it's perfectly allowable. Where's the outrage there?

Re:SI units

[darthflo]

The problem with storage examples is the limitation they have - basically, there's only three kinds of storage: Hard drives, internal electronic (as in Flash, RAM etc) and removable devices. But since you pointed out that all storage is computed in base 2 by referring to the second kind, I found myself forced to check the capacities of some removable media: (Mkt. C stands for Marketed capacity, Real C for the actual Capacity)

Product| Mkt. C | Real C .| Result
Name
DVD . .| 4.7 GB | 4.3 GiB | bse 10
CD . .| 700 MB | 703 MiB | base 2
BluRay | 50. GB | 46. GiB | bse 10
HD-DVD | 51. GB | 47. GiB | bse 10
DLT . .| **. GB | **. GiB | bse 10 -- see explanation below

I couldn't confirm the DLT claim, but according to a Quantum manual [quantum.com], DLT seems to be based on base-10 as well. Ultrium and co will (probably) be similar.

Four out of five samples of removable storage media seem to be using base-10 for their claims which kinda positions two of three storage types in the base-10 camp which could be interpreted as RAM being the evil, non-uniform group. So let's ask them to add the tiny little "i", eh?