Creative Sued for Base-10 Capacities On HDD MP3 Players

Dorkz brings news of a class-action settlement from Creative Labs over the capacity of their HDD MP3 players. Evidently they calculated drive capacity in base-10 (1,000,000,000 bytes per GB) instead of base-2 (1,073,741,824 bytes per GB). The representative plaintiff is entitled to $5,000, and everyone else who bought one of the HDD MP3 players in the past several years gets a 50% discount on a new 1GB player[PDF]. They can also opt for a 20% discount on anything ordered from Creative's online store. Creative has made available all of the necessary legal forms. Seagate lost a similar lawsuit late last year.

Comments

50%?

[Jizzbug]

50% off for less than 10% less space?

Re:50%?

[wizardforce]

they misrepresented the capacity of their products knowingly, this is a warning to any other company stupid enough to lie like this.

Re:50%?

[mrbluze]

...they misrepresented the capacity... I'm sure they thought they were just being creative.

Re:50%?

[iamacat]

According to the most accepted definition K == 10^3, M = 10^6, G = 10^9, T = 10^12. Why should consumer product manufacturers use definitions only understood by technical professionals, that will confuse their average customer and are unflattering to the product?

Re:50%?

[Anonymous Coward]

Because it is part of the trade, and if you don't understand the rules and definitions in the trade tough shit, you should learn them before getting involved.

Nobody is suing lumber manufacturers because 2x4s aren't 2 inches by 4 inches. Everyone in the trade understands the real dimensions. If you want to get involved in construction you have to learn things like that.

Re:50%?

[aleph42]

Because when your OS displays the empty space on your device, it uses powers of 2.

You don't have to be a "technical professional" if you OSãtranslates for you.

Re:50%?

[d0mokun]

they misrepresented the capacity of their products knowingly, this is a warning to any other company stupid enough to lie like this. Stupid enough to lie? Don't jump the gun or anything! What if they actually did make a mistake? Hey.. even NASA made mistakes with thier calculations.. It can happen..

Re:50%?

[Anonymous Coward]

They should use those definitions because they are definitions. A kilobyte is not 10^3 bytes, so they should not say it is. For memory sizes, K = 2^10, M = 2^20, G = 2^30, T = 2^40. If the manufacturer actually put 2^30 bytes in a 1 GB product, the packaging of that product could remain the same, the average customer would not be any more confused, geeks would be satisfied, and no body would go wtf? when they plugged in their drive and it says it is 954 MB.

Re:50%?

[Grant_Watson]

Don't jump the gun or anything! What if they actually did make a mistake?

Drive makers have been doing this for a long time-- it's not an accident.

Heck, the drivemaker's kilobyte shrinks four bytes each year for marketing reasons. :-P

Re:50%?

[__aaclcg7560]

If they wanted to be creative, they should've used octal base.

Re:50%?

[zippthorne]

A kilobyte is 10^3 bytes. You are thinking of a Kibibyte (kilo_binary-byte) where kilo_binary takes advantage of the fact that 2^10 is very close to 10^3. Unfortunately it gets worse for mega, giga, etc, which is why SI finally ruled on the standards.

I believe it's customary to include a relevant xkcd

Re:50%?

[beav007]

Yes, but is it a 1GB player in Base2 or Base10?

Re:50%?

[uvajed_ekil]

I'll pass. The prescribed relief for having bought crap is a discount on more of the same company's crap? No thanks, I'll just buy from somebody else next time.

Re:50%?

[PAjamian]

What kind of "warning" is that? That if you misrepresent your product you might have to sell another product to affected consumers at around the wholesale price (where the manufacturer still makes a profit)? And that people might flock to your online store to buy loads of stuff because they think they're getting a great deal with a 20% discount? The only real fine here is $5000, pocket change.

Re:50%?

[BigAssRat]

No, my computer has 3.25GB of RAM because I installed two 2GB sticks. I should be able to get one hell of a discount from MS then heh?

Re:50%?

[zippthorne]

I was under the impression that 2x4s are, in fact, actually 2in by 4in when cut wet, but shrink to the standard size when seasoned. Or, at least, did at some point in their history of manufacture.

Re:50%?

[calebt3]

Nah, you just need to fork over some more cash for the 64-bit version.

Re:50%?

[iamacat]

So should they also take your filesystem overhead into account to somehow come out with 2^30 bytes of formatted capacity? If not, how do you know it is not 2^30 bytes internally, including 70MB for remapping bad sectors?

Re:50%?

[i.of.the.storm]

Yeah, 32 bit OSes can't see the full 4GB due to the way memory addressing works. Check wikipedia or something for more info.

Re:50%?

[MadnessASAP]

32 bit OSs CAN see 4 GB but anything beyond that your out of luck with the exception of a few VERY ugly hacks that came out shortly between the need for more the 4 GB and 64 bit processors.

Re:50%?

[LilGuy]

If they truly made a mistake I would wonder if the mp3 player would really function at all then. I would think it would constantly crash trying to access memory it didn't have.

Re:50%?

[Psychotria]

Which definition of GB? The 32-bit OSs can only address 4GB because it's 2^32 (i.e. 4294967296 bytes)...

Re:50%?

[pembo13]

Does your mechanic do similar?

Re:50%?

[Anonymous Coward]

Or http://xkcd.com/394/

Re:50%?

[mrsteveman1]

mmm seasoned 2x4

Re:50%?

[i.of.the.storm]

No, 32 bit OSes can address 4GB TOTAL memory, that's including graphics RAM and various caches. Thus, the total RAM addressable by a 32 bit OS is somewhere around 3-3.5GB depending on the configuration of the computer.
This forum post explains it in greater detail, people were asking this so often that they eventually just stickied the post. http://www.maximumpc.com/forums/viewtopic.php?t=71236

Re:50%?

[MadnessASAP]

Well thank you very much for correcting me I did not no that till today.

Re:50%?

[AuMatar]

No, he means kilobyte which is 2^10 bytes. Kibibytes are made up words which have no meaning.

Re:50%?

[Graff]

I was under the impression that 2x4s are, in fact, actually 2in by 4in when cut wet, but shrink to the standard size when seasoned. Partially true. A 2x4 is rough cut to 2" by 4", then is dried (seasoned) and planed (smoothed), reducing its dimensions to approximately 1.5" by 3.5". The shrinkage due to drying and the material removal due to planing is what reduces the boards to their current commonly-found dimensions.

Re:50%?

[i.of.the.storm]

No problem, it's a common misconception. It's just that it wasn't very common until recently, and even now most people don't run into it. But you can see it in the RAM configurations in many computers these days; many companies are putting 3GB RAM into their machines instead of using 4GB and installing a 64 bit OS, because although there aren't many incompatibilities, when something does happen the consumer is going to blame either the PC maker or Microsoft.

(posted from a Windows Vista x64 machine)

Re:50%?

[thefekete]

Electrons go from negative to positive, but since some jackass got it wrong at the beginning we affectionately refer to current as "the flow of positive charge". In fact, there are better reasons for base 2 kilobytes than the current direction convention.

In EE 101 you learn that this is how it's done and to quit bitching about it. Same applies here, since the beginning kilobyte has meant 1024 bytes, mega - 1024^2, and so on. However, for quite some time, hard drive manufacturers have sold hard drives in base ten.

If you don't know enough about this industry's sales gimmicks and conventions, you probably aren't going to miss those few extra bytes (until some properties window tells you about it).

Re:50%?

[wolferz]

such as seagate, western digital, samsung, toshiba, ibm... yeh every single HDD manufacturer there is with no exceptions.

Re:50%?

[Daniel Phillips]

I was under the impression that 2x4s are, in fact, actually 2in by 4in when cut wet, but shrink to the standard size when seasoned. 2x4 when cut rough, before planing, actually. Or it is supposed to be. Lumber companies argue that planing + shrinkage takes 1/4 inch off each side, resulting in 1 1/2 x 3 1/2, a 35% reduction in volume. I have a trouble believing this, personally I believe this is more about saving money by scamming consumers by getting the official standard changed, who knows how.

Re:50%?

[phantomfive]

In the old days, a 2x4 was actually 2 inches by 4 inches. However, if you cut the boards smaller, you can get more usable wood from a given tree. So, they shrunk the size of the boards and changed building codes to make studs closer together in walls. If you look in an old house, you can still see 2x4s that are actually 2x4. Or so the old salt I framed houses with one summer told me.

Re:50%?

[FishWithAHammer]

You are incorrect. All Intel processors after the Pentium Pro except for a few Pentium M chips, as well as most Athlons and later AMD processors, support PAE, which can address up to 64GB of RAM. There are 36 address lines on all PAE-capable memory buses; 4 are just unused by Windows except for use with the XD bit.

Wiki article

Re:50%?

[FishWithAHammer]

If you're actually seeing bad sectors, your disk is on the way out anyway.

Re:50%?

[i.of.the.storm]

Yeah, but most versions of Windows don't use PAE. In fact, you can even try to enable the boot-time switch but it won't do anything. Also, from what I've heard the reason it's disabled is because it's buggy and not really worth it. Out of curiosity, do you actually use PAE or are you just being pedantic? I don't really know much about how Linux does it, but in any case it's not use much and anyone who might need PAE might as well use a 64 bit OS.

Re:50%?

[FishWithAHammer]

Yeah, but most versions of Windows don't use PAE. Irrelevant. When you have to say "yeah, but", you're wrong. You were incorrect.

In fact, you can even try to enable the boot-time switch but it won't do anything. The boot-time switch isn't needed in the version of Windows I run. PAE is enabled by default on versions that aren't crippled.

Also, from what I've heard the reason it's disabled is because it's buggy and not really worth it. It is buggy because driver writers are generally ignorant of the capabilities of the platform on which they code. Microsoft made no effort to rectify the situation because anything that drives hardware sales helps them out, too, and shoveling people to 64-bit before it was ready (XP x64, I'm looking at you) didn't hurt them one bit.

Out of curiosity, do you actually use PAE or are you just being pedantic? I use it. I've used it for quite some time, before a viable 64-bit Windows existed. One of my older Linux machines also exceeded 4GB of RAM.

I don't really know much about how Linux does it, but in any case it's not use much and anyone who might need PAE might as well use a 64 bit OS. Unless they don't feel a need to upgrade hardware.

Re:50%?

[i.of.the.storm]

Ah, I see. My main sources of information about PAE are forums and this article: http://www.dansdata.com/askdan00015.htm which basically says that PAE was disabled from doing anything in Windows XP SP2. I'm assuming you use a server version of Windows or something then? That's pretty interesting. XP x64 was pretty stupid, but Vista x64 is just as good (or bad, depending on your opinion) as Vista x86. Driver support is definitely better than it was for XP Pro x64, since driver makers have to make both versions of a Vista driver for it to be certified or whatever Microsoft does to approve drivers.
I still don't understand why anything pre 64 bit would require over 4GB of RAM, but then I don't have much experience with servers or anything of that sort.

Re:50%?

[Amouth]

most mother boards get screwy when you get above 3.5GB of ram installed. like the gp said it is total.. base+extended+video.

something to try is IF your board supports it is PAE.. it allows the OS to address all the physical memory (up to the 4GB and beyond i belive) but it does some screw things in how it does it..

been so long that i can't remember exactly.. i just use it and move on - but feel free to read up on PAE.. i remember it being intresting.. but then again i have been up for 30 hours so intresting could be completely wrong

Re:50%?

[English French Man]

That's hardly the point here...

The point is even unformatted à 100 GB drive only contains 100,000,000,000 bytes at most, where most systems calculates disk size in powers of 2, where 100 GB is roughly equal to 93.1 billions of bytes, which is quite different than space used for the allocation table and all that stuff...

Re:50%?

[hvm2hvm]

And according to what everyone thinks when you talk about memory sizes, KB=1024 Bytes, MB=1024 KiloBytes, GB=1024 MegaBytes, etc. The companies that create harddisks and other storage devices know this and take advantage of it. Actually I'm impressed that someone sued them and actually won. I thought it was a lost cause but apparently it isn't. Still, I haven't seen a HDD offer that expresses the size in 'real' GigaBytes.

Re:50%?

[penguin king]

Isn't the Angstrom commonly used in biology/chemistry (and biochemistry too)?

One could say that not all scientists really follow SI, because Angstrom isn't SI. It's be far more SI to say 0.1nm.

Adaptation isn't really necessary, those in the business know what they're talking about. It's the people who interface with general public who need to be explicit about what they mean. Just like a scientist might commonly say that something measures 10 Angstrom and specify (or 0.000000001m) for the general public why don't they just say this product has 80Gb capacity*


*85 899 345 920 bytes

It's all abount communication (too much communication).

Re:50%?

[MichaelSmith]

I was under the impression that 2x4s are, in fact, actually 2in by 4in when cut wet, but shrink to the standard size when seasoned. Or, at least, did at some point in their history of manufacture. Thats strange. If I buy 90*45 timber in Australia then it really is 90*45, to within a millimetre or so.

Re:50%?

[maxume]

Calling it a scam is a bit much, it has been well understood by most people for at least 30 years.

Re:50%?

[MichaelSmith]

Electrons go from negative to positive, but since some jackass got it wrong at the beginning we affectionately refer to current as "the flow of positive charge". Please don't be too hard on Ben Franklin. He had a 50% chance of being right. But come to think of it, why is a proton "positive" and an electron "negative"? Isn't it just convention?

Re:50%?

[Gerzel]

Except 2x4s ARE 2x4 inches, as all wood is measured when green because wood shrinks and expands with humidity anyway any standard representation would be false at certain points.

Also when the definition was put into effect the fact that green wood shrunk after it dried, and that most wood was milled while green would have been well known by the general public.

In this case Creative was deviating from a well set industry standard w/o notifying the consumer.

Re:50%?

[sorak]

Another way of thinking about it is that the product wasn't quite what they advertised, so rather than having to give the customer what creative _DID_ advertise, creative is telling the customer that, for 50% extra, they can get what they originally paid for.

Re:50%?

[claytonjr]

Oblig: http://xkcd.com/394/

Re:50%?

[mikael]

Commonly known as the "2x4 of motivation" by senior management.

Re:50%?

[Shinobi]

No, they did not "knowingly misrepresent the capacity", they actually conformed to a standard. SI Prefixes are EXPLICITLY defined as base-10. Just because some shit-for-brains people started applying it to base-2 does not mean it's a correct usage. SI Prefixes are explicitly defined as Base-10, that's the standard, and has been since the 1890's, now start conforming to the bloody standard.

Re:50%?

[X3J11]

No, 32 bit OSes can address 4GB TOTAL memory, that's including graphics RAM and various caches. Thus, the total RAM addressable by a 32 bit OS is somewhere around 3-3.5GB depending on the configuration of the computer.

Splitting hairs, I know, but the bold text is in error. Regardless of how much hardware you install and how the system maps the memory, 4GB is still the total addressable memory. The system configuration only affects the accessible amount of memory.

tl;dr: s/addressable/accessible/

Re:50%?

[electrictroy]

What exactly did they misrepresent?

Giga == base 10
Gibi == base 2

If a customer assumed that 100 gigabytes actually meant 100 gibibytes, then isn't that the customer's fault? I say yes. giga == base 10. gibi == base 2. Not the same thing.

Re:50%?

[hasdikarlsam]

That is just convention, but it doesn't really matter.

The *direction* of current flow is what he got wrong, and it'd be wrong even if we switched the labels.

Re:50%?

[sm62704]

Also, it's 50% off the purchase of a new player. "Please don't throw me into the briar patch, brer fox" say brer Creative.

Only a corporation can make money by losing a lawsuit!

Re:50%?

[electrictroy]

Yeah but, imagine if a new "fad" takes hold, and everybody starts building their own homes or porches. I'm sure it would lead to a lawsuit over 2x4 boards only measuring 1.5 x 3.5 inches, due to lack of experience among the general population.

i.e. Same thing that happened when the computer "fad" fell out of hobbyists and into the hand of Joe "duh" Smith.

Re:50%?

[CastrTroy]

When you think about it though, 90mm X 45mm is 3.54" X 1.77". Which is roughly equal to the size of a standard 2x4 (4x2). All they did was in the conversion to metric was measure a standard 2x4, an start calling it a 90mm X 45 mm.

Re:50%?

[maxume]

The smaller boards generally cost less, and they seem to be strong enough.

It's just the same with hard disks, the per byte price isn't affected a great deal by the base-2 base-10 confusion, and the things do what they are supposed to(well, usually for at least 15 months anyway).

Re:50%?

[easyTree]

they misrepresented the capacity of their products..

According to this http://physics.nist.gov/cuu/Units/binary.html page we 'should' really be using GiB, MiB, KiB etc instead of GB, MB and kB.

e.g.

one mebibyte 1 MiB = 220 B = 1 048 576 B
one megabyte 1 MB = 106 B = 1 000 000 B

Re:50%?

[easyTree]

uhh, that should have been:

one mebibyte 1 MiB = 2^20 B = 1 048 576 B
one megabyte 1 MB = 10^6 B = 1 000 000 B

Re:50%?

[arb phd slp]

Or it is supposed to be. Lumber companies argue that planing + shrinkage takes 1/4 inch off each side, resulting in 1 1/2 x 3 1/2, a 35% reduction in volume. I have a trouble believing this, personally I believe this is more about saving money by scamming consumers by getting the official standard changed, who knows how. My brother built his house with unplaned lumber that he bought directly from the sawmill. The materials were very consistently 2"x4", 2"x6" and 2"x10". Planing does take 0.25" off each surface.

Re:50%?

[binaryspiral]

According to the most accepted definition K == 10^3, M = 10^6, G = 10^9, T = 10^12. Why should consumer product manufacturers use definitions only understood by technical professionals, that will confuse their average customer and are unflattering to the product? Being most accepted and being correct are two different things. When your computer reports sizes in base 2, then damn it - sell me stuff with capacities written in base 2.

If I go to the store and buy a 100GB hard drive. Why should I have to convert my actual storage capacity will be?

It's just lame - always ways, always will be.

Re:50%?

[zippthorne]

Yeah, but the conversion factor from angstroms to nanometers is a factor of ten. 1 deca-angstrom = 1 nm. The conversion factor from ergs to joules is a similar constant factor.

The conversion factor between "look at me, I'm a clever engineer binary almost-prefixes" and standard SI prefixes varies depending on the prefix. And the difference is only going to get worse as we increase the number. For instance, a yottabyte in the binary notation of 2^10n is 2^240 = 1,766,847,064,778,384,329,583,297,500,
742,918,515,827,483,896,875,618,958,121,606,201,292,619,776
  but, unfortunately, 2^239 = 883,423,532,389,192,164,791,648,750,371,459,257,913,741,948,
437,809,479,060,803,100,646,309,888
  more closely approximates the SI yottabyte, so which one will you use?

Further, only OS and RAM designers have any history of using the binary prefixes. In every other area, SI prefixes were already established as standard. Processor speed in megacycles? Communications industry already used SI prefixes, for about a hundred years.

We get it. They were a clever little trick that joked about computers using binary arithmetic. That in no way means they should be accepted as a standard for commerce. It's about communication, as you say, and it is better for communication not to have to explain domain specific use of words that have standard use.

Re:50%?

[EsbenMoseHansen]

At work we use it, and it works fine.

Re:50%?

[TheRaven64]

When I connect the device to a computer, which operating systems will give you the size in base-10 units? Even if the buyer's don't know how big one GB is, they will know what their OS is reporting their music library size to be. If they use iTunes, then the bottom of the playlist display shows you the size of the files in the current playlist. Having a device that is advertised as being 1GB and doesn't store a 1GB playlist is misleading advertising.

Re:50%?

[HeroreV]

Hard disk drive manufacturers were using base ten long before other people came along and started using base two. Why should the HDD people have to change just because others decided storage capacity should have it's own standard?

In any case, the "byte" unit is be replaced with the "octet" unit to undo this mess. Base-10 megabytes (MB) become megaoctets (Mo) and base-2 megabytes (MB), aka mebibytes (MiB), become mebioctets (Mio).

Base-10 was first, even for storage capacity, and the switch from "bytes" to "octets" will make it the standard again.

Re:50%?

[zippthorne]

Well there's an easy way to find out. Take a calipers with you and ask the closest lumber mill to give you a tour. Then, measure the rough-cut 2x4s.

Re:50%?

[Alpha830RulZ]

2x4's are definitely 1.5" x 3.5" as purchased. It's close enough to exact as to not matter (construction grade lumber is not a tight precision material). They -have- to be pretty close to this spec, or the plans won't work out. Framers don't even measure them to check. They've been this way as long as I have built stuff, which is coming up on 40 years.

Around the turn of the century, a 2 x 4 was definitely a 2x4. I had an older house that used them. However, the studs were still on 16 inch centers.

I don't buy any planing and shrinking argument. The turn of the century boards didn't shrink, and they were cut from douglas firs, same as the modern ones I use. I think the industry just wanted to get more boards out of a tree. Interestingly, they are still priced by the board foot, which is calculated on the nominal, rather than the actual measurements. This is true for 2 x 6's, 2 x 8's, 2 x 12's, etc, as well.

Re:50%?

[Intron]

So have disk sizes.

Re:50%?

[ps236]

Not really..

On that assumption, an 8 bit 'OS' would only be able to address 256 bytes.. Now, we all know that most 8 bit PCs could access up to 64kb (2^16 address space). Similarly a 16 bit 'OS' (eg Windows 3.1) could access 1MiB (2^20 address space)

The 'bitness' of the processor doesn't determine the size of the address space that can be used, but the size of the 'normal' word of data which is processed. It just so happens that for 32 bit Windows & Linux, they can access a 2^32 address space.

BTW - 64 bit processors can't access a 64 bit address space. Currently, 64 bit Intel processors can access a 2^40 address space.

So, address space is only related very loosely to 'bitness' of processor/OS

Theoretically, a 32 bit processor could have an address space of may TiB - they don't, but it's possible in the same way that 8 bit processors could easily access 64kiB, and that didn't involve trickery like swappable blocks of memory (such as OS9 level 2 had)

Re:50%?

[OwnedByTwoCats]

They represented the capacity of their products accurately, and the lawyers make out like bandits.

Re:50%?

[beckerist]

or Gb instead of GB... x8!

Re:50%?

[sexconker]

Computer science has always used powers of 2.
They use powers of 2 because computer systems are binary and quantum (logically). All forms of counting in computer science are based on this - be it storage, throughput, calculations, etc.

The uses of K M G etc in classical sciences are different. They measure things continuously - distance, size, mass, etc. Computers measure things discretely. There is no deci centi milli micro nano pico femto etc. in computer science because of this.

The reason they are different is because they are two different types of measurements. And if you want to bitch about which one is correct, guess what - the computer science one is correct. Classical measurements are continuous, but we all know that the universe is quantum (not necessarily binary, though you could always represent it as such). Since the universe is quantum, it is in fact the discrete measurements which are more fitting. How many possible states/configurations can this clump of particles take? Guess what, you'll be involving that funny little ^ symbol to count them.

Manufacturers of computer peripherals know what KB means in the computer world. They know what it will show up as when the user hooks it up to their pc. It was marketing that said "HAY - lets use 1000 instead of 1024! We can advertise larger capacity!!".

The terms KiB, MiB, GiB and such are a travesty.

Don't get me started on sexadecimal vs hexadecimal.

Re:50%?

[jedidiah]

K = base^3

The fact that the proper terms of are are "unflattering" is
exactly the problem. Vendors shouldn't be able to get away
with essential lying to you because the information is
"inconvenient".

KiloByte is simply the closest number to the "real kilo". It's
a practical solution to a problem rather than something that
just forces "easily confusable" consumers to remember ever more
jibberish and jargon.

It used to be that metric was supposed to be the "practical" solution.

Re:50%?

[sumdumass]

The problem is that binary has two values, on and off. This is why there is a base to association with it. Now without getting into the complexity if bits and bytes or spacing and addressing, you have to understand that program space is calculated in base two not base ten math. That is how the industry solved the issues of accurately reporting sizes in order to programming to accurately fit into the spaces it needs to run/be in.

Now, when you transliterate the numbers to base 10 because others are used to doing it and it makes your numbers look bigger and more attractive, it misleads the ignorant which amounts to a scam. You ask why when people are used to doing X is Y used. That answer is because Y is how it started and how it continues. If the people accustomed to X know enough to know what X means when buying a hard drive, then it wouldn't matter if it was a base 2 math system. They would know that Y is the amount, the higher the number the higher the space which would be true for X or Y. But seeing how Y reflects the actual usable space, the inflation of X can only be seen as an attempt to mislead those who aren't in the know and there lies the problem.

If they would have started using a base 10 calculation system in the beginning, there would be no problem at all with it in the reporting of drive sizes. But they switched in order to make it appear that they were selling more then they were and to mislead people into thinking they were getting a deal.

And more importantly, why don't you ask yourself why they don't size memory in the same system. Why isn't a company allowed to use 10^6 for a mega instead of 2^20. Your game that says it need 64 megs wouldn't run right. Your OS would be slower and so on. So what they did was use base 2 math because of the binary issue and called the multiple that just surpasses the Base 10 amount equivalents as the K, M, G, and so on. But your going into a specific field that has specific definitions for specific reasons so you have to use those definitions and so on. I mean you can't transpose all SAE (traditional English wrench sizes) to metric sizes and have everything fit. Even with half numbered sizes like the 8.5 millimeter you don't have an exact fit. You even have situations where thread size and counts are different making metric bolts incompatible with SAE applications. I have to have two fastener bins and two sets of wrenched and sockets in my garage just to work on my cars and lawn equipment because they don't convert properly all the time. So when you goto an application, you have to stick with the use and defined terms of the application. When you goto a computer, you have to use base 2 math on the parts that use base two math.

Re:50%?

[Cervantes]

Nobody is suing lumber manufacturers because 2x4s aren't 2 inches by 4 inches. Everyone in the trade understands the real dimensions. If you want to get involved in construction you have to learn things like that. +1
I couldn't have said it better myself.

Re:50%?

[Yvan256]

The customer doesn't really know how big the hard drive capacity is, he just goes with whatever Windows/Mac OS X/Linux is telling him.

Let's sue Microsoft, Apple and... erm... Linus?

Re:50%?

[Lost Engineer]

Wrong. Windows (and Linux?) separate the address space into kernel and process space as a matter of convenience (no need to translate addresses in the kernel region). Other operating systems do not necessarily have this problem.

Re:50%?

[FishWithAHammer]

I use Windows 2003. Works great; playing games takes a bit of hackery, but nothing special. Direct3D had its moments of sheer blinding hatemongering (would silently close out D3D apps) until I found a page that explained why there were problems and how to get around them. About the biggest complaint I had was that Photoshop CS3 got itself confused and thought that Windows 2003 was older than XP SP2, so I had to download a Microsoft tool and make the installer think I was running on XP SP2; the applications run just fine without the same treatment.

Re:50%?

[u38cg]

Possibly your old 2x4s were cut from lumber that was dried and seasoned as a log - this was certainly common for hardwoods that went into musical instruments at that time. Presumably if it had been, no considerations over shrinkage would apply.

Re:50%?

[hb253]

This is correct. While upgrading the bathroom in my parent's house (built in 1916), I noticed that the 2x4's were actually 2 inches by 4 inches. I'm curious as to when 2x4's became 1.5x3.5's.

Re:50%?

[nuzak]

I'd be fine for moving to SI measurements, but every OS out there still measures files using the binary prefixes.

In other words, Creative sold a device that claimed it would store 1 gigabyte, but wouldn't actually store 1 gigabyte as reported by the OS. They did so knowingly with intent to deceive. This is the very definition of false advertising.

Re:50%?

[MichaelSmith]

When you think about it though, 90mm X 45mm is 3.54" X 1.77". Which is roughly equal to the size of a standard 2x4 (4x2). All they did was in the conversion to metric was measure a standard 2x4, an start calling it a 90mm X 45 mm. Oh for sure. But a litre of milk really is a litre and I am surprised that US timber manufactures get away with incorrect labelling.

Re:50%?

[springbox]

they misrepresented the capacity

No, it clearly says GB and not GiB. Other storage manufacturers do this too and they are actually using the units correctly!

Re:50%?

[bhtooefr]

PAE is enabled by default, but Microsoft has it all set to keep everything within the 4 GiB barrier anyway.

Unless, by "most versions of Windows," you mean Windows 2000 Server, Windows Server 2003, and Windows Server 2008.

Re:50%?

[Lehk228]

they didn't misrepresent anything, software miscalculates storage space for convenience,

Re:50%?

[i.of.the.storm]

Note that I said RAM, not memory...

Re:50%?

[Alpha830RulZ]

I think it's unlikely, but don't really know. This was in a house in Portland, OR, dated 1911. The area was bustling at the time, and logging was a major industry. My belief is that the lumber was locally cut, and fairly fresh. It was very roughly hewn. There are large toothed circular saw marks. That's not a blade that you'd use with hard, seasoned wood, and is typical of western sawmills.

(disclaimer: my father used to work for a northwest timber company, Crown Zellerbach, so I claim some knowledge, probably without justification). At any rate, to the general point, modern timber is cut to dimension and then kilned, in which it reaches its final dimensions. There isn't too much variation in the results ( 1/32 in), and they don't mill it further after kilning. That means that the process is fully planned - there can't be a lot shrinkage involved, or the result would be more variable. So the lumber companies can make lumber whatever dimension they want from the first cut of the log.

Re:50%?

[iowannaski]

2x4 is a unitless board size. It's hardly incorrect labeling, as inches aren't even implied as the unit of measure. Besides, it's not like a lumber company selling 1.75" x 3.75" 2x4s would be doing their customers any favors - they would just be selling worthless non-standard sized sticks.

Re:50%?

[Endo13]

The "shrinking" reason is bogus. Planing is what reduces the size. As someone else posted, if you go to a sawmill where they prepare lumber for the lumberyard, a 2x4 is actually 2"x4" before they plane it. There isn't necessarily any reason you *have* to plane off a quarter inch on each side, but they did it so lumber sizes would still be standardized, and a quarter inch is generally enough to make sure each side will end up smooth and splinter-free.

Re:50%?

[GrievousMistake]

Actually, they did use the binary units, but forgot to tell us that they were using 7-bit bytes.

Re:50%?

[Graff]

Actually, a 2x4 is the right width to place in a wall that is 4" thick. There are other layers that go on top, sheetrock for example, so that the studs have to be somewhat smaller. It's strange that a bunch of lumber companies don't use the same reasoning that you do:
Arnold Lumber
Advantage Lumber
Howe Lumber

Just to name a few.

Yes, by putting sheetrock on the lumber you bulk the size of the wall out a bit but the true reason for lumber being a bit less than it's nominal size is how it is finished.

Re:50%?

[PSdiE]

Because the consumer has chosen which capacity MP3 player to go for based on how many MP3s they want to fit on. MP3 sizes are measured in MB, which is Base-2. Hence the capacity of devices for storing MP3s should also be expressed in Base-2 - it's not complicated. A consumer purchasing a 1GB MP3 player expects to fit 1GB of MP3s onto it, as reported by their operating system (i.e., Base-2). If they have a folder containing 1GB of MP3s on their PC, but cannot fit it onto the device, they have been mislead on its capacity.

What's 73MB between friends?

[BadAnalogyGuy]

Although I think it is a stupid abbreviation scheme, "mebibytes" and "gibibytes" as represented by "MiB" and "GiB" could really catch on due to this kind of confusion. In every industry except computing, base-10 is assumed to be the standard. It is basically only in this industry that base-2 is the standard, and the confusion over storage sizes becomes an issue.

So either the general public needs to learn about MiB and GiB or storage makers need to start labelling their products as holding amounts measured in base-10.

The former probably won't happen, so we'll most likely see tons of ink wasted on the longer small print.

Re:What's 73MB between friends?

[anethema]

Mebibytes and gibibytes will NEVER catch on. The geeks would have to spearhead something like this, and they never will because the words sound stupid and you sound stupid and unprofessional using them.

Storage makers ALWAYS label their products as base-10 amounts.

My 74GB WD Raptor HDD is 69.2 GB. They are getting sued because for 40-60 years now, maybe longer, storage/ram etc in computing has been base-2 (it being a binary system). Now they are switching to base 10 because they can advertise a higher space than what is actually sold. They are getting sued for this and rightfully losing. Seagate lost a similar law suit and now creative. The companies are going to have to learn to ignore the dolts in marketing and just tell the truth or start losing money through similar law suits.

One judgement and one settlement in favour or the plaintiff starts setting a pretty big precedent and future judgements will now almost certainly go in favour of the plaintiff.

Re:What's 73MB between friends?

[TapeCutter]

Your right, it hasn't happened already so it probably won't. Instead of a bunch of legal ink they could simply write out the whole thing.

Eg: Instead of 250GiB they could write something like : 2.0^E13 useable magnetic toggles : and let the consumer translate it into what they see in the O/S.

Re:What's 73MB between friends?

[hairyfeet]

Or the HDD manufacturers could just,oh I don't know,actually give them a little extra instead of screwing them? My last Maxtor was advertised as a 200Gb and it came out as 203Gb when formatted. needless to say I had no problem with getting an extra 3Gb. With the stuff getting so cheap would it REALLY kill them to actually let the consumer come out ahead by some tiny amount?It would be smart if for no other reason than to ensure that they don't have to deal with stupid lawsuits like this and would have the side benefit of letting the customer feel like they got a good deal. But that is my 02c,YMMV.

Re:What's 73MB between friends?

[TapeCutter]

I agree that a few cents on each HDD sale is worth far less than a happy customer and tip my hat to Maxtor. The post was intended as sarcasm.

You also answered a question I put in another post ie: Is there any manafacturer who still advertises in base 2?

Re:What's 73MB between friends?

[i.of.the.storm]

That's interesting. I noticed that my 80GB regular hard drive was actually 73.something GiB, so I assume Raptors were actually giving their size in GiB when they said 74GB. You learn something every day...

Re:What's 73MB between friends?

[hairyfeet]

I know it was sarcasm,and I apologize if I came off as snarky,it just really irks me that all these manufacturers will screw their customers,cause bad feelings to be generated from said customers,and risk lawsuits for a savings of what? 02c on a HDD to at most 06c of a flash drive? When they could have happy customers by simply raising their prices across the board by 15c and actually letting us come out a little ahead. I just bought another Sandisk M2XX series MP3 player(I love how reliable they are and how you can carry spare AAAs instead worrying about recharging) and I would have happily paid an extra 25c to $1.00 to have it say 4.03Gb instead of 3.8Gb when I first plugged it in.

And as for the Maxtor,I know some folks really hate them,but I've never had a bit of trouble out of them. In my years of experience dealing with HDDs I've found a little extra cooling,especially with Maxtor or any of the cheapy Asian brands like Exelstor goes a long way,otherwise they run too hot and tend to die early.I just slapped a $3 HDD fan from Newegg on and hadn't had a bit of trouble since.And I was really surprised by the extra 3Gb since it wasn't even their top drive,but a DiamondMax 9 I got for a dirt cheap $20 on Black Friday. But after finding that I actually got more instead of less when I need an upgrade or a drive fails I'll be looking at the Maxtor lines first. IMHO the manufacturers would earn a LOT of good will and make a nice bit of extra profit if they would just raise the price a few cents and let us have a little over instead of under. But that's my take on the subject,YMMV.

Re:What's 73MB between friends?

[Jesus_666]

Mebibytes and gibibytes will NEVER catch on. The geeks would have to spearhead something like this, and they never will because the words sound stupid and you sound stupid and unprofessional using them.

Except for those geeks who think that sticking to ambiguity because the unambiguous prefix doesn't pass arbitrary coolness standards is unprofessional. "I don't support it because the name isn't cool enough" is not a best practice.

Even though when talking I still use "megabyte" for both MB and MiB (prefixing it with "decimal" or "binary" if neccessary), when writing I do take care to differentiate between MB and MiB because they describe different things and things are clearer when using them.

What the hard drive manufacturers should do is give the capacity in both GB and GiB. That way there is no confusion and everyone is happy. Well, except for those who think that the GiB figure isn't stylish enough.

Re:What's 73MB between friends?

[TapeCutter]

"I apologize if I came off as snarky"

No problem, I wasn't offended in the least. I've been a software developer for 20yrs and a hobbyist since the Apple II days but I can't say I can single out any particular brand of HDD that has caused me more greif than any other. I do however recall discovering (and being a bit miffed by) the base ten thing with the first HDD I bought, but nowadays I just assume it's base 10.

BTW: That first HDD was a 20MB western digital I purchased second hand for $200 - roughly half the price of a new one!

Re:What's 73MB between friends?

[hairyfeet]

Never had any IBM Deathstar drives I take it? But seriously, as someone who has worked in repair shops since Win95,and built and sold pc rigs long before that,I can tell you that most of the dead drives I have seen are TOTALLY the users fault. They buy some cheapy mini tower and then have somebody squeeze in a second drive so the the things aren't even half an inch apart and then are shocked when their new drive bites the farm and takes all their data with them.

Of course the idiot who designed the circa 2000 HP Pavilion "knuckle buster" mini tower should be shot,as all the fans in the world won't cool that hell beast down. I am actually typing this on one I got for cheap because the owner got tired of overheats and after trying every fan combo I could think of I finally had to use my old bosses "white trash cooling special" which consists of removing the side and sticking a Wal Mart box fan against it,LOL.

But while I can understand where the original change came from(and for those who don't, the race to reach 1Gb was a point of pride for the HDD manufacturers back in the day) I still say that now that space is just so damned cheap it wouldn't kill them to raise the price a few cents and let the customer come out ahead for once. And while a few cents more cost wouldn't bug most consumers,I for one would have been happy to give them a little extra money to have the three albums worth of extra space on my flash based MP3 player. But that is my 02c,YMMV

Re:What's 73MB between friends?

[TapeCutter]

"white trash cooling special"

Hah, I like the term. I live in Australia I've and used this myself to stop a cpu from complaining about the heat and shutting down, that was after I bumped up the temprature setting in the bios. And yeah, if you work in a repair shop your going to the one who sees dead drives daily and cops all the "where's my missing 73meg" agro.

Re:What's 73MB between friends?

[Nicolay77]

Do you mean 203.912'880.128 bytes?

That's 203GB but it is 189GiB (as reported by windows).

I have a Maxtor Quickview disk.

Re:First!

[Jizzbug]

Anonymous coward loses! (#23270978)

1 GB drive?

[icydog]

"and everyone else who bought one of the HDD MP3 players in the past several years gets a 50% discount on a new 1GB player"

Is anyone else wondering how many bytes are in this thing?

Seriously though, it would be nice for storage vendors to use binary prefixes rather than decimal. We probably don't mind too much other than for the sake of being pedantic, but it must be pretty annoying to Joe Sixpack who buys a 1 TB drive only to find that he's missing one hundred billion bytes!

Re:1 GB drive?

[prockcore]

Seriously though, it would be nice for storage vendors to use binary prefixes rather than decimal.

No, it would be nice if everyone (including your braindead OS) would start using decimal prefixes.

You do it in your head anyway. Every single person who looks at a file that is 1,500,000 bytes says it's "1.5 megabytes".

There is absolutely NO reason the consumer should be exposed to binary math. NONE.

how about dell dj's?

[neersign]

the dell dj's are just rebranded creative mp3 players, so are they covered?

Hard Drives

[BillOfThePecosKind]

So can I sue Samsung or whomever for doing the same thing? Or is this different?

Re:Hard Drives

[Kingrames]

Of course it's different. they haven't been sued yet.

Re:Hard Drives

[wal9001]

Seagate was. The rest still haven't, so you're free to go after WD, Hitachi, or whoever else seems like they might have some money for you.

Re:Hard Drives

[anethema]

Seagate settled and it was a class action suit. If you havent made any claims against the class action settlement, you are also free to sue Seagate for the same reason.

Punish corps by giving them money...

[pintpusher]

Completely orthagonal to the whole stupid debate over base10/base2 gi(bi|ga)bytes or whatever....

I really hate this trend. A corporation loses a case and the punishment is that consumers get to spend more money with them. I fully believe that they will at least break even if not make money on this settlement. WTF. They should be forced to refund everyone who bought one of these players an amount equivalent to the proportion of storage space the "lost".

I'm a class action settlement "Winner" in my business and my prize? I get 20% off products that are outside my usual purchase contract with the company. How lame is that! They get to keep charging me the same ripoff prices as before *and* I get to spend more money with them. And if I mess up filling out the little coupons, then they are invalid, no recourse. </rant>

Re:Punish corps by giving them money...

[TapeCutter]

"I really hate this trend."

Ditto! However since the base10 thing has been standard industry practice for eons I think that in this case that type of 'winning' is entirely appropriate.

I'm waiting for the spam headline - "Disk prices double, get a 50% discount by signing up now!"

Re:Punish corps by giving them money...

[plowfunkel]

The "stupid debate over base10/base2 gi(bi|ga)bytes or whatever" is exactly the point. As you imply with that statement, the differences between a base 10 and a base 2 computation of hard drive space are not significant, and so the payout to consumers shouldn't be significant either.

Re:Punish corps by giving them money...

[martin-boundary]

It ought to be standard procedure for courts in class action suits to mandate that all lawyers' fees and awards be settled in the same way as their clients'. So if the clients get 20% rebates, then lawyers should be paid exclusively in 20% rebates.

Re:Punish corps by giving them money...

[tokul]

Only if they have bigger than 50% profit margin.

Re:Punish corps by giving them money...

[sorak]

I really hate this trend. A corporation loses a case and the punishment is that consumers get to spend more money with them. I agree 100%. The customer is entitled to say "these people misrepresented themselves and I'm not doing business with them any more", but, unfortunately, our legal system has decided that BOGO coupons are legal tender in this country now.

WTF. They should be forced to refund everyone who bought one of these players an amount equivalent to the proportion of storage space the "lost". I'd say they need to take it a step further. Since the correlation between storage capacity and product cost does not seem to be correlated (I.E. the price per gig skyrockets on new technology), the only fair way to do it is to either have the company send the customer a nearly identical unit with a satisfactory amount of disk space, or to give them a refund. (At least, it is fair if you assume that the company is being required to make ammends for a mistake or illegal action.)

Re:Punish corps by giving them money...

[Jarik_Tentsu]

Considering that almost all data storage companies doing this...would this mean anything significant in the industry? I mean, everything from USB keys, to internal and external HDD storage as well as devices using HDDs (MP3 players) advertise their stuff in GB (10^9) instead of GiB (2^30). Or is this case specific and will not affect across the board?

~Jarik

Re:Punish corps by giving them money...

[Jekler]

Even if the profit margin isn't greater than 50% (or 20% on other products), they still gain market share or at least an opportunity to retain a significant portion of their market share.

Re:Punish corps by giving them money...

[Threni]

> I really hate this trend.

I hate this trend. People trying to make a fast buck out of their own ignorance as to how much something can store. If you don't know, work it out - don't just jump on a bandwagon someone else started.

Re:Punish corps by giving them money...

[CmdrPorno]

Yes, they should be required to write checks. This type of "settlement" is a sham.

I remember the Iomega Zip Drive "click of death" "settlement." As a party who owned one of the affected drives (although mine hadn't failed yet), I was entitled to purchase a new zip disk from Iomega at a price of $20, at a time when independent retailers were selling the same disk for less than $20.

Re:Punish corps by giving them money...

[fredmosby]

You didn't win, creative paid off your lawyers with millions of dollars and gave you cupons to try to keep people from realizing what is going on.

Re:Punish corps by giving them money...

[glitch23]

They should be forced to refund everyone who bought one of these players an amount equivalent to the proportion of storage space the "lost".

Creative thought of this and opted to pay people using dollars in base 10 since that would be cheaper but the court wouldn't allow it.

Re:Punish corps by giving them money...

[DiEx-15]

Does anybody remember that this was done with the original Nintendo Entertainment System/Famicom? Yeah, it's lame that their punishment is to have the victims "spend" money. However when the legal water is uncharted, I guess the law can only get away with a slap on the wrist.

Re:Punish corps by giving them money...

[LangDaLa]

When I got the e-mail for this I almost marked it as spam, I thought it was a phishing e-mail. Nice to see it somewhere else, now I know it's legit.

Innumeracy

[Detritus]

The court should have awarded each of the plaintiffs a calculator and a boot to the head.

Re:Innumeracy

[statemachine]

The court should have awarded each of the plaintiffs a calculator and a boot to the head.

It's been quite a while since I've heard that skit.

1GB = 10^9 Bytes

[phizix]

1GB = 10^9 bytes, by definition. It is ridiculous Creative Labs has to pay this settlement.

Re:1GB = 10^9 Bytes

[91degrees]

Fighting in court = 10^9 dollars :P

Well, okay - maybe not that much but I imagine they're not going to lose out too much from the settlement. Most people are going to go for the 20% off option (they already have a creative mp3 player). I wouldn't be surprised if Creative break even there.

Re:1GB = 10^9 Bytes

[DurendalMac]

Only if they had the little annotation and footnote that indicated how they were measuring the space. If they had that, then screw this lawsuit. If not, then screw Creative. Can't say they didn't have it coming after that retarded patent suit against Apple for what? Column view on an MP3 player?

Oh for fuck's sake...

[NonSequor]

When will we computer geeks get over this obsession with binary memory measurements.

Using the binary units makes referring to RAM capacities easier and makes many other things (storage capacities and file sizes) clumsier to deal with. I suppose that OS internals also use 1024 bytes as a basic organizational unit, but that hardly seems relevant to the issue of whether a file labeled as 8GB should actually be 8 billion bytes or 8.6 billion bytes.

Everyone around here seems to hate tradition for tradition's sake unless it's a computer related tradition. Congratulations, you've become what you hate and you didn't even realize it.

Re:Oh for fuck's sake...

[NonSequor]

The fact that this thing still lingers in end user software after decades of usability research is two-thirds of what makes this so annoying.

The other one-third is the people who keep on defending it.

Re:Oh for fuck's sake...

[nmb3000]

Everyone around here seems to hate tradition for tradition's sake unless it's a computer related tradition.

I don't know if that statement is accurate as "get off my lawn!" posts are so common, but in any case I think change for change's sake is worse.

The usage of K to mean 1024 has been around the computing industry essentially since there was a need for the term. It was simply understood that because computer hardware doesn't do base 10 (it can, but just creates waste) in computer circles and reference the SI prefixes were applied to a binary system. The only reason ambiguity entered into it is because of marketing. By playing pretend and going against decades of precedent they realized they could sell less bang per buck without consumers knowing the difference.

This thread from the other story does a great job of explaining this idea. In the computer industry, K, M, and G stand for slightly different values than their normal SI meaning. That's just how it is and snobby new prefixes (that are still ignored in all practical use) won't change it. I'm glad we're finally sticking it to manufacturers (though I hardly think Creative was punished here) for intentionally misleading consumers.

Re:Oh for fuck's sake...

[Bill, Shooter of Bul]

I don't hate tradition for tradition's sake. And I don't think its done for tradition's sake either. Its easier to count things based on boolean logic with a base two system. Its not like they made it base two because they didn't know any better. Its not the equivalent of defining current as the flow of positive charge and then discovering years later that its usually the negative charges that move.

Re:Oh for fuck's sake...

[prockcore]

In the computer industry, K, M, and G stand for slightly different values than their normal SI meaning.

Except when it doesn't. Network speed is always base10. So is CPU speed.

Re:Oh for fuck's sake...

[zippthorne]

Wrong question. Why is it that everyone assumes the hard drive manufacturers are doing the dumb thing. They are using the prefixes as SI intended, after all.

They should really sue the OS distributors for under-reporting both the size of the disk, and the size of the files on the disk.

Re:Oh for fuck's sake...

[MortimerV]

I have a problem with disks being sold in base 10 but my OS displaying everything as base 2. It's just confusing. I look at the solution from the other side, though: Why doesn't my OS display everything in base 10? As a basic end-user, it makes no difference to me.

Instead of 5MB mp3s I could have 5.24. Instead of a 3GB collection, it's 3.22. And now my 100GB drive really is 100GB, rather than 93.13.

People selling memory get to use their inflated numbers, all of our files get bigger, and the confusion is gone for the basic user.

Re:Oh for fuck's sake...

[NonSequor]

I understand that internally many things are aligned along 1024 byte boundaries. I don't see why that should be considered relevant to the end user. The machine exists to serve me and any time it displays information in a format that is inconvenient for me, the machine has failed.

I'll only change my opinion on this if it can be demonstrated that displaying file and storage capacity sizes in binary measurements benefits the user.

People can call 1024 byte units whatever they want when they are working with things that are organized into 1024 byte units, but please just let the rest of us who are working with data in all sorts of odd sizes call a million bytes a MB.

Re:Oh for fuck's sake...

[kurt555gs]

Except when using Vista.

Re:Oh for fuck's sake...

[NonSequor]

So the exact same file should change sizes depending on if it is read into RAM or not? Pretending not to understand someone else's argument doesn't make it invalid.

Re:Oh for fuck's sake...

[SanityInAnarchy]

They started out using the binary units, actually, like everyone else. At some point, some manufacturer figured out that, yes, it's technically correct to use them for 1000 instead of 1024, so they did that -- and it seems like they did that to deliberately misrepresent it as being bigger than it is.

Kind of like broadband providers which sell "unlimited" connectivity at a certain amount of "burst" bandwidth, with no guarantees whatsoever.

Now, if they had actually attacked the OS distributors back then, they might have a case.

Regardless, what part of what I just outlined is difficult?

Re:Oh for fuck's sake...

[drfireman]

Everyone around here seems to hate tradition for tradition's sake unless it's a computer related tradition. Congratulations, you've become what you hate and you didn't even realize it. That's because consistent use of units of measurements is not tradition for tradition's sake, it's so that we can tell how much stuff will fit on a storage device. As luck would have it, we have a decent system for that, and it's based on round numbers in base 2. I don't think the system has any real drawbacks, although apparently you find it somehow clumsy. But as soon as you introduce a second system, suddenly it becomes difficult to tell how many bytes a 120GB drive will hold, unless you happen to keep track of exactly when each manufacturer switched over, or in general to measure things precisely. Probably if the entire industry agreed to switch all at once, it could be settled in about ten years. But why bother? Nobody really cares if a gigabyte is this size or that size, the only thing we really care about is that it always means the same thing. And this really has nothing to do with computers, but fortunately the companies that distribute orange juice and things like that haven't yet figured out a way to give me less (while still maintaining some asinine argument about why it's still a "quart"). Actually, for all I know they have. The size of a quart is much more of an arbitrary tradition than the size of a gigabyte. If I find out they've been doing that, I'll be a little irked. Not because of tradition, but because I don't like being cheated.

Re:Oh for fuck's sake...

[__aawfbm2023]

I always thought K denoted degrees Kelvin.

Re:Oh for fuck's sake...

[AcidPenguin9873]

You obviously didn't read the thread that GP linked to. Things that computers address are measured in powers of 2, like RAM. Measurements of things like network speed or CPU frequency can use base-10 because computers don't need to address every bit in a network transmission or every cycle in a CPU clock. Read this post from that thread if you're still confused.

Re:Oh for fuck's sake...

[Oggust]

Exactly. The only thing that ever used the base-2 units was memory, because it actually made sense back when. (Since memory capsules come in base-2 sized units.) And the error was small.

• * All the ethernet (and other networking)standard use real (base-10) megabits/gigabits.
• * Disks, DVDs and tapes. Yes, they all do this. Because it makes sense.
• * CPU frequency. I get 3.6 GHz, and those are real "G"s.
• * Some of the stats on memory chips are even base-10, like the latency and clock rate, for example.

Also, at some point this needs to get fixed. The error gets bigger for every generation:

difference between:

• Kib and KB: 2.4 %
• MiB and MB: 4.9 %
• GiB and GB: 7.3 %
• TiB and TB: 10%
• PiB and PB: 13%

...and so on. This has to stop. It makes it a royal PITA to use these sizes together with anything else.

It's useful to be able to figure how long it takes to transfer a 10 "TB" file across a 10Gb ethernet connection without having to know about the obscure 10% fudge factor.

/August.

Re:Oh for fuck's sake...

[pembo13]

It's not an obsession it's how it is. Do you have no respect for things at all?

Re:Oh for fuck's sake...

[Lord Kano]

Congratulations, you've become what you hate and you didn't even realize it.

I sir, will NEVER become a Democrat.

LK

Re:Oh for fuck's sake...

[jimicus]

Speak for your own OS. Linux (along with the various GNU utilities) has been moving towards using SI definitions when it does the translation for some time now.

For an example of the confusion, download the latest Knoppix (it hasn't been updated in a while). ls -lh uses binary units, df -h uses SI units.

Re:Oh for fuck's sake...

[91degrees]

No, it represents Kelvin. The temperature is an absolute scale so there are no degrees.

Re:Oh for fuck's sake...

[Znork]

In the computer industry, K, M, and G stand for slightly different values than their normal SI meaning.

No they don't. They are sometimes used as an approximation, or as a lazy you-know-what-I-mean in cases where mistaking one for the other basically doesn't matter much anyway (like, for example, storage sizes!).

Actually believing K means 1024 is, in my experience, a fairly new (last 10-15 years), and something that would have gotten you a big 'F' in any case where it actually mattered.

I'm glad we're finally sticking it to manufacturers for intentionally misleading consumers.

Oh, please. Consumers were intentionally misleading themselves. If a seller is selling you something clearly labeled as being measured in US gallons but you usually use UK gallons to measure it, you getting less liquid than you thought is your own fault.

The label on the storage device is correct. The flaw is in your interpretation of that label; if that creates a problem for you, you would more appropriately sue your school district.

Thankfully, progress is being made in stamping out this error; the last thing we need is _yet_ another unit ambiguity causing everything from rockets blowing up to building collapses.

Re:Oh for fuck's sake...

[EdIII]

The "Obession" is not for one way or another. I, personally, could give a shit which way we choose to represent it, as long as it CONSISTENT.

That is the problem here. They sold a car marketing it in Miles per hour, but the consumer is told ALL other pertinent information in Kilometers per hour. How can a consumer make an informed decision when they are unable to understand the units being used, their relation to each other, and what it means to them in the first place?

If we use Base-10 fine. Base-2 would be equally fine. Sure, there are some arguments to be made about which is easier, more efficient, etc. The only important part is that the same units representing the space are the same units representing the file size.

The same problem exists with ISP's and advertised bandwidth. Just about every single UI I have seen represents speed in KiloBYTES per second, yet ISP's represent their speeds in MegaBITS per second. WTF? They expect the average person can easily convert between Bytes, Bits, Kilo, and Mega that easily in their head?

As for your question of a file is labeled as 8 Gigs being 8 billion bytes or 8.6 billion bytes is 100% RELEVANT. That is a 7% difference. 7% MORE MP3's can be stored in that space. The larger the size, the MORE relevant the question becomes, as that 7% starts representing quite a number of complete MP3's. Remember, that is 600 Megabytes of difference you are talking about there. That can represent approx. 100 6 Megabyte MP3's. That is the difference between say 7 to 9 albums.

If you mean that it is not important whether or not the internal OS represents storage space as 1K being 1000 Bytes or 1024 Bytes, I somewhat agree. However, they then need to represent the size of the files the SAME way. Otherwise the size of all the files does not accurately represent the actual amount of space taken up on the hard drive.

If there is any tradition that we "computer geeks" like to stick to, it is being farking consistent with the technology we are using. After all, wasn't it NASA that created a new crater on Mars because they could not get the units right?

Re:Oh for fuck's sake...

[jhol13]

Base-2 would be equally fine. Until you make any calculations (how many, how much money, how long a time, ...).

Re:Oh for fuck's sake...

[zippthorne]

At what point were hard disks ever sold in clever-binary-kludge units?

Re:Oh for fuck's sake...

[smallfries]

Keep your opinion if you want, but your line of argument is entirely redundant. This is not a usability issue and I think that you know that. The end user does not really care if sizes are displayed in base 2 prefixs or base 10. It doesn't matter.

A file size is some number. If this other file is about 10x larger then the number will be about 10x larger. That's about as concrete as it gets from a user interface point of view.

For a more technical user it does matter, because they may have issues related to how many addressing units (blocks) the file spans. So the actual matter of which is used is completely irrelevant for the vast majority of users from a usability perspective; but vital to the usability of a certain segment of users.

Of course the choice of binary is not arbitrary and there is the basic choice of whether most convenient sizes should be integer (with base-2 prefixes) or non-integer (with base-10). I'm sure that making most storage sizes integer makes the system simpler, buy hey maybe you want to argue differently.

Re:Oh for fuck's sake...

[slcdb]

The only reason ambiguity entered into it is because of marketing. Uh, I believe the ambiguity was introduced by computer programmers who decided to misuse standardized prefixes by reassigning them non-standard meanings.

By playing pretend and going against decades of precedent they realized they could sell less bang per buck without consumers knowing the difference. And actually it was these silly computer programmers who were playing pretend and went against a couple centuries of precedent when they started using "kilobyte" to mean a value other than exactly 1,000 bytes.

But the more important point that you are ignoring is that this ambiguity has caused confusion. And this confusion isn't going to go away until something is changed. There is no official standard anywhere that says that addressable units in computing are always base-2 and that any other units in computing are base-10. And there probably never will be. The KiB and MiB family of prefixes is a very logical solution to this problem.

A logical person would believe that 1MB of data transferred over a 1MB/s link would take precisely one second. And it should, but in many cases it won't. And that's just stupid because doing it any other way just complicates such calculations. If we use a more sensible unit, like MiB, then:

      1MB at 1MB/s = one second

and

      1MiB at 1MiB/s = one second

Wouldn't that be convenient?

Re:Oh for fuck's sake...

[cydnub]

Actually believing K means 1024 is, in my experience, a fairly new (last 10-15 years), and something that would have gotten you a big 'F' in any case where it actually mattered. No, it's been that way for at least 23 years. My compsci class in 1985 had 1K=1024Bytes.

Re:Oh for fuck's sake...

[prockcore]

It was simply understood that because computer hardware doesn't do base 10

Utter nonsense. For my OS to say I have "4.6G" of disk space available, instead of "4.9G" requires the exact same amount of math.

As soon as the OS started using decimals to report file sizes it was using base 10 math.

got wood?

[unt0ld]

Does this mean I can sue Home Depot because 2x4 studs do not measure out to be exactly 2 inches by 4 inches? They are actually 1.5 x 3.5. That's a lot of missing wood.

Re:got wood?

[Raineer]

lol, very good point actually.

Re:got wood?

[kidgenius]

No, you cannot. The wood was 2 x 4 in its rough and green state, and after its been finished and dried, it is now smaller than before. That's just the way wood is, and it's understood. For instance, if you buy 4/4 wood, it should be 1" thick, but you'll only get that if its rough. When you buy it and it's been surfaced on two sides, you'll actually have something that is 7/8. It's just something that is understood in the wood world.

Re:got wood?

[DarkOx]

Its been a standard practice in the storage world for as long as I CAN REMEMBER to report base 10 capacities. This whole gibi and mibi vs giga and mega is very new. The fact that a 200 gigabyte disk holds is only 190 some odd gigs in capacity is understood in the storage world.

Re:got wood?

[Dave Walker]

Actually, they're down to 1 7/16" x 3 7/16". And 3/4" plywood is actually 23/32", just enough to require a special router bit if you want tight joints.

Re:got wood?

[rts008]

"They are actually 1.5 x 3.5."

They should only be 1.5 x 3.5 inches in the warmest AND driest part of the year where you are at. More often than not, they will measure 1 5/8" x 3 5/8".

The term "2 by 4" is a base standard for reference purposes. That is the dimension the rough lumber is cut to from the still 'green' (wet) tree. The reason for this is that 'green' wood has a reasonably consistent moisture content wherever you cut it, but it will 'season' (dry and stabilize) in predictable (but different) ways according to your climate.

It's the same principle with the speed of sound. The speed of sound only has meaning if you also include present conditions like altitude, barometric pressure, temp, and humidity. Most assuredly the speed of sound is quite different at sea level compared to 40,000 feet altitude.

So it's similar to a 2 by 4, under 'standard' conditions, it is 2" x 4", but will most certainly change with local conditions. This has been known, accepted, and dealt with by carpenters and craftsman for many centuries.

Next time try a car analogy, as 2 by 4's don't mix well with computers and computer tech.(Fire and some Building codes) You should be using metal studs instead. (these are marked as 1 5/8" x 3 5/8" because they are not subject to change by local conditions unlike wood is)

Not quite.

[raehl]

Creative didn't rip anybody off, but some snarky lawyer thought he could make some legal fees by suing them for using standard definitions.

So they decided to make some lemonade and sell some units.

Not losing much is only fair when they didn't do anything wrong to begin with.

Re:Not quite.

[anethema]

These definitions haven't been 'standard' in this industry for over half a century now. Just because some storage company's marketing departments have now decided otherwise, does not mean this is legal. Every piece of software including the OS (not to mention all software system requirements, oops!) measures in base 2.

Companies are unsurprisingly getting sued, and equally unsurprising, are settling these suits as quickly as they can because they know they will lose.

http://www.betanews.com/article/Seagate_Settles_Suit_Over_Gigabyte_Definition/1194025700
http://www.harddrive-settlement.com/notice.htm

Re:Not quite.

[Waffle Iron]

Just because some storage company's marketing departments have now decided otherwise

You mean "now", as in as long as anybody can remember? Because that's how long hard drives have been specified in base 10 by manufacturers. That makes sense, since unlike RAM, the size of a hard drive, based on cylinders and sectors, has nothing to do with the number 2.

Floppy drives have always been specified in an even more insane system than RAM, where "MB" == 1000*1024. That's not base 2 either.

I can't believe that after all these years of hard drives being sold in SI notation, that anybody is surprised by that. It's probably because one single application program, Windows Explorer, happens to confusingly report sizes in pseudo-binary. (Which BTW makes it really hard for people to interpret and work with the displayed numbers properly because of the extra needed math to convert between KiB, MiB and GiB). That's a software bug, not a problem with disk drives or their marketing.

Re:Not quite.

[phagstrom]

Right you are. This is the case of the rest of the world being wrong. Just because a bunch of people have decided to use the SI standard in the wrong way, does not mean that companies should be sued for still using it the right way.

So now kilo now longer means 1000, it just means...somewhere around 1000, depending on what the judge rules.

Re:Not quite.

[inasity_rules]

And everyone always forgets that the minimum allocation size on a hard disk is 512bytes so it actually makes sense to report the space in powers of 2. Even for the clueless consumer. And it therefore makes less sense for the drive manufacturer (who designs their drives to allocate in 512byte blocks) to report the sizes in powers of 10. It is highly inconsistent.

Re:Not quite.

[Waffle Iron]

And everyone always forgets that the minimum allocation size on a hard disk is 512bytes so it actually makes sense to report the space in powers of 2.

No, it doesn't. All that means is that the capacity is divisible by 2^9 bytes. Since all higher multipliers of capacity have no dependence on 2, it means that 2 is not a useful radix for dealing with drive capacities. In particular, a drive has absolutely zero relationship to the quantities for MiB (1024^2) or GiB (1024^3).

The drive I'm using right now has 30401 logical cylinders and 16065 blocks/cylinder. Those numbers are not even divisible by a single 2, much less powers of 2.

Re:Not quite.

[inasity_rules]

From the operating system's (and therefore the user's) point of view the smallest unit on the drive is 512bytes. You are therefore suggesting that we take this "unit" of 512bytes and multiply it by powers of ten...???

The logical structure of any particular drive has nothing to do with the case. Whichever system is used, the end user will be presented with fractions.

Now to clarify for you; disk drives are divided up into sectors of 512bytes. It is unpractical and dificult to write to a fraction of a sector, therefore a whole number of sectors is always used to store a file. This intricatly ties the number 512 with the size of all files on any disk(and therefore free space by the way). 512 is 2^9, or half a "kilobyte". Now either we use powers of two and retain consistency or we don't and loose it. It is actually that simple. Powers of two give us an informative and useful indication of disk space. They may not be as "user-friendly" as you would like, but they are certainly more practical.

But I guess computing power and storage space are cheap enough for people to waste them making conversions for the users that "cant understand powers of two", although I reckon it would cause more confusion then it would solve. Most people don't even know about this issue or care, but changing now would cause problems with anyone who took the time to try and understand.

Re:Not quite.

[Waffle Iron]

The problem is that they don't use just powers of two. They mix together odd multiples of 1024, 1048576 and 103741824 in the same program, and then present this mix of incompatible numbers to you in one listing. How many 100 MiB files fit on a 4.38 GiB DVD? You can't tell unless you get out a calculator. (Hint: the answer isn't 43.8).

Nobody cares how many bytes are in a sector. Hard drives use 512 byte sectors, but filesystems use different numbers. Sometimes one filesystem type uses different sized blocks (1024, 4096) based on the size of the media. That's totally irrelevant to the user and just gets tacked into the filesystem overhead along with space for directory entries, timestamps, filenames and security attributes. Some filesystems can even pile multiple small file fragments within a single block, rendering your argument totally invalid. None of these low-level implementation details need to show up in a directory listing, especially if it makes it needlessly impossible to do mental computations on the displayed numbers.

Re:Not quite.

[inasity_rules]

..Sometimes one filesystem type uses different sized blocks (1024, 4096) based on the size of the media. That's totally irrelevant to the user and just gets tacked into the filesystem overhead along with space for directory entries, timestamps, filenames and security attributes. Some filesystems can even pile multiple small file fragments within a single block, rendering your argument totally invalid. None of these low-level implementation details need to show up in a directory listing, especially if it makes it needlessly impossible to do mental computations on the displayed numbers.

Ummm... last time I checked 1024!=1000... Also, the free space on any given disk is always a multiple of the block size which is always a multiple of 2.

The fact remains that computers are completely based around a binary number system. While you can hide that, you can't run away from it. There are abnormal people in the world who are good at power of two math, but I grant you the point that most people are better at base 10 math.

It occured to me that this is all moot, since the average user just drags files into k3b/nero until he runs out of space. But to answer your origional question: about 43. The exact answer of 44(you can't have part files or your question is irrelavent) is a nice bonus - which is probably why people sued creative. 43 is close enough for any practical purposes of estimation. Noone has had to manually calculate disk space for ages. You may choose to, but you don't have to.

But if we hide all the low level details from you you'll start complaining that your group of 10 000 234888 byte files is taking more disk space than it should. Perhaps if the user is "too dumb" to see these low level details perhaps the user should just get a pie chart like in say, windows?

I wonder if the rep. plaintiff will complain...

[harmony7]

... upon receiving his $5k, that he should have gotten $5,120 ?

Re:I wonder if the rep. plaintiff will complain...

[jollyreaper]

... upon receiving his $5k, that he should have gotten $5,120 ? He should ask for more. $640k ought to be enough for anybody.

gibibytes?

[ihatethetv]

I got a card in the mail for this yesterday. I personally feel its ridiculous. Sure they're using the industry standard way of marketing hard disks...every other company does the same thing. Why does creative get hit with this? Why not sue apple for it...they certainly have more money to win. Are they going to go on to sue all the hard disk manufacturers too? Are we going to see federally mandated explanations of digital storage capacity? Will they use gibibytes? btw: I've been very happy with my creative muvo^2 mp3 player. It's ugly, but sounds great and still works perfect despite two years of abuse...has a standard dc 5v jack on it so i can continue to use it after the lion dies...

Re:gibibytes?

[anethema]

Cant sue everyone! And this is a start.

Re:gibibytes?

[Enleth]

Well, Apple covered its ass by writing "1GB = 1 billion bytes; formatted size is smaller" on the box.

Except when it doesn't

[davidwr]

When dealing with RAM, for decades KB has meant 1024 and MB 1024^2 and so on.

It's been a bit fuzzier with hard disks, since the hard drive manufacturers have actually gotten away with claiming 1MB=1 million for so long.

If the MP3 players were solid-state, the rules of RAM apply.

When it comes to lawsuits over advertising, the rule of thumb is "what would the average consumer believe."

Well, if you don't count the average non-techno consumer who is only concerned about how many songs or how many minutes of music he can fit, you are left with techno consumers who will look at the device, think "solid state," and assume 1 GB=1024^3 bytes.

If the manufacturers would list the actual usable capacity in bytes they would be in the clear. By "usable" I exclude hardware-overhead for spare blocks and the like but include "filesystem overhead" since the space could in principle be used for user data if the device were used as a raw device.

Re:Except when it doesn't

[phizix]

When dealing with RAM, for decades KB has meant 1024 and MB 1024^2 and so on. Creative Labs should not have to pay for decades-old abuses of SI prefixes.

Re:Except when it doesn't

[jhol13]

If the MP3 players were solid-state, the rules of RAM apply. No, they do not. There is no "raw" device in a MP3 player. Just for an example even my USB memory stick has "only" 4073432 usable 1024 byte blocks.

I would assume that the manufacturers cannot state the actual usable capacity as it vary between devices ("bad" blocks).

The lawsuit is completely idiotic. Well, it would have been anywhere except in the USA.

Re:Except when it doesn't

[anethema]

Every software lists their requirements in base 2 and even your OS shows it in base 2. It has been an SI prefix for a long time granted, but it HAS been industry standard to use base 2 in this industry for over half a century.

I wouldn't say these suits are meritless.

Re:Except when it doesn't

[prockcore]

My OS gets it right. GB = base 10, GiB=base 2. Pretty much all linux tools use SI prefixes.

You're using base-10 numbers when you say "20 gigs".. you should use base-10 prefixes.

I know it's an unpopular viewpoint here, but we programmers are *wrong*. The user should NEVER be exposed to base 2.

Re:Except when it doesn't

[OverlordQ]

And their usage as such has been deprecated for the last decade by the relevant standards bodies.

Your point being what exactly?

Re:Except when it doesn't

[calebt3]

It seems to me that it would be more proper to be upset at the OSes because they have been misrepresenting the size of our drives. Not that anyone could sue any of them over it. Maybe the drive makers can.

Re:Except when it doesn't

[Martin Blank]

nslookup has been deprecated since around 2001, as near as I can tell, and yet it shows no signs of dropping out of use. Its usage still commonly shows up in a very wide variety of training materials for all manner of operating systems. Sure, dig is slowly replacing it, but by the time it's completely gone, it will be because DNSSec has finally come into serious use and nslookup no longer functions sufficiently.

The old terms aren't going away, no matter how much a small group of geeks demands it to be so. They're too ingrained in the population, and that population doesn't like terminology to change once they're used to it. They then pass it on to the next group that they teach (be that children or colleagues), and they become comfortable with it, and the cycle continues.

Attempts to force the change through pedantry are only going to lead to frustration of and ridicule by the general computer-using populace.

Re:Except when it doesn't

[Jesus_666]

But the lawsuits shouldnt't be against Seagate, Creative et al. but rather against Microsoft, Apple and RAM manufacturers. I think the OSes are actually misleading the user by constantly lying about what a (kilo|mega|giga|tera)byte is. Telling the user that a megabyte is 1,048,576 bytes might have been okay before the SI created the binary prefixes, but today it's simply wrong.

And don't tell me that "but we always used the short prefixes" is an excuse. We're the IT industry. Five years ago is to us what fifty years ago is to other industries. We don't write hand-optimized assembler code because that's how we used to do it somewhere in the past - so why should be stick to an obsolete mis-definition of a standard prefix when the issue was fixed ages ago?


I say that someone (maybe Seagate) should slap the OS manufacturers with a lawsuit for telling users that 1 MB is not equal to 1,000,000 bytes and thus causing harm (thorugh unneccessary legal battles) for the hard drive manufacturers. The OS manufacturers arguably are at fault here.

Re:Except when it doesn't

[AdamWeeden]

Not here in Nautilus under Ubuntu Hardy.

Re:Except when it doesn't

[anethema]

I think it is mostly because the SI binary prefixes just sound so terrible. A mebibyte? It sounds like a five year old child trying to speak about computers.

If I said that to my boss at work (I run some backend ISP servers as part of my job) I would probably be laughed at right then and there. Being laughed at in the work place is not exactly something most people crave.

Re:Except when it doesn't

[MBGMorden]

Like it or not, that's what the prefixes were used for. As such, 1GB = 2^30 bytes, by definition. Creative claims 10^9 because they get to claim higher GB numbers for less storage space, using the crutch that "well the definitions shouldn't have been defined that way".

Tough stuff. 10^9 bytes as a measurement of computer storage is a number that doesn't break down into any nice little unit. As close as you're going to get is 0.93 GB.

Contents may have settled during shipment

[ihatethetv]

Right on! Can I sue McDonalds for not filling my "44oz" soda to the top?

Re:Contents may have settled during shipment

[Digestromath]

You can argue that your 44oz cup is always full. It's just a matter of what it is full with (save a vacuum). They could argue that you just got extra carbonation and it's floating on top.

Re:Ughh.. again...

[packeteer]

We don't use the Si"ssy" system in the god damn US of A.

base-10

[dazlari]

Class Action to Creative: All your base-10 are belong to us.

Re:Ughh.. again...

[corsec67]

What is wrong with saying that there are "2 kilomiles" between Chicago and Los Angeles?

Stupid liars

[TapeCutter]

Misrepresented? - I thought this had all been settled in the early 90's and the entire industry knew it was base 10. To emphasise this point can you (or anyone else) find a HDD manafacturer who still advertises in base 2?

I don't think anyone is lying and I'm not sure who's 'stupid' here, is it the courts, the plantifs, or the manafactures? In any case uninformed is a much better word to describe leagal pedants who complained.

Re:Stupid liars

[TapeCutter]

Bad form to answer ones own post, ect, but hairyfeet has pointed out in a reply to another post of mine that Maxtor use base2. Kudos to Maxtor, any others?

Re:Stupid liars

[Nicolay77]

Well, Maxtor used Base10 in the disk I have here, so I don't think that's true, or that's true only recently.

Re:Stupid liars

[TapeCutter]

Hairyfeet claims to work in a repair shop and I have no reason to doubt that, he also said it was a drive he bought recently so it's quite possible that it's a recent thing. Perhaps you could look him up with the link I provided and look at the conversation. ;)

My own experience over some 20 odd years has been base 10 all the way, thus my original challenge for someone to come up with a base 2 example. The fact that he came up with the example in a reply to a different post of mine tends to add weight to his claim. Anyone else got a shinny new maxtor?

So....

[Donniedarkness]

Now that I've bought an MP3 player from Creative, I can get a discount on an MP3 player from Creative...

Deprecated for quite a while now

[Mathinker]

"part of the trade"? The use of metric prefixes for binary powers of two has been deprecated by professional standard bodies for almost 10 years now!

That doesn't mean that I don't agree that the manufacturers shouldn't have to print the full number, and it's representation in both SI and binary prefix units. If they did that, then people like you might start to become aware that the misuse of SI prefixes based on context is stupid.

Re:Deprecated for quite a while now

[Toonol]

Yeah, but honestly, everybody ignores the standard bodies on this issue. Does your computer have 512 megabytes of ram or 537 mb? It's very rare that anybody refers to a memory measurement based on a power of 10, and it's obviously going to be pretty unanimously misinterpreted if printed that way on product labeling.

Printing both clearly would be fine, though.

Re:Deprecated for quite a while now

[hobbesmaster]

RAM has different standards.

Re:Deprecated for quite a while now

[h4rm0ny]

SI prefixes are there to make things simpler. They don't do that in the case of KB, MB, etc. because it has different needs. Only people who lack understanding of what the numbers mean find it confusing, whilst using base 10 is more awkward for those who do know what they mean. Why should technical terms be biased toward those who know less? A very long history of usage determines the meaning of the word and the re-definition came solely from marketing departments deliberately trying to cause confusion to profit from it. Thank about it: if everyone used the same terms, there are no preferential terms you can exploit to make your product sound better than it is, but when there is confusion over terms, you can use the preferential one and rely on people's expectation that it is another meaning. The only profit is in confusion. And then once the hard drive marketing departments had instigated this confusion, it was sustained by people who liked to be able to correct others.

Re:Deprecated for quite a while now

[iamacat]

I don't see one hard drive manufacturer who ignores the standard bodies here. Even if they did, I don't see how confirming to the standard warrants a lawsuit, even if it is to your advantage. Furthermore, the difference between 2^30 and 10^9 is too small to make a realistic difference to potential users, while confusion for users who didn't learn math in a C.Sci class is considerable when powers of 2 are used as a measurement unit.

Re:Deprecated for quite a while now

[SL Baur]

Well I'm sorry, but not really. When you say 1k in the context of computers, it means 1024. When you say 1$ it means something different in every country that uses $ and dollar to denote their currency. USA, Canada, and Singapore to name three.

I find the notations that stupid people have derived to work around the base two versus base ten issue confusing. Better to put it in stark terms. A disk drive with capacity 50GM base 10 is like 50 US dollars. A disk drive with capacity 50G base 2 is like 50 Canadian dollars. Got it?

Re:Deprecated for quite a while now

[SL Baur]

50GM base 10 s/50GM/50G/

typo, sorry.

Re:Deprecated for quite a while now

[Oscaro]

Also, 'b' is for 'bit' (used especially for transmission rates, like 54 kbps) and 'B' is for 'byte'.

Re:Deprecated for quite a while now

[MichaelSmith]

A millibit would be funny, like write only memory.

Re:Deprecated for quite a while now

[CastrTroy]

If it is too small of a difference for most users who don't know the terminology to notice, then why not just use the correct units. People who have no idea what a GB is will go on their way, and those of us who do understand the difference will be satisfied.

Re:Deprecated for quite a while now

[Smidge204]

Furthermore, the difference between 2^30 and 10^9 is too small to make a realistic difference to potential users. 7.3% is not what I'd call a small difference. The difference is only small if you have small capacity to begin with, but once you're up in the "GB" range the difference really starts to show. "1GB" is really only 930MB, and 70MB is still a decent amount of data.

On a 50GB iPod the difference is big enough to hold about six full albums worth of music or a full length movie.
=Smidge=

Re:Deprecated for quite a while now

[OwnedByTwoCats]

Huh? 2^30 is 7% greater than 10^9. It's enough of a difference for the lawsuit...

Re:Deprecated for quite a while now

[Hognoxious]

When you say 1k in the context of computers

This isn't about computers, it's about dippy teens not being able to get all the Britney songs on their mp3 players and wanting a lifelong meal ticket because of it.

Re:Deprecated for quite a while now

[Intron]

"Well I'm sorry, but not really. When you say 1k in the context of computers, it means 1024."

Shouldn't you say that 1k = b10000000000? Saying that using base ten doesn't make sense and then saying that it is 1024 is kind of hypocritical.

Re:It's not geeky, its simply correct

[NonSequor]

If something says a file is n GB then in a just and reasonable world the actual size of the file would always be a 10+int(log n) digit number that is closer to n*10^9 than it is to (n+1)*10^9 or (n-1)*10^9.

I do not think I am unreasonable for expecting this and by extension, I believe that you are unreasonable for suggesting that this should not be so.

slashdot broken... again

[Hans Lehmann]

37 comments, my threshold's set to 3, and only one comment shows up. Hey Taco! What's the deal with the new code? Forgot to beta test it first?

Re:slashdot broken... again

[Tacvek]

No, slasdot is not broken. There was only one post rated 3 or higher at the time of your post. There are only a few more than that at the moment

Re:slashdot broken... again

[mgblst]

When I reply using the new Ajax interface, I find that I no longer get a karma bonus, so all my posts are at 1, where presumably noone reads them. I can't imagine this is just happening to me. (Still have excellent karma, btw).

Re:slashdot broken... again

[Xtravar]

Click on options, uncheck "No karma bonus".

The stupid thing is that the options button under your post is for global options, so it's not just "No karma bonus (for this post)" it's "No karma bonus (until you realize that you aren't getting one anymore and uncheck this box)". That's different than the old functionality, where you could just check it for one post. Poor UI decision...

Just trying to clarify

[Tangamandapiano]

http://www.xkcd.com/394/

Getting sued for profit?

[Paranoid times]

I have some trouble seeing why Creative would be unhappy about all of this. So they pay out a million dollars for court fees (Maybe). But how much do they get back in all that free advertising? And of course there is the detail of they don't pay anything out to most people.

already happen?

[conan1989]

didn't something similar already happen to seagate and WD?

Re:Ughh.. again...

[hobbesmaster]

Yes, we do.

Re:It's not geeky, its simply correct

[Asmor]

You know, I've never actually measured the amount of liquid in a gallon of milk (or water, or any other product, for that matter). It's entirely likely that the unopened gallon of milk in my fridge right now is only .9 gallons, for all I know.

they don't think anything of the kind

[HappyEngineer]

The average non-techy person isn't going to assume 1GB=1024^3 bytes.

They're going to look at the 10GB one and think that it's twice as large as the 5GB model which is 5 times the size of the 1GB model which they currently own.

If they currently have X songs filling up the 1GB model then they'll look at the 10GB model and think "wow! X*10 storage!"

The whole GB vs GiB thing is an argument that is relevant only to geeks and lawyers. The only important thing to average consumers is that all MP3 players be consistent with what a GB is so that they can make a relative comparison.

I personally happen to think that RAM should simply be labelled as GiB instead of GB. You may think it looks weird, but we'd all eventually get used to it just like everyone soon (within weeks) got used to the name "Wii".

Re:they don't think anything of the kind

[shutdown -p now]

The whole GB vs GiB thing is an argument that is relevant only to geeks and lawyers. The only important thing to average consumers is that all MP3 players be consistent with what a GB is so that they can make a relative comparison.

It doesn't have to be consistent only with other players; it also has to be consistent with how OS reports file sizes (and possibly with how online music stores report those). If I right-click on my music folder in Windows to see its total size, find out that it's "10 Gb", go buy a player with such declared capacity, plug it in, and then find out that I can't copy all my music because the player actually "9.9... Gb" as far as Windows is concerned - yes, I would say that would be a valid reason to be annoyed at someone.

Re:they don't think anything of the kind

[HappyEngineer]

Well, you got a 5 while I got a 2, so apparently the people have spoken. *smile*

I still think it's an asinine thing to sue over. Even if someone is confused by this, they won't be confused more than once because all manufacturers of portable devices are going to use the SI standard because the OS standard will make them look smaller.

Re:they don't think anything of the kind

[HappyEngineer]

Windows reports both the prefixed value and the exact number of bytes in the form: 9.29MB (9,749,149 bytes). Perhaps it would be best if, instead of advertising hard drives as 1TB they should list them as 1,123,456,789,012 byte hard drives.

Anyway, I think you're being a bit harsh. It's my understanding that every hard drive ever produced has used metric units to identify the size. There is nothing in the construction of a hard drive which implies that the amount of information on it will be a power of two. Heck, if you want to get into the details I'm sure that a unit system in base PI would be more relevant due to the circularity of the discs.

Knowing this, operating systems were still written to display sizes using the base 2 units. It seems to me that if anyone should be sued it's the operating systems people who knowingly used base 2 units even when ever hard drive was sold using the base 10 units.

Honestly, this argument is ridiculous. Years ago some asses decided to commandeer a few perfectly good metric prefixes rather than invent new ones of their own. What would people say if someone tried that now? Standards are king.

I'm all for sticking it to the Man

[vsage3]

...but these settlements seem to come down to stupidity rather than a legitimate claim. This claim only extends back to 2001; I'm fairly sure I knew about this discrepancy the first time I booted up my first 386 computer in '95 when I noticed that 500 base 10 megs was actually 466 base 2 megs, and that was FAR before 2001.

Mark me redundant, but I just don't feel like these lawsuits represent a good use of the legal system.

Re:I'm all for sticking it to the Man

[Dunbal]

Mark me redundant, but I just don't feel like these lawsuits represent a good use of the legal system.

      Sure they do. They lawyers made money, didn't they?

Re:GB is Base 10

[ghostdancer]

Byte is just another form of measurement for storage.

Commonly, 1 Byte = 8 Bits.

What kind of punishment is this?

[ryanisflyboy]

Let me get this straight. If you got "screwed" (somewhat debatable), you get a choice between 50% off a 1GB MP3 player (so it'll cost you about $30), or you can get a coupon for 20% off at the over-priced company store? What's to stop creative from upping the prices 5% to offset the 20% 'discount'? This isn't a punishment, it's a marketing campaign. You get a better deal going to their sales and clearance sections!

http://us.creative.com/shop/shopcategory.asp?category=720

Go Creative Labs. You must have very good lawyers.

Re:What kind of punishment is this?

[mgblst]

I am no mathematician (actually I am, but it sounds better), but uping prices by 5% still means you are going to get a deal.

Of course, it is the rip-off company store, so maybe not.

Re:Gibibyte.

[Toonol]

Good for you. I'm teaching my children to never use the term, also. Not that they would.

Re:Ughh.. again...

[TypoNAM]

Nearly every programmer, network administrator, and what have you I've talked to in this I.T. industry do NOT recognize IEC has an authority figure over redefining of nearly forty years of well known and understood meaning of byte size modifiers (kilobyte, etc..) as a base of two, not base of ten. It was the marketing departments of various storage vendors that started this confusion mess and for IEC to basically give in to these vendors (I wonder how much it cost them to pay off IEC) and IEC's reason of "confusion" is not founded and they can basically go fuck themselves as for as we care.

That's all I have to say about it.

They should have offered to settle for $10000

[MrSteveSD]

...in base-2 money :)

Well you can argue the OS is wrong

[Sycraft-fu]

The metric prefixes predate OSes by a long time and as the GP pointed out, they are very well established. Computers decided to coopt kilo and use it to mean 2^10 instead of 10^3 since they were close. However now that we are dealing with 2^30 vs 10^9, the difference is quite a bit bigger.

Personally, I think the things like HDs, network gear, and such are correct. We need to use the metric prefixes for base 10 for base 10. If we want to talk base 2, use the base-2 prefixes.

Re:Well you can argue the OS is wrong

[shutdown -p now]

PCs have been around for more than 20 years, so it is firmly established by now that kilobytes, megabytes etc are meant to be in 2^10. The OS may be wrong in a pedantic sense, but it does behave as expected by the end user.

Re:Well you can argue the OS is wrong

[swillden]

The metric prefixes predate OSes by a long time and as the GP pointed out, they are very well established. Computers decided to coopt kilo and use it to mean 2^10 instead of 10^3 since they were close.

Even computers didn't do this at first. The first computers with storage capacities large enough to bother with prefixes used base 10 units. The first ever hard disk drive (IBM 350) had a capacity of five million characters. 5 MB, base 10. That was consistent with the units used by the machine it was built for (IBM 305 RAMAC), which had a 3500-character drum memory, plus a 100-word core memory buffer. Notice, all multiples of 10, not two.

The first computer to seriously use base 2 in its memory sizing and addressing (the IBM System/360, IIRC), didn't show up until ten years and a half-dozen generations of hard drives later. The drives still used base 10 notation for their capacities, though, both because that was the established notation and because there wasn't (and isn't!) any engineering reason to use base two. Unlike inside the computer, where CPU memory addressing makes it very convenient to have power-of-two memory sizes, inside the hard drive there is no reason to use such strange units. Similarly, all data communications technologies measure bandwidth with base 10 units.

The argument that HDD manufacturers label drives with base 10 capacity in order to mislead buyers implies that at some point they did use base two. If not in their published capacities, at least in their internal calculations. That is simply not the case. The logical units for engineers designing HDDs to work in are base 10. They design a drive and then pass the specs on to marketing, in the natural units. I suppose marketing is guilty of not telling the engineers "Wait a minute, it may make sense for you to work in base 10, but modern computers will divide the space up in power-of-two block sizes, so we should advertise it on that basis, to be honest to our customers, even though it'll make our drives look smaller than all of our competitors', since they're using the traditional measure."

The situation is a little less clear for flash memory devices, though, since they're directly addressed by CPUs, just like DRAM, and are therefore usually constructed in power-of-two sizes. Choosing to construct a device sized in powers of 10, against industry norms, so that you can advertise it as 1 GB while saving a few bucks is arguably underhanded.

I think the real answer is quite clear: Device manufacturers should be required to use correct units on their capacity statements, either GB or GiB. It is not beyond the capacity of the average person to understand that these are different, and that GiB is a little bigger than GB. I know the binary units piss off a lot of slashdotters, but, really, they just make sense. I can't figure out how anyone can think it's a good idea to use ambiguous units and then try to guess which meaning is intended.

Re:Well you can argue the OS is wrong

[cpotoso]

Actually, I should have written that I'd give 1 kilo pieces of gold (meaning 1,000, NOT 1,024 you greedy bastards!)... Oh, well. It is better to put the joke late rather than never!

Re:Well you can argue the OS is wrong

[derflammenhund]

Kelvin bytes sounds like it should be a unit of Internet work to me.

Has anyone started to develop ways to quantify how much we're doing with computers these days? If not, I recommend we start soon.

Re:Well you can argue the OS is wrong

[Xtravar]

You aren't used to how things are done around here (the world), are you?

If we say the Queen of England is Kylie Minogue, she better damn well be, because that's who we'll be signing treaties with.

Re:Well you can argue the OS is wrong

[shutdown -p now]

WTF does it have to do with Americans? I'm not an American, I'm Russian. I don't know anyone here who would strongly believe that kilobyte is not 1024 bytes. In fact, most local books of a "Learn to use PC in 7 days" kind explicitly state that kilobyte = 1024 bytes. Metric doesn't have anything to do with it - yes, it was used for a long time for SI units, but byte is not a SI unit! And, for as long as byte was used as a unit of measurement, "kilobyte" has meant "1024 bytes" for the majority of users, like it or not.

Re:Well you can argue the OS is wrong

[orkysoft]

The situation is a little less clear for flash memory devices, though, since they're directly addressed by CPUs, just like DRAM, and are therefore usually constructed in power-of-two sizes. Choosing to construct a device sized in powers of 10, against industry norms, so that you can advertise it as 1 GB while saving a few bucks is arguably underhanded.

Actually, my 1GB SD-card has about 963MiB of space on it. And no, they are not directly addressed like DRAM.

Re:Well you can argue the OS is wrong

[jedidiah]

All these shenanigans have done is shown me that the previous pro-SI
propaganda was all just a big fat pile of bullsh*t and that perhaps
the old school measurments had more practical value after all.

Understanding something and buying into something are entirely different
things.

This wouldn't be the first or the last French scheme that either
the Americans or the British would look at with a reasonable degree
of skepticism.

Infact, the English seem to be suffering from their own form of
metric induced fraud right now. The old shorthand (but "archaic")
quanties were taken away in favor of some 3 digit number that no
one has any intuitive grasp of.

Re:Well you can argue the OS is wrong

[Yvan256]

Flash memory cards (and Flash chips) usually have bad blocks, that's part of the manufacturing process. If you were to buy two identical flash cards and compare their exact available space (in bytes), you'd see different numbers.

Re:Well you can argue the OS is wrong

[swillden]

Actually, my 1GB SD-card has about 963MiB of space on it. And no, they are not directly addressed like DRAM.

The flash memory in the embedded systems I've worked on were directly addressed like DRAM. There was obviously some extra work to be done when writing to them, but reading was dereferencing a pointer.

What the?!

[XMode]

Wait just a second here... Standards are good unless they mean you get less? Kilo is 1000 not 1024.. Yes the computing industry has been using it as 1024 since the beginning of time, but thats incorrect and has always been incorrect. Kilo, Mega, Giga, Terra etc are all base 10, they always have been and always will be.

Do I get so sue because my Linux distro that I just hacked up uses K=10000 instead of K=1000 so all the drives are an order of magnitude smaller on my OS..

Re:What the?!

[Dunbal]

Wait just a second here... Standards are good unless they mean you get less? Kilo is 1000 not 1024.. Yes the computing industry has been using it as 1024 since the beginning of time, but thats incorrect and has always been incorrect.

      But isn't it the industry that gets to define the "standard"? After all, what exactly is a "horsepower"? What if my horse is stronger than your horse? Who has the right horse? Then someone comes up with something completely new - called "brake horsepower" and decides to use a pony as a horse... what does it all mean then?

      Kilo in computer terms has always meant the closest approximation available in powers of 2, because digital things can only have 2 states - on or off. If memory chip producers can get it exactly right - to the bit, how come hard disk manufacturers get a free ride?

Re:What the?!

[AmigaHeretic]

>>Kilo in computer terms has always meant the closest approximation available in powers of 2, because digital things can only have 2 states - on or off. If memory chip producers can get it exactly right - to the bit, how come hard disk manufacturers get a free ride?


They could do one of two things. Make it so the HD actually make the HD a little bit bigger and make it base2 size like ram (since that is how computers store data GOD FORBID) OR they could put the ACTUAL size that we are going to SEE.

My DATA is binary. When my I have a 4GB file on my HD it is not 4 billion BYTES!!!!!! I can't fit 100 4GB files on a 400GB drive because they are measuring base10!! It's not right.

Re:What the?!

[Jesus_666]

But isn't it the industry that gets to define the "standard"? After all, what exactly is a "horsepower"? What if my horse is stronger than your horse? Who has the right horse? Then someone comes up with something completely new - called "brake horsepower" and decides to use a pony as a horse... what does it all mean then?

"Horsepower" is a new unit. The car manufacturers didn't decide to use "joule" in a nonstandard way, the defined a new unit. I'm fine with that as it avoids confusion. When you make up a new unit you can be as arbitrary as you want as long as your unit doesn't interfere with an already established one.

Kilo in computer terms has always meant the closest approximation available in powers of 2, because digital things can only have 2 states - on or off.

Except when you're talking about frequencies. Or bandwidths. Or just about everything that isn't a RAM chip or a file. Don't create the impression that the binary "kilo-" is universal in IT; it's only relevant in a few cases. Certainly not enough to decide that we can arbitrarily overrule the SI.

Re:What the?!

[JesseMcDonald]

Kilo in computer terms has always meant the closest approximation available in powers of 2, because digital things can only have 2 states - on or off.

Except when you're talking about frequencies. Or bandwidths. Or just about everything that isn't a RAM chip or a file. Don't create the impression that the binary "kilo-" is universal in IT; it's only relevant in a few cases. Certainly not enough to decide that we can arbitrarily overrule the SI.

One problem with this line of reasoning: the byte is not an SI unit of measurement. SI prefixes only apply to SI units; outside of that domain they have no standard interpretation. Or you you often complain that microcomputers aren't exactly 1/1000 the size of minicomputers?

Face it, bytes were consistently measured in powers of two right up until the HDD manufacturers decided to get creative with their marketing. Compound units (e.g. bytes/second) and pure SI units (e.g. Hz) follow the SI base-10 prefix scheme, and certain external storage formats have since adopted the non-standard HDD storage units, but precedent is solidly in support of kilobyte meaning 2^10 bytes, megabyte 2^20 bytes, etc.

Re:What the?!

[Jesus_666]

In that case we shouldn't mix SI prefixes with non-SI units. When we use an SI prefix we tell people that our unit behaves like an SI unit as far as prefixes are concerned. It doesn't matter if we mean something entirely different; that is what we tell them. If we want the prefixes to mean something else we use different prefixes.

As for microcomputers: "Microcomputer" is not a unit of measurement. Nobody expects it to behave like one, much like nobody expects a microscope to be one millionth of a scope. And it tells a lot about your knowledge of the SI system of prefixes that you can't differentiate between "mini" and "milli".

Re:What the?!

[Lehk228]

Kilo in computer terms has always meant the closest approximation available in powers of 2, because digital things can only have 2 states - on or off. If memory chip producers can get it exactly right - to the bit, how come hard disk manufacturers get a free ride?

unless of course you are doing something as daft as transferrring information over a network, measuring the clock frequency of anything, or outputting tones to a sound card, in which case the computer uses base 10.

Re:It's not geeky, its simply correct

[gardyloo]

In a just and reasonable world you'd specify a base-10 log there, rather than leave it ambiguous :)

Re:SI

[Reziac]

And this isn't the 2nd time it's happened; it's the third. Western Digital got bit by the first (that I know of) of these stupid class action lawsuits.

In the case against W.D., the plaintiff was asking something like $8 MILLION dollars (which was more than W.D.'s income for that year; such an award would have put them out of business). Fortunately the judge saw through it, awarded something like $250k in legal fees (half what the lawyer was asking just for HIS "expenses") and a few grand to the main plaintiff. (Members of the "class" got a free download of some backup software.) And the judge then severely lectured the plaintiff's lawyer for being a sleazebag.

W.D. now carefully labels HDs with both binary and decimal units.

As to the so-called "merits" of these cases -- since when do computers run on anything but binary? Why is an industry standard notation "cheating" the customer?? Why are YOUR CHOSEN OS's filesystem's formatted capacity and filesystem overhead the MANUFACTURER'S problem?? These are the points that the class action plaintiff, and MOST of the posters here today, are wilfully misunderstanding.

In these types of class action suits, the result is always the same: the lawyers make out like bandits, ONE person (the primary plaintiff, I forget the legal term) gets some substantial part of the "award", and ALL of the other "class members" wind up with something trivial, like a coupon to buy more of whatever.

And then prices go up on all of the company's other merchandise, because the shortfall caused by having to pay out the "award" has to be made up from SOMEWHERE. And that means more money out of OUR pockets.

Re:Ughh.. again...

[MadnessASAP]

yes but bits are different then bytes, The whole system for referring to sizes, speeds and capacities in computers is horribly messed up. bits, bytes, base-2, base-10, base-16, MIPS, FLOPS, Hz it can all get a bit overwhelming for people who don't have a fair amount of knowledge about the differences.

Breath of fresh air

[LilGuy]

One of the very few times a car analogy wasn't the first choice on slashdot.

Re:It's not geeky, its simply correct

[Bill, Shooter of Bul]

That may depend upon you local laws. I'm pretty sure in the us, we buy our milk by the quart/gallon (although the legal measurement is mL/L) and have a whole scientific agency dedicated to weights and measures. They really do have to sell the stated quantity of milk (and not the volume of the container) within a certain tolerance. Otherwise, they very well could be sued -- just like creative.

Re:OS's fault

[Psychotria]

I don't think it's the OS programmers fault. All OSs I use do convert, but what is the use of that after you've bought a product?

Re:SI

[pembo13]

And outside the USA, high schoolers are taught that data sizes in computing are based around base-2. I would assume people within the USA are also capable, but I can assure you that those of us outside the USA are fully capable of maintaining that concept.

Rough vs finished

[cheros]

I bet the ones you bought are finished product. They do start out rough as 2x4, though :-)

My Ruler shows 8"

[AmigaHeretic]

Sure I could use my own "special" ruler that shows it's 8", but once I stick it in I'm going to get the same complaints the HD manufactures are starting to get now and everyone will know the real size in a matter of seconds. But hey, by then it will already be in, that's all the HD companies care about, right???

All that matters is the real size and that is what shows up in the OS (all OSes since the beging of time) and how much "binary" data I can put in it. ( I don't know if any deciaml data out there )

Re:My Ruler shows 8"

[Ant P.]

Please, don't give them any ideas! We don't want drives measured in terms of how many binary-coded-decimal digits they can store... it's bad enough with POSIX.

What?

[mosb1000]

The plaintiff is the one who is wrong, any idiot who knows anything about numbers should know that. A gigabyte is exactly one billion bites, hence the name "Giga". How can you win a case centered around a claim that is so obviously false?

Re:What?

[AmigaHeretic]

>> A gigabyte is exactly one billion bites, hence the name "Giga". How can you win a case centered around a claim that is so obviously false?

Simple question then. Why don' they lable HD's in the "ONLY" way they can be read by a computer then. i.e. binary?? GB vs. Gibi, fine. Computers can't read Gigabytes then. So why the hell would they label it a size it can't understand?? Can you explain?

Re:What?

[pelrun]

It's like going to buy a TV, but you get home to find that the seller didn't sell you a television, but the letters T and V printed on a bit of paper in a box. Sure, technically they described what it was correctly, but they're still ripping you off.

Re:What?

[karmatic]

A gigabyte is exactly one billion bites, hence the name "Giga".

Um, no. A kilobyte is exactly 1024 bytes. It's "close enough" to 1000, hence the term "kilo". Why bother making up a whole bunch of new prefixes when there's one that already exists, especially given that it's blindingly obvious that the original meaning makes no sense in context?

Since measurements are conventions of man to begin with, they mean what people define them to mean. The original definition of kilobyte was 1024, since 1000 bytes is an largely useless number in base 2. If we were working in base 10, it would make sense. Terms are redefined based on context all the time - after all, what does a "metric tonne" have to do with pounds, anyway? As in this case, there was a term (ton) that was "close enough" to what they wanted, so they used it.

Once they were (wrongly or not) used and understood to mean 1024, saying "kilo always means 1000, and can never mean anything else" is senseless pedantry.

Creative (and other companies) attempted to redefine a commonly used and understood term, and did so for the purpose of making their product appear to be larger than it actually was. It was not done because it provided a more meaningful, or useful measurement. It made the companies that did it look bigger than the companies that didn't, and everyone decided to play "keep up with the joneses".

Re:What?

[kaizokuace]

better yet, You buy an HDTV that says its 720p but turns out its only 700 pixels high! Or something stupid like that resulting in less pixels for your buck.

Re:What?

[vrt3]

kilo is 1000 in any scientific or technical context and it has been that way since long before the introduction of computers. If decimal prefixes make no sense in a specific field, fine, then don't use them. But don't create confusion by using with the wrong meaning. Imagine that a kilogram is 976 gram, a kilometer is 988 meter, a kiloNewton is 981 Newton, a kilovolt is 1011 volt. Not very practical. The beauty of the system is that kilo means thousand, independent of the unit that follows it. That makes, for example, 1 km * 1 N = 1 m * 1kN. 1kB != 1000 bytes is a serious wart in the system, a mistake that we computer people never should have made. But we did, and now we should try to correct it as soon as possible.

I've never understood why exactly in the field of computers, where the tiniest error in input can make the difference between a system working and not working, people are so sloppy with the meaning of terms.

Now I know that the hard drive manufacturers use the decimal prefixes because it makes their drives look bigger, but it's wrong to blame them for doing the right thing, even if they do it for the wrong reasons. It's not the manufacturers who redefined a commonly used and understood term; it's the computer people who did so.

Re:What?

[Fourier]

Um, no. A kilobyte is exactly 1024 bytes. It's "close enough" to 1000, hence the term "kilo". Why bother making up a whole bunch of new prefixes when there's one that already exists, especially given that it's blindingly obvious that the original meaning makes no sense in context?

Base-2 measurement does not offer any particular advantages when quantifying the storage of magnetic media; the capacity is a function of drive geometry and bit density, and has nothing to do with the number of logic gates on the board. So "in context" it is not blindingly obvious what number should be used.

Because context is not always enough to resolve ambiguity, reusing the SI prefixes was clearly a mistake. This is why we have binary prefixes.

Re:What?

[cpotoso]

Your post is so full of nonsense... There are quite a few replies stating the obvios (kilo = 1000, NOT 1024, is the well established standard). Of course you have to continue with your non-sense by writing:

Terms are redefined based on context all the time - after all, what does a "metric tonne" have to do with pounds, anyway?

Precisely not much. In fact, you will find that 90% of the world does not use "metric tonne" but simply tonne or ton (without distinction) and it means 1000 kilograms. I guess that the trouble with the kilobyte and other nonsensical definitions really arise (once again) from the americans' (I'd bet the kilobyte definition was invented here in the us) disregard for standards accepted by everyone else: metric system, that torture is not acceptable (in civilized societies), death penalty (ditto), evolution vs. creationism (50% of americans), respect for international agreements, and the list goes on and on. Sure, it used to be that the US was the only major player in town, so why bother with anyone else (no longer true, as seen lately on many many fronts), and the difference between 1024 and 1000 was not that horrible (but at the GB or TB level things get comparatively worse). Sloppy standards make for nasty results at the end of the day.

Re:What?

[ps236]

Not quite

With Creative (and other HDD manufacturers), you totally get what you pay for. You just don't get what you THOUGHT you'd paid for. There is a difference.

As other people have said, RAM is about the only thing where it makes sense to measure it in 2^n numbers. That's because it is nowadays generally addressed using a certain number of binary address lines, so it will nearly always going to be a power of 2. Hard disks on the other hand can be any number of 512 byte sectors. There is no binary addressing involved, so using standard 10^n numbering is best. I suppose you COULD sue Microsoft & Linus for displaying file sizes in 2^n sizes rather than 10^n sizes - that would make more sense..

If HDD purchasers can't be bothered to find this out, it's not the HDD manufacturer's fault.

BTW, How big is your car engine? A lot of '2 litre' engines are just 1968cc, and so on. Are you going to sue Ford?

Re:What?

[kaizokuace]

my car's engine is around 3150cc but thats cuz its bored and stroked and a lot of head work. Is it possible to "bore and stroke" an HDD? Thats why computers arent cars. I understand your point tho HDD manufacturers need to be for open with this base 10 labeling. I always feel jipped when I buy a new HDD and it has like gigs missing. I understand their marketing dept likes to make sure they can write whole large numbers for the capacity on the box and that the company wants to make more money by lowering production costs. But c'mon at least don't take that feeling of having an even number of gigs! How many OCD types have had their lives ruined by these crazy HDD companies!

Re:What?

[Palinchron]

No, your 640GB drive stores 640GB AKA 596GiB of files. It's just that your crappy operating system is misleading you by calling it 596GB.

Re:SI

[pesc]

... or how many bytes does one second of a 128kbit/s MP3 song occupy on a Creative MP3 player? :)

Re:Double Standard

[AmigaHeretic]

>>Gigabytes ARE base-10... Gibibytes are base-2.

I use Windows 98. Gibibytes didn't even exist until after the year 2000. So I am granfathered in. The HD manufactures screwed me!! I didn't even know about Gigbytes until just now! Where's my lawyer!!

Re:It's not geeky, its simply correct

[meringuoid]

How would you feel if you bought a 2 litre bottle of milk, only to find out that the farmer actually filled it with 1.9 litres of milk becuase "he counts litres differently to the normal method". Wouldn't stand up would it?

I'd be upset at the farmer, until I discovered that the manufacturer of my measuring cylinder had actually calibrated it in llitres instead of litres, but labelled it litres anyway.

Yes Virginia

[wiredlogic]

Yes Virginia. The world really is run amok by idiots. We have an SI standard for this nomenclature now. No matter what idiot lawayers want to argue they can't deny the fact that GB is defined for base-10 usage and GiB is base-2. It isn't that difficult to get a grip on by anyone except the hordes of innumerate Americans. The days of being able to lazily apply base-2 counting because "it's only off by 1024KB or 1024MB" are long gone when the difference can be hundreds of MB nowadays. It's unfortunate that the storage industry is being punished for doing the right thing and properly declaring capacities in conventional units in accordance to international standards which the US has ratified as the official means for measurement.

Who wants to repeat this in the EU?

[say]

Bah:

You are a member of the plaintiff class if you purchased in the United States from a retail store in the United States (including Creative's and others' on-line retail stores) a new Creative brand hard disc drive MP3 player between May 5, 2001 and April 30, 2008.

Does this mean...

[Nozsd]

we can stop using those ridiculous kibi, mebi, and gibi prefixes?

Re:Does this mean...

[Jesus_666]

I say we also get rid of our ridiculous "byte" units. "Meter" sound much nicer and that way we can sell MP3 players with capacities of 10 kilometers.

"Misrepresented", really ?

[this great guy]

Everytime a base-2 vs. base-10 debate takes place, I like to demonstrate that, contrary to common belief, even in the computer industry, power-of-10 prefixes (as used by hard disk manufacturers) are more commonly used than power-of-2 prefixes. As seen in the list below, power-of-10 prefixes apply to things varying from bytes to bits to Hz to FLOPS to baud. They also apply to many areas (not "only bitrates", as some claim): storage capacity, clock frequency, stream bandwidth, baud, pixel numbers, data throughput, processing power, etc.

• A 4.7 GB single-layer DVD is 4.7 * 10^9 byte (power of 10)
• A 50 GB dual-layer Blu-ray Disc is 50 * 10^9 byte (power of 10)
• A 2.5 GHz processor is 2.5 * 10^9 Hz (power of 10)
• A PC3200 (as in 3200 MByte/s) memory stick is 3200 * 10^6 byte/s (power of 10)
• A 25.6 GFLOPS CPU core is 25.6 * 10^9 FLOPS (power of 10)
• A 128 kbit/s audio stream is 128 * 10^3 bit/s (power of 10)
• An 8 kbaud V.92 modem is 8 * 10^3 baud (power of 10)
• A 6 Mpixel digital camera is 6 * 10^6 pixel (power of 10)
• A 4000 MB/s HyperTransport link is 4000 * 10^6 byte/s (power of 10)
• A 480 Mbit/s USB2 link is 480 * 10^6 bit/s (power of 10)
• A 2.5 Gbit/s PCI-E lane (after 8b/10b encoding) is 2.5 * 10^9 bit/s (power of 10)
• A 250 MB/s PCI-E lane (before 8b/10b encoding) is 250 * 10^6 byte/s (power of 10)
• A 1 Gbit/s ethernet card is 1 * 10^9 bit/s (power of 10)
• A 54 Mbit/s 802.11g network is 54 * 10^6 bit/s (power of 10)
• A 3.0 Gbit/s SATA link (after 8b/10b encoding) is 3.0 * 10^9 bit/s (power of 10)
• A 300 MB/s SATA link (before 8b/10b encoding) is 300 * 10^6 byte/s (power of 10)
• A 6 Mbit/s DSL line is 6 * 10^6 bit/s (power of 10)
• A 2 GByte USB flash drive is 2 * 10^9 byte/s (power of 10)
• Curiosity: a 1.44 MByte floppy disk is 1.44 * 1000 * 1024 byte (mix of power of 10 and 2)
• And of course, a 750 GByte hard disk drive is 750 * 10^9 byte (power of 10)

People have this misconception that binary prefixes "are more common" and "should be the norm", but the only few places where they are used are to refer to RAM capacities and file sizes. I am surprised Creative's lawyers weren't able to explain this.

Re:"Misrepresented", really ?

[inasity_rules]

A PC3200 (as in 3200 MByte/s) memory stick is 3200 * 10^6 byte/s (power of 10)

One Byte being 8 bits... Or two nibbles..

Also note that RAM is measured in powers of 2 -due to the way the processor accesses it. It would be ludicrous from both the technical and consumer point of view to do otherwise; it is something that can not be changed. So why not change what we can (e.g. hard drive sizes) change to be consistent with what we cant change?

In any case Operating systems use use powers of two to measure disk space for the same reason we use powers of two for RAM: it makes more sense. Did you not realize that disk space is allocated in a minimum of a 512byte (2^9) size block?

Off topic

[KlaymenDK]

First of all, lumber is not my business. Second, I did hold an apparently inaccurate belief that 2x4's were indeed 2" by 4".

Pray tell, where can I find a reference or explanation of the true dimensions? I'd be very interested!

Re:Off topic

[threephaseboy]

Wikipedia to the rescue

Re:Off topic

[KlaymenDK]

Ah, great. :) I had no idea what search term to input, my best bet would have been Random Page! ;-) Thanks!

Your claim is incorrect.

[mosb1000]

Hard drive manufacturers have always used the term gigabyte to mean one billion bytes. Your (implied) claim that they haven't is completely out of touch with reality.

Since it is the convention across all mass storage manufacturers, the claim of false advertising has no basis in reality, and is completely incorrect in every meaningful way.

Re:Your claim is incorrect.

[mosb1000]

A kilobyte is not a gigabyte, and this lawsuit is over gigabytes.

Re:Your claim is incorrect.

[Kjella]

Hard drive manufacturers have always used the term gigabyte to mean one billion bytes. Yes, and they've always been wrong. A kilobyte is 1024 kilobytes. Deal with it. If they've "always been wrong", they've always been right since they would have an unbroken line of use dating back to the creation of the SI units which is far far older than any computer scientists that decided to abuse it.

Re:Your claim is incorrect.

[richlv]

they have always done what ?
surely they haven't. i remember the outrage when it was first discovered that some of them started marking drives with 10 base units (or even mixed 2/10 base units when calculating size).
granted, that was at the time when drives were still measured in megabytes. yes, megabytes.

Re:Your claim is incorrect.

[mariushm]

No, they have not. Initially, drives were using Base 2.

See for example the image with the drive on this page:

http://www.pctechguide.com/tutorials/HardDrive_Reasons.htm

or the hard drive on the left in this image:

http://www.divideby0.com/photos/vaio-f280/pics/DSCN1323.JPG

Re:Your claim is incorrect.

[prockcore]

i remember the outrage when it was first discovered that some of them started marking drives with 10 base units

no, you remember when you first figured it out.

The very first harddrive by IBM could store exactly 5 million characters. Not 5,242,880, but 5,000,000.

Base-2 scumbags deserved to lose... whos next?

[Ed Avis]

Im glad the lying scum at Creative got what they deserved. Next, time to sue those wicked Ethernet card manufacturers claiming one gigabit per second. Whos with me?

Re:It's not geeky, its simply correct

[red crab]

Looks nice that at last somebody has taken notice of this. Nobody bothered when storage sizes were small. So if 1 GB was pared down by a few bytes by the 1000/1024 calculation, it didn't matter much. But when 40 GB actually came down to near 38 GB and 80 GB to 76 GB, you felt sort of cheated. Rounding off, if necessary, should always done to the lower value. For instance, my old P-II processor machine actually ran at 352 MHz but its advertised speed was 350 MHz.

SD/Flash/Whatever memory cards next?

[Ilgaz]

I don't say people should sue them but we gotta find a way to stop Flash card manufacturers using the same trick.

Their sizes are much more lower such as MicroSD card (2GB) I purchased for my Nokia E65. Device says: Capacity=1920 MB . I wonder if that high end card manufacturer would go out of business if they shipped it as 2048 MB. They already sell it more expensive and people like me who got horrified already by the filesytem in use (FAT16!) buys them to make extra sure. At least to get spared from bad memory block problems.

If you didn't know already, Memory manufacturers started same trick via SD business.

It has to be said..

[goombah99]

...Less space than a Nomad, Lame.

Now are those Imperial Gigabytes or Metric Gigabytes.

And when they say 50% off do they really mean 48.8%?

Re:It has to be said..

[electrictroy]

What exactly did they misrepresent?

Giga == base 10
Gibi == base 2

If a customer assumed that 100 gigabytes actually meant 100 gibibytes, then isn't that the customer's fault? I say yes. giga == base 10. gibi == base 2. Not the same thing.

Re:It has to be said..

[yog]

precisely. They did NOT misrepresent the storage size; in fact, they represented it very precisely.

Presumably the trial will take place in West Texas, where similar litigation almost always takes place and goes in favor of the plaintiff, like that bogus lawsuit against Toshiba a few years back.

Re:It has to be said..

[jedidiah]

No, they just used the same sort of fraud you're perpetrating right now.

They ignored how a word is used as a term of art in a particular context.
They ignored this context and decades of history surrounding it. They did
so to exploit a non-existent "ambiguity".

There's probably some sort of legal idea that when such an "ambiguity" exits
that it goes against the person making the representation or contract.

IOW, you can be self-serving when you try to abuse language.

No we get stuck with some pseudo-metric quantities that can't even be "divided by ten" in the relevant framework.

Re:It has to be said..

[yog]

No, they just used the same sort of fraud you're perpetrating right now.
Christ, man, take a deep breath. You're elevating the use of a numbering base to a crime. Hardly a credible basis for a lawsuit in a sane society. We're going to sue you for representing the size of your hard disk as a power of 2 rather than a base 10 quantity--seriously? What part of gigabyte do you not understand? This brings to mind that lawsuit against Apple a few years ago over, what, the number of color bits in their display, because they used interpolation or something?

This kind of lawsuit has nothing to do with the consumers, who generally neither know nor care about such trivia, and everything to do with some crafty lawyers who sense an opportunity to make a quick buck. Your attempt at a rationalization of the plaintiff's point of view merely demonstrates that you are not an impartial observer and that you have little regard for the impact of such frivolous suits on the industry and on society.

In a lawsuit-happy nation, where every little thing becomes an actionable offense against someone out there, we are quickly devolving into an overly cautious, fearful, politically correct stodginess that will melt down in the face of less litigious groups like the Chinese and Japanese. You can take that to the bank, and if you don't like it, sue me!

Re:It has to be said..

[Z34107]

"Gibi" is a prefix invented by Wikipedia. For those of you who have been foiled, I supply a conversion chart.

Some people are angry that their precious SI prefixes were usurped. I'd say "understandably angry," but I'm afraid it's not. Memory has been measured in kilo-mega-giga-tera-et al. since at least the time that IBM made PCs, and probably since 5000 years ago when Adam and Eve rode dinosaurs to church every Sunday.

Case in point: Go to newegg.com's memory page. See any memory modules sold by the "gibibit"? Consult your motherboard manual; I doubt they'll support a 512 mebibit SDRAM stick, but it maybe, just maybe might support that 512 MEGABYTE module.

Gibi? Might as well measure memory in millionths of a square furlong chip area times a density coefficient.

Remember booting any computer made since the '70s? The BIOS POST would always report memory in "K" - which God^H^H^HIBM did not intend to mean metric kirbybits or whatever nonsense.

Moderators, I humbly suggest modding any "gibi" references as "troll." It's what's right for America!

Re:It has to be said..

[Lost Engineer]

They ignored how a word is used as a term of art in a particular context. No, they didn't. Hard drives and the like have had capacities measured in multiples of 10 bytes as long as I can remember, which is long enough to establish a precedent.

Re:It has to be said..

[Dolda2000]

Some people are angry that their precious SI prefixes were usurped. Pray tell, did you at no point in your fired rage stop to think that some people may actually be angry (or merely annoyed) at the fact that units like kB, MB and GB mean completely different things depending on who says them? If so, did you, by any chance, also consider that it is an issue that may actually be fixable by calling the two different uses of them by different names? Like, say, "GiB"?

Sure, I may agree that "Gibibyte" sounds really ridiculous, and that whoever invented the name may have been better served by naming it slightly differently, but I don't see there being anything wrong with at least using the acronym "GiB" instead of "GB". God knows I do when I write programs (except, of course, when I actually mean 1e9 bytes).

Re:It has to be said..

[Z34107]

I wasn't going for "rage", more of "flamebait" + "funny". (I got informative!) I was hoping "kirbybits" and "dinosaurs" would peg people's sarcasm meters. (Why did I get informative?)

We only need one system: 1 KB = 1024 bytes. 1 MB = 1024 KB. 1 GB = 1024 MB. 1 TB = 1024 GB. And 1 TB ought to be enough for anyone ^.^

That's the way it's always been, which of course is in no way a logical fallacy. But, those units had their own well-understood meanings within the technology community until hard drive manufacturers out to make a buck supplied their own meanings.

The duplicity of the One Prefix comes from storage manufacturers trying to gyp consumers - and guess what, they're getting sued for it! The system works.

Re:It has to be said..

[Dolda2000]

We only need one system: 1 KB = 1024 bytes. 1 MB = 1024 KB. 1 GB = 1024 MB. 1 TB = 1024 GB. And 1 TB ought to be enough for anyone ^.^ While I may agree that we only need one system, the fact is that we have two systems. And not only that, but you can't really say that people are in any way "wrong" when they all they do is use the SI prefixes in the exact same way that they are being used by everyone else outside of the computer industry.

Thus, since we have two systems (whether you want it or not), and one of them is the actual correct usage of the term in question, would it not be quite reasonable to invent new terms for the new system?

Re:It has to be said..

[Z34107]

I don't think so. I'd advocate "consumer education" - maybe posters about "1024 not 1000!" and class-action lawsuits. ^.^

I don't like the 1000^3 system used by marketers to inflate drive sizes, and I don't want to grant legitimacy to it gimping the name of the "true" system. If we're going to be pendantic (no, I don't know what that word means) force Create to use "imperial decabytes." As long as consumers know that Seagate is gypping them out of ~93GB on their terabyte drive.

Re:It has to be said..

[flibuste]

"Memory has been measured in kilo-mega-giga-tera-et al. since at least the time that IBM made PCs".

This is not exactly accurate.
The "K" as in "Kb" was never supposed to mean "kilo" (1000 units) but really "1024 units". Historically the k was an unfortunate/poor choice of terms to designate "1024 bytes". And you normally don't say "Kilo-Byte" but "K-byte" if you really want to point out the difference.

This has been lost over the time with computers going to the general public. As for mega, tera and the like, it's just an extension of that poor choice of words/letters.
At the time when people used to say "K-bytes", 640K of them were considered enough for anyone (Hi Bill!). The usage of mega or tera came later as computer capacities increased. And the mistake got mainstream .

As for Creative, it really looks like they were trying to play with the confusion and mislead consumers.
Or maybe it was a mistake made by someone who should have known MUCH better.

Re:It has to be said..

[dublin]

"Gibi" is a prefix invented by Wikipedia.

I call bullshit! Check the /. archives (around 2000/2001, IIRC), and you'll find that this was led by a real standards organization to avoid confusing the correct base-10 prefixes with their binary-overloaded counterparts. (I think it was IEEE, since this sounds too useful to have been a Euro-weenie ISO endeavour...)

Creative, Seagate, and the others were correct - mega means 10^6, and giga means 10^9. Tort lawyers and hopeless computer dweebs (who don't even know you start counting at *one* for crying out loud!) don't get to redefine numbers just because they say so. This case was wrongly decided, it's too bad they decided it was easier to capitulate with coupons than actually fight this crap...

Re:It has to be said..

[eldepeche]

So could they get sued for misrepresenting the size of their flash-based players?

Re:It has to be said..

[JLF65]

You clearly have a VERY short memory. Hard drives were measured in base 2 units for more than a decade before switching to base 10 units. It wasn't until drives were more than a gigabyte in size that they switched. In fact, the first lawsuits over the switch were what prompted companies to produce that abortion known as kibi/mibi/gibi in an effort to fend of further lawsuits. That very switch is why the companies are losing their cases today. The precedent is the other way around.

For many decades, kilo, mega, and giga had VERY WELL UNDERSTOOD MEANINGS to the computer scientist. It was MARKETING by greedy corporations that introduced the "confusion" in an effort to rip people off. Trying to force programmers to use those asinine terms they invented to try to get around fraud laws is an insult to any programmer over the age of twenty.

Re:It has to be said..

[Palinchron]

We only need one system: 1 KB = 1024 bytes. 1 MB = 1024 KB. 1 GB = 1024 MB. 1 TB = 1024 GB. And 1 TB ought to be enough for anyone ^.^ The problem is that that system conflicts with the even more respected SI system, which states that the k, M, G, and T prefixes mean 1e3, 1e6, 1e9, and 1e12, respectively, NO MATTER THE POSTFIX.

Re:It has to be said..

[Anonymous Coward]

Gibi is not invented by wikipedia, it's proposed by IEC: http://physics.nist.gov/cuu/Units/binary.html. The SI prefixes are not usurped and never will be.

K doesn't always mean 1024 even in computers. When you say the computer in the 70s booted in K, I don't think you remember right. There's some huge history I read a while back, you can google if you want.

Re:It has to be said..

[Pooua]

The "K" as in "Kb" was never supposed to mean "kilo" (1000 units) but really "1024 units".

I have a lot of electonic textbooks from over the last 30 years that say otherwise. The same with my college and trade school work through the '80s and early '90s. I never saw any sign that "k" in digital electronics meant anything other than "1024" until the late '90s, when hard drive manufacturers started chirping about "1 gigabyte = 1,000 megabytes." It was never an issue prior to the introduction of "gigabyte" hard drives. What is more, the computer science classes that I am taking now still refer to the SI prefixes as powers of 2.

Re:It has to be said..

[tsm_sf]

Yeah, STFU and code you stupid kids!

Re:It has to be said..

[tsm_sf]

Pray tell, did you at no point in your fired rage stop to think that some people may actually be angry (or merely annoyed) at the fact that units like kB, MB and GB mean completely different things depending on who says them? It's because the conceit was dreamed up by the marketing drone, the programmer's natural enemy.

Re:It's not geeky, its simply correct

[Leafie]

Run that gallon thing by me again:

1 UK gallon = 1.03205 US gallons

How can Creative be so stupid?

[freaker_TuC]

Or smart? Either this is a nice marketing ploy to sell more products with 20%/50% reduction;

Or they are really stupid not forwarding this case towards the real con-artists, the HDD manufacturers printing the wrong information on their products?
This playing-with-numbers-charade could end right now if they add a few large HDD manufacturers to the process.

Read your references

[Mathinker]

Did you even bother to read the article you linked to (in its entirety)? It quotes the standard (emphasis mine):

The definitions of kilo, giga, and mega based on powers of two are included only to reflect common usage. IEEE/ASTM SI 10-1997 states "This practice frequently leads to confusion and is deprecated." Further confusion results from the popular use of a "megabyte" consisting of 1 024 000 bytes to define the capacity of the familiar "1.44-MB" diskette. An alternative system is found in Amendment 2 to IEC 60027-2: Letter symbols to be used in electrical technology â" Part 2

Re:Read your references

[fbjon]

The fact is after nearly ten years IEC has failed to get their standard adopted by the majority so it loses. You're right, it loses, but it shouldn't. There's no logical reason whatsoever why we should use powers of two, except tradition. If it causes confusion and isn't useful, why keep doing it?

It seems a lot of geeks try to "defend" using powers of two, as if it were somehow the "correct" way, without thinking whether or not it is really correct and logical (*).

(*)Yes, it makes sense for memory, but nobody's confused about memory either.

Re:Read your references

[CastrTroy]

And as far as Memory goes, we should all follow Nintendo, and go with "blocks" which is ambigious enough that nobody could ever sue them. I don't understand why they just can't equate blocks with KB, or MB, or something else that's remotely useful, instead of 256 KB.

Re:Read your references

[koollman]

How can it makes sense for memory but not disks ?

On my computer, disks sectors are 512 bytes, and the most commonly used memory block size is 4096 bytes. which is also the block size of my fs. now, what happens if the blocks on disks and in ram are not multiple of each others ?

Should I use non-aligned storage in ram when reading the fs or use non-aligned blocks on my hard drive?
And how should I calculate the hard drive cache size ? with powers of ten ? And how about DMA ?

The point is : connected pieces of hardware should use the same basic units. and since it really makes sense to use power of two for some of these pieces, these basic units really should use powers of two.

Re:Read your references

[rah1420]

we should all follow Nintendo, and go with "blocks"

If we're going to go down that road, why don't we measure disk space in tracks and cylinders, like the IBM mainframes I work on? :)

Re:Read your references

[binaryspiral]

we should all follow Nintendo, and go with "blocks"

If we're going to go down that road, why don't we measure disk space in tracks and cylinders, like the IBM mainframes I work on? :) Screw that... let's go with KLOCs and nibbles.

Re:Read your references

[fbjon]

I don't see how any of that relates to the common use of MB, GB, and so forth, which is used and free disk space.

Re:Read your references

[smallfries]

OK, I'll take that challenge.

Most quantities that we measure are base-neutral so we default to base-10 because it is the standard counting system. But when we measure storage we are talking about a volume of information. And information in digital form is inherently binary, both when stored, and when manipulated.

So the only base that it makes sense to talk about amounts-of-information in is binary. Hence decades of engineers using the correct, i.e most logical measurements.

Now on a tangent, but if I think (way back) to my school days I seem to remember being taught kB, mB and gB. The idea being that the lower case prefix would prevent confusion with SI prefixs. But I'm way too lazy to look for some sort of citation for that, and yes, only engineers would think that reduces confusion.

Re:Read your references

[Smidge204]

I thought the standard was "Libraries of Congress" for data measurement!

=Smidge=

Re:Read your references

[fbjon]

You're confusing two things in your argument there: the amount of information, and the shape of the information. The shape is the base, which is two for bits. But the amount of information is the number of bits, which makes sense to count in base ten, since everything else we count, we count in base 10. Furthermore, for hard disks we use bytes, which are a specific number of bits long (8 today) which is not binary. The byte itself happens to have a range of 0-255, which corresponds to 2^8 since it's composed of bits (and not trits), but that's still not relevant to the total number of bytes.

Cue posts on block sizes, sector sizes, which are still not relevant to the number of bytes in a file.

Re:Read your references

[smallfries]

Cue posts on block sizes, sector sizes, which are still not relevant to the number of bytes in a file. Not quite :)

You're confusing two things in your argument there: the amount of information, and the shape of the information. Ah, but then I don't really see any difference between the two, so it makes it easier to confuse :) But then I'm used to dealing with information that is self-referential

Re:Read your references

[fbjon]

Ok, just to clarify then: suppose we start using computers that have trits as their fundamental unit. Three states of each "bit", 100 follows 022, three voltage levels, etc. Would it then make sense to count filesizes in base three?

Or suppose we use septs: 100 follows 066, seven voltage levels... would it then make sense to count filesizes in base seven?

No, of course not. All countables make sense to count in one base, and that base is 10 by convention. Bits, bytes, digits and apples are all countables, regardless of their internal representation.

I duly note that I still haven't seen an argument for using base two. :)

Re:Read your references

[TheSeventh]

No, let's use examples from Star Trek:

The borg 'One' has assimilated 47 billion Terraquads of data . . .

Re:Read your references

[jedidiah]

I never knew anyone that was ever confused about this before 1997.

This sounds like nothing more than a petty beaurocrat who can't get any
getting to think of himself as a big man by making things needlessly
difficult for others.

Re:Read your references

[jedidiah]

Yes, because when something breaks it would make far too much sense to be
able to sort it all out in nice round numbers like the original advocates
of the SI system were pushing for. Expressing binary quantities in decimal
is what's completely unecessary here.

As long as people who want to cheat you don't lie to you then a person
could go their entire life working with computers and not know what
the decimal representation of a megabyte looks like.

Re:Read your references

[fbjon]

That's still not a reason for using powers of two. I'm looking for a logical, sound argument why powers of two are better for bytes and filesizes than powers of 10.

As long as people who want to cheat you don't lie to you then a person could go their entire life working with computers and not know what the decimal representation of a megabyte looks like. That doesn't mean base-2 representation was ever a good idea to begin with.

You say in your sig:

SI units are meant to be computationally convenient, not arbitrarily assigned. So which is computationally more convenient, counting in base 2 or counting in base 10? Provide an example if you can. Note that computers don't care which base you use for representing numbers, only other humans do, so we're talking about convenient for humans.

Re:Read your references

[Smauler]

Erm, don't we already express binary quantities in decimal? When asked how many kilobytes are in a megabyte, do you really answer "10000000000"? This is partially tongue in cheek, but it does go to show that we definately do work better in decimal than binary. I will always still think a MB is 1024KB though.

Re:Read your references

[edp]

"And information in digital form is inherently binary, both when stored, and when manipulated."

While common methods of storing information may be inherently binary, the amount of information is not. However, since memory also has to be addressed, and the addresses are also in binary form, that leads to the amount of memory being a power of two, or small multiples of a power of two. E.g., it is easier to design a system into which you can plug some number of one-mebibyte memory chips because then 20 bits can select the byte within each chip and additional bits can select which chip. If the chips were one-megabyte, it would be harder to design electronics that could select the chip given an address in a continuous address space.

However, that reasoning does not apply when there is a single device involved. If you had only one memory chip, it would not make much difference whether it were one-mebibyte, one-megabyte, or some other number--all the address bits involved would select a byte within the chip, and the operating system would simply never use an address larger than those in the chip. The same applies to disk drivesyou do not need to fit together the bytes or blocks of multiple disks into a seamless sequential number scheme, so there is no particular reason for the sizes to fall into any particular pattern.

"... kB, mB and gB. The idea being that the lower case prefix would prevent confusion with SI prefixs.

I have never seen any such convention. In fact, the SI abbreviation for kilo is k, in lowercase, unlike M for mega and G for giga.

Electronics has a long history of correct use of SI units for quantities of information, including device sizes and communication speeds. The use of binary units is newer, and the old and correct uses should not be required to change to the new uses that violate the formal, specific, and long-standing SI definitions.

Re:Read your references

[JebusIsLord]

I don't think metric (base 10) is really the best number system. It's based on our finger count, but isn't nearly as robustly useful as base-2, especially in digital systems. Think about inches; we usually think of them in halves, quarters and eighths. The inch is sort of a like a byte, in this sense. Metric gives us ugly values like 0.125 to represent 1/8th. Easy for humans, hard for machines. This debate stems from this tension.

Re:Read your references

[the.Ceph]

I always thought of it as being in base two because of how we utilize it. For instance a 32 bit processor can address 4GB of memory or 2 ^32 = 4,294,967,296 bits. Additionally various file systems have space constraints which mostly have to do with how addresses are stored, and since addresses are stored in binary, the limitations are generally a power of 2. For that reason counting it in base 2 always made sense to me.

Re:Read your references

[avoision]

Binary addressing makes sense for RAM, since there's a certain number of addressing lines and it would be a waste for some of the lines to not be fully used. But for hard drives (and flash drives) with 512-byte sectors, it does not make more sense than any other addressing scheme. My hard drive has (thanks to fdisk -l), 255 heads, 63 sectors/track, 3,648 cylinders, which gives 8,225,280 bytes per cylinder or a total of 30,005,821,440 bytes. Those are the logical numbers, not the physical ones, but that only helps my point; that a given mass-storage device's capacity is an arbitrary number of 512-byte sectors. In this case, I paid for 30,000,000,000 bytes and I got that and a little extra. So anybody's lawsuit should only be able to extract money for the difference between the stated number of gigabytes and some integer times 512 bytes (but this is moot, since they always add extra like in the case above).

Furthermore, for the past two thousand years, the greek prefix giga has meant one billion. Just because we have binary computers doesn't mean we should change that for the purposes of lawsuits.

Lastly, you know damn well after reading the fine print on any mass storage device in the last 10 or 15 years that it says that when the listed capacity is x gigabytes, that means x * 1,000,000,000 bytes. And that fine print is on the outside of the box, so you know before you buy.

The National Institute of Standard's has suggested that for the useful binary numbers (2^10, 2^20, 2^30), we use the different prefixes KiB, MiB and GiB to show that they refer to the binary versions. See NIST's recommendation on this.

I'm happy to buy my ram with capacity listed in GiBs and my hard drives with capacity listed in GB or GiB, but let's not confuse the two prefixes.

Re:Read your references

[Lost Engineer]

When I can buy burritos with Wii points that will be cool.

Re:Read your references

[fbjon]

First of all, 32-bit addressing means that 2^32 bytes can be addressed. But memory is a different design and circumstance than disk space, so I don't think it's comparable. Also, there's not really any confusion about RAM size, now is there?

Second, "counting in base 2" is a bit of a misnomer, since we're really counting in decimal, except with powers-of-2 prefixes. It's mixed counting, in other words.

Third, and most important, the whole issue is about how to express the length of a file, and how to express the length of a filesystem and the underlying block device. We're talking specifically about length here, there aren't really any other dimensions in filesystems of today. The hardware itself may have some structure to it that rely on powers of two, like 512-byte sectors, but that is a hardware issue. Likewise for filsystems.

The whole thing boils down to the user interface: do you divide by 1000 or 1024 when presenting the numbers? Does either way provide any advantages, or disadvantages?

Here's a disadvantage for using 2^n prefixes: try calculating 10 GiB - 500 MiB. How long did it take? Contrast this with 10^n prefixes, i.e. 10 GB - 500 MB. 10^n prefixes are easier to grasp and figure out. Are there any particular situations where using 2^n prefixes is an advantage?

Re:Read your references

[smallfries]

I would say so, although it is more efficient to use binary than any other base so I don't think it will come up. When I mentioned self-referential data I meant things that contained addresses of other pieces of information. Whenever that is the case, the fact that the address is being stored in binary means that it makes sense to use units based on powers of two. But what I'm arguing is just a variant of the old memory addressing argument.

The basic argument is that when storing anything it makes sense for the units to be "round" numbers (in the everyday sense), and where those addresses are being manipulated in binary, base-2 units make sense. Of course, not everyone stores and manipulates the same kind of data so YMMV.

Re:Read your references

[JLF65]

It seems a lot of geeks try to "defend" using powers of two, as if it were somehow the "correct" way, without thinking whether or not it is really correct and logical (*).

It IS correct. All computer science is built on base-two math. Period. There are no base-10 computers. It takes a fairly complicated routine to even convert numbers into base-10 representative strings (and vice versa). So it's both correct and logical. It doesn't take a geek, only someone has had a minimum of one class on boolean logic (which you clearly have not).

Re:Read your references

[fbjon]

And that routine is used regardless, if you think for a just a little longer. The difference is with the prefixes, not the string of decimal numbers. So why are base-2 prefixes correct for filesystems again?

Re:Read your references

[tsm_sf]

THANK YOU

Re:Read your references

[Mathinker]

I personally heard of at least one university test (in the '80s) where people got a question wrong because the professor intentionally used the same SI prefixes in the same question but in two different "contexts" (data transmission rate and memory size) so that you had to know the different "conventions" to get the answer right.

Although from your sig I see I preach to the converted...

Re:Read your references

[Mathinker]

> connected pieces of hardware should use the same basic units

Your computer is probably connected to a network cable, so I suppose that the manufacturer of the cable should use the power-of-two convention when he calculates the bandwidth of the cable in megahertz, then, right?

And in 2009, US television transmissions will go digital, so Americans will have to recalculate the frequencies of the TV station transmissions, I suppose, no? Good luck trying to figure out if your microwave interferes with the WiFi in your house!

Using the same notation for two different things is an invitation for disaster.

> And how should I calculate the hard drive cache size ? with powers of ten ?

Far from it. Use kibi-, mebi-, and gibi- bytes to your heart's content! I do! Just don't call them kilo-, mega-, and giga- ...

Re:Read your references

[Mathinker]

> no logical reason whatsoever why we should use powers of two

Of course there is, as you correct yourself --- memory units should be power-of-two. You're missing the whole point, which is that you just shouldn't call the power-of-two units the same names as the power-of-ten units --- which is why there is kibi-, mebi-, and gibi- ...

Re:Read your references

[Mathinker]

> only engineers, mathematicians, and scientists would think that reduces confusion

There, fixed that for you.

> But I'm way too lazy to look for some sort of citation for that,

kibi- = 1024, mebi- = 2^20, gibi- = 2^30

But I suppose you're also too lazy to actually use them. It does require effort nowadays, like any kind of pioneering.

Re:Read your references

[smallfries]

You didn't actually have anything to back up your point then? Just a quote out of context and a quick jibe. You call me lazy?

Re:Read your references

[brokenbeaker]

furthermore, 'k' is the SI prefix for kilo. K, in SI means kelvin.

Re:Read your references

[Mathinker]

You called yourself lazy. Yes, I am also lazy --- too lazy to try to investigate you over and above how you characterize yourself.

1) You contradict yourself in your own post. You state that the engineers have always been using the one correct or logical way to measure information, which is in base-2 units (in order support your point), but then you deride them for wanting to differentiate the base-2 units they use from previously established base-10 units.

I do agree with you and disagree with fbjon that base-2 units are the most logical for memory capacity, but you, like him, miss the major point of the argument --- that these base-2 units need new unambiguous names.

2) By your logic, when the US moves to digital TV transmission, the frequencies of the TV channels, which up to now have been measured in megahertz = 10^6 hertz, should start to be measured in base-2 units, which would be mebihertz = 2^20 hertz but anyway you want to call those mebihertz frequencies megahertz because base-2 is the only logical way we'd talk about information transfer. Good luck trying to figure out if your microwave oven interferes with your WiFi or one of those digital TV channels.

3) Look at what happens when you try to figure out how long it will take to transfer the contents of the RAM of your computer over your Internet connection (the same engineers who you claim have been always using the "logical" SI units for memory capacities have also always been using the real (i.e., base-10) SI units for data transfer rates).

Re:Read your references

[smallfries]

Yikes, you need to read less into what you think people have said. Just because I didn't add a smiley you shouldn't assume that my post was serious. I thought that it was obvious I was just pissing around from context, and besides we both know I'm too lazy to type the extra two characters :)

Point 1 - fair enough. Point 2 - not what I said. Point 3 - not what I said. Basically for storage of digital information it makes sense to me (as a programmer and a computer scientist) to do it in binary. YMMV. For data transmission or anything to do with frequencies it would be messy.

Now take a deep breath and repeat after me: "Not everyone on slashdot is having a go at me..."

Re:Read your references

[Mathinker]

smalfries: "Not everyone on slashdot is having a go at me..."
mathinker (takes deep breath): "Not everyone on slashdot is having a go at me..."
smalfries: "Not everyone on slashdot is having a go at me..."
mathinker: "Not everyone on slashdot is having a go at me..."

I am calm. I am very calm. I... what was I going to say?

Oh, yeah... Sorry if I was too reactive, there, no offense...

I remember reading about a study made long ago about electronic text communications; the researchers came to the conclusion that it tended to spiral out of control a lot easier than face-to-face. I think we just experienced that...

Re:Read your references

[smallfries]

I remember that study - was probably duped on slashdot a few times. The one where their point was that we always assume an emotive stance for what we read because that's what we're used to in real life.

Anyways, no worries, it was fun and I learnt a couple of things which is always nice :)

Re:Let me get this straight...

[Hognoxious]

Seems they aren't the only ones.

This is a storm in a teacup. Everybody who's technically minded knows the numbers are approximate (and where/how to find the true byte count). Anyone who isn't won't understand the difference anyway, and would find it much more confusing having to deal with fractions resulting from the base conversion.

Re:This is stupid

[Nullav]

Here's a nice idea: State what you mean on the box. "2TB (SI)" Drive manufacturers and the like will still use SI kilobytes for the sake of larger numbers, but at least we can all stop arguing about this stuff and put that suing power to a better use. Also, I will never ever say 'tebibit' aloud.

Where are the ethernet lawsuits?

[lpontiac]

"100 Mbit/second" ethernet operates at 100*10^6 bits/sec, not 100*2^20 bits/sec. If every hard drive manufacturer under the sun is going to get hit up for adhering to their common convention (whether you agree with it or not, it's the same for every hard drive), why not Realtek, Intel, 3com and Cisco?

I hate these settlements

[Propaganda13]

I remember Iomega's settlement for the click of death was "Sorry, we built a poor quality product that was supposed to back up your data, but lost it instead. How about you buy another product from us at a reduced priced"

Give a check for $3.50 instead, but don't give me a discount on the same manufacturer's products.

I haven't looked lately, but I thought a lot of manufacturers used GB*.
*GB refers to 1,000,000,000 bytes. on their packages.

Re:SI

[Kalriath]

Ask Verizon.

The OS _is_ wrong - Long live bits and base10!

[PMBjornerud]

Personally, I think the things like HDs, network gear, and such are correct. We need to use the metric prefixes for base 10 for base 10. If we want to talk base 2, use the base-2 prefixes. Burn these foul aberrations begotten by the kilobyte! This, my dear IT fellows, is the path of evil and only horrors lurk ahead.

Computers function in the realm of magic. Behold! 500MB plus 500MB! The sum not a full, but strangely a 0.97 of a gigabyte. The remaining 3 percent gone, - a sacrifice to evil!

Don't even get me started with base 2. The byte itself is not even a 1, but itself an 8. Thus, the kilobyte is really 2^13 bits, and a megabyte is 2^23. The whole system is ludicrous. This happened because a useful technical shortcut have been kept alive for too long, and made its way into the real of the end-user.

Stop this madness and see the light of the network engineers. Behold! The wonder of the Mbps. 1Mbps is a wonderful, intuitive 1,000,000 full bits per second. This is stuff I can explain my mother - and she'll understand.

Re:The OS _is_ wrong - Long live bits and base10!

[dublin]

Stop this madness and see the light of the network engineers. Behold! The wonder of the Mbps. 1Mbps is a wonderful, intuitive 1,000,000 full bits per second. This is stuff I can explain my mother - and she'll understand.

Amen. You almost touched on the key difference - The binary crap is by computer scientists (many of whom make their living as high priests by spreading obfuscation), while networks (which have to work in actual practice, regardless of theory) were built by engineers.

My money is on the engineers every time...

Re:The OS _is_ wrong - Long live bits and base10!

[chopper749]

Much easier. At 1Mbps, how many bytes are transmitted per second? 125,000? Nope. ...but at least she'll understand!

copy of my settlement email

[poetmatt]

Just as a humorous reference, this is what I got in the mail from creative:

SUMMARY EMAIL NOTICE

To: (I'm not putting my email like that on slashdot)

From: Settlement Claim Administrator

Subject: Notice of Creative Hard Disc Drive MP3 Player Class Action and Proposed Settlement

SUMMARY EMAIL NOTICE

If you purchased in the United States between May 5, 2001 and April 30, 2008 from a retail store in the United States (including Creative's and others' on-line retail stores) a new Creative brand hard disc drive MP3 player ("Creative HDD MP3 Player"), a proposed class action settlement may affect you. A hearing has been scheduled in United States District Court, Central District of California to approve the settlement. Under the settlement, you may have the right to make a claim for a discounted MP3 player or a discount certificate. You also may choose to exclude yourself from the settlement. Alternatively, you may file written objections to the settlement or seek to intervene and appear (or have your own attorney appear) at the court hearing. If the settlement is approved and you do not exclude yourself, you give up the right to sue for the claims the settlement resolves, and you will be bound by the terms of the settlement. To learn more about or exercise any of your rights, please read below and visit www.creativehddmp3settlement.com.

The lawsuit is Talwar v. Creative Labs, Inc., United States District Court, Central District of California, Case No. CV 05 3375 FMC. In the suit, plaintiffs allege that in the sale and marketing of its hard disc drive MP3 players Creative stated that purchasers of the drives would receive approximately 7% more usable storage capacity than they actually received and misrepresented the number of songs and number of hours of music the players could hold. Creative has denied and continues to deny each and all of plaintiffs' claims, and denies that anyone has been harmed or deserves compensation. The Court has not made a decision on the merits.

You are a member of the plaintiff class if you purchased in the United States between May 5, 2001 and April 30, 2008 from a retail store in the United States (including Creative's and others' on-line retail stores) a new Creative brand hard disc drive MP3 player.

As part of the settlement, Creative will make certain disclosures regarding the storage capacity of its hard disc drive MP3 players.

In addition, if you submit a valid claim, you will receive either a 50% discount off the price of a new 1 GB MP3 player, or a discount certificate good for 20% off the price of any single item purchased at www.us.creative.com. To receive the discount player or discount certificate you must submit a claim form available at www.creativehddmp3settlement.com by August 7, 2008. You may submit a claim for each Creative HDD MP3 Player you purchased.

If the settlement is approved, plaintiffs' counsel will apply for an award of attorneys' fees and expenses not to exceed $900,000, plus incentive awards for the two representative plaintiffs in the amount of $5,000 each, to be paid separately from and in addition to the relief available to plaintiff class members.

All claims of plaintiff class members which were or could have been asserted in the litigation, based upon the facts alleged in the litigation will be released. This means that if you do not exclude yourself from the plaintiff class, you will give up the right to sue for the claims the settlement resolves, and you will be bound by the terms of the settlement.

You need not take any action. If you wish to exclude yourself from the plaintiff class, you must submit an exclusion request to plaintiffs' counsel: Brian R. Strange/Gretchen Carpenter, Strange & Carpenter, 12100 Wilshire Boulevard, Suite 1900, Los Angeles, CA 90025. If you exclude yourself, you will not receive the benefits of the settlement, and you cannot object to the settlement or intervene.

If you wish to object to the settlement, intervene or appear (or have

Re:Absolutely.

[Kalriath]

Except in Computing. Lookee here:

http://www.google.com/search?hl=en&q=one+gigabyte+in+bytes&btnG=Google+Search

Re:missing _____

[TaoPhoenix]

Now we know where those White House emails went!

"We kept 1 Million Bytes of Emails".

"But Sir, a megabyte is larger than that. Where are the missing emails?"

"I don't know where then!"

Really?

[rennerik]

I thought everyone knew that by now all drive manufacturers list capacities in ISO units instead of the traditional binary method? I don't think I've ever gone to the store and bought a 250gb drive expecting to get exactly 250gb.

Do we have to start putting disclaimers now for different filesystems too? "Depending on the filesystem used on this drive, and the block size specified when formatting, actual capacity of this drive may be affected."

I can see it now... "Microsoft accused of 'favoring' hardware industry by making 1 byte files take up nearly 1kb! Oh the conspiracy!"

Argh.

Countable blocks vs theoretical performance bits/s

[argent]

You're comparing a theoretical maximum performance measurement (one that is also used by disk manufacturers without raising any eyebrows there either) with the total measurable and usable capacity of an actual object that can hold an actual number of blocks of data.

A hard drive is a physical object storing data in chunks that are sized in power-of-two units. A hard drive contains multiples of 512, 1024, or 2048 byte blocks, each containing 8 bits per byte, and the data read from the drive is frequently copied block-for-block to physical RAM that is sized in 2^10, 2^20, and 2^30 byte units. Operating systems use these same units when reporting available disk space. These are natural sizes for talking about storage, and there is a reasonable expectation that you can use all of that storage.

The speed of a network is a theoretical measure of performance in ideal conditions, based on the transmission of data in variable sized chunks up to a maximum that depends on the characteristics of the hardware and protocols, typically 1500 bytes or less per packet. Packets are normally processed after being received... decompressed, decrypted, and unpacked... and so the number of bits or bytes transmitted per second is only loosely related to the size of the data finally received by the user. In addition the rated speed of a network is further reduced by per-packet overhead and by protocol inefficiencies such as collisions in the original 10 Mbit/s ethernet. There is no standardized fixed-size "quantum" that network bandwidth chunks into, and there's no expectation that you can actually fill up all 100 million bits every second outside a lab.

Yes, RLY.

[argent]

drive manufacturers list capacities in ISO units instead of the traditional binary method?

The fact that binary is traditional for storage is a pretty good argument for NOT assuming that everyone knows it. Particularly when operating systems continue to report capacity that way for good practical reasons.

"Depending on the filesystem used on this drive, and the block size specified when formatting, actual capacity of this drive may be affected."

When you bought drives in unformatted capacity (which you only do for floppies any more) they did in fact have language like that in their fine print. These days the block size on a given drive is fixed so formatted capacity is used.

Re:"creative sued"?

[sm62704]

Ewe muss bee knew hear.

Byte me.

[argent]

We have an SI standard for this nomenclature now. No matter what idiot lawayers want to argue they can't deny the fact that GB is defined for base-10 usage and GiB is base-2.

Given that even people who are advocating this obscure terminology can't get it right (it's IEC, not SI, that defined this standard, and if you want to use SI units, you should be calling them "Octets" not "Bytes"), and that virtually nobody (not even the drive manufacturers) actually uses it outside legalese and fine print, I think you are making unreasonable assumptions and unreasonable demands.

Re:SI

[lubricated]

in the us we don't use metric and a byte is still 8 bits and not 10.

Re:It's not geeky, its simply correct

[Jesus_666]

Except that Creative did sell the stated quantity; the customers simply expected that number to have a different meaning. It's as if in the automobile industry one gallon of motor oil (but not gasoline) equals 1.2 regular gallons and then someone sues a motor oil manufacturer for selling one regular gallon of motor oil as "one gallon of motor oil".

Yes, I know that motor oil isn't sold by the gallon. Substitute "0.1 gallons" or whatever an appropriate smaller amount would be.

Base-2 Computers

[Kenoli]

Computers operate in base-2. It only makes sense to represent different amounts of data in base-2.
As I understand it, the only reason the terms use SI-like prefixes in the first place is because 1000 and 1024 are "close enough".
Presumably anyone actually dealing with data would be able to understand these units, but nowadays the average user has no idea how data is stored internally. They'll believe "kilobyte" means "1000 bytes". Why wouldn't they?
But that creates some problems. Even if manufacturers decide to use base-10, computers are still going to be base-2, and users are going to see a discrepancy between what box and the OS tell them, and they aren't going to be happy with it.

Re:Base-2 Computers

[prockcore]

Computers operate in base-2. It only makes sense to represent different amounts of data in base-2.

For the love of god. You're using base 10 numbers. Unless you're going to say "I have 1000000000b megabytes of ram" instead of "512 mebibytes of ram".

After all, you should use base 2 since computers use base 2. Don't covert to base10 numbers.

Defend the Binary Gig!!

[rayk_sland]

Yeah!

Base 2

[henrik.falk]

I'm loving this. I hate that all OS manufacturers use base 2, but all storage manufacturers use base 10. And it's not like there is a debate, base 2 is obviously the correct one since the storage is binary.

SI Units?

[EmagGeek]

Ummm I don't remember kilo EVER meaning 1024. Nor do I ever remember mega meaning 1048576. I certainly never remember Giga meaning "times 2 to the 30."

Kilo, Mega, Giga: these have always been base 10, they are now base 10, and they will always be base 10.

The fact that people stopped giving a shit about learning this in grade school doesn't mean they are suddenly right and the scientific establishment is wrong.

I missed the boat...

[stewbacca]

Where was this sort of lawsuit when I bought a 160GB Maxtor HD, only to find out that WinXP could only recognize the first 120GB?

Re:Apple?

[Yvan256]

http://www.apple.com/ipodshuffle/specs.html

Capacity
- 1GB or 2GB USB flash drive (1)

Footnotes:
1. 1GB = 1 billion bytes; actual formatted capacity less.

They should be writing 1GiB or 2GiB instead. And their footnote only makes matter worst as it's basically stating that 1GB = 1GiB...

Justice denied -- the legal system is BROKEN

[intnsred]

This really offends me and IMHO is yet another example of the broken nature of the US "justice" system.

Let's recap: Some company lies to me. They screw me out of bits and bytes by playing the base-10 game. In short, I am ripped off.

But what is our "justice" system's remedy?

They give me a discount coupon that forces me to go back and to do business with the company that just lied to me and screwed me!

That isn't justice, that's a marketing gimmick to increase sales. Companies should use these tactics deliberately (if they're not already!), since their upside of new sales far outweighs their risk from the justice[sic] system.

How about instead if the courts pulled some of these sleazy companies' corporate charters (in effect giving them the corporate death penalty)? I bet this behavior would stop real quickly then!

But wait -- our gov't doesn't give corporations the death penalty. The death penalty is only reserved for selective human beings...

Re:Countable blocks vs theoretical performance bit

[JesseMcDonald]

More concisely, bits use base-10 prefixes whereas bytes use base-2 prefixes. "Theoretical maximums" have nothing to do with it. Bits are treated as though they were an SI base unit, although they aren't listed among the official SI bases. Bytes are not SI (or even pseudo-SI) units, and do not follow the standard SI prefixes. Compound units tend to use SI prefixes (e.g. MB/s -> bytes/microsecond or bytes*MHz -> 10^3 bytes/second).

Still have your receipt from 2002?

[TPJ-Basin]

According to the claim forms, you must show proof of purchase for your '40GB ZenXtra' that you purchased prior to 2004. Good luck with that.

This isn't news..

[bturcotte]

I think it is safe to say that everyone who posted a comment here is well aware about how storage capacities are presented. Why so much surprise?

Re:Ughh.. again...

[Bill_the_Engineer]

Who is we?

Jeez, chill man.

Maybe (just maybe) the confusion began when the computer electronics industry applied SI prefixes to industry units. I mean come on, what harm could come from using base-10 prefixes with base-2 units? After all, it's only the folks in the hardware industry using this right?

Oh wait... Now that we have more users outside of the hardware industry using these machines and they are more familiar with the correct (base-10) usage of SI prefixes who could blame them for the confusion over quantity?

It is hard for us to imagine what a lay person comprehends when they see 1MB. For all they know, they received some free bonus memory after the initial 1,000,000 bytes.

Of course unit confusion can happen to people actually in the computer industry. I've lost count of the number of times I had to correct someone because they were off by a factor of 8 because they aren't accustomed to hardware capacities of memory components being measured in bits and not in bytes (octets)...

Anyway back to the topic of the article and this is not necessarily address to the parent post.

I'm old enough to remember when everything was advertised using base-2 units, and remember when hard-drive capacities began to be advertised using base-10 units. This happened a long time ago, and long before MP3 players made their appearance. So I am sceptical that any consumers where confused by the use of base-10 to sell hard drive capacities.

Besides most of the advertisements and the packaging of the retail versions of these drives have fine print stating the capacity is given in base-10 and the amount reported by the OS may be lower...

Units shmunits, they tried to decieve

[BlueParrot]

Whenever this discussion crops up people start talking about which definition is "more appropriate" or whatever. This is in fact irrelevant because it is outright obvious WHY device manufacturers use base-10 when every other part of the IT industry uses base-2, they are trying to deceive their customers selling devices with less storage space than the customer thinks they get. As a physicist I have a lot of respect for the SI units, but we also use other units every now and then because it makes sense in context. We use electron volts rather than joule when we talk about photon energies, we use barns rather than square meters when talking about nuclear cross sections, and in particle physics we often use units where c = 1 because it makes the algebra easier. This works because everybody knows what to expect from the context ( well most of the time ).

What happened with hard drives was that some manufacturers got greedy and decided to screw with a long established trend inside the IT sector because they wanted to fool the customers. You can argue about which unit is better but does anybody seriously believe device manufacturers started using base-10 because they had a burning desire to be consistent with the SI system? It is plain obvious what happened, they deliberately tried to deceive their customers to get a competitive advantage, and the courts decided this was illegal. Quite right if you ask me.

Do we all get to sue Intel and AMD now?

[ThreeGigs]

How fast is your processor? 3.73 GHz? Is that in binary or decimal Gigas? Do we all get to sue Intel and AMD now because their stuff is in a computer and therefore should be using base 2 notations? Creative got sued over something that wasn't even a computer! What's next, "I thought my petaflop supercomputer would be doing 1,125,899,906,842,624 operations per second, not a measly 1,000,000,000,000,000!" Do I get to sue Cisco too, because my gigabit ethernet seems to be 73 million bits per second short?

Is there even an ISO standard on what Kilo, Mega, Giga, etc. mean when applied to digital quantities?

Re:Countable blocks vs theoretical performance bit

[argent]

Bytes when a measure of throughput are also measured in 10^n units.

Bits or words, when used as a unit of storage (and yes that does happen, occasionally) tend to be measured in 2^n units.

Why? Because it's useful to measure storage that way, because it naturally tends to come in 2^n chunks or multiples of 2^n chunks, for the reasons described above, while transfer rates rarely such natural division by powers of two.

Re:HDD IS memory

[Oggust]

remember back when 20MB was HUGE?

Yessir.

Well, adverts for PC's said "has 20MB of memory" meaning "has a 20MB HDD". /

Yeah, well, I can't say I remember such ads, but sure, maybe people did that. It was obviously wrong, and they've stopped doing that. What's the problem?

And it was powers-of-two megs.

Kind of shot yourself in the nuts with that one.

Huh? No. Even if your argument had made any sense, those disks were using power-of-ten sizing, at least mostly. Of course, the difference was smaller then, so you'd notice it less.

And when it comes to telecoms, it was all baud. Which WAS NOT 1 bit/sec. And powers-of-ten when it was used was accepted because you had 8 bits, a stop bit and a parity bit. 10 bit bytes.

No.

Well, yeah, baud isn't the same thing as bps, that's true. Everything anybody cared about was the bps number though, even though many people (wrongly) called that baud. Ie V.32bis (14.4 kbps, )(and those are actual "k"s - 1000) used 2400 baud.

And it's true that you could figure the transfer speed in bytes to be about 10:1. But that's completely unrelated to the question about power-of-2/10. I assure you that a kbaud also would have meant 1000 baud. (baud is deprecated, btw, mostly because people keep getting this wrong. "Symbols per second" is the new hotness.)

Please show me where hard drives use 10 bit bytes...

Huh?

Except For One Thing

[maz2331]

Your analysis ignores the fact that storage has ALWAYS been measured in Base-10 nomenclature. Thus, the attack on the "greedy manufacturers" should fail simply because the units used are understood historically to mean 1k = 1000, 1M = 1,000,000, etc.

In other words, the entire argument is based on a fallacy that ignores history and customary use of the units.

Memory has been customarily measured in Base-2 (basically, KiB, MiB, GiB, etc). Although "common use" confuses the units, it does not change historical usage.

This is akin to a lawsuit over, say, gold sales. Gold is measured in "Troy Ounces" not "Avoirdupois" ounces like nearly everything else, and has been for centuries. They are not identical, but in common use gold is quoted per "ounce".

Re:Except For One Thing

[BlueParrot]

Not in the IT industry. Drives used to come in sizes measured in the same units as system memory. Now while the latter cannot easily be adjusted, some manufacturers decided to CHANGE how they described their drives to get a competitive advantage. What I am arguing is that the manufacturers in this case knew people would misinterpret it, they knew this would make their drives look bigger, and THAT IS WHY THEY DID IT. Does anybody seriously believe they decided to change the measurements of their drive size from what system memory uses to SI units because they care about the SI system? Please, give me a break. They did this because they figured they could sell more drives by making people think they get more storage than they do. It is wrong and it should be punished accordingly.

Creative makes mp3 players?

[seanonymous]

Huh, didn't know that. This site is very informative.

Re:Ughh.. again...

[toddestan]

What is wrong with saying that there are "2 kilomiles" between Chicago and Los Angeles?

Because that's only 2000 miles, not 2048 miles.

I don't see where there's a problem here

[rfc1394]

It has been a standard practice for years that while memory manufacturers measure in 1024^2 megabytes and 1024^3 gigabytes, disk drive manufacturers have been measuring in 1000^2 megabytes and 1000^3 gigabytes. I have an article on my blog about this, how my new computer has a certain amount claimed on the box, but Windows shows a lower amount, which when you use the two different figures it does match, while the box on my new computer says it's a 400gb drive, it is correct if you use 1000^3 gigabytes, but it is only 373gb if you use 1024^3 gigabytes.

For decades, TV (and computer monitor) manufacturers have announced the size of the picture tube diagonally, which gets you a larger number than either horizontal or vertical measurement. And they have generally openly admitted that their measurements are diagonal; every ad for TV specifically says that.

I suspect that the drive manufacturers are getting in trouble and losing lawsuits because they do not explicitly state that they are using 1000^3 sized gigabytes rather than the 1024^3 sized that people normally expect.

It's an attempt to cheat people when you're not honest about what you're doing. I mean, I may not particularly like them using the slightly sleazy method of the 6% shortage by using 1000^3 size but as long as they're straightforward about it, I have no problem with them using that method.

--
Paul Robinson — My Blog

Re:I don't see where there's a problem here

[HeadlessNotAHorseman]

The diagonal measurement of TV tubes actually makes a lot of sense. For a start, tvs tubes were originally almost round anyway. But how do you measure the width and height of a screen in which the edges are curved? The manufacturer would naturally be measuring at the widest point, which is useless because the wider the curve, the bigger you could say the tv screen was. This would just encourage manufacturers to maintain large curves so that they could market tvs as having larger screens than they actually do. If you measure by the diagonal, then the best way to make the screen size bigger according to measurement whilst minimising the increase in area is to straighten the edges. Of course, you could cheat and give the edges a little concavivity*, but that would look silly and few people would buy it.
 
*I thought that I invented this word, but it turns out that at least three other people have used it before me!

And the winner is...

[t33jster]

The lawyers.

If the settlement is approved, plaintiffs' counsel will apply for an award of attorneys' fees and expenses not to exceed $900,000, Also ran: A$$hole plantiffs who were either too dumb or too ignorant to bother to understand industry standards.

plus incentive awards for the two representative plaintiffs in the amount of $5,000 each, to be paid separately from and in addition to the relief available to plaintiff class members. Also also ran: Creative for settling a class action lawsuit in such a way that they may actually increase their revenue.

In addition, if you submit a valid claim, you will receive either a 50% discount off the price of a new 1 GB MP3 player, or a discount certificate good for 20% off the price of any single item purchased at www.us.creative.com.

Re:And the winner is...

[poetmatt]

yep, everyone gets screwed.

I want to object to the class action, but I wouldn't even know where to get started.

I think this is why the companies do class actions - the only one who makes anything is the lawyers.

Re:Let me get this straight...

[Bastard of Subhumani]

That's different. If Apple had to put the exact size on their machines, logos & adverts it would spoil the appearance. Which looks and sounds more elegant, iPod Nano 8Gb or iPod Nano 7.450580596923828125Gb?

Gibi = garbage

[Gonoff]

IT has always calculated in powers of 2. The "gibi" nonsense was invented by dodgy salesmen to talk up their equipment.

Even Microsoft gets it right. I am sitting at a machine with a HDD of 60,011,606,016 bytes. It was sold as 60GB but Windows reports it correctly as 55.8. Why should people be misled because some suit wearing sales wheasel decided to invent a series of rubbish words beginning in gibi?

We need more court cases until this misrepresentation ends. Have you noticed that flat screen monitor sizes are correct, where CRT ones dishonestly used to include bits of the tube you couldn't even see...

Re:Gibi = garbage

[bh_doc]

What is with idiots moderating up factually incorrect nonsense about gibi being "invented" by salesmen or wikipedia? It's an International Electrotechnical Commission (IEC) Technical Committee proposal. At best these posts are funny, trolls at worst.

Re:Gibi = garbage

[GrievousMistake]

No, Microsoft does in fact get it wrong.
Using the standard SI-prefixes for something almost but not quite SI is practical in many contexts, but it should never have been allowed to propagate to a consumer product. It's a colloquial usage, really, IT slang.

This kind of ambiguity is unacceptable in engineering as well as marketing, where you need to be able to communicate with clearly defined terms.
You can't just say that mega- means 1024 "with computers", because with a lot of things (networks, hard drives, frequencies), it doesn't. That so many people here seem to think that hard drives used to use the 1024-based system until quite recently, just goes to show that confusion is rampant, even among the techies.

If you were leading a project where size was a critical factor, wouldn't you want to specify it with precise terms? This is the kind of problem that crashes Mars orbiters.

Re:Gibi = garbage

[tsm_sf]

What is with idiots moderating up factually incorrect nonsense about gibi being "invented" by salesmen or wikipedia? It's an International Electrotechnical Commission (IEC) Technical Committee proposal. At best these posts are funny, trolls at worst. Did you even read the page you linked? Here's an interesting bit (aha ha):

Historical context* Once upon a time, computer professionals noticed that 2^10 was very nearly equal to 1000 and started using the SI prefix "kilo" to mean 1024. That worked well enough for a decade or two because everybody who talked kilobytes knew that the term implied 1024 bytes. But, almost overnight a much more numerous "everybody" bought computers, and the trade computer professionals needed to talk to physicists and engineers and even to ordinary people, most of whom know that a kilometer is 1000 meters and a kilogram is 1000 grams. So, basically you're wrong.

Re:Gibi = garbage

[bh_doc]

So, basically you're wrong.

Oh, I didn't realise the IEC was a salesman consulting group. Why not quote the rest of that passage while you're at it?

Then data storage for gigabytes, and even terabytes, became practical, and the storage devices were not constructed on binary trees, which meant that, for many practical purposes, binary arithmetic was less convenient than decimal arithmetic. The result is that today "everybody" does not "know" what a megabyte is. When discussing computer memory, most manufacturers use megabyte to mean 2^20 = 1 048 576 bytes, but the manufacturers of computer storage devices usually use the term to mean 1 000 000 bytes. Some designers of local area networks have used megabit per second to mean 1 048 576 bit/s, but all telecommunications engineers use it to mean 10^6 bit/s. And if two definitions of the megabyte are not enough, a third megabyte of 1 024 000 bytes is the megabyte used to format the familiar 90 mm (3 1/2 inch), "1.44 MB" diskette. The confusion is real, as is the potential for incompatibility in standards and in implemented systems. Faced with this reality, the IEEE Standards Board decided that IEEE standards will use the conventional, internationally adopted, definitions of the SI prefixes. Mega will mean 1 000 000, except that the base-two definition may be used (if such usage is explicitly pointed out on a case-by-case basis) until such time that prefixes for binary multiples are adopted by an appropriate standards body.

So a technical committee came up with a solution for the confusion *already present in the marketplace*. That is not the same as "gibi" being "invented by salesmen".

Re:HDD IS memory

[prockcore]

Please show me where hard drives use 10 bit bytes...

If 8bit bytes is really why people use powers of 2, wouldn't it make MUCH more sense to use octal?

Re:Apple?

[prockcore]

No, you've got it backwards.. 1GiB = 1,073,741,824 bytes, 1GB = 1,000,000,000 bytes.

I have two questions:

[stewymcstewstew]

What Operating Systems do Creative support?

How does that Operating System display drive size?

Also, which measurement do they use for "130MB of free space?"

Ok, maybe three questions.

You think this is bad...

[autophile]

...just wait till a geek takes economics in college and his head asplode when he finds that economists use the "M" prefix to indicate one thousand!

--Rob

Re:It's not geeky, its simply correct

[Lehk228]

it would be like suing a baker for giving you 12 when you ordered a dozen, because you thought he should give you 13

Bandwidth?

[Mathinker]

a person could go their entire life working with computers and not know what the decimal representation of a megabyte looks like. Until he has to transfer the information from his computer to another one via transmission (as opposed to moving media around).

Hmm, do I smell a class action suit against Comcast? Bandwidth and transmission rates are traditionally stated in real SI units.