How big is a gigabyte in hard drive advertising and packaging?

How big is a gigabyte in hard drive advertising and packaging?

From another newsgroup, I suppose it's true:

Coverage of a recent New Mexico Supreme Court case said that
states were getting bold in finding excuses to void arbitration
agreements under general state contract law and the U. S. Supreme
Court was consistently refusing to review the state decisions.

The New Mexico case was a class action suit against Dell for
rating hard drive capacity in decimal instead of binary gigabytes.
The class included people who ordered computers from Dell's web
site. Dell requires users of their web site to agree to Texas law
and arbitration of individual claims only. The artibtration clause
and class action waiver are legal under Texas law. New Mexico law,
the court decreed, includes a fundamental right to bring a class
action. Class action arbitration was not possible under the rules
in effect (though that may change) therefore the case could be tried
in New Mexico state court.

(It was one of those cases where I wanted both sides to lose.
Dell, for making it practically impossible to get relief if
they had actually done something wrong. The plaintiff, for
caring about the trivial difference between 2^30 and 10^9.)
-- end quote --

But someone else says the normal manner of business is to use
1,000,000,000 bytes for a gigabyte.

I thought that formatted drives had fewer bytes capacity than
unformatted drives, because of the overhead used by formatting** but
that on the box and in the advertising for a hard drive, a gigabyte
meant 1024^3. Is that true?


**Space allowed for directory information is the only overhead I can
think of. Is there more?



If you are inclined to email me
for some reason, remove NOPSAM :-)

On 07/07/08 04:59 PM, mm wrote:
> How big is a gigabyte in hard drive advertising and packaging?
>
> From another newsgroup, I suppose it's true:
>
> Coverage of a recent New Mexico Supreme Court case said that
> states were getting bold in finding excuses to void arbitration
> agreements under general state contract law and the U. S. Supreme
> Court was consistently refusing to review the state decisions.
>
> The New Mexico case was a class action suit against Dell for
> rating hard drive capacity in decimal instead of binary gigabytes.
> The class included people who ordered computers from Dell's web
> site. Dell requires users of their web site to agree to Texas law
> and arbitration of individual claims only. The artibtration clause
> and class action waiver are legal under Texas law. New Mexico law,
> the court decreed, includes a fundamental right to bring a class
> action. Class action arbitration was not possible under the rules
> in effect (though that may change) therefore the case could be tried
> in New Mexico state court.
>
> (It was one of those cases where I wanted both sides to lose.
> Dell, for making it practically impossible to get relief if
> they had actually done something wrong. The plaintiff, for
> caring about the trivial difference between 2^30 and 10^9.)
> -- end quote --
>
> But someone else says the normal manner of business is to use
> 1,000,000,000 bytes for a gigabyte.
>
> I thought that formatted drives had fewer bytes capacity than
> unformatted drives, because of the overhead used by formatting** but
> that on the box and in the advertising for a hard drive, a gigabyte
> meant 1024^3. Is that true?
>
>
> **Space allowed for directory information is the only overhead I can
> think of. Is there more?
>
>
>
> If you are inclined to email me
> for some reason, remove NOPSAM :-)


This is a recurring confusion, and it has been a mix-up for many years now.

There have been attempts at changing the prefixes, but those are not so
widely spread or used. How many uses Kibi- or Gibi-bytes in specs today?

IEC:s definition: http://www.iec.ch/zone/si/si_bytes.htm

(If we only had eight fingers, I'd think that problem would not exist.)

/Rolf

mm <NOPSAMmm2005@bigfoot.com> wrote:

> How big is a gigabyte in hard drive advertising and packaging?

The SI standard decimal value.

> From another newsgroup, I suppose it's true:

> Coverage of a recent New Mexico Supreme Court case said
> that states were getting bold in finding excuses to void arbitration
> agreements under general state contract law and the U. S. Supreme
> Court was consistently refusing to review the state decisions.

Like it or lump it.

> The New Mexico case was a class action suit against Dell for
> rating hard drive capacity in decimal instead of binary gigabytes.

Which hasnt got a hope in hell of succeeding even in the completely ****ed US legal system.

> The class included people who ordered computers from
> Dell's web site. Dell requires users of their web site to
> agree to Texas law and arbitration of individual claims only.

They cant 'require' anything of the sort, legally.

Thats pure bluff that only bluffs the legally pig ignorant.

> The artibtration clause and class action waiver are legal under Texas
> law. New Mexico law, the court decreed, includes a fundamental right
> to bring a class action. Class action arbitration was not possible under
> the rules in effect (though that may change) therefore the case could
> be tried in New Mexico state court.

> (It was one of those cases where I wanted both sides to lose.

More fool you.

> Dell, for making it practically impossible to get
> relief if they had actually done something wrong.

Wrong.

> The plaintiff, for caring about the trivial difference between 2^30 and 10^9.)
> -- end quote --

> But someone else says the normal manner of business is to use
> 1,000,000,000 bytes for a gigabyte.

And they are right. And its the SI standard too.

> I thought that formatted drives had fewer bytes capacity than
> unformatted drives, because of the overhead used by formatting**

Irrelevant to what is being discussed, how the size before formatting is specified.

> but that on the box and in the advertising for a
> hard drive, a gigabyte meant 1024^3. Is that true?

Its always true of the datasheet, it may or may not be true of the box and the advertising.

> **Space allowed for directory information is the only overhead I can think of. Is there more?

Yes, other stuff like the FAT etc.

Rolf Blom <all.spam@round.bin> wrote:
> On 07/07/08 04:59 PM, mm wrote:
>> How big is a gigabyte in hard drive advertising and packaging?
>>
>> From another newsgroup, I suppose it's true:
>>
>> Coverage of a recent New Mexico Supreme Court case said that
>> states were getting bold in finding excuses to void arbitration
>> agreements under general state contract law and the U. S. Supreme
>> Court was consistently refusing to review the state decisions.
>>
>> The New Mexico case was a class action suit against Dell for
>> rating hard drive capacity in decimal instead of binary gigabytes.
>> The class included people who ordered computers from Dell's web
>> site. Dell requires users of their web site to agree to Texas law
>> and arbitration of individual claims only. The artibtration clause
>> and class action waiver are legal under Texas law. New Mexico law,
>> the court decreed, includes a fundamental right to bring a class
>> action. Class action arbitration was not possible under the rules
>> in effect (though that may change) therefore the case could be tried
>> in New Mexico state court.
>>
>> (It was one of those cases where I wanted both sides to lose.
>> Dell, for making it practically impossible to get relief if
>> they had actually done something wrong. The plaintiff, for
>> caring about the trivial difference between 2^30 and 10^9.)
>> -- end quote --
>>
>> But someone else says the normal manner of business is to use
>> 1,000,000,000 bytes for a gigabyte.
>>
>> I thought that formatted drives had fewer bytes capacity than
>> unformatted drives, because of the overhead used by formatting** but
>> that on the box and in the advertising for a hard drive, a gigabyte
>> meant 1024^3. Is that true?
>>
>>
>> **Space allowed for directory information is the only overhead I can
>> think of. Is there more?
>>
>>
>>
>> If you are inclined to email me
>> for some reason, remove NOPSAM :-)
>
>
> This is a recurring confusion, and it has been a mix-up for many
> years now.
>
> There have been attempts at changing the prefixes, but those are not
> so widely spread or used. How many uses Kibi- or Gibi-bytes in specs
> today?
>
> IEC:s definition: http://www.iec.ch/zone/si/si_bytes.htm
>
> (If we only had eight fingers, I'd think that problem would not exist.)

Feel free to cut off two of yours and find that it still does.

Previously mm <NOPSAMmm2005@bigfoot.com> wrote:
> How big is a gigabyte in hard drive advertising and packaging?

> From another newsgroup, I suppose it's true:

> Coverage of a recent New Mexico Supreme Court case said that
> states were getting bold in finding excuses to void arbitration
> agreements under general state contract law and the U. S. Supreme
> Court was consistently refusing to review the state decisions.

> The New Mexico case was a class action suit against Dell for
> rating hard drive capacity in decimal instead of binary gigabytes.
> The class included people who ordered computers from Dell's web
> site. Dell requires users of their web site to agree to Texas law
> and arbitration of individual claims only. The artibtration clause
> and class action waiver are legal under Texas law. New Mexico law,
> the court decreed, includes a fundamental right to bring a class
> action. Class action arbitration was not possible under the rules
> in effect (though that may change) therefore the case could be tried
> in New Mexico state court.

> (It was one of those cases where I wanted both sides to lose.
> Dell, for making it practically impossible to get relief if
> they had actually done something wrong. The plaintiff, for
> caring about the trivial difference between 2^30 and 10^9.)
> -- end quote --

> But someone else says the normal manner of business is to use
> 1,000,000,000 bytes for a gigabyte.

1G<something> = 1'000'000'000<something>. That is SI and the law
almost everywhere. The 1024*1024*1024 = 1GiB, but that is
just a newer IEC standard, and not the law. HDD manufacturers
are doing this right. A large fraction of the computing industry
is doing it wrong, including Microsoft.

The reason that lawsuit was ''sucessful'' is the general
incompetence of juries and the emssed up state of the US
legal system. 1GB is 10^9 bytes and has allways been that.
The inaccurate value is 2^30.

> I thought that formatted drives had fewer bytes capacity than
> unformatted drives, because of the overhead used by formatting** but
> that on the box and in the advertising for a hard drive, a gigabyte
> meant 1024^3. Is that true?

Not anymore. Also HDD vendors hal allways used the correct units.

> **Space allowed for directory information is the only overhead I can
> think of. Is there more?

Actuually there is suync info, trach herade4r, sechor headers, sector
ID fields, etc.. To give you an idea, a 2MB floppy formats to
1.44MB.

Arno

mm wrote:

> I thought that formatted drives had fewer bytes capacity than
> unformatted drives, because of the overhead used by formatting** but
> that on the box and in the advertising for a hard drive, a gigabyte
> meant 1024^3. Is that true?

As others have said, HD vendors use the SI (and NIST) standard definitions,
including 1GB = 1,000,000,000B.

Yes, formatting consumes capacity. There is confusion in the M$ world
because formatting means more than one thing. Low-level formatting of
PC HDs is a factory-only operation that lays down tracks (with their
embedded servo info) and sectors. High-level formatting, done by the
DOS command FORMAT (or the equivalent, such as XP's Disk Management),
creates the structures of a file system on a HD's partition. And,
partitioning a HD (the prerequisite to FORMAT'ing) could be called a
mid-level formatting operation.

Every layer of formatting consumes space. The standard used by PC HD
vendors is that the advertised space is the guaranteed minimum number
of sectors after low-level formatting, times 512 (Bytes/sector). This
number is frequently rounded down to get something simple; i.e., the
large WDC Raptor is listed as 150GB, but it actually has user space of
150,039,945,216 Bytes (plus some spare sectors).
--
Cheers, Bob

Bob Willard wrote in news:446dnX1oX-oLze7VnZ2dnUVZ_rTinZ2d@comcast.com
> mm wrote:
>
> > I thought that formatted drives had fewer bytes capacity than
> > unformatted drives, because of the overhead used by formatting** but
> > that on the box and in the advertising for a hard drive, a gigabyte
> > meant 1024^3. Is that true?
>
> As others have said, HD vendors use the SI (and NIST) standard definitions,
> including 1GB = 1,000,000,000B.
>
> Yes, formatting consumes capacity.

> There is confusion in the M$ world

Well, apparently not just there ...

> because formatting means more than one thing.

> Low-level formatting of PC HDs is a factory-only operation

No, it's not.

> that lays down tracks (with their embedded servo info)

That's servo track writing.

> and sectors.

And that's Low Level Formatting.

> High-level formatting, done by the
> DOS command FORMAT (or the equivalent, such as XP's Disk Management),
> creates the structures of a file system on a HD's partition. And,
> partitioning a HD

> (the prerequisite to FORMAT'ing)

Wrong again.

> could be called a mid-level formatting operation.

Yeah sure, why not.
Others idiots already use it to describe Low Level Formatting in the field.

>
> Every layer of formatting consumes space. The standard used by PC HD
> vendors is that the advertised space is the guaranteed minimum number
> of sectors after low-level formatting, times 512 (Bytes/sector). This
> number is frequently rounded down to get something simple; i.e., the
> large WDC Raptor is listed as 150GB, but it actually has user space of
> 150,039,945,216 Bytes (plus some spare sectors).