Sign in with
Sign up | Sign in
Your question

Question about the new Western Digital EARS Disks

Last response: in Storage
Share
December 31, 2009 2:28:11 AM

Hi;

So WD said with their new disk & advanced format system we should have 7-11% more space, so i'm asking the current owner, after you format the disk (1 full partition) how much do you have? With "old" disk, 1To gives 931.5Go...

:hello: 
December 31, 2009 2:51:43 AM

Hmm would there be any difference at all? I thought 1Tb would always give around 931Gb in windows due to their "1024" system. 1 Petabyte would give 909 Terabyte and 1 Exabyte would give 888 Petabyte and so on.... The new system (which I've never heard of) should show more than 1Tb if that is the case, instead of showing 1Tb while having more space at the same time.
a b G Storage
December 31, 2009 10:53:37 AM

Hard drives will still format to the same size but since each file you store will take up less space with the new 4k system, you will be able to store more files. This will not change the 1000=1024 byte "problem" we have seen.

Each file has overhead info needed to accurately tell the controller where it is on the drive. This is what they changed. It now uses less overhead info, thus freeing up space for other files.
Related resources
December 31, 2009 11:28:42 AM

This is beyond my capacity of understanding. I saw screen capture of a guy who has these disk and the old one too, both shower 931Go so it's really same, i can't understand how you can store more file with same space, sorry...
a c 415 G Storage
December 31, 2009 5:52:23 PM

Interesting - this is the first I've heard of these new drives, although I've been wondering for quite some time how long the 512-byte sector size would last in a multi-terabyte world.

This isn't going to make any difference in raw capacities for a disk with a given advertised size. An "advanced format" disk that the manufacturer markets as a 1TB drive should still contain about 1,000,000,000,000 bytes, which would be around "931GB" as reported by Explorer. That conversion between decimal "TB" and power-of-two "GB" has nothing to do with the on-disk format.

The classic reason to use a smaller sector size is to reduce space wastage. With 512-byte sectors, there would be close to an average of 256 wasted bytes per file, since the last block of the file may be nearly empty or nearly full. With 4096-byte sectors, the average wasted space per file would be about 2048 bytes.

But since the default cluster size for NTFS is 4K bytes anyway (for any disk over 2GB), it's not going to make any practical difference at all in the amount of data you can store on a drive rated at a given size (1TB, for example).

So, aside from potential boot disk incompatibilities with some BIOSes and a slight improvement in transfer performance, this is going to be completely transparent to pretty much everyone. You can think of it like the choice between a 1TB disk with 2 500GB platters or 4 250GB platters - it's an internal design decision made by the drive manufacturer.
January 1, 2010 12:32:21 AM

Ok so WD fooled us. When news website announce "7-11% space gain" and finally, it's all the same, i call this marketing lie :fou:  So if it's so, i'll buy the EADS version since it's cheaper.
a c 114 G Storage
January 1, 2010 12:48:39 AM

They been lying about size for for years....you expected different ? :) 
January 1, 2010 3:05:44 AM

I couldn't really imagine we'll have 11% more but a little yes, the "old lie" is really old so i was never surprised of it but now they announced that new thing so i really expected... F**** off WD!
a b G Storage
January 4, 2010 5:17:06 PM

....curious....why has this feature been applied only to the 'Green Drives'?

Something is a foot here in these woods!

BTW...

The problem with hard disk drive capacity is that manufacturers assume that kilobyte (KB), megabyte (MB), gigabyte (GB) and terabyte (TB) are different things from what they really are, making you to have a hard disk drive with less capacity than advertised. This problem is known by several names, like “rounding”, “formatted capacity vs. unformatted capacity”, etc. Some people even wrongly assume that the operating system is the villain behind the vanishing of space, but the truth of the matter is that the hard drive manufacturers are the one to blame, as they announce their products with a capacity higher than the real drive capacity.

Unit Symbol Base2 Base10

Kilo K 2^10 10^3
Mega M 2^20 10^6
Giga G 2^30 10^9
Tera T 2^40 10^12
Peta P 2^50 10^15
Exa E 2^60 10^18

For example, hard disk drive manufacturers assume that 1 GB equals to 1 billion (10^9) bytes, while in fact 1 GB equals to 1,073,741,824 (2^30) bytes.

Let’s take a real example, Seagate/Maxtor DiamondMax 21 hard disk drive with “250 GB”. It is announced as being a 250 GB hard disk drive, having 488,397,168 sectors. With this number of sectors we can easily find out that the capacity of this hard disk drive is of 250,059,350,016 bytes, or 232.88 GB and not 250 GB. So here is why your 250 GB hard drive is only formatted with 232 GB: it IS a 232 GB hard drive!

That's 250,059,350,016 / 1,073,741,824 = 232.88GB

Feel free to read the rest of the article here:

http://www.hardwaresecrets.com/article/482/1

:p 

a c 415 G Storage
January 4, 2010 9:48:55 PM

elmo2006 said:
hard disk drive manufacturers assume that 1 GB equals to 1 billion (10^9) bytes, while in fact 1 GB equals to 1,073,741,824 (2^30) bytes.
The metric system has been around a lot longer than computers have - in fact 1K really does equal 1,000, 1M = 1,000,000 and 1G = 1,000,000,000.

The problem is that the OS manufacturers insist on reporting disk sizes in binary powers of 2 rather than even powers of 10. Yet transfer rates and network speeds are reported based on power-of-10 numbers. For all the "user-friendliness" that companies like Microsoft claim to espouse, this is a really stupid decision IMHO. It leads to numbers that have no obvious consistency, and is the result of endless posts by people wondering where their disk space went.

I want to make it clear that I've been a techhead for over 3 decades, am completely fluent with binary and hexadecimal numbers, both signed and unsigned, and still think that having a "user-friendly" OS report disk and file sizes using binary powers of 2 is stupid.

To be technically accurate, if you really want to use power-of-2 notation then you should use the KiB, MiB, GiB or TiB suffixes and not KB, MB, GB or TB - see: this Wikipedia article
a b G Storage
January 5, 2010 12:37:57 PM

:hello:  Sminlal...

You have made some really good points however it would appear our perception or perhaps understanding of who's at fault may not be aligned. I have found that healthy and respectful debate on contrasting views is typcially benficial to all interested parties.
Personally speaking I believe that the HDD manufacturers are to blame as it would be more cost effective to 'label' the drives via the binary concept than it is to redesign the modern computer.

:lol: , it would be odd to have a label on a HDD read 232GB instead of 250GB!

Some history....

The smallest amount of transfer is one bit. It holds the value of a 1, or a 0. (Binary coding). Eight of these 1's and zero's are called a byte.

Why eight? The earliest computers could only send 8 bits at a time, it was only natural to start writing code in sets of 8 bits. This came to be called a byte.

A bit is represented with a lowercase "b," whereas a byte is represented with an uppercase "b" (B). So Kb is kilobits, and KB is kilobytes. A kilobyte is eight times larger than a kilobit.

How many bytes are in a kilobyte (KB)? One may think it's 1000 bytes, but its really 1024. Why is this so? It turns out that our early computer engineers, who dealt with the tiniest amounts of storage, noticed that 2^10 (1024) was very close to 10^3 (1000); so based on the prefix kilo, for 1000, they created the KB. (You may have heard of kilometers (Km) which is 1000 meters). So in actuality, one KB is really 1024 bytes, not 1000. It's a small difference, but it adds up over a while.

The MB, or megabyte, mega meaning one million. Seems logical that one mega (million) byte would be 1,000,000 (one million) bytes. It's not however. One megabyte is 1024 x 1024 bytes. 1024 kilobytes is called one Megabyte. So one kilobyte is actually 1024 bytes, and 1024 of those is (1024 x 1024) 1048576 bytes. In short, one Megabyte is really 1,048,576 bytes.

All of this really comes into play when you deal with Gigabytes, or roughly one billion bytes. One real Gigabyte is actually 1024 bytes x 1024 bytes x 1024 bytes...1,073,741,824. However, most people like to simplify this by simply saying that one Gigabyte is only 1,000,000,000 (one billion) bytes; which makes sense because the prefix Giga means one billion.

...taken from....

http://www.smartftp.com/support/kb/bits-bytes-mega-giga...

:bounce: 
January 9, 2010 2:17:53 AM


elmo2006 is right, the K's, M's, G's... in the computing world these are base2 because of a reason. In that sense, the French were very smart to name their "byte", "octet" (or set of 8) since the very beginning - so they have Ko, Mo, Go... As opposite to the English language that came up recently with a new nomenclature to distinguish base2 quantities from the standard base10 quantities by adding an 'i' to distinguish binary nomenclature from decimal, and in an attempt on end the eternal discussion about storage marketing.
Now to address the claims of WD, if you read properly, they're not claiming that the physical storage of a given platter will increase by using the new sector size of 4KiB (4096Bytes), but the actual useful capacity of the drive will increase. Some one earlier in this thread already tried to explained this. If you read through WD white papers and do a proper search on the net, you'll find a lot of literature that supports this claim, and not only from WD. So, I wouldn't call WD a lier, in fact, I'll be buying a couple of these new drives and start checking them out (I'm sure data transfers will also improve by the fact that there is less overhead.
a c 415 G Storage
January 9, 2010 2:31:32 AM

trancos said:
elmo2006 is right, the K's, M's, G's... in the computing world these are base2 because of a reason.
The use of the K/M/G metric multipliers in the computer world is inconsistent. MHz/GHz are powers of ten. MByte/sec, GBit/sec are powers of 10. It's only for RAM and disk storage that power-of-two notation is commonly used.

RAM sizes are always powers of 2 because physical RAM chips have binary address connections. Therefore the capacity of each RAM chip is always a power of 2.

That limitation doesn't apply to disk drives. Manufacturers cram as much space as they can onto each track and platter with no regard for it being an exact power of two. So there's no compelling argument for stating disk capacity in binary powers of two. And it causes a lot of confusion - so why do it at all?
a b G Storage
January 11, 2010 1:54:57 AM

elmo2006 said:
:hello:  Sminlal...

You have made some really good points however it would appear our perception or perhaps understanding of who's at fault may not be aligned. I have found that healthy and respectful debate on contrasting views is typcially benficial to all interested parties.
Personally speaking I believe that the HDD manufacturers are to blame as it would be more cost effective to 'label' the drives via the binary concept than it is to redesign the modern computer.

:lol: , it would be odd to have a label on a HDD read 232GB instead of 250GB!

I disagree. As Sminlal pointed out, the metric system came first, not the other way around, and it is already used widely as powers of 10 (including in other portions of the computer). There's no reason at all to report disk capacities in binary, and it only causes confusion (and the confusion only increases with larger capacity drives).

elmo2006 said:

Some history....

The smallest amount of transfer is one bit. It holds the value of a 1, or a 0. (Binary coding). Eight of these 1's and zero's are called a byte.

Why eight? The earliest computers could only send 8 bits at a time, it was only natural to start writing code in sets of 8 bits. This came to be called a byte.

A bit is represented with a lowercase "b," whereas a byte is represented with an uppercase "b" (B). So Kb is kilobits, and KB is kilobytes. A kilobyte is eight times larger than a kilobit.

How many bytes are in a kilobyte (KB)? One may think it's 1000 bytes, but its really 1024. Why is this so? It turns out that our early computer engineers, who dealt with the tiniest amounts of storage, noticed that 2^10 (1024) was very close to 10^3 (1000); so based on the prefix kilo, for 1000, they created the KB. (You may have heard of kilometers (Km) which is 1000 meters). So in actuality, one KB is really 1024 bytes, not 1000. It's a small difference, but it adds up over a while.

This is your first error. A kilobyte is defined as 1000 bytes. This is the standard use of the metric prefix "kilo", and it iremains consistent here. Although operating systems use kilobyte to mean 1024, they are incorrect. 1024 bytes is actually a kebibyte (KiB), which can also be written as 2^10 bytes.

elmo2006 said:


The MB, or megabyte, mega meaning one million. Seems logical that one mega (million) byte would be 1,000,000 (one million) bytes. It's not however. One megabyte is 1024 x 1024 bytes. 1024 kilobytes is called one Megabyte. So one kilobyte is actually 1024 bytes, and 1024 of those is (1024 x 1024) 1048576 bytes. In short, one Megabyte is really 1,048,576 bytes.

All of this really comes into play when you deal with Gigabytes, or roughly one billion bytes. One real Gigabyte is actually 1024 bytes x 1024 bytes x 1024 bytes...1,073,741,824. However, most people like to simplify this by simply saying that one Gigabyte is only 1,000,000,000 (one billion) bytes; which makes sense because the prefix Giga means one billion.

...taken from....

http://www.smartftp.com/support/kb/bits-bytes-mega-giga...

:bounce: 

The remainder of this is wrong for the same reason. Megabytes are indeed 10^6 bytes (1 million). Gigabytes are indeed 10^9, and terabytes are in fact 10^12. Mebibytes (MiB) are 2^20, Gebibytes (GiB) are 2^30, and Tebibytes (TiB) are 2^40, which correspond to the numbers you listed above.
a b G Storage
January 11, 2010 12:05:15 PM

:hello:  @cjl

cjl said:
I disagree. As Sminlal pointed out, the metric system came first, not the other way around, and it is already used widely as powers of 10 (including in other portions of the computer). There's no reason at all to report disk capacities in binary, and it only causes confusion (and the confusion only increases with larger capacity drives)..


LOL, I did not dispute the metric system and what came first, however; the binary system is used internally by all modern computers. In most cases the same "kilo" prefix continues to be used whether the meaning is a power of ten or a power of two.

cjl said:
This is your first error. A kilobyte is defined as 1000 bytes. This is the standard use of the metric prefix "kilo", and it iremains consistent here. Although operating systems use kilobyte to mean 1024, they are incorrect. 1024 bytes is actually a kebibyte (KiB), which can also be written as 2^10 bytes.

The remainder of this is wrong for the same reason. Megabytes are indeed 10^6 bytes (1 million). Gigabytes are indeed 10^9, and terabytes are in fact 10^12. Mebibytes (MiB) are 2^20, Gebibytes (GiB) are 2^30, and Tebibytes (TiB) are 2^40, which correspond to the numbers you listed above.


Usage of these terms (Kib, Mib, Gib...) was intended (introduced by the IEC) to avoid the confusion, common in describing storage media, as to the ambiguous meaning of "kilobyte". Thus the term kibibyte has been defined to refer exclusively to 1,024 bytes.

HOWEVER, as the IEC Prefix was introduced back in 2000, today; 10 years later, storage devices still use the metric system to describe storage capabilities. Looking through WD's or Seagate's website, there is no mention of HDD's that are 'marketed' as 250Gib but instead use 250GB (powers of 10) and herein lies the problem.

It's all about marketing and as such, the terms Gigabyte and Terabyte sounds plenty better to the average consumer than Gibibyte and Tebibyte. And until this is corrected, these debates will continue to exist.

Though you have brought about some great responses, it is still in my opinion that the HDD and/or storage manufactures are at fault.

These are all great responses and I welcome dialogue of this nature as it allows for the sharing of information and to allow others to critique. Please keep the comments/debates going...

:lol: 

a b G Storage
January 11, 2010 3:56:11 PM

trancos said:
elmo2006 is right, the K's, M's, G's... in the computing world these are base2 because of a reason. In that sense, the French were very smart to name their "byte", "octet" (or set of 8) since the very beginning - so they have Ko, Mo, Go... As opposite to the English language that came up recently with a new nomenclature to distinguish base2 quantities from the standard base10 quantities by adding an 'i' to distinguish binary nomenclature from decimal, and in an attempt on end the eternal discussion about storage marketing.
Now to address the claims of WD, if you read properly, they're not claiming that the physical storage of a given platter will increase by using the new sector size of 4KiB (4096Bytes), but the actual useful capacity of the drive will increase. Some one earlier in this thread already tried to explained this. If you read through WD white papers and do a proper search on the net, you'll find a lot of literature that supports this claim, and not only from WD. So, I wouldn't call WD a lier, in fact, I'll be buying a couple of these new drives and start checking them out (I'm sure data transfers will also improve by the fact that there is less overhead.


Please keep us informed if and when you do purchase these drives. From what I have read it does make sense and it's pretty smart from WD's standpoint - get rid of the extra overhead and re-use that space for storage!
January 11, 2010 4:15:07 PM

@elmo, its actually the OS's fault, they should be reporting 931GiB (notice the i), that is the base 2 equivalent of the metric system where KiB = 1024B
a b G Storage
January 11, 2010 4:29:00 PM

mindless728 said:
@elmo, its actually the OS's fault, they should be reporting 931GiB (notice the i), that is the base 2 equivalent of the metric system where KiB = 1024B

:hello:  mindless728

Please refer to http://www.hardwaresecrets.com/article/482/1

LOL, this is where all the confusion is some say 'toematoe' and some say 'tomato'!

:pt1cable: 
January 11, 2010 4:35:16 PM

elmo2006 said:
:hello:  mindless728

Please refer to http://www.hardwaresecrets.com/article/482/1

LOL, this is where all the confusion is some say 'toematoe' and some say 'tomato'!

:pt1cable: 


it's not tomato and toematoe, its right and wrong, i do believe the SI standards are KB = 1000B and KiB = 1024B, since HDD manufacturers use GB they are conforming to the base 10 SI standard where as windows uses GB (means GiB) as the base 2 standard, btw i do believe this is not a problem in *nix (including OSX) as when they report GB they are using base 10, so really microsoft just has to change there GB to GiB or start reporting the base 10 GB equivalent

EDIT: if this seems any way rude, sorry it wasn't meant that way
a b G Storage
January 11, 2010 4:46:17 PM

mindless728 said:
it's not tomato and toematoe, its right and wrong, i do believe the SI standards are KB = 1000B and KiB = 1024B, since HDD manufacturers use GB they are conforming to the base 10 SI standard where as windows uses GB (means GiB) as the base 2 standard, btw i do believe this is not a problem in *nix (including OSX) as when they report GB they are using base 10, so really microsoft just has to change there GB to GiB or start reporting the base 10 GB equivalent


Great response...but please refer to the following for further reading ....http://en.wikipedia.org/wiki/Kibibyte and http://en.wikipedia.org/wiki/Gigabyte

The latter link is really interesting....

Please note:

The difference between units based on SI and binary prefixes increases as a semi-logarithmic (linear-log) function—for example, the SI kilobyte value is nearly 98% of the kibibyte, a megabyte is under 96% of a mebibyte, and a gigabyte is just over 93% of a gibibyte value. This means that a 300 GB (279 GiB) hard disk is indicated only as 279 GB. As storage sizes increase and larger units are used, this difference becomes even more pronounced. Some legal challenges have been waged over this confusion such as a suit against Western Digital.[5][6] Western Digital settled the challenge and added explicit disclaimers to products that the usable capacity may differ from the advertised capacity.



;) 
January 11, 2010 4:51:53 PM

my bad, the base 2 is the IEC standard (not SI like i said), but yes, the difference gets worse as you go higher prefixes

though at the same time, i don't see this as a problem, its just the annoying questions we get about not seeing the full 1TB (only getting 931GB)
a b G Storage
January 11, 2010 4:54:32 PM

mindless728 said:
my bad, the base 2 is the IEC standard (not SI like i said), but yes, the difference gets worse as you go higher prefixes

though at the same time, i don't see this as a problem, its just the annoying questions we get about not seeing the full 1TB (only getting 931GB)


+1
I agree wholeheartedly!

:D 
a c 415 G Storage
January 11, 2010 4:57:36 PM

OK, so let's say that the world is a perfect place and the hard drive manufacturers continue to use (decimal) GB (which makes perfect sense for them because it produces a number that looks bigger), and the OS manufacturers actually display the "GiB" suffix to indicate that their numbers are based on power-of-two multipliers.

Technically speaking, in such a world nobody would be "wrong".

BUT - it's still going to be confusing for your typical non-technical user. You're still going to get questions about why their 1GB flash drive only holds 956KiB, they're still going to scratch their heads when they see the same file reported with three apparently different sizes:

1,021,758 KiB -> Explorer "Details" view
997MiB (1,042,282,240 bytes) -> Explorer "Properties" dialog

Again I say - for companies like Microsoft who boast about user friendliness, this makes no sense!. If you have a file that's 1,000,000 bytes long, then why on earth would you choose to display it as "976KiB"?
January 11, 2010 5:03:14 PM

sminlal said:
OK, so let's say that the world is a perfect place and the hard drive manufacturers continue to use (decimal) GB (which makes perfect sense for them because it produces a number that looks bigger), and the OS manufacturers actually display the "GiB" suffix to indicate that their numbers are based on power-of-two multipliers.

Technically speaking, in such a world nobody would be "wrong".

BUT - it's still going to be confusing for your typical non-technical user. You're still going to get questions about why their 1GB flash drive only holds 956KiB, they're still going to scratch their heads when they see the same file reported with three apparently different sizes:

1,021,758 KiB -> Explorer "Details" view
997MiB (1,042,282,240 bytes) -> Explorer "Properties" dialog

Again I say - for companies like Microsoft who boast about user friendliness, this makes no sense!. If you have a file that's 1,000,000 bytes long, then why on earth would you choose to display it as "976KiB"?


my guess, because it is simpler and faster to use bit shifting (dividing by power of 2) than actual division by powers of 10

personally i would love for microsoft to use base 10 so both software and hardware are in line with each other
a c 415 G Storage
January 11, 2010 5:19:43 PM

Here's the acid test, IMHO:

Let's say that in Windows 7 SP1 there's an Explorer option called "Display file and drive sizes using decimal multipliers". How many people do you think would take advantage of that? I sure would. And I'm guessing the most people, even those who think that the binary multiplers are "correct", would choose it as well.

Let's face it, if I'm going to copy a file whose reported size is 500KB to a flash drive that reports having 0.5GB free, I really don't want to have to do number conversions in my head to know whether it's going to fit or not.
January 11, 2010 5:23:55 PM

oh i agree totally, and make it default behaviour (ie needing to set it away from Decimal to see the binary size

btw, has any large group of people try to get microsoft to change this
a b G Storage
January 11, 2010 6:58:34 PM

sminlal said:
OK, so let's say that the world is a perfect place and the hard drive manufacturers continue to use (decimal) GB (which makes perfect sense for them because it produces a number that looks bigger), and the OS manufacturers actually display the "GiB" suffix to indicate that their numbers are based on power-of-two multipliers.

Technically speaking, in such a world nobody would be "wrong".

BUT - it's still going to be confusing for your typical non-technical user. You're still going to get questions about why their 1GB flash drive only holds 956KiB, they're still going to scratch their heads when they see the same file reported with three apparently different sizes:

1,021,758 KiB -> Explorer "Details" view
997MiB (1,042,282,240 bytes) -> Explorer "Properties" dialog

Again I say - for companies like Microsoft who boast about user friendliness, this makes no sense!. If you have a file that's 1,000,000 bytes long, then why on earth would you choose to display it as "976KiB"?


:heink:  I'm still trying to figure that one out, you would think that 1 000 000 would actually equate to 1,048,576 bytes
a b G Storage
January 11, 2010 7:04:19 PM

sminlal said:
Here's the acid test, IMHO:

Let's say that in Windows 7 SP1 there's an Explorer option called "Display file and drive sizes using decimal multipliers". How many people do you think would take advantage of that? I sure would. And I'm guessing the most people, even those who think that the binary multiplers are "correct", would choose it as well.

Let's face it, if I'm going to copy a file whose reported size is 500KB to a flash drive that reports having 0.5GB free, I really don't want to have to do number conversions in my head to know whether it's going to fit or not.


One would have to use the value listed to the right; right mouse click on the drive ---> properties ---> refer to the 'full' value instead of the value in bytes.
Thus for a 4GB drive you would have 4 066 770 944 bytes and 3.78GB, go with the 3.78GB.

On another note, now there are three listings; the capacity marketed on the drive/flash, then two other values one in bytes and one in Gibibytes!

All this talk about bits and bytes is making me hungry! :p 
a c 415 G Storage
January 11, 2010 7:31:38 PM

elmo2006 said:
:heink:  I'm still trying to figure that one out, you would think that 1 000 000 would actually equate to 1,048,576 bytes
See how confusing it is? 976KiB is 976 x 1024 = 999,424 bytes. To be more precise, 1,000,000 bytes = 976.5625 KiB.
a b G Storage
January 11, 2010 7:35:17 PM

sminlal said:
See how confusing it is? 976KiB is 976 x 1024 = 999,424 bytes. To be more precise, 1,000,000 bytes = 976.5625 KiB.


'Clear as mud'!

:??: 
a b G Storage
January 11, 2010 7:44:20 PM

elmo2006 said:
:hello:  @cjl



LOL, I did not dispute the metric system and what came first, however; the binary system is used internally by all modern computers. In most cases the same "kilo" prefix continues to be used whether the meaning is a power of ten or a power of two.



Usage of these terms (Kib, Mib, Gib...) was intended (introduced by the IEC) to avoid the confusion, common in describing storage media, as to the ambiguous meaning of "kilobyte". Thus the term kibibyte has been defined to refer exclusively to 1,024 bytes.

HOWEVER, as the IEC Prefix was introduced back in 2000, today; 10 years later, storage devices still use the metric system to describe storage capabilities. Looking through WD's or Seagate's website, there is no mention of HDD's that are 'marketed' as 250Gib but instead use 250GB (powers of 10) and herein lies the problem.

It's all about marketing and as such, the terms Gigabyte and Terabyte sounds plenty better to the average consumer than Gibibyte and Tebibyte. And until this is corrected, these debates will continue to exist.

Though you have brought about some great responses, it is still in my opinion that the HDD and/or storage manufactures are at fault.

These are all great responses and I welcome dialogue of this nature as it allows for the sharing of information and to allow others to critique. Please keep the comments/debates going...

:lol: 

You seem to be implying here that hard drive manufacturers (for example WD and Seagate) are marketing their 250GB drives as 250GiB. You say that there is no mention on their website of hard drives that are marketed as 250GiB but instead use 250GB. The problem with this statement is that nowhere do I see hard drives advertised as 250GiB (or 500GiB, or whatever). All of them, without exception (including the SSDs) appear to market it as 250GB, with the standard base 10 prefix. They contain exactly what they are marketed as containing. Besides, this system makes perfect sense, and since our entire number system is based around base ten, this also makes the mental math quite simple for the average user.

Besides, as sminlal said, I would bet quite a bit that if everyone here had the choice between either changing the drive labels to base 2 or changing the windows readouts to base 10, they would change the readouts. It simply makes more sense to say that 1000 files of 500MB each should fill a 500GB drive, rather than the current situation where it would take 1024.
a b G Storage
January 11, 2010 8:06:27 PM

cjl said:
You seem to be implying here that hard drive manufacturers (for example WD and Seagate) are marketing their 250GB drives as 250GiB. You say that there is no mention on their website of hard drives that are marketed as 250GiB but instead use 250GB. The problem with this statement is that nowhere do I see hard drives advertised as 250GiB (or 500GiB, or whatever). All of them, without exception (including the SSDs) appear to market it as 250GB, with the standard base 10 prefix. This makes perfect sense, and since our entire number system is based around base ten, this also makes the mental math quite simple for the average user.

Besides, as sminlal said, I would bet quite a bit that if everyone here had the choice between either changing the drive labels to base 2 or changing the windows readouts to base 10, they would change the readouts. It simply makes more sense to say that 1000 files of 500MB each should fill a 500GB drive, rather than the current situation where it would take 1024.


No, the manufactures are not but in my humble opinion they should. If I implied otherwise then I errored and apologize for the confusion. We all agree that the metric system has been around prior to the actual 'desktop pc'. However we have to understand that computers comunicate via binary; bits and bytes.

I welcome change, however I would 'opinion' that HDD manufactures should re-brand or re-label their drives to accomodate the binary language or at least specifiy the actual storage space - list the Metric and maybe list the actual availability in brackets via the IEC standards. Seems much simplier than to redesign the computer! But this is just my humble opinion.

It seems that there are those that blame the software companies and then there are others that blame the HDD manufactures. Who is right or wrong is beyond the scope of this thread and I'm certain there is no easy answer to provide and I'm also sure that this topic will be debated for years to come!

One thing is for certain, I am thankful for an arena that provides so much insight without insult. I am greatful for allowing me to converse on this topic and for you all to provide positive and informative feedback.

Kudos...

;) 

February 5, 2010 12:42:47 AM

sminlal said:
That limitation doesn't apply to disk drives. Manufacturers cram as much space as they can onto each track and platter with no regard for it being an exact power of two. So there's no compelling argument for stating disk capacity in binary powers of two. And it causes a lot of confusion - so why do it at all?


A very simple reason - because the drives are being used in computers where the environment and user expectation supports binary numbers.

Think of it this way - if you were buying a car, and the manufacturer quoted fuel consumption in ounces per metre rather than in miles per gallon, how would you react? Everyone that buys the car is used to mpg and changing the rules simply confuses the customer. Given that it will cost no more to quote mpg rather than ounces per metre, the customer may quite reasonably assume the reason the unusual measure is used is just that - a blatant attempt to confuse the customer.

Why should hard drive manufacturers use 1000 for KB in an industry which expects it to mean 1024. No other reason to confuse, since it costs no more for an employee producing a pamphlet on a hard drive to divide by 1024 rather than 1000.

Best Regards,
Richard Pillay
a c 415 G Storage
February 5, 2010 1:50:38 AM

RichardPillay said:
Why should hard drive manufacturers use 1000 for KB in an industry which expects it to mean 1024. No other reason to confuse, since it costs no more for an employee producing a pamphlet on a hard drive to divide by 1024 rather than 1000.
Because in a commodity market like disk drives it makes sense to use the system that lets you put the biggest number on the box.

The "industry" is by no means 100% standardized on 1024 multipliers. As I mentioned above, frequencies and transfer rates are quoted using decimal multipliers. Why shouldn't I expect it to take 10 seconds to transfer a 1GByte file over a 100MByte/sec connection? And when I'm doing the math, why the heck should I be concerned about whether 1GB is equal to, larger than, or smaller than 10 x 100MBytes?

The problem isn't in the disk manufacturers using decimal sizes, it's in the OS vendors using binary sizes. You can't tell me that it makes sense to highlight 11 files whose size is shown as "100KBytes" and see that the total size is "1.07MBytes".

I'll ask it again: if Windows Explorer offered you one of these two options:

-> Show a 1,000,000 byte file as "1MB" and a 1,000,000,000 byte file as "1GB", OR
-> Show a 1,048,576 byte file as "1MB" and a 1,073,741,824 byte file as "1GB"

...which would you choose? I've been a computer professional for 3 decades, am fully fluent in binary and hexadecimal after writing more assembly programs that I can count, and would still choose the first option. And I'm quite certain that most other people would as well, whether they're in the computer industry or not.
February 8, 2010 6:02:06 PM

Because it doesn't make sense to use "MB" and "GB" when these already mean something specific.

It should say "100 million bytes/sec" and "1.5 trillion byte drive"

Takes a few more letters, but at least the meaning is clear. Kilobyte, Megabyte, Gigabyte, Terabyte all have specific binary numbering meanings.
a c 415 G Storage
February 8, 2010 9:37:30 PM

Master Saul said:
Kilobyte, Megabyte, Gigabyte, Terabyte all have specific binary numbering meanings.
No they don't. 10MByte/sec always means 10,000,000 bytes per second. And in other computer usage of the same prefixes, 2GHz always means 2,000,000,000 cycles/second.
a b G Storage
February 8, 2010 11:10:11 PM

Well, maybe you can talk Linus Torvalds into that.
!