[TriLUG] RHAS---300gig?

Aaron S. Joyner aaron at joyner.ws
Thu Jul 22 09:27:01 EDT 2004


Jeremy Portzer wrote:

>On Thu, 2004-07-22 at 07:13, Brian Henning wrote:
>  
>
>>>Most companies see 300000mb as 300GB, but OSs see 300000 as 300000/1024 or
>>>      
>>>
>>293GB..
>>
>>Actually (for the sake of hair-splitting...), it's worse than that.
>>Marketers see 300,000,000,000 bytes as 300 GB, while OSes will see it as
>>300,000,000,000 / (1024 * 1024 * 1024) = 279.4 GB.
>>
>>I remember reading somewhere that it's the difference between "300GB" and
>>"300 GB" but I'm not sure if that's true, and if it is I can't remember
>>which is which.
>>    
>>
>
>The unambiguous way to resolve this is to use different prefixes when
>referring to the 1024 multiples.  For examples, 300 GB = 279.4 GiB.
>1 KiB = 1024 bytes, 1 MiB = 1024 KiB, 1GiB = 1024 MiB.
>
>The Ki,Mi,Gi, prefixes are not officially part of the metric or US units
>system but are beginning to see wider acceptance.  The original document
>on this subject is here:  http://physics.nist.gov/cuu/Units/binary.html
>.  If you do a Google search on "Kibi Mebi Gibi" or something like that
>you'll see many more informative articles.
>
>I have to conclude that the drive manufacturers are "correct" when they
>use the metric prefixes exactly as defined, K=10^3, M=10^6, G=10^9. 
>
>--Jeremy
>
>  
>
Indeed - the drive manufacturers are "technically" in the right.  The 
catch is with computers it's always easier to deal in powers of 2 - 
hence the 1024, being 2^10.  Back when it was 1024 bytes, or 1024 
kilobytes, people just assumed that it was "close enough" - but as this 
discussion shows, when you scale that up a few orders of magnitude, the 
difference is much more significant.

Aaron J.



More information about the TriLUG mailing list