Atom 330 drives only enough pins for 32-bit physical addresses

[Christopher Browne]

On Mon, Nov 8, 2010 at 1:26 PM, Lennart Sorensen
<lsorense-1wCw9BSqJbv44Nm34jS7GywD8/FfD2ys at public.gmane.org> wrote:
> On Sat, Nov 06, 2010 at 10:55:11PM -0400, Kevin Cozens wrote:
>> Just like in the days of the hard drive wars. You could buy a drive that
>> states a capacity of 100GB on the box. You had to check the fine print to
>> find out if that is the formatted or unformatted capacity of the drive.
>
> Wrong.  You have to check the fine print for how they define GB.
> There is no such thing as formatted versus unformatted capacity.

The thing I got bit on, some years ago, was the problem of
applications defining disk requirements in terms of GB =
1024x1024x1024, when the others involved (e.g. - those selling disk
space) chose to define them in terms of 1000x1000x1000.

Every time we increase to another level (e.g. - K --> MB --> GB -->
TB), we wind up losing another 2.4% to the measurement difference.
With TB, there's very nearly a 10% difference between "binary
measurements" and "decimal measurements."

That is, to the vendors, 1TB = 1000000000000 bytes.

But to many of the rest of us, 1TB = 1099511627776 bytes, which differs by 90GB.

The one time I built an SAP server, the 7.5% difference at the GB
level bit me, because the machine was spec'ed in decimal, whereas
requirements were expressed in "binary" (ala 1024^3), and I didn't
have enough disk space :-(.  It taught me to make sure this sort of
thing was described with better specificity!

-- 
http://linuxfinances.info/info/linuxdistributions.html
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists