Kjetil Torgrim Homme wrote:
> [j.goerke@mobilcom.de]:
>> 1024 KB = 1 MB
>> 1024 MB = 1 GB
>> 1048576 KB = 1024 MB = 1 GB
>
> please get your units right. there is no such thing as "KB", it's
> either kB (1000 bytes) or KiB (1024 bytes).
>
> 1024 B = 1 KiB
> 1024 KiB = 1 MiB
> 1048576 KiB = 1 GiB
>
> 1000 B = 1 kB
> 1000 kB = 1 MB
> 1000000 kB = 1 GB

You might also want to inform others in the industry such as Intel, AMD,
IBM, Sun, etc. of this convention. For example, Intel lists their
Pentium-4 chip as having 512 KB of L2 cache. Other examples of major
vendors getting it wrong are easily found.

Doug
--
Dr. Douglas O'Neal
Manager, Bioinformatics Center
Delaware Biotechnology Institute
(302) 831-3456

"Richard B. Gilbert" <rgilbert88@comcast.net> writes:
>Kjetil Torgrim Homme wrote:
>> [j.goerke@mobilcom.de]:
>>
>>> 1024 KB = 1 MB
>>> 1024 MB = 1 GB
>>> 1048576 KB = 1024 MB = 1 GB

>> please get your units right. there is no such thing as "KB", it's
>> either kB (1000 bytes) or KiB (1024 bytes).

I don't understand why people that are so militant about making the
kilo prefix have a small k to match SI unit conventions then insist
that the kibabyte have a capital K breaking with SI conventions. If
you are going to allow that, why not allow and acknowledge that KB has
been in common use for say the last 50 years?


>> 1024 B = 1 KiB
>> 1024 KiB = 1 MiB
>> 1048576 KiB = 1 GiB
>>
>> 1000 B = 1 kB
>> 1000 kB = 1 MB
>> 1000000 kB = 1 GB

>This convention if, in fact, it is a convention, is a new one to me.
>Perhaps I've been doing it wrong for the last forty years?

Its only been a convention that started after hard drives got up into
the Gigabyte size and marketing claims that that 1000000000 bytes = 1GByte
and people found less disk space than they thought.

Wikipedia claims the KiB convetion to be inuse in less than 0.3% of
web references, but still, probably important enough to know about.

In comp.unix.solaris Kjetil Torgrim Homme <kjetilho@kaksi.ifi.uio.no> wrote:
> [j.goerke@mobilcom.de]:
>>
>> 1024 KB = 1 MB
>> 1024 MB = 1 GB
>> 1048576 KB = 1024 MB = 1 GB
>
> please get your units right. there is no such thing as "KB", it's
> either kB (1000 bytes) or KiB (1024 bytes).
>
> 1024 B = 1 KiB
> 1024 KiB = 1 MiB
> 1048576 KiB = 1 GiB
>
> 1000 B = 1 kB
> 1000 kB = 1 MB
> 1000000 kB = 1 GB

There certainly is such thing as KB--it's a common use, and has been for
decades. The fact that it's not an SI unit is only one factor.

K and k have been used interchangeably in computing for a long time, and
both have commonly meant 1024, unless you're a weaselly hard drive marketer.
Distinguishing between them is unnecessary, because there aren't any other
'k' prefices (except see below). In comparison, the distinction between
m and M is significant, since both are common prefices, meaning milli- and
mega- respectively.

SI specifies k as 1000, K as the unit Kelvin, and does not address the
powers-of-two at all. IEC invented the Ki/Mi/Gi notation quite recently,
and it hasn't caught on well primarily because it's annoying and stupid.
OK, that may be a _bit_ of a personal opinion, but it's disingenuous and
wrong to suddenly declare common use, standard practice, and existing
literature incorrect.

Besides which, "kibi" as a prefix just plain sucks.

Colin

Doug McIntyre wrote:
> "Richard B. Gilbert" <rgilbert88@comcast.net> writes:
>
>>Kjetil Torgrim Homme wrote:
>>
>>>[j.goerke@mobilcom.de]:
>>>
>>>
>>>> 1024 KB = 1 MB
>>>> 1024 MB = 1 GB
>>>> 1048576 KB = 1024 MB = 1 GB
>>>
>
>>>please get your units right. there is no such thing as "KB", it's
>>>either kB (1000 bytes) or KiB (1024 bytes).
>>
>
> I don't understand why people that are so militant about making the
> kilo prefix have a small k to match SI unit conventions then insist
> that the kibabyte have a capital K breaking with SI conventions. If
> you are going to allow that, why not allow and acknowledge that KB has
> been in common use for say the last 50 years?
>
>
>
>>>1024 B = 1 KiB
>>>1024 KiB = 1 MiB
>>>1048576 KiB = 1 GiB
>>>
>>>1000 B = 1 kB
>>>1000 kB = 1 MB
>>>1000000 kB = 1 GB
>>
>
>>This convention if, in fact, it is a convention, is a new one to me.
>>Perhaps I've been doing it wrong for the last forty years?
>
>
> Its only been a convention that started after hard drives got up into
> the Gigabyte size and marketing claims that that 1000000000 bytes = 1GByte
> and people found less disk space than they thought.
>
> Wikipedia claims the KiB convetion to be inuse in less than 0.3% of
> web references, but still, probably important enough to know about.
>
>

Actually, the confusion in disk sizes goes back much further than that;
to the earliest "Winchester" disks at least! Disk marketers liked to
claim disk sizes in megabytes meaning multiples of 10^6 whereas real
computer people thought of megabytes as 2^20. The marketing terminology
made disks seem bigger than they were.

"Richard B. Gilbert" <rgilbert88@comcast.net> writes:

>Kjetil Torgrim Homme wrote:
>> [j.goerke@mobilcom.de]:
>>
>>> 1024 KB = 1 MB
>>> 1024 MB = 1 GB
>>> 1048576 KB = 1024 MB = 1 GB
>>
>>
>> please get your units right. there is no such thing as "KB", it's
>> either kB (1000 bytes) or KiB (1024 bytes).
>>
>> 1024 B = 1 KiB
>> 1024 KiB = 1 MiB
>> 1048576 KiB = 1 GiB
>>
>> 1000 B = 1 kB
>> 1000 kB = 1 MB
>> 1000000 kB = 1 GB
>>

>This convention if, in fact, it is a convention, is a new one to me.
>Perhaps I've been doing it wrong for the last forty years?

They are SI Units; they are generally ignored when it comes to computers
except by harddrive manufacturers.

Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.