WD 160gig Drive Formats to 148gig -- Normal?

[jim evans]

I just bought a Western Digital 160 gig drive. After
installing/formatting it came out to 148 gig. Seems like a big loss.
Is this normal?

jim

 

[Rod Speed]

I just bought a Western Digital 160 gig drive.
After installing/formatting it came out to 148 gig.
Seems like a big loss. Is this normal?

Yep, just two different ways of defining a GB.

The hard drive manufacturers use the decimal GB, 10^9 bytes.

Its stated in binary GBs in other areas, particularly in the OS,
1024*1024*1024 byte GBs.

 

[jim evans]

Yep, just two different ways of defining a GB.

The hard drive manufacturers use the decimal GB, 10^9 bytes.

Its stated in binary GBs in other areas, particularly in the OS,
1024*1024*1024 byte GBs.

Yes, of course. <blush>

Thanks.

jim

 

[John .]

My WD 160gb drives shows as 149.05 gb in Windows XP Disk Management.
I believe the explanation is due to difference in nomenclature, for
example 1k=1024 bytes, not 1000 bytes. etc.

 

[John .]

Following up on my last message:

1.024 x 1.024 x 1.024 x 149.05 = 160.04121

149.05 is the capacity reported by XP
160. is the capacity reported by Western Digital

 

[Arno Wagner]

My WD 160gb drives shows as 149.05 gb in Windows XP Disk Management.
I believe the explanation is due to difference in nomenclature, for
example 1k=1024 bytes, not 1000 bytes. etc.

Actually some vendors are still ignorant of ISO and IEC standards.

1000 bytes = 1kB
1024 bytes = 1kiB

See here:

http://physics.nist.gov/cuu/Units/binary.html

HDD manufacturers do it correctly. There is one special case:
RAM. Since it grows only by potences of 2, it is not really specified
in measurement units, but more informally. However at some time
ther legal requirement to use the correct units will be enforced across
the board. The people doing this have just nit (yet) noticed the problem.

Arno

 

[Arno Wagner]

Following up on my last message:

1.024 x 1.024 x 1.024 x 149.05 = 160.04121

149.05 is the capacity reported by XP
160. is the capacity reported by Western Digital

Actually that would be 149.06 GiB = 160 GB. XP just gets the unit wrong.

Arno

 

[Eric Gisin]

The only entities using KB/MB/GB incorrectly are UNIX and Windows. KB=1024 is
required by POSIX.2, and Windows just went along with what UNIX did 20 years
before.

Is someone going to make them conform to standards rather than existing
practices?

 

[Rod Speed]

The only entities using KB/MB/GB incorrectly are UNIX
and Windows. KB=1024 is required by POSIX.2, and
Windows just went along with what UNIX did 20 years before.

Bullshit it did. The use of binary KBs was
seen in DOS before Win was even thought of.

Is someone going to make them conform
to standards rather than existing practices?

There is quite a bit of use of decimal values in
Win now and quite a few where both are used.

 

[Lil' Dave]

And all this time I thought the unit of measure was bytes, not bits...
Dave

 

[J. Clarke]

And all this time I thought the unit of measure was bytes, not bits...

Huh? 149.06 GiB is 149.06 gigabinary bytes, 160 GB is 160 decimal
gigabytes. Where do you see bits?

 

[Arno Wagner]

The only entities using KB/MB/GB incorrectly are UNIX and
Windows. KB=1024 is required by POSIX.2, and Windows just went along
with what UNIX did 20 years before.

Is someone going to make them conform to standards rather than existing
practices?

Actually ISO/IEC is the legal standard. In Germany, e.g., you can be
fined for using non-legal units when selling something. RAM gets away
beacuse the size is not really a measurement. HDDs are different.
There the size can have any (almost) value and is therefore required
to use a legal unit.

POSIX.2, while nice, is not a standard for units and measurements in
any legal sense.

Arno

 

[Arno Wagner]

The only entities using KB/MB/GB incorrectly are UNIX and Windows. KB=1024 is
required by POSIX.2, and Windows just went along with what UNIX did 20 years
before.

Is someone going to make them conform to standards rather than existing
practices?

P.S.: A sample from Linux 'ifconfig':

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:15460 errors:0 dropped:0 overruns:0 frame:0
TX packets:15460 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2696519 (2.5 MiB) TX bytes:2696519 (2.5 MiB)

See the 'MiB'?

'ls' on the other hand gets it partially right. It gives 'K','M','G'
as prefix. The 'K' is for binary units only, ISO says 'k' for 1000.
However the 'i' is missing, so 'ls' does it mostly wrong.

Arno

 

[Arno Wagner]

And all this time I thought the unit of measure was bytes, not bits...
Dave

B = byte
b = bit

The base unit is 'bit'. But byte is so common it is also allowed.
Note however that networking People only use 'bit' and call a
byte an 'octett', i.e. an 8-tuple of bits.

Arno

 

[John .]

Back in 60's, I never heard it called "octett". It was simply octal,
that is a numbering system based on 8. So the octal address of for
example 0014 was actually 1*8 plus 4 or 12 decimal.

No one ever dreamed of a computer with mega, much less giga!

 

[Folkert Rienstra]

B = byte
b = bit

The base unit is 'bit'. But byte is so common it is also allowed.

Parallel buses use bytes, serial buses use bits.

 

[J. Clarke]

Back in 60's, I never heard it called "octett". It was simply octal,
that is a numbering system based on 8. So the octal address of for
example 0014 was actually 1*8 plus 4 or 12 decimal.

An "octet" is different from "octal". An "octet" is 8 bits, which used to
be different from a "byte" which was the unit of granularity of the memory
addressing--some machines had 6 bit bytes, some had 12, etc. An octet can
represent 2 hexadecimal digits or 2 and a fraction octal or decimal.

Internet addresses are formed from four octets to use a common example.

 

[Rod Speed]

Back in 60's, I never heard it called "octett". It was
simply octal, that is a numbering system based on 8.

Those two are completely different.

Yes, the octet wasnt that common in the 60s.

 

[Arno Wagner]

John . wrote:

An "octet" is different from "octal". An "octet" is 8 bits, which used to
be different from a "byte" which was the unit of granularity of the memory
addressing--some machines had 6 bit bytes, some had 12, etc. An octet can
represent 2 hexadecimal digits or 2 and a fraction octal or decimal.

Internet addresses are formed from four octets to use a common example.

Interesting. Was before my time obviously. That explains why some
communities still insist on "octet".

Arno