memory reported less than installed

Qiuli Han ivyharry at gmail.com
Sat Oct 11 00:49:23 UTC 2008


So, I have 8G ram installed on my machine,
htop shows 7995MB, free shows 8186940 in K

my laptop, 4G ram,
htop shows 3892MB, free shows 3985452 in K

both bios tells proper size in G, which i dont know if i should say B or b

IF, it is just BIOS counts the mem in 1000, and the kernel counts it in
1024, where is my 5M gone on my workstation?

on the other hand, my laptop has dedicated video card, i am missing 118M if
it counts 1000, and 204M if counts in 1024.

And, as far as I know, memory should go with 1024, not 1000

First i thought it is an aprx 2% memory reservation thing, but it does not
seems on my laptop, it is 4.98%

arrrghhhhhh, math......





On Tue, Aug 12, 2008 at 4:02 PM, NoOp <glgxg at sbcglobal.net> wrote:

> On 08/12/2008 11:05 AM, Johnny Rosenberg wrote:
> > 2008/8/11 NoOp <snip>
> [snip]
> >>
> >> Regarding memory manufacturer's, this is interesting:
> >> http://www.edgetechcorp.com/support/technical-glossary.asp
> >> <quote>
> >> MEGABIT
> >> Amount of memory equal to 1,048,576 bytes of data.(abbreviated MB)
> >>
> >> Megabyte
> >> A megabyte is composed of 1024 kilobytes or 1,048,576 bytes. Megabyte is
> >> commonly abbreviated using 'M' or 'Mb'. NOTE: In many cases, an megabyte
> >> is incorrectly stated as being 1 million bytes.
> >
> >
> > No, that's not incorrect.
> > 1 MB=1000000 bytes
> > 1 MiB=1048576 bytes
> > This is called IEEE
> > 1541<
> http://freedos-32.sourceforge.net/showdoc.php?page=standards#ieee1541>
> > .
>
> That was my point... and why I pointed to the NIST references which is a
> much more authoritative source. See:
> http://physics.nist.gov/cuu/Units/bibliography.html
> NIST is the US National Institute of Standards and Technology
> (http://nist.gov/) and in their bibliography they point to Le Système
> International d'Unites (SI)
> However I was wrong with regard to the semiconductor industry (see JEDEC
> info below).
>
> The IEC has a nice writeup here & demonstrates the differential problem
> as the number of bytes increases. http://www.iec.ch/zone/si/si_bytes.htm
>
> That said, Rashkae is correct in that the semiconductor industry
> continues to use the binary metric. The JEDEC has, for legal reasons,
> place a 'clarifier' in their standard - from JEDEC Standard No. 100B.01
> (http://www.jedec.org/download/search/JESD100B01.pdf):
>
> <quote - modified for ASCII to show ^ for powers>
> giga (G) (as a prefix to units of semiconductor storage capacity): A
> multiplier equal to 1 073 741 824 (2^30 or K^3, where K = 1024).
>
> NOTE 1 Contrast with the SI prefix giga (G) equal to 10^9, as in a
> 1-Gb/s data transfer rate, which is equal to 1 000
> 000 000 bits per second.
> </quote>
>
> The mega (M) section contains further clarification regarding the
> semiconductor useage and the SI prefix.
>
> So given the JEDEC standard, I wonder if the changes from 'MB or GB'
> (Gutsy) to  'MiB or GiB' to reflect _memory_ in the Hardy System Monitor
> is correct. As you said, it certainly would be so much easier if
> everyone followed a single published standard. :-)
>
>
>
> --
> ubuntu-users mailing list
> ubuntu-users at lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/ubuntu-users/attachments/20081010/7d0050f3/attachment.html>


More information about the ubuntu-users mailing list