Searching \ for '[OT] new units for bits and bytes' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=new+units+bits+bytes
Search entire site for: 'new units for bits and bytes'.

Exact match. Not showing close matches.
'[OT] new units for bits and bytes'
1999\03\13@144243 by

while fussing around a bit on an official-looking web site about official
units, i found the following page regarding kilobits and magebytes:
http://physics.nist.gov/cuu/Units/binary.html

it says basically that, to end the confusion between, say, mega = 10^6 or
2^20 or 10^3*2^10, the ieee and the iec and iso seem to have decided on the
following units:

k (kilo) is =only= 10^3
M (mega) is =only= 10^6

2^10 is going to be "Ki" ("kibi", pronounced "keebee")
2^20 is going to be "Mi" ("mebi", pronounced "mebee")
2^30 is going to be "Gi" ("gibi", pronounced "geebee")

it'll take some time to take over, i guess... :)

ge

|it says basically that, to end the confusion between, say, mega = 10^6 or
|2^20 or 10^3*2^10, the ieee and the iec and iso seem to have decided on the
|following units:

|k (kilo) is =only= 10^3
|M (mega) is =only= 10^6

Note that, in abbreviated form, the issue had already been resolved by
the use of the "K" vs "k" prefix.  Unfortunately, all the other larger
prefixes are already uppercase even in decimal form...

|2^10 is going to be "Ki" ("kibi", pronounced "keebee")
|2^20 is going to be "Mi" ("mebi", pronounced "mebee")
|2^30 is going to be "Gi" ("gibi", pronounced "geebee")

|it'll take some time to take over, i guess... :)

What do they want to call 1,024,000 of something?  Or for that matter,
10^3*2^20 or 10^6*2^10?

At 11:09 03/19/99 -0600, John Payson wrote:
>|k (kilo) is =only= 10^3
>|M (mega) is =only= 10^6
>
>Note that, in abbreviated form, the issue had already been resolved by
>the use of the "K" vs "k" prefix.  Unfortunately, all the other larger
>prefixes are already uppercase even in decimal form...

i don't think that the use of "K" is consistent, but my memory might be
wrong here. in any case, it is not "official" :)

>|2^10 is going to be "Ki" ("kibi", pronounced "keebee")
>|2^20 is going to be "Mi" ("mebi", pronounced "mebee")
>|2^30 is going to be "Gi" ("gibi", pronounced "geebee")
>
>What do they want to call 1,024,000 of something?  Or for that matter,
>10^3*2^20 or 10^6*2^10?

they don't appear in the proposal. i guess there's no real use (and no
solution either :) to associate each number =somebody= could come up with
in =some= context with a quantity prefix... :)  i think if they (the
marketing guys :) want to use 10^3*2^10, they should use it that way,
otherwise they better stick to commonly used prefixes... it is simply and
only misleading to call 10^3*2^10 "M"

ge

<rant>
My opinion is that the entire proposal is laughable.

KiloBytes have been 1024 bytes in this context for too long for any
dictate from "on high" to have any practical influence.  The horse is
gone- close the barn door if you like.  Don't these people have
something better to do? (Like promote adoption of their existing
standards, mayhap?)
</rant>

Has anyone here EVER been unsure from the context whether a Mega- or
a Kilo-something was 1024 or 1000 multipliers?

The only case I can think of is (probably deliberate) specsmanship by
hard disk vendors, where (unlike DRAM) capacities are not naturally
binary progressions.

And besides, can you pronounce MiByte ("me-bee-byte"???) without
laughing?  Maybe-bytes?  That's the RAM where Windoze puts my data
files, right?  Sigh.

------------
Barry King, KA1NLH
Engineering Manager
NRG Systems "Measuring the Wind's Energy"
Hinesburg, Vermont, USA
barrynrgsystems.com
"The witty saying has been deleted due to limited EPROM space"

At 16:01 03/19/99 -0500, Barry King wrote:
>My opinion is that the entire proposal is laughable.

that might not be a bad thing -- there wouldn't be much to laugh in these
spheres if we hadn't the bureaucrats :)

>Has anyone here EVER been unsure from the context whether a Mega- or
>a Kilo-something was 1024 or 1000 multipliers?

i guess in communications it is common to talk about 10^6 bits per second
as Mbps, since they're closer to MHz and (decimal) statistics than to
binary calculations.

the common 1.44MB floppy seems to have a capacity of 1.44*1000*1024 bytes,
if i'm not mistaken.

i guess it's the same as with "v" or "V" for volt or "mA" or 'Mamps" for
milli amperes: as long as you stay within a relatively small context, there
is not much ambiguity to bother about, and if it is -- you always can ask.

>And besides, can you pronounce MiByte ("me-bee-byte"???) without
>laughing?  Maybe-bytes?  That's the RAM where Windoze puts my data
>files, right?  Sigh.

some work-out for your face muscles won't hurt you -- imagine how much sun
this brings into your life if you laugh every time you talk or write about
kiwibytes or maybebytes :)

while i don't think that "maybe-bytes" are exactly a catchy solution, i
think =a= solution would be nice, to make it clear once and for all what
kind of mega one is talking about. (there's only two kinds of kilo and
three of mega, but there are already 4 of giga, and because the next
windows version will require terabyte disks, we better get used to the fact
that there will be 5 versions of tera around... :)

ge

kbits, kBytes, Mbits, MBytes is a pure poor usage of the
metric system to something that has nothing to do with.

It is the same to say "hundred bits" refering to 128 bits,
or something worse.

If we don't find a new denomination for those quantities,
we will be doomed forever.

Probably in the year 2300, some people will be blaming us
almost something like the "thousand million = billion"
and imperial vs metric system...

We have this ability to create confusion and think that
somebody someday will fix it... this somebody is just us,
and this someday is just today.

Binary quantification is just 2^n and that's it.
You couldn't say 2^k, 2^M, 2^T, but you can say
kilo in decimal it means 10^k, Mega is 10^M.

So, first, the expression kbit is wrong, it doesn't fit
in any aspect, except if you are saying 1000 bits.

I don't see any problem to understand  2^10 as 1024 bits,
or 2^20 as 1048576 bits...

Even a child would understand that a memory of 2^25 bytes
is bigger than other with 2^24 bytes.  That same child
will understand that doubling size means square gain,
than it wouldn't be 2^25 but a jump from 2^24 to 2^26.

Now, how you explain it in plain kilos, Megas...terrible.

It is so much easier to see 1048576/2 as 2^(20/2), it
makes much more sense about the digital point of view
and any programming routine.

I am sorry for the ignorance, but a little bit of study
is necessary to deal with electronics or computers, and
understand those numbers is at least pre-qualifying to
be able to work in this area.

I don't expect any grandmother with her 86 years old to
be able to understand what means this, but for sure she
would not understand what means her computer has 16MB
either...

Ok, for those who want to keep playing the wrong game,
tell me, what will be called a memory with 2^60 bytes?
How long do you think it will take to have this memory
at the market? 5, 10 years? Are we ready for the
strange new name?
Isn't easy to say 2^60? what's wrong with that?

Which the following is easier to understand?

1) Which is bigger?   2^36 or 2^37
2) Which is bigger?   6871947674 or 13743895348
start to count digits at the numbers above...
how would you do it? counting two by two digits,
or three by three...are you sure you didn't skip one?

If you think the second is easier, get out of here.

Ok, you can say, it is easy, it is 6.8Giga!!!
and what you do with the 71947674 bytes? trash?

By the way, the bigger is the 13.745895348 GigaSomething

Wagner.

If you want to create more problems, just lets change the byte from the
Then a byte will have 10 bits, and will represent 1024 combinations,
than, we would be a little bit closer to what te caotic system created
with kbits and kBytes.

Then you can say that 1kByte is exactly 10^4 cells, that a MByte is
exactly 10^7 cells.

So, two bytes addressing system would access the infame actual "1MByte"

It would take byte counts to means 2^10 new steps in addressing,

1 byte  = actual infame 1kByte
2 bytes = actual infame 1MByte
3 bytes = actual infame 1GByte
4 bytes = actual infame 1TByte

It would make more sense.  I vote for that.. lets make this a complete
caos... Y2K problems would be small peanuts compared to this new
"The world is waiting the Windows2000 with bytes in decanos, not
octatos"... lets make him worried a little bit^10.

By the way, PIC doesn't use 14 bits? hmmm, it makes me think about
something...

hey, that's just a joke.

Wagner.

1999\03\19@221437 by
If we are smart enough to easily convert from teaspoons to tablespoons
to cups to ounces to pints to quarts to gallons (and back again) we
certainly should be equipped to handle kilos of 1,000 and Kilos of 1024.

On a little more serious note, I propose that we trash the decimal
system entirely.  If man were not genetically predisposed to doing most
things irrationally, being bipedal and having eight fingers would have
led him to naturally use a base 16 counting system ... and the base 10
system from the Orient and the base 12 system from the Brits would never
have contaminated our minds.

- - - Nick - - -

Nick Taylor wrote:
>
> If we are smart enough to easily convert from teaspoons to tablespoons
> to cups to ounces to pints to quarts to gallons (and back again) we
> certainly should be equipped to handle kilos of 1,000 and Kilos of 1024.
>
> On a little more serious note, I propose that we trash the decimal
> system entirely.  If man were not genetically predisposed to doing most
> things irrationally, being bipedal and having eight fingers would have
> led him to naturally use a base 16 counting system ... and the base 10
> system from the Orient and the base 12 system from the Brits would never
> have contaminated our minds.
>
>  - - - Nick - - -

it means old Brits had 6 fingers in each hand?
uhum, we are able to do several things, we can go to the Moon, to Mars,
to convert from teaspoons to tablespoons, sometimes it took some time,
even when you need to check tables and handy conversion calculators,
like 6"and 5/32 is... in millimeters...??? ahhh, here is the handy table
supplied by that screw&bolts company... let me see... hmmm, this
measurement isn't at the table... hmmm... if I take 6", is .. 6 times
2.5 = 12+3, plus 0.04 times 6 = 0.24, so its 15.24, plus 5*2.54/32, let
me see... hmmm...where is the calculator??? hmm, somewhere in between
1/6 of an inch, then... 15.24 + 0.4?? = aprox 15.64mm? oh my, I need to
print a 6 pages conversion table and stick on the wall aside of... let
me see... miles//kilometers?... no, hmmm, ounces/grams?...or perhaps I
move this 4 pages conversion table of ¡F/¡C to the right and then it
will fits.. no, it will overlap the conversion table between
Japanese/American transistors... ahh, I know, put it below the screw
conversion table... do I need to know what is #4 screw in millimeters?
just stick it over the PSI/Inches of Water table.
Done. Now I am ready to convert anything, anything at all. :)  Nautical
Knots? If 1 bit = 1mm, how many foot of memory do your computer has?
yards? miles?  if 1 clockcycle = 1 gram, how many teaspoons are your PIC
running?
Ok, at 24¡C (75.2¡F) each 15 destiled water drops make one milliliter,
so in one ounce (1 liter = 35.27oz) we have aprox 425 drops, now, if the
faucet drops 1 per second, how many time to a gallon?  Looks like that
seconds (time measurement) is the only thing everybody agree, huh?

Come on, we have better things to do instead crazy conversions.
Wagner.

> What do they want to call 1,024,000 of something?

For 3.5" floppy disks.  2,880 sectors, each of which is 1/2 KB
(where K = 2^10 = 1024 = 2 * 512).  So a standard, high density
PC floppy is 1.44 * 1,024,000 -- originally written as 1,440 KB
and now commonly written as 1.44 MB where M is (2^10 * 10^3).

Alternately, floppy is 1.475MB (M = 10^6) or 1.406MB (M = 2^20).

Lee Jones

> For 3.5" floppy disks.

Since we are off-topic already, let me mention some thing about the 3.5"
floppy disk.

It originated in Japan, as the 90mm-disk (Japan is metric territory).

When it made its way to the US, it was called the 3.5" disk (analogy
to the previous 5.25" disk).

As such it came to Europe, including Germany.

In Germany exists a law that enforces advertisers to use the metric
system. It is supposed to help customers compare products easily.
The 3.5" disk as around in Germany for roughly a decade, until
some clever brain found out that he can sue every computer shop
for advertising a disk as "being 3.5" large" (ignoring the metric
system). This is really a stupid idea of course, as this is not
really a size measurement but more a product type.

But nevertheless, in Germany you can make money from that: by
accusing the offender and taking a fee for not suing. Indeed there
are companies that live from browsing the ads daily and cashing
in on any offense to ad regulations ("Abmahnvereine").

Within weeks the 3.5" disk turned into the "8.89cm disk" all
over Germany.

That's really funny - nobody cared about the actual size of the disk,
which they were supposed to advertise after all (by a ridiculous
reason though).  Neither the advertisers, nor the "Abmahner".

They just translated inch to cm:  3.5 * 2.54 = 8.89

I wonder what the inventors of the 90mm disk feel deep inside them
when they see it offered as the "8.89cm disk"..

Is it April 1. already? How time flies...

/holger

{Quote hidden}

On Fri, 19 Mar 1999, Gerhard Fiedler wrote:

> At 16:01 03/19/99 -0500, Barry King wrote:
> >My opinion is that the entire proposal is laughable.
> that might not be a bad thing -- there wouldn't be much to laugh in these
> spheres if we hadn't the bureaucrats :)
PTM: Wasn't that the main reason behind EU, or am I mistaken ?

> >Has anyone here EVER been unsure from the context whether a Mega- or
> >a Kilo-something was 1024 or 1000 multipliers?
before and after the formatting gives you a good laugh. There is something
in 10^3 and something in 2^20=1048576 and what is left is in
(10^3)x(2^10)=1000x1024=1024000

> the common 1.44MB floppy seems to have a capacity of 1.44*1000*1024 bytes,
> if i'm not mistaken.
>
> i guess it's the same as with "v" or "V" for volt or "mA" or 'Mamps" for
> milli amperes: as long as you stay within a relatively small context, there
> is not much ambiguity to bother about, and if it is -- you always can ask.

PTM: I have eight AA accu on my table. They are advertised to have 1.5V
1200MAh. Wait until I get it connected to my car ! Let's meet in Salt Lake
City !
Good luck my multimeter has ranges 1MA..200MA, and 700V..10MV.
--------------------------------------------------------------------------
PTM, ptmustautu.fi, http://www.utu.fi/~ptmusta                 OH1HEK
Lab.ins. (mikrotuki) ATK-keskus/Mat.Luon.Tdk                    OI7234
Lab.engineer (PC support) Computer Center
Mail: Turun Yliopisto / Fysla,  20014 Turku
Pt 02-3336669, FAX 02-3335632 (Pk 02-2387010, NMT 0400-555577)
--------------------------------------------------------------------------

At 10:17 03/22/99 +0200, Pasi T Mustalahti wrote:
>On Fri, 19 Mar 1999, Gerhard Fiedler wrote:
>> that might not be a bad thing -- there wouldn't be much to laugh in these
>> spheres if we hadn't the bureaucrats :)
>PTM: Wasn't that the main reason behind EU, or am I mistaken ?

it's been =laughable= all the time, but it is a lot =funnier= since i'm not
living there anymore :)

>PTM: I have eight AA accu on my table. They are advertised to have 1.5V
>1200MAh. Wait until I get it connected to my car ! Let's meet in Salt Lake
>City !
>Good luck my multimeter has ranges 1MA..200MA, and 700V..10MV.

so you've got the right equipment for your accus... mind sending me the
supplier of the accus? i'd like to drop a few of these in my car -- need to
exchange the engine anyway :)

ge

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...