Actual hard drive space?

[Robert Heiling]

Base 16 only that each hex character is made up of four bits (2^4).

Bytes were created out of convenience. A byte has 256 possible values
which allowed them to represent the entire alphabet, integer numbers,
special characters and still have room for control characters.

Yes. That's all quite well known. The term came into widespread use with the
introduction of the IBM System 360 and was tied into the word size of that
system. Systems from other manufacturers often tended to use octal.

Bob

 

[Rod Speed]

Robert Heiling wrote

Base 16 only that each hex character is made up of four bits (2^4).

Must be one of those rocket scientist stupid children.

Bytes were created out of convenience.

Meaningless waffle.

A byte has 256 possible values which allowed them to represent the entire alphabet, integer
numbers, special characters and still have room for control characters.

You can do that with 7 bits too.

 

[Rod Speed]

just the puerile shit thats all it can ever manage.

 

[Cal Vanize]

You can do that with 7 bits too.

Useless bullshit answer.

So listen little shithead child. You might just learn something before
you go back to the third grade for another year.

The use of half-bytes for packed numeric data doesn't work very well
with seven bits, does it? Also didn't allow for much use of parity for
those 7-bit characterizations back then, did it. Six bits plus parity
wasn't quite enough, was it dickmouth.

 

[jameshanley39]

(e-mail address removed) wrote






More fool you.


Corse it isnt, the multiplier is completely independant of the unit being multiplied.


Wrong, as always. Most obviously with comms speeds which have never done that.


They dont do that exclusively. There have been plenty that didnt
even use bytes at all and which had none byte mulitple word sizes.

Bytes are a relative recent thing computer wise.


Irrelevant to other collections of bytes which arent referenced by
x bits, most obviously with hard drive capacity and comms speeds.


Wrong again, it is in fact used quite recently with some flavours of linux.


Wrong, most obviously with comms speeds where they never did.


Only the pig ignorant fools with hard drive capacity and comms speeds.


Pathetic really.


There was no 'new terminology' with comms speeds, the decimal
multipliers were always what was used with bytes and bits.


Only with SOME memory, memory that is binary organised and that isnt even all memory.-

interesting.

Would you say that the convention goes by the organisation(whether
binary or not). And Convention follows that.

I notice that windows measures hard drive capacity as 2^x.

I don't know whether windows measures communication speed as 2^x? I
always assumed it does.

I can imagine that windows would just use 2^x because it wants to use
one uniform convention.

Is windows wrong? What of other operating systems? and partitioning
programs?

 

[jameshanley39]

IMO kony has been on the losing side of this whole argument. I'm somewhat
surprised.


In fact, it is usually represented by 2 hexadecimal characters (base 16), not to
be confused with decimal (base 10).



It became convenient to adopt those terms by convention, but that doesn't change
their true meaning. The burden in any discussion is upon people to define their
terms, either implicitly or explicitly. When speaking of data storage, the
original meaning of those terms is in effect and there is no convention in
effect to have them mean otherwise. If some people misunderstand that, then it's
like any similar situation where they need to educate themselves.


Whenever I have used those terms in discussions with fellow software types, it
has always been clear as to how the terms were being used in any given context.
If not, it's pretty easy to clarify by asking a question.



And likewise for Kilobytes in data transmission.


Depends om who's talking and if they are using a certain convention. Unless they
have agreed to use 1024 for kilo as a convention, then kilo is 1000 as in
kilogram, kilometer, etc and also Kilobytes, as in KBps.

Bob-

interesting. I wasn't aware of the data transmission convention
regarding Bytes , being standard SI.

As Word size grows (and i'm not sure if Word size, is size of data bus
or size of cpu registers). But if we're talking "data bus", then I
suppose that'd be standard SI too ?

What would you say is the convention for HDDs..
If the convention comes from designing them then it may not be not so
well known generally because the average person isn't designing them?!
If designers go by the apparently non binary organisation. And use
10^x, then it may be better known.

Most peoples' usage of the term Byte with HDDs goes by how the OS
defines it. Windows uses the 2^x form. If other OSs and software
does the same then perhaps 2^x should be the convention for HDDs.

 

[Robert Heiling]

interesting.

Would you say that the convention goes by the organisation(whether
binary or not). And Convention follows that.

I notice that windows measures hard drive capacity as 2^x.

Windows *reports* *filesystem* space in 2^x units of measure.

I don't know whether windows measures communication speed as 2^x? I
always assumed it does.

I can imagine that windows would just use 2^x because it wants to use
one uniform convention.

It doesn't. Windows reports communication speed base 10, as in 100 Mbps.

Is windows wrong? What of other operating systems? and partitioning
programs?

It's the responsibility of the reader/user to know if a convention is being used
or not used. There is no conspiracy in regards to size of HD's, only poorly
informed buyers.

Bob

 

[Robert Heiling]

interesting. I wasn't aware of the data transmission convention
regarding Bytes , being standard SI.

I wouldn't be so quick to toss in the word "standard", but it;s the way that
most of the world seems to report it.

As Word size grows (and i'm not sure if Word size, is size of data bus
or size of cpu registers). But if we're talking "data bus", then I
suppose that'd be standard SI too ?

What would you say is the convention for HDDs..
If the convention comes from designing them then it may not be not so
well known generally because the average person isn't designing them?!
If designers go by the apparently non binary organisation. And use
10^x, then it may be better known.

No need to even get into that as you're dealing with the finished product. The
manufacturer reports the size in base 10. Windows reports in base 2. It's a
surprise only to the uninformed.

Most peoples' usage of the term Byte with HDDs goes by how the OS
defines it. Windows uses the 2^x form. If other OSs and software
does the same then perhaps 2^x should be the convention for HDDs.

Shrug.

Bob

 

[jameshanley39]

:

No need to even get into that as you're dealing with the finished product. The
manufacturer reports the size in base 10. Windows reports in base 2. It's a
surprise only to the uninformed.

when you say "base 2" and "base 10" you mean base as in
base^exponent, rather than binary and decimal, right?

As i'm sure we agree,
We could use any number system (binary,decimal,octal) to represent
either number.. A different number system doesn't change the
quantity.
The quantitiy is changing due to convention for kilo,mega e.t.c. Not
due to a change of number system.

<snip>

 

[jameshanley39]

Windows *reports* *filesystem* space in 2^x units of measure.



It doesn't. Windows reports communication speed base 10, as in 100 Mbps.

maybe for Ethernet.


But what about download and upload speeds. ISP's talk of 512KB
download, 256KB upload.
512 and 256 are those numbers that are written with a flick of a
binary digit. But is the count for Kilo using 2^10? As your reasoning
might indicate looking at the numbers. Or does it use 10^3 as your
reasoning went that data comm uses 10^x.

Suppose at one point your download drops from 128KB to 100KB/s I
wouldn't then conclude that before Kilo was 1024 and now Kilo is
1000.

The count doesn't give it away at all.

I suppose comm speeds would encompass speed of bytes on the data bus
(which may one day be in KB), on an IDE or SATA cable, CDROM speed,
ADSL line e.t.c. I'm not sure that all comm speeds would use the
decimal convention. I don't know. But I take issue with the
reasoning you use behind your claim that data comm is all 10^x.

It's the responsibility of the reader/user to know if a convention is being used
or not used. There is no conspiracy in regards to size of HD's, only poorly
informed buyers.

With technologies there are battles between engineers and marketers.
Or worse, engineers go with the market and there is not much
division. It would be naiive to think there is no politics in
choosing conventions or terminology. To dismiss the politics as "a
conspiracy" is silly (on 2 levels - since you imply that conspiracies
should be dismissed- a silly implication to make!).

I don't know if it was an innocent choice, or marketting. But the only
way I might assume it wasn't marketting, is if Rod is right that this
(i'll say, choice of convention) is to do with whether a thing uses
what he calls "a binary organisation " or not.

 

[GT]

Can't have it both ways, either 1000 is limited to a decimal

only definition, or it is invalid in binary.

Are you seriously saying we can only use the number 1000 if we count in
decimal?!? Funny, but I can count 1000 units of anything in any base!!

In the end, use of a different base to express a quantity
MUST NECESSARILY be expressing the same quantity, not
something rounded off (unless it is expressly stated to be
rounded off).

Well, stop rounding it then - you're the one saying 10^3 = 1024!

Using decimal (base 10) to count any quantity is valid. There is nothing
wrong with writing 1000 bytes - this is a quantity. Given that 1000 is
entirely equivalent to 10^3, we can express the exact same quantity (no
rounding) as 10^3 bytes. Given that kilo is DEFINED as 10^3, then we can
express the exact same quantity, without rounding, as 1 (base10) kilobyte.
Therefore, 1 kilobyte = 1000 bytes.

I failed to clarify something in an earlier post as I was distracted by
talking about the number 1000 rather than the prefix kilo. The term kilo
means 10^3. This is the mathematical definition of Kilo and is independant
of base. In any base, 10^3 is valid - it means 1 unit from the second
'column' multiplied by itself 3 times. We are all familiar with the decimal
usage, kilo means 1 unit from the second column (base10) 10 cubed = (base10)
1000. In binary this still means 1 unit from the second column (base2) 10
cubed = (base2) 1000, which in decimal is 8. In hexidecimal this means 1
unit from the second column (base16) 10 cubed = (base16) 1000, which in
decimal is 4096.

So with all this in mind, please show us your calculation that demonstrates
1 kilobyte = 1024 bytes, without rounding?

 

[GT]

How many bits, according to you, are meant by kilobits?

1000.

Robert, Kony, everyone:

If 1 kilobits = 1000, then 8 kilobits = 8000. There are 8 bits in a byte, so
8 kilobits = 1 kilobyte, so how bytes in 1 kilobyte? If you say 1024, then
where did you get the extra 24 from?

 

[Robert Heiling]

when you say "base 2" and "base 10" you mean base as in
base^exponent, rather than binary and decimal, right?

As i'm sure we agree,

It's the same. I was attempting to be unambiguous by using the mathematical term
since we are dealing with mathematics. Binary is base 2, octal base 8, decimal
base 10, and hexadecimal base 16. See the Oxford Dictionary:
Base noun Mathematics a number used as the basis of a numeration scale.

We could use any number system (binary,decimal,octal) to represent
either number.. A different number system doesn't change the
quantity.
Correct.

The quantitiy is changing due to convention for kilo,mega e.t.c. Not
due to a change of number system.

<snip>

Just to be picky, because it's needed for the topic, the quantity doesn't change
at all, but the quantity *reported* does.

Bob

 

[Robert Heiling]

maybe for Ethernet.

Yes, either on LAN or WAN. I didn't realize that my use of the word
"communication" would be misunderstood.

But what about download and upload speeds. ISP's talk of 512KB
download, 256KB upload.
512 and 256 are those numbers that are written with a flick of a
binary digit. But is the count for Kilo using 2^10? As your reasoning
might indicate looking at the numbers. Or does it use 10^3 as your
reasoning went that data comm uses 10^x.

Perhaps this quote from a popular and respected source will help:
" Kilobit = 1000 bits per second (#14472)
For the purposes of presenting speed test results we adopt the
data-communications convention of k = 1000, not k = 1024. For example, 28.8k
modems ran at 28800 bits per second. 56k modems ran at 56000 bits per second.
http://www.dslreports.com/faq/14472

Suppose at one point your download drops from 128KB to 100KB/s I
wouldn't then conclude that before Kilo was 1024 and now Kilo is
1000.

Changing reporting method in midstream? That's unreal.

The count doesn't give it away at all.

I suppose comm speeds would encompass speed of bytes on the data bus
(which may one day be in KB), on an IDE or SATA cable, CDROM speed,
ADSL line e.t.c. I'm not sure that all comm speeds would use the
decimal convention. I don't know. But I take issue with the
reasoning you use behind your claim that data comm is all 10^x.

I made no such "all" claim.

With technologies there are battles between engineers and marketers.
Or worse, engineers go with the market and there is not much
division. It would be naiive to think there is no politics in
choosing conventions or terminology. To dismiss the politics as "a
conspiracy" is silly (on 2 levels - since you imply that conspiracies
should be dismissed- a silly implication to make!).

I don't know if it was an innocent choice, or marketting. But the only
way I might assume it wasn't marketting, is if Rod is right that this
(i'll say, choice of convention) is to do with whether a thing uses
what he calls "a binary organisation " or not.

As far as that section goes, I can't respond when words are being put in my
mouth, so discussion is ended.

Bob

 

[Robert Heiling]

Robert, Kony, everyone:

If 1 kilobits = 1000, then 8 kilobits = 8000. There are 8 bits in a byte, so
8 kilobits = 1 kilobyte, so how bytes in 1 kilobyte? If you say 1024, then
where did you get the extra 24 from?

That muddies the discussion and has a major error of omission. Here's a
corrected version.

If 1 kilobits = 1000, then 8 kilobits = 8000.
If 1 kilobits = 1024, then 8 kilobits = 8096.

You can't switch conventions in midstream.

Bob

 

[GT]

How many bits, according to you, are meant by kilobits?

That muddies the discussion and has a major error of omission. Here's a
corrected version.

I am not muddled. Using base 10 (decimal). kilo = 10^3 = 1000 not 1024. Kilo
never means 1024.

If 1 kilobits = 1000, then 8 kilobits = 8000.
If 1 kilobits = 1024, then 8 kilobits = 8096.

You can't switch conventions in midstream.

I didn't! The point I was trying to make is that we can't change the
convention at all, or this entire problem arrises. Let me try again...

We seem to all agree that 1 kilobit (1Kb) is 10^3 bits = 1000 bits. So 8
kilobits is 8x10^3 = 8000 bits. Given that there are 8 bits in a byte, then
8 kilobits must equal 1 kilobyte. So 8Kb = 1KB = 8000 bits = 1000 bytes. 1KB
= 1000 Bytes. It can't be 1024. It isn't 1024. 1024 has nothing to do with
it. This is my point. my question is where does everyone keep getting the
extra 24 from? The fact that 2^10 = 1024 is unfortunately close to 1000, but
is totally irrelevant - the term kilo means 10^3, which in decimal is 1000.
FULL STOP.

 

[kony]

Why are you ducking my question, kony?

I was getting tired of the thread frankly, how many times
can one rehash this discussion? It's already a giant thread
and in the end it won't cover anything that wasn't covered
in one of the dozens it succeeds.

"Bits", as you know, is a computer term and its usage in the computerese jargon
predates "bytes". If "bytes" has some sort of effect on the meaning of kilo,
because it is computerese, then "bits" would also have that same effect
following your logic.

How many bits, according to you, are meant by kilobits?

1024, even if someone does similar to a HDD manufacturer and
tries to round down for misleading specs.

Was it supposed to be a trick question? There are several
references to it,
http://www.google.com/search?hl=en&q=define:kilobit

In the end there are only two camps - thouse who know how to
use the binary system, and those who either accidentally or
deliberately misused it.

 

[kony]

IMO kony has been on the losing side of this whole argument. I'm somewhat
surprised.

The problem with opinions is that in the end, no matter what
the opinion is, the actual quantity expressed MUST
NECESSARILY be the same quantity expresed in a different
system. In any such base change where the quantity does not
remain the same, the original base expression is valid but
the one with changing quantity is not.

Since, as I wrote in my last post, there is nothing new here
just a rehash of the same old arguments (a waste of time),
I'm done wasting time on the thread.

 

[Cal Vanize]

If 1 kilobits = 1000, then 8 kilobits = 8000.
If 1 kilobits = 1024, then 8 kilobits = 8096.

You can't switch conventions in midstream.

Bob

Not exactly. If 1 kilobits = 1024, then 8 kilobits is 8192.

 

[Robert Heiling]

I am not muddled.

I didn't suggest that you were, only the material that was presented.

Using base 10 (decimal). kilo = 10^3 = 1000 not 1024. Kilo
never means 1024.

It does mean 1024 if the group that is using the term has agreed to have it mean
1024 for their purposes. The reality of the matter is that the 1024 usage is
already in place and it isn't about to change.

con·ven·tion
5. a rule, method, or practice established by usage; custom: the convention of
showing north at the top of a map.
6. general agreement or consent; accepted usage, esp. as a standard of
procedure.
http://dictionary.reference.com/browse/convention
Find your own if you don't like that one.

I didn't! The point I was trying to make is that we can't change the
convention at all, or this entire problem arrises. Let me try again...

We seem to all agree that 1 kilobit (1Kb) is 10^3 bits = 1000 bits. So 8
kilobits is 8x10^3 = 8000 bits. Given that there are 8 bits in a byte, then
8 kilobits must equal 1 kilobyte. So 8Kb = 1KB = 8000 bits = 1000 bytes. 1KB
= 1000 Bytes. It can't be 1024. It isn't 1024. 1024 has nothing to do with
it. This is my point. my question is where does everyone keep getting the
extra 24 from? The fact that 2^10 = 1024 is unfortunately close to 1000, but
is totally irrelevant - the term kilo means 10^3, which in decimal is 1000.
FULL STOP.

I give up.

Bob