Window is stealing my HD size

[dew]

dew wrote

Initially, HD capacity was stated in characters;

Not with the PC.

and, the choice of SI or M$ for defining KB/MB/GB/etc.

It wasnt just MS that used binary MBs,
the original hard drive for the PC did too.

did not matter, since IBM spec'd HD sizes in exact (or maximum) number of characters.

No they didnt with the PC.

For ref, see the IBM Journal of R&D, Vol.1, No.1, of 1957. The IBM 350 (the first,
AFAIK, commercially offered HD) held 5,000,000 characters: 500 chars/track, 100 tracks/
surface, 2 surfaces/platter, and 50 platters on the one shaft.

Irrelevant to the PC being discussed.

And, the latter (by a decade or so) IBM 1301 supported two character sizes, 6-bit and
8-bit; IBM did not, even then, refer to those storage entities as bytes -- or to HD
capacities in MBs or GBs. (The IBM 1301 held up to 2840 6-bit or up to 2205 8-bit
characters on each of its 10,000 data tracks.)

Irrelevant to the PC being discussed.

IIRC, the major scam in HD sizing for many years was that HD vendors delivered HDs
unformatted, and quoted HD capacity in unformatted bytes -- knowing full well that the
required act of formatting took a big bite.

That is just plain wrong with the HDs, only seen with floppys.

The formatted v. unformatted capacity difference was far more significant than the
relatively trivial SI v. M$ definition of a GB.

The original hard drive for the PC, the ST506,
was sold as being 5MB, and that is binary MBs.
The unformatted capacity was stated as being 6.38MB
http://www.seagate.com/support/disc/specs/mfm/st506.html

And the use of binary values preceeded MS too.

Personally, I do not recall HD vendors ever sticking to the
powers-of-two sizes that were near-universal for RAM.

The earliest drives on the PC did use binary values.

My first home HD was a 20 MB unit,

That was well after the initial drives.

Even the ST125 21MB drive still used binary units.

latter upgraded to a whopping 30MB; seemed adequate, since that computer did not run
anything from M$.

Must be why Gates stuck pins in his Bob Willard doll.

 

[dew]

Since using google seems to be too much effort for you, here is one:

http://www.plantboard.org/weights measures law.htm

Seems to be from Arkansas.

Pity it says

The International System of Units (SI) and the system of weights and measures
in customary use in the United States are jointly recognized, and either one (1) or
both of these systems shall be used for all commercial purposes in the state.

Clearly SI units are NOT mandated in Arkansas.

You might also want to look here:

http://lamar.colostate.edu/~hillger/laws/ncwm-uniform-laws.html

Pity it says

Among other things, it recognizes the International System (SI) and the
customary system for use in the state, and adopts the various uniform laws
and regulations, including those mentioned below (a state may, of course,
pick and choose which ones it incorporates into its laws).

Clearly SI units are NOT mandated.

Where you can also find the "Uniform Packaging and Labeling
Regulation" and a document giving the adoption status of it and the
the Uniform Weights and Measures Law (5th link from the top). If I
read this correctly, every state besides Rhode Island has adopted the
Weight and Measurement law, however Rhode Island uses it as guideline.

Pity it does NOT mandate that SI units must be used.

Here is something else, which provides a neat
summary of why being metric has advantages:

http://www.law.cornell.edu/uscode/html/uscode15/usc_sec_15_00000205---a000-.html

Irrelevant to your claim about SI units being mandated by law.

They arent anywhere in the US.

Personally I think that should be more than enough.

Fraid not.

You may continue this discussion without my participation
from here onwards. You obviously have no interest in the
truth, merely in maintaining your own misconceptions.

Looks rather like the pot calling the kettle black.

Otherwise you could have easily found all the above by yourself.

Pity they dont substantiate your original claim that SI units are mandated by law.

 

[Folkert Rienstra]

Which makes sense. The raw storage units on a HD are sequential by nature,
and there's nothing particularly binary about them.

The minimum unit -the sector size- is.

The raw storage units of RAM, however, are accessed through a defined set
of binary address lines, usually located in component cases where size and
pin count is important and minimized, and therefore (so far) the usual sizes
are binary potencies.

As far as a memory slot goes.
Then it becomes linear, just as with diskdrive sectors, platters etc.

And even in a memory chip there doesn't necessarily be a full complement of
smaller units (eg the MoSys memory chips came in odd sizes using a technique
called Multibank DRAM) or some units are set aside as spares -similar to
disk drives- leaving less than a full complement of units for actual storage.

It's quite possible that RAM storage goes the same way disk storage has
gone and becomes simply a quantity, not necessarily a binary one,
disconnected from specific case requirements -- and at that point, the more
common decimal potency multipliers may start being used.

Which wouldn't be a bad thing... it's about time that 1 MB of data in RAM
fits in 1 MB of storage space on disk

It does actually. Using exactly 2048 512-byte sectors.

 

[J. Clarke]

The original PC hard drive, the ST506, used binary not decimal MBs.
http://www.seagate.com/support/disc/specs/mfm/st506.html

Run the numbers and the actual capacity isn't 5 MiB nor is it 5 MB, it
appears that they just rounded off to the closest even number.

So did quite a few drives after that too.

Did they now, or did they just round off?

<snip>

 

[J. Clarke]

Not with the PC.

So? Hard disks were around long before PCs.

It wasnt just MS that used binary MBs,
the original hard drive for the PC did too.


No they didnt with the PC.

Which has exactly what to do with the history of the nomenclature?

Irrelevant to the PC being discussed.

Who was discussing "the PC"?

Irrelevant to the PC being discussed.

Which PC is that?

That is just plain wrong with the HDs, only seen with floppys.

Nope. 'Fraid not. It's really the only honest way to do it since the drive
manufacturer has no way of knowing what filesystem the user is going to be
putting on the disk.

The original hard drive for the PC, the ST506,
was sold as being 5MB, and that is binary MBs.
The unformatted capacity was stated as being 6.38MB
http://www.seagate.com/support/disc/specs/mfm/st506.html

And the use of binary values preceeded MS too.


The earliest drives on the PC did use binary values.


That was well after the initial drives.

So was the first Seagate drive.

 

[J. Clarke]

Interesting. That was before my time.


Incidentially, bit and byte are not SI units. Not too surprising,
since they do not need any physical basis. They can simply be
counted. Only the prefixes are SI. For binary prefixes and definition
of bit and byte, I find only IEEE and ICE standards. IEEE and IEC have
'B' for byte, but IEEE uses 'b' for bit while IEC uses 'bit'.

Now, I'm curious as to your opinion on the use of "ksi", commonly used in
American engineering as shorthand for "thousand pounds per square inch".
Of course kilo to refer to 1,000 precedes the SI by several thousand years.

 

[dew]

Run the numbers and the actual capacity isn't 5 MiB nor is it 5 MB,

Its 5MiB rounded.

it appears that they just rounded off to the closest even number.

No they didnt, the capacity is stated as being 5.0MB

Even clearer with the other drives of the same era which
also used binary MBs, most obviously with the ST225.
http://www.seagate.com/support/disc/specs/mfm/st225.html
The stated 20.4MB is clearly MiB, its actually 21.4MB

Did they now, or did they just round off?

Nope, clearly they didnt, the capacity was
clearly stated to one decimal place, in MiBs.

 

[Rod Speed]

The minimum unit -the sector size- is.

As far as a memory slot goes.
Then it becomes linear, just as with diskdrive sectors, platters etc.

No it doesnt, its still binary organised.

And even in a memory chip there doesn't necessarily be a full
complement of smaller units (eg the MoSys memory chips
came in odd sizes using a technique called Multibank DRAM)
or some units are set aside as spares -similar to disk drives-
leaving less than a full complement of units for actual storage.

Completely off with the fairys now on ORGANISATION.

 

[dew]

So?

So that comment is irrelevant to MY statement about PCs.

Hard disks were around long before PCs.

Duh.

The change from using binary MBs to decimal
MBs happened with PC hard drives.

Which has exactly what to do with the history of the nomenclature?

Everying to do with my original comment about PCs that he was commenting on.

Who was discussing "the PC"?

I clearly was in the post he commented on.

Which PC is that?

The one I referred to in the post he commented on.

Nope. 'Fraid not.

Yep, fraid so.

It's really the only honest way to do it since the drive
manufacturer has no way of knowing what filesystem
the user is going to be putting on the disk.

Wrong with a hard drive where the user
doesnt get to specify the sector size.

So was the first Seagate drive.

Not with the PC that I clearly specified in my original.

 

[Arno Wagner]

Arno Wagner wrote:

Now, I'm curious as to your opinion on the use of "ksi", commonly used in
American engineering as shorthand for "thousand pounds per square inch".
Of course kilo to refer to 1,000 precedes the SI by several thousand years.

No problem with that. SI is a standard. That does not mean it is
the origin of any of the terms defined by it. In fact it is not
the origin of most of what it defines. However it defines units
and prefixes in a uniform way and removed all imprecisions or
ambiguities.

Here it is customary to speak of kCHF (for 1000 CHF).

SI prefixes are not limited to being used with SI units.

Arno

 

[J. Clarke]

Its 5MiB rounded.


No they didnt, the capacity is stated as being 5.0MB

And how is that not the closest even number? Or more precisely the closest
whole number?

Even clearer with the other drives of the same era which
also used binary MBs, most obviously with the ST225.
http://www.seagate.com/support/disc/specs/mfm/st225.html
The stated 20.4MB is clearly MiB, its actually 21.4MB

??? Where is 20.4 MB "stated"? They state 21.4 MB and the actual capacity
is 21,411,840 bytes.

Nope, clearly they didnt, the capacity was
clearly stated to one decimal place, in MiBs.

You mean like the 20.4 MB capacity for the ST225 that you made up?

 

[J. Clarke]

So that comment is irrelevant to MY statement about PCs.


Duh.

The change from using binary MBs to decimal
MBs happened with PC hard drives.

You seem to be having trouble proving that.

Everying to do with my original comment about PCs that he was commenting
on.



I clearly was in the post he commented on.


The one I referred to in the post he commented on.



Yep, fraid so.

Then howcum we still have people complaining every day that their drives are
not delivering the stated capacity.

Wrong with a hard drive where the user
doesnt get to specify the sector size.

ROF,L. Format a disk with 5 different file systems and get back to me with
the capacity of each post formatting.

Not with the PC that I clearly specified in my original.

<snip>

 

[J. Clarke]

No problem with that. SI is a standard. That does not mean it is
the origin of any of the terms defined by it. In fact it is not
the origin of most of what it defines. However it defines units
and prefixes in a uniform way and removed all imprecisions or
ambiguities.

Here it is customary to speak of kCHF (for 1000 CHF).

SI prefixes are not limited to being used with SI units.

Nor is their creeping into the vernacular to be taken as evidence that such
use is mandated by law.

 

[dew]

And how is that not the closest even number?
Or more precisely the closest whole number?

You wouldnt state it as 5.0MB if that was
the case, it would be stated as 5MB instead.

??? Where is 20.4 MB "stated"? They state 21.4 MB
and the actual capacity is 21,411,840 bytes.

Yeah, I mangled that one considerably.

You mean like the 20.4 MB capacity for the ST225 that you made up?

I didnt make it up, I just mangled it.

 

[dew]

You seem to be having trouble proving that.

Nope. In spades with the miniscribe MS1006
http://marina.mfarris.com/theref/hard_drives/h_mini_ms1006.html
and the Micropolis 1302
http://marina.mfarris.com/theref/hard_drives/h_m1302.html

Then howcum we still have people complaining every day
that their drives are not delivering the stated capacity.

For a different reason, they see the size reported in GiBs in
Win whereas the manufacturer states the size in decimal GBs.

ROF,L. Format a disk with 5 different file systems and
get back to me with the capacity of each post formatting.

Irrelevant to the capacity of the physical drive as reported in Win.

 

[Arno Wagner]

Arno Wagner wrote:

Nor is their creeping into the vernacular to be taken as evidence that such
use is mandated by law.

If you start to misuse them in business transactions, you will
find out about that. If it looks like an SI prefix, then it has
to be one.

Arno

 

[Folkert Rienstra]

in message news:[email protected]
Nor does it state that in the first place ....

Actually, it's truncated to 5.0MiB.

And then added a zero for decimal precision?

You wouldnt state it as 5.0MB if that was
the case, it would be stated as 5MB instead.

Yeah, I mangled that one considerably.

Which is just fancier words for
'I wanted to prove myself so badly that I ****ed up in doing it'.

I didnt make it up, I just mangled it.

Seeing things that weren't there, ie 'made up'. In denial, as always.

 

[Folkert Rienstra]

Yep.

In spades with the miniscribe MS1006
http://marina.mfarris.com/theref/hard_drives/h_mini_ms1006.html

and the Micropolis 1302
http://marina.mfarris.com/theref/hard_drives/h_m1302.html

from that site:
Formatted Capacity: 20.4MB
Cylinders: 830
Sectors/Track: 17
R/W Heads: 3

830*3*17*512:1024:1024 = 20,668,950 bytes = 20.7 MiB

What's in spades is that that is obviously edited info from a resellers
website.

You can lookup info from that site and go to another resellers site
and find different info there.

"As always" with you, this proves absolutely nothing.

[snip]

 

[Gerhard Fiedler]

Clearly SI units are NOT mandated.

That was not really the issue. The issue was that /if/ one uses an SI unit
or prefix (as for example the "giga" prefix in the size spec of harddisks),
it is defined by law what it means.

If you don't believe it, try selling a disk that has 500 GB as having 800
GB and claiming that your 'GB' means 625000000 bytes... No go, and you'll
have at least small claims court all over you -- based on law, of course.

Gerhard

 

[Gerhard Fiedler]

The minimum unit -the sector size- is.

Right. And -- notwithstanding the discussion dew and John seem to have -- I
think that was one of the reasons why originally disk sizes were given in
binary units or something similar. But as the disk size move far enough
away from the sector size, the sector size became irrelevant for the disk
use (not for the controller firmware, of course).

As far as a memory slot goes.

Exactly -- but for the commerce, that's the important part. There's still
some way to go until the memory that goes into the slot becomes as detached
from the underlying binary structure as the disks are, but I think it will
come. And as you say, it may be already underway.

Gerhard

Then it becomes linear, just as with diskdrive sectors, platters etc.

It does actually. Using exactly 2048 512-byte sectors.

1 MB of disk space, at least according to harddisk vendors, is not 2048
512-byte sectors, it is less than 1954 512-byte sectors. And 1 MB of RAM
doesn't fit in there...

Gerhard