Talk:Binary prefix

From Wikipedia, the free encyclopedia

Contents

[edit] Wow, much talk

Wikipedia EditThisPage complains:

"This page is 127 kilobytes long. This may be longer than is preferable; see article size."

I very nearly didn't contribute, once I had decided I would not read that much for context.

192.42.249.130 01:11, 11 August 2006 (UTC)

[edit] Decimal bits, binary bytes

Apple may be ducking this controversy by counting decimal bits. Relevant to the article's claim "Operating systems usually report disk space using the binary version" is the reference "Apple Publications Style Guide" of January 2006, available in Google. That guide claims:

"kilobyte (1024 bytes) KB computer memory"

"kilobit (1000 bits) kbit computer memory"

[edit] Binary or decimal context implied by electronic memory

In the page, I see the quote: Electronic memory such as RAM and ROM always uses the binary versions, because the physical structure of the device makes it naturally come in sizes that are powers of two.

This isn't true, but I don't yet know how to fix it concisely, help.

For example, flash is an electronic memory, and ECC-protected RAM is an electronic memory, but their capacities aren't naturally powers of two.

A concise true alternative I saw said recently was: The RAM and ROM folk gave us this confusion by behaving as if memory were reliable. That is, by converting to precise binary prefixes from loose decimal approximations, they left no room for ECC, etc.

In more detail ...

All that's naturally a power of two in memory is the count of raw cells in a single layer.

Yes, P * Q cells appear in a rectangular array. P * Q is a power of 2 because P and Q are powers of 2. P and Q are powers of 2 because pins cost much money. If you make N pins into an address bus, then you can address 2^N rows or columns. Yes. So far so good.

But electronic memory today - 2005 - often contains a lot more than a single layer of raw 1 or 0 cells. As in modems, so now in memory cells, the discrete stored voltage may be multilevel, e.g., representing 0 or 1 or 2 or 3 or 4, not always binary. The chip may contain more than one layer. Part of the chip may be dedicated to ECC, or left free for wear-leveling.

Consequently, the capacity of flash memory in particular is now "continuously variable", and their physicists follow the HDD physicists by using the standard metric units to count bits.

Memory controllers on motherboards always address whole bytes (which includes any parity or ecc bits) or whole words (whole multiples of whole bytes) in chunks measured in binary multiples. This is because binary math is the natural form for practically all processors. The presence of ECC in RAM makes no difference. The ECC bits are part of the byte, even though the ECC bits are not sent to the CPU, the memory controller reads the extra bits when it reads the byte. RAM modules are always measured according to how many binary multiple bytes are available no matter how wide the data path is.

"Flash memory" today refers to non-volitile storage devices that are accessed exactly like hard drives. They use virtual cylinder, head, and sector specifications for the computer system to address the device and even though they contain memory that is accessed in a "random" (non-serial) method, in discreet rows and columns, it is not considered "RAM".

Yes, Flash often uses multilevel cell technology, but the data is converted to a binary representation. So even though one cell can hold more than a simple on/off representation, the absolute capacity is still a function of bits and bytes. And the memory used by the CPU through the memory controller does not yet use multilevel cell technology. DDR and RAMBUS are still ones and zeros.

Flash manufacturers have followed the hard disk drive path to designate capacity, decimal. No matter how many ECC cells or spare cells for wear-leveling are allocated, the capacity is based on decimal multiples and does not include the spares and ECC.

So I guess I would say, instead of, "Electronic memory such as RAM and ROM always uses the binary versions, because the physical structure of the device makes it naturally come in sizes that are powers of two. I propose, "RAM memory modules always designate capacity in binary multiples because that type of math is "native" to binary processors. or " because CPU's use RAM in binary multiples and building RAM modules according to that system is most logical."

JJLatWiki 18 July 2005 09:11 (Pacific Daylight)

' "Flash memory" today refers to non-volitile storage devices that are accessed exactly like hard drives.'
Large flash units used for bulk storage are certainly treated like hard drives but bare flash chips and flash used in microcontrolors for program storage is far more like a more conviniant to use variant of EPROM (reads out like rom, requires special actions for programming) Plugwash 23:53, 16 June 2006 (UTC)

[edit] Discussion of page name

This page needs a rename as part of the "Kill your friendly neighbourhood stub" campaign. Any suggestions? At the risk of sounding a little audacious, I suggest a renaming from "Byte/Prefixes" to "Byte prefixes" ;-) -- Tarquin

I find it a bit confusing to have the first table listing the deprecated usage of the prefixes. Maybe it's better to have the first table listing the current standard, then have the colloquial usage listed much later in the article... --Bob03:33 Oct 14, 2002 (UTC)

Better yet, kill the first table altogether. The bottom half is superfluous anyway. I have never seen "yotta" being used meaning 2^80, except by a handful of nerds at Wikipedia... ;-)
Herbee 02:52, 2004 Feb 21 (UTC)

Change "Byte" to "Binary", because these prefixes are also applied to other units such as bits and words (as the introductory text clearly says!) -- Dwheeler 21:26 21 May 2003 (UTC)

[edit] The mythical nona- and dogga-

User:81.63.111.215 added the phrase "as well as nobi- and dogbi-", presumably based on the assumption that there are legitimate decimal-based prefixes nona- for 1027 and dogga- for 1030. They are not SI prefixes, or at least NIST knoweth not of them. I'm perfectly prepared to be convinced of their existence, but I want to see some evidence that some recognized standards organization is promulgating them. I have so far found nothing but loose assertions in Google searches. Dpbsmith 00:27, 21 Mar 2004 (UTC)

The official list of prefixes is maintained by the Bureau International des Poids et Mesures. Consider that evidence against this nonsense.
—Herbee 02:37, 2004 Mar 21 (UTC)
Thanks for keeping up on all of this, dpb. +sj+

[edit] From VfD, re: Zebi/Yobibyte

  • According to what I hear at Wikipedia, IEC has officially coined kibibyte through exbibyte, but not zebibyte or yobibyte. Any significance of these 2 articles?? 66.245.22.210 16:53, 3 Aug 2004 (UTC)
  • Delete both - dicdefs, and also delete if they are not official terms - Tεxτurε 17:20, 3 Aug 2004 (UTC)
  • Weak vote to delete. If they are not the official terms, then they should go. If they ever get certified, they can be recreated. -- DropDeadGorgias (talk) 17:33, Aug 3, 2004 (UTC)
  • Delete. Redirect to Binary prefix where they are mentioned and properly identified as speculative. Sans such a qualification, these articles are misinformation, because the terms have not been endorsed by any standards organization, are not in any dictionary, and are not in significant use (because as of 2004 they are too big for there to be any actual need for them). This is very similar to the issue with Nonabyte and Doggabyte which were deleted per VfD some time ago. Nerd oneupmanship. If these are not deleted, I shall contribute my own fine articles on "bajillion skillion gazillion antidisestablishmentillion," and "mobidikkabyte" (=the binary equivalent of one melvillion bytes) [[User:Dpbsmith|dpbsmith (talk)]] 20:24, 3 Aug 2004 (UTC)
  • Keep. Zetta- and yotta- are SI prefixes [1]. Some prefixes have modified forms for naming powers of 2 [2]. Zetta- and yotta- don't appear to have official binary forms but it's easy enough to see that zebi- and yobi- would be those forms if they did, and every now and then somebody makes that obvious extrapolation, e.g. [3]. Zebibyte and yobibyte should state explicitly that they're unofficial terms. I'd do that but I'm feeling lazy at the moment. Wile E. Heresiarch 15:06, 4 Aug 2004 (UTC)
Hmph. Such extrapolations can surely be made, but they're not encyclopedic. Wikipedia is not a crystal ball. I feel very confident that Kerry will take Massachusetts in the 2004 U. S. Presidential election, but that does not mean that I should write an encyclopedia article that states this as fact. Wikipedia is accumulating enough of this sort of garbage to be of at least mild concern. [[User:Dpbsmith|dpbsmith (talk)]] 16:18, 4 Aug 2004 (UTC) P. S. Too many nerds have obviously gotten too many pats on the back for being able guess "what comes next" in a sequence...
  • Modify and keep as outlined by Wile. - UtherSRG 15:44, 4 Aug 2004 (UTC)
  • Modify and keep. Spiff 17:12, 4 Aug 2004 (UTC)
Well, I've done a modify since if the articles are going to be, ugh, kept I want them to be accurate. With regard to Wile E. Heresiarch's citation [4] I do not see "zebibyte", "zebi-", "yobibyte", or "yobi-." I indeed see the abbreviations ZiB and YiB, and the (puzzling) explanation that "`Zi'[sic] is a GNU extension to IEC 60027-2." Presumably that's a typo for ZiB. Ditto for YiB. If accurately described and cited, I wouldn't object to this factoid being added to the article, since IMHO the FSF does count as some kind of authority—perhaps as an external link identified as "Use of abbreviations ZiB and YiB in GNU software." [[User:Dpbsmith|dpbsmith (talk)]] 17:32, 4 Aug 2004 (UTC)
  • Delete. Redirect these to Binary prefix for now. If and when they become standard, they can have their own articles. -- WOT 18:21, 4 Aug 2004 (UTC)
  • Delete. Checking google, most uses seem to either be wikipedia or reference wikipedia as the source of this info. The words just aren't useful.
  • Delete speculation. -- Cyrius|✎ 19:19, 6 Aug 2004 (UTC)
  • Delete. Newbiebyte. But you can't fool Mr. Metric. Patent nonsense. Denni☯ 01:41, 2004 Aug 7 (UTC)
  • Delete... oh, I'm too late. Please delete, all the same. As dpb says, WP is not a crystal ball. In particular, we will now be used by others as the source confirming that these two variants are the binary equivalents of the SI forms, something WP we should under no circumstances be. +sj+ 17:36, 11 Aug 2004 (UTC)
  • Delete. I'd also like to point out that they currently redirect to Zettabyte and Yottabyte, which are distinctly different. Suggest changing to RFD. Betbest1 20:43, 26 December 2005 (UTC)

[edit] Hexadecimal billion

The contents of this section "Hexadecimal billion" were transferred on a "archieved" page called Closed talk: section Hexadecimal Billion in Talk:Binary prefix Talk:Binary prefix/Hexadecimal Billion.

Started from a different of opinion concerning the numerals corresponding to the values of the multiples of unit, like the prefix "Tebi" or "Tera binary", i.e. a "billion" according to Michael, and Ian who deleted while he considered this too ambiguous. Afterwards we talked about the significance of the terme "one billion hexadecimal".

To make place for other discussions, Ian and Michael decided to not continue this discussion at this place more, also because we had moved away, finally, too much from "Binary prefix".

If somebody would be interested to continue this discussion, that's now at  User talk:Michael Chuquet/Hexadecimal billion.

Just for the record: the original archive was in the main article namespace, which is undesirable because the random page feature can send a reader to it. I've moved the archive to a subpage of the talk namespace: Talk:Binary prefix/Hexadecimal Billion. • Benc • 04:48, 9 Oct 2004 (UTC)

[edit] Hard disk sizes

My Maxtor 40 GB is actually 38.2 GB (dd tells me it has 80022600 sectors, or 38.1577 MB). That's 39,073.5352 MB, 40,011,300 K, 40,971,571,200 bytes. OTOH, my Maxtor 160 GB has 320173056 sectors, or 152.67 GB. That's 156,334.5 MB, 160,086,528 K, 163,928,576,512 bytes. Clearly, if GB meant 1000*1024*1024 bytes, then they'd be giving me less than claimed (and I could sue them!). If GB meant 1000³ bytes, they'd be giving me almost a gig more for my 40 GB, and almost 4 gigs more for my 160 GB. They probably use 1 GB = 1000*1000*1024 bytes.

But my Quantum Fireball advertises "4.0 GB" and actually has that much (assuming Windows is accurate), 4.0062 GB, 4102.32 MB, 4200776 K, 4301594624 bytes. It also says "4.3AT" and has "43" in the model number, which I assume refers to 4.3 decimal GB.

Then, my Seagate 40 GB has 78163247 sectors, or 37.2711 GB, 38165.6479 MB, 39081623.5 K, 40019582464 bytes. It should have 78165360 sectors (according to the specs), no idea why it has less. It'd be false advertising unless GB meant 1000*1000*1000 bytes. Western Digital seems to use 1 GB = 1000*1000*1000 bytes, as well (which means it was worth paying more for my Maxtor). Hitachi seems to use 1000*1000*1024 bytes.

So we have one company using 1024³ (except I have no recent HDs from Quantum, and they don't seem to make them anymore), two companies using 1000²*1024, and two more using 1000³.

Elektron 22:25, 2004 Aug 30 (UTC)

Also, Toshiba uses 1 GB = 1000³ bytes, so does Fujitsu. Samsung says they use 1000³, and not-so-official sector counts appear to confirm this. Since Quantum doesn't make hard disks anymore, for current hard disks, we have two companies using 1000²*1024, and five using 1000³ Elektron 23:36, 2004 Aug 30 (UTC)

[edit] Consolidate all the little articles

I propose that all the little articles:

SI, bits SI, bytes IEC, bits IEC, bytes
Kilobit Kilobyte Kibibit Kibibyte
Megabit Megabyte Mebibit Mebibyte
Gigabit Gigabyte Gibibit Gibibyte
Terabit Terabyte Tebibit Tebibyte
Petabit Petabyte Pebibit Pebibyte
Exabit Exabyte Exbibit Exbibyte
Zettabit Zettabyte — —
Yottabit Yottabyte — —

should be consolidated into this article (without destroying any useful information). If they are not, then someone should at least go through them all and lowercase all of them. Unit names are always lowercase. For instance:

"The Gibibyte is closely related to the Gigabyte, which is..." should be

"The gibibyte is closely related to the gigabyte, which is...", etc. - Omegatron 03:44, Sep 24, 2004 (UTC)

I concur, but really, the SI "gigabyte" isn't really a 'binary prefix'. One day I think I'll have to do this though. Elektron 20:01, 2004 Nov 1 (UTC)
Why not? That's what this article is about; using giga and gibi as binary prefixes. The individual "small" articles are getting bigger, though. - Omegatron 20:17, Nov 1, 2004 (UTC)

I still think all the little articles should be combined into this article. Kibibyte, for instance, has info that is not in Mebibyte. It's really silly to have a separate article for each one, that basically only contains a single number and it's relation to two other numbers. All of that info should be in this article, and then the info that isn't doubled everywhere will be in one place. - Omegatron 18:57, Dec 21, 2004 (UTC)

Nobody objected, so go ahead and do it if you want. [[User:Smyth|– Smyth\talk]] 19:36, 21 Dec 2004 (UTC)
Nobody said it was a good idea, either. Plus it will take some significant work with the bigger articles like megabyte. Don't wanna change it if everyone is suddenly going to say "no that was bad" and revert it. - Omegatron 14:22, Dec 29, 2004 (UTC)

I thought this sort of thing was already done with other articles like kilowatt, attogram, and so on, except now that I look, they all have little articles for each. Hmm... Has there been discussion about this for other units? - Omegatron 22:21, Dec 30, 2004 (UTC)

I agree that gigabyte shouldn't redirect here... it's not a binary prefix, and that would just confuse things. However, it wouldn't be a bad idea to go through all the articles and strip out all the information that belongs in more general articles (like this one) and replace it with a quick summary and link to the more comprehensive discussion, per Wikipedia:Summary style. The amount of duplication is distressing. 68.81.231.127

Good point. gigabit/byte should not redirect, i guess, even though they are binary prefixes in this instance (according to this article, giga, mega, etc. can be used as both binary prefixes and SI prefixes, though I think they should only be considered SI prefixes. but according to this very article, they ARE binary prefixes...) I already added the see also binary prefix to each article, but no one notices and just adds stuff to each individual article. - Omegatron 20:07, Jan 13, 2005 (UTC)
It's even worse.. those are SI prefixes, but they are not SI units because B is not an SI symbol for byte (or even Bel, at least not yet I think :). And then there are special case like the 1,024×1,000 megabyte variant....
Anyway, a lot of articles seem to link to the kilo-, mega-, and giga- variants (150–250 for the -bytes series), while tera- is only linked about 50 times, and peta- and higher have <20 each. So at least the big three or four are natural links, and might justify their own articles. There are also enough special cases in the big three (which industry uses it, etc) that they'd still need separate sections in a big article, so it's probably easier to keep them separate. (I don't think it matters either way, but if the larger units are turned into redirects, they should probably be pointed to byte [or bit].)
Byte unit (SI prefix)
Other: kilo- | mega- | giga- | tera- | peta- | exa- | zeta-
Related bit unit: Megabit
Related binary prefix unit: Mebibyte
A bigger problem is reducing duplication of effort. There is very different text saying very similar things on different pages. A Wikipedia:Navigational templates could help with all the related units, while using a standard paragraph and pointer to the main articles here and at the SI prefix page using would clear out a lot of the rest (using Wikipedia:Summary style again... there is an example at Gigabyte#Distinction between 1000 and 1024 megabytes, though it needs a lot of work). The navigation template to the right needs some work, but really like simple over cluttered... ahem... Afrotropic is an abomination :).
I'm just playing with ideas... I'm not even sure this is the right approach. 68.81.231.127 00:09, 14 Jan 2005 (UTC)

Alright. If we're not going to consolidate all the little articles, we should add a template to all of them with all of the others on it. Anons search for "petabyte", find our article, and then start adding info about gigabytes, etc. We need to indicate that there already are articles about each particular value and also about this article. - Omegatron 01:16, Mar 24, 2005 (UTC)

[edit] Template for Binary prefixes

I have created a template available at:

Decimal prefixes

 

 Binary prefixes

edit
SI prefixes
powers of ten		 Binary prefixes
powers of two		 IEC std prefixes
powers of two
Prefix Symbol Multiple Prefix Symbol Multiple Prefix Symbol
yotta
Y
1024
 ≈ yotta
Y
280
      
zetta
Z
1021
 ≈ zetta
Z
270
      
exa
E
1018
 ≈ exa
E
260
 = exbi
Ei
peta
P
1015
 ≈ peta
P
250
 = pebi
Pi
tera
T
1012
 ≈ tera
T
240
 = tebi
Ti
giga
G
109
 ≈ giga
G
230
 = gibi
Gi
mega
M
106
 ≈ mega
M
220
 = mebi
Mi
kilo
k
103
 ≈ kilo
k or K
210
 = kibi
Ki
Notes:
  • The SI prefixes have similar values to, but are different from, the corresponding Binary prefixes. The IEC prefixes were proposed to distinguish these meanings.
  • As of 2004, the IEC prefixes have not been widely taken up

which I would like to insert into this article. I would hope that this could enable aome rationalisation of the existing tables, and could perhaps eventually be inserted into all the individual binary prefix articles, e.g. kibi, mebi, etc (assuming that they are not merged beforehand).

Does anyone have any objection or wish to edit the template beforehand (see edit link in top line of template)? Ian Cairns 21:47, 15 Nov 2004 (UTC)

Looks good. pretty big though. maybe list only as powers of 10 and powers of 2? - Omegatron 22:55, Nov 15, 2004 (UTC)
Very good, but not 800x600 friendly. Alternating table row backgrounds would be nice though. --Delicates 03:09, 24 Nov 2004 (UTC)
I've reduced the overall width by removing the number values. Is this any better, particularly for 800x600? (The alternating backgrounds will have to wait for another day...) Ian Cairns 18:46, 24 Nov 2004 (UTC)
Now I really don't like the middle section, and don't think it deserves a place under the sun due to it being inappropriately erroneous, and making the whole table ambiguous and confusing. Might want to put word "standard" as well into the SI heading. The IEC section lacks the 2n column which is more appropriate there. --Delicates 22:13, 24 Nov 2004 (UTC)
Thanks for that. I'm concerned that you think it's erroneous. Please can you indicate where the errors might be? The SI prefixes are known as such in Wikipedia, so I didn't see the need to include the word 'standard'. If they become known as SI standard prefixes, then I'm happy to change the template. Regarding the 2^n column, I was told that the table was too wide as it was previously. I was trying to avoid duplicating the same column. I have been wondering whether it might be possible to remove one of the 'kilo' columns to reduce the width further without generating confusion, but I don't think this can be done easily. Please can you indicate where you find the template confusing? Alternatively, be bold with your editing and let others review your changes. Thanks, Ian Cairns 23:35, 24 Nov 2004 (UTC)
The problem I have with it is that it has two identical columns that contradict eachother which is both erroneous and confusing in context of the same table. I think the middle section should be taken out alltogether, because it is redundant in the presence of “≈” symbol, which makes the association of the SI prefixes with the binary meaning pretty clear. The only problem that with this is the inconsistent use of ‘k’ and ‘K’ for “kilobyte”. The people who find this page through a search engine won't be aware of SI prefixes being known in Wikipedia as standard. It is good to be consistent so that people don't have doubts when they see something applied to one thing but not to another. --Delicates 02:13, 25 Nov 2004 (UTC)
Given that Delicates thought the template was 'very good' at 03:09, I've added the template to the existing article, to see it in context - without thinking yet whether any of the existing tables can now be rationalised / avoided. Ian Cairns 00:22, 25 Nov 2004 (UTC)

pebi, mebi, etc. just redirect to this article, you know. - Omegatron 22:25, Dec 30, 2004 (UTC)

[edit] Do not include -zebi and -yobi

There are no such prefixes. They should not be included in a table that purports to be documenting the IEC standard. Everyone understand, and the article states explicitly, that they are the logical extension of the IEC system, but until and unless the IEC chooses to extend the system they should not be included. Dpbsmith (talk) 19:35, 30 Jan 2005 (UTC)

  • PLEASE DO NOT reinsert these values into the table without discussion. They are not "unofficial extensions." They are not real terms at all. They are purely speculative; someone's guess as to what names would be used if these units had names. I believe, based on Google hits, that they are mostly a sort of nerdish urban legend propagated by people copying tables from teach other. Unless you can convince me otherwise by showing a good citation from a serious, authoritative source that shows the names "zebibyte" and "yobibyte" are in real use in the computer field, I will continue to feel that these should not be in the table of binary prefixes at all. I don't mind the sentence pointing out that -zebi and -yobi are the obvious continuations. I'm not at all sure it's necessary to give the numeric values of 270 and 280, but you wanted them there and I didn't see a problem with it. Change them to the 1000n×1.#### format if you think that provides more insight. But they should not go in the main table. Dpbsmith (talk) 13:34, 4 Feb 2005 (UTC)
I'm just not sure what harm there is in adding them to the table, properly identified as speculative. Why the allergy? As for giving the values, yes they are required because they're not easily calculated (i.e. they must be done by hand, unless you have some peculiar calculator available). Ah well, this is not worth an edit war, obviously.
Urhixidur 00:51, 2005 Feb 5 (UTC)
What's so hard about 1024 × = = = = = = = = on the calculator that comes with Windows and many other calculators? Gene Nygaard 01:02, 5 Feb 2005 (UTC)
The Calculator handles it, because it takes special care with large numbers. Excel, on the other hand, uses normal integers and thus fails to carry enough precision.
Urhixidur 15:47, 2005 Mar 27 (UTC)

[edit] B is for bel

In editing the article, Urhixidur commented " B is Bel, b recommended for byte".

Big deal. B is also for boron.

But you'd need to change that last b to an m to describe someone who imagines a likelihood of even as much confusion between bytes and bels (lowercase, of course) as there would be between boron and bels.

  1. Why should bels get preference in any case? They aren't an SI unit, are only listed as acceptable for use with SI.
  2. Bels and bytes are used in completely different fields of activity.
  3. The bel is never used standing alone, and never used with any prefix other than "deci-".
  4. Bytes are never used WITH the prefix "deci-".

Those points are, of course, applicable even setting aside the reason I changed Urhixidur's edits in the first place; "b" is used as the symbol for "bits" in this article, and it would be silly to use the same symbol for "bytes" as well. Gene Nygaard 23:44, 30 Jan 2005 (UTC)

One more point about SI: In SI, the radian (symbol rad) is a derived unit. The rad (symbol rad) is among the units in Table 10 of the BIPM brochure and in Table 9, temporarily accepted for use with the SI, in the NIST brochure. Gene Nygaard 00:01, 31 Jan 2005 (UTC)

« Why should bels get preference in any case? »
Because of the well established rule that upper-case symbols tand for units named after people, in this case Alexander Graham Bell.
« Those points are, of course, applicable even setting aside the reason I changed Urhixidur's edits in the first place; "b" is used as the symbol for "bits" in this article, and it would be silly to use the same symbol for "bytes" as well. »
There is not a single occurrence of "b" standing for bit anywhere in the article.
Urhixidur 01:30, 2005 Feb 5 (UTC)

[edit] We need a template for the little articles

I want to compromise the conciseness and prettiness of this:

Byte unit (SI prefix)
Other: kilo- | mega- | giga- | tera- | peta- | exa- | zeta-
Related bit unit: Megabit
Related binary prefix unit: Mebibyte

with the info in this:

Decimal prefixes

 

 Binary prefixes

edit
SI prefixes
powers of ten		 Binary prefixes
powers of two		 IEC std prefixes
powers of two
Prefix Symbol Multiple Prefix Symbol Multiple Prefix Symbol
yotta
Y
1024
 ≈ yotta
Y
280
      
zetta
Z
1021
 ≈ zetta
Z
270
      
exa
E
1018
 ≈ exa
E
260
 = exbi
Ei
peta
P
1015
 ≈ peta
P
250
 = pebi
Pi
tera
T
1012
 ≈ tera
T
240
 = tebi
Ti
giga
G
109
 ≈ giga
G
230
 = gibi
Gi
mega
M
106
 ≈ mega
M
220
 = mebi
Mi
kilo
k
103
 ≈ kilo
k or K
210
 = kibi
Ki
Notes:
  • The SI prefixes have similar values to, but are different from, the corresponding Binary prefixes. The IEC prefixes were proposed to distinguish these meanings.
  • As of 2004, the IEC prefixes have not been widely taken up

and create template that we can put in all the little articles like kilobit and pebibyte so newcomers realize each article has its own info already and stop trying to write the same info into one of the little articles.

I basically just want to make this:

SI, bits SI, bytes IEC, bits IEC, bytes
Kilobit Kilobyte Kibibit Kibibyte
Megabit Megabyte Mebibit Mebibyte
Gigabit Gigabyte Gibibit Gibibyte
Terabit Terabyte Tebibit Tebibyte
Petabit Petabyte Pebibit Pebibyte
Exabit Exabyte Exbibit Exbibyte
Zettabit Zettabyte — —
Yottabit Yottabyte — —

into a prettier navbox with a little more info. Ideas? - Omegatron 17:22, Mar 25, 2005 (UTC)

[edit] A little better

Decimal SI prefixes
powers of ten		 Binary IEC prefixes
powers of two
Prefix Symbol Multiple Prefix Symbol Multiple
kilo k 103 ≈ kibi Ki 210
mega M 106 ≈ mebi Mi 220
giga G 109 ≈ gibi Gi 230
tera T 1012 ≈ tebi Ti 240
peta P 1015 ≈ pebi Pi 250
exa E 1018 ≈ exbi Ei 260
zetta Z 1021
yotta Y 1024
These prefixes are often applied to byte and bit.

How about this? Feel free to come up with a better/smaller phrase/sentence to put at the bottom. Variants I thought about: "applied to", "added to", "used with", remove word "prefixes", add word "terms", and so on... What do you think? Also, if we nuke all those dictdef articles we should make sure any interesting info like in petabyte is preserved into the common byte article. Oh and we gotta think of the name for the template. Delicates 19:44, 16 Apr 2005 (UTC)

  • I like it. — Sebastian 07:11, 2005 Mar 27 (UTC)
Hmm.. That's better, but I wanted to link directly to articles like yottabyte, not yotta
Basically I want this made into a smaller, prettier navbox: - Omegatron 16:40, Mar 27, 2005 (UTC)
SI, bits SI, bytes IEC, bits IEC, bytes
Kilobit Kilobyte Kibibit Kibibyte
Megabit Megabyte Mebibit Mebibyte
Gigabit Gigabyte Gibibit Gibibyte
Terabit Terabyte Tebibit Tebibyte
Petabit Petabyte Pebibit Pebibyte
Exabit Exabyte Exbibit Exbibyte
Zettabit Zettabyte — —
Yottabit Yottabyte — —
I think all those dinky little stub articles are nonsense which should be eliminated, most of them wouldn't even be worth a "dictionary" entry. They certainly are not encyclopedic. I could see having an article for each prefix, and linking to that. For the rest, one article on Orders of magnitude for bits and for bytes would be more than sufficient--even that's too much IMHO, this article is enough for both. Gene Nygaard 17:33, 27 Mar 2005 (UTC)
    • I agree completely. There is no reason whatever for these quantities to have articles which can never be more than a dictdef. As valid "index entries" they should all redirect here. Dpbsmith (talk) 17:43, 27 Mar 2005 (UTC)
ARGH! That's what I said in the first place! Talk:Binary prefix#Consolidate all the little articles - Omegatron 21:54, Mar 27, 2005 (UTC)
A good idea. Well spotted. Bobblewik  (talk) 23:18, 29 Mar 2005 (UTC)
Can we please do this soon? See Petabyte#See also for example. They are just getting bigger. - Omegatron 20:29, Apr 3, 2005 (UTC)

Alright so almost everyone likes the template idea. Do we have one yet? - Omegatron 12:19, May 19, 2005 (UTC)

[edit] Vote vote vote!

This vote is over – we decided to create a template that links all the little articles together.

Little articles like pebibit keep getting stuff about other units added to them, get altered so they aren't consistent with the others, etc. by anons who probably don't realize there is an article for each individual prefix and unit, like megabit/mebibit/megabyte/mebibyte.

So what should we do?

[edit] Create a template that links them all to each other and put it in every one

  • Support. I'm sure we can all work together to keep the articles to an acceptable consistency, and a template project, along with other minor changes along the way, would be the best way to maintain them without losing information on each article or making this one too central and overpowering. --Alexwcovington (talk) 04:47, 19 Apr 2005 (UTC)
  • Support. Good Idea. I want to find Exabyte or Terabyte if i search for it. --Mononoke 09:25, 19 Apr 2005 (UTC)
  • Support. The template should contain prominent links to Binary prefix, Bit and Byte, and the articles themselves should contain nothing else except a brief definition of the word's meaning(s), and a handful of usage examples. The only exceptions that come to mind are Megabyte and Gigabyte, which have interesting debates about their meanings for floppy and hard disks respectively. – Smyth\talk 16:39, 20 Apr 2005 (UTC)
  • Support, and I do believe I owe an apology. I deleted most of the merge templates because I found the discussion on Talk:Kilobyte dated June 2004 and I assumed they all dated that far back...then I found the discussion here. What to do? My rationale:
    • Each article removes the need to link munge (i.e., kilobytes per second) since I think it's extremely bad practice
    • Each article has its own "What links here" and it'd be much easier to see what links to each unit (instead of one page (which is currently limited to 500 max) for all units)
    • Each article would *explicitly* state what each unit means instead of forcing the visitor to hunt down a table
  • Again, my apologies on the merge template removal! Cburnett 21:45, Apr 20, 2005 (UTC)
    You missed all the kibibit/mebibyte merge entries.  :-) - Omegatron
  • Support - This makes sense to me. - Omegatron 22:52, Apr 20, 2005 (UTC)
  • Support - Sounds good. --Pmsyyz 19:20, 17 August 2005 (UTC)

[edit] Merge them all into this article or into byte and bit or something like that

  • no votes

[edit] Merge all but kilo/mega/gigabyte that have significant info on their own

  • Support. It's A Good Thing™ that someone finally called for a vote on this. With the present scheme we'd surely be swamped with boring micro-maintenance to keep things consistent at all times. --Wernher 05:09, 18 Apr 2005 (UTC)
  • Support Dpbsmith (talk) 09:57, 18 Apr 2005 (UTC)
  • kinda support i think it would be a good idea to move the IEC stuff to their corresponding standard prefix pages. there will likely never be more to say about gibli-libbli-bit or whatever silly thing we are calling it than currently is written. i don’t support killing all the articles such as tera which will be growing as time passes on. the template is great and kudos for to the people who wrote this page, its clear and concise. Cavebear42 23:41, 19 May 2005 (UTC)
  • Support It's too repetive. raylu 04:07, August 17, 2005 (UTC)

[edit] Do nothing

  • no votes

[edit] Onward

It's been a month since that poll was last updated, and there seems to be support for the template idea. (Not a consensus, true.) I've adopted the above table into two: one for -byte pages and one for -bit pages.

Table for pebibit, petabit, etc articles:

Decimal names
powers of ten		 Binary names
powers of two
Name Symbol Multiple Name Symbol Multiple
kilobit kb 103 ≈ kibibit Kib 210
megabit Mb 106 ≈ mebibit Mib 220
gigabit Gb 109 ≈ gibibit Gib 230
terabit Tb 1012 ≈ tebibit Tib 240
petabit Pb 1015 ≈ pebibit Pib 250
exabit Eb 1018 ≈ exbibit Eib 260
zettabit Zb 1021
yottabit Yb 1024

Table for petabyte, pebibyte, etc articles:

Decimal names
powers of ten		 Binary names
powers of two
Name Symbol Multiple Name Symbol Multiple
kilobyte kB 103 ≈ kibibyte KiB 210
megabyte MB 106 ≈ mebibyte MiB 220
gigabyte GB 109 ≈ gibibyte GiB 230
terabyte TB 1012 ≈ tebibyte TiB 240
petabyte PB 1015 ≈ pebibyte PiB 250
exabyte EB 1018 ≈ exbibyte EiB 260
zettabyte ZB 1021
yottabyte YB 1024

I suggest that at the bottom of each -bibyte pages we have a see also with links to: the corresponding decimal-prefixed version, the corresponding -bibit, the binary prefix article, and the orders of magnitude (data) article. Similarly for the -bibit and decimal-prefixed pages.

I'll wait for a couple days for comments. If there are no objections, I'll turn the above into real templates and go about changing all the relevant pages. One-dimensional Tangent (Talk) 19:44, 19 May 2005 (UTC)

Looks good to me. Just one minor point: there is no sense in linking each -bit article to the corresponding -byte, as the two have no connection.
I also suggest that the table headers link to the primary articles, as follows:
Multiples of bytes
Decimal prefixes Binary prefixes
Name Symbol Multiple Name Symbol Multiple
kilobyte kB 103 ≈ kibibyte KiB 210
megabyte MB 106 ≈ mebibyte MiB 220
gigabyte GB 109 ≈ gibibyte GiB 230
terabyte TB 1012 ≈ tebibyte TiB 240
petabyte PB 1015 ≈ pebibyte PiB 250
exabyte EB 1018 ≈ exbibyte EiB 260
zettabyte ZB 1021
yottabyte YB 1024
And of course, there should be a third template for the bare prefixes, as in #A little better above.
– Smyth\talk 21:19, 19 May 2005 (UTC)
Looks good! I don't think enough people care about this for it to reach a true consensus in the next millenium. I say we just be bold and add it.
I agree that kilobit should link to kilobyte and vice versa.
Where would the third template go, Smyth? - Omegatron 21:27, May 19, 2005 (UTC)
I agree that we won't get a consensus; the reason for the delay was to gather any input others might care to offer. I like the title on Smyth's version, but I'm not sure whether to label the two columns "prefixes", since they show the prefixed words rather than just the prefix. ("Prefixed" perhaps? But that sounds as though something were about to be broken.) One-dimensional Tangent (Talk) 22:08, 19 May 2005 (UTC)
Created Template:Quantities of bytes and Template:Quantities of bits. My schedule just changed, so I'm going to start changing pages now. One-dimensional Tangent (Talk) 22:16, 19 May 2005 (UTC)
We can always change the titles. I'm just glad we finally got around to doing this.
By the way, "standard" capitalization is the first word capitals and then the rest lowercase, so it's "See also", not "See Also". It's absurdly pedantic of me to even bring it up, though. :-) - Omegatron 22:59, May 19, 2005 (UTC)
Ah. Now I know. Well, I don't feel like changing that now that I'm done, but I'll keep it in mind for future articles. Thanks for pointing that out. :) One-dimensional Tangent (Talk) 23:05, 19 May 2005 (UTC)

Uh oh. Do we need to do the same thing with megabit per second, kibibit per second, etc.? I see everyone is misspelling "mebibit" as "mibibit". - Omegatron 23:49, May 23, 2005 (UTC)

I made another one for the bit rates:

Bit rates
Decimal prefixes (SI)
Name Symbol Multiple
kilobit per second kbit/s 103
megabit per second Mbit/s 106
gigabit per second Gbit/s 109
terabit per second Tbit/s 1012
Binary prefixes
(IEC 60027-2)
kibibit per second Kibit/s 210
mebibit per second Mibit/s 220
gibibit per second Gibit/s 230
tebibit per second Tibit/s 240

Byte rates don't deserve their own articles at this point, and are just merged into bit rates. I would like if someone could double check them all. I fixed a lot of errors, but missed a few I'm sure.

For the record, it's mebibit (Mib) and mebibyte (MiB), not mibibyte or MeB. Everyone gets those mixed up. - Omegatron 23:11, Jun 19, 2005 (UTC)


[edit] decimal prefixes wrong?

Discussion moved from Talk:Gigabyte since it's much more appropriate here. I didn't actually move the whole conversation, since I don't want to overstep anything. - Omegatron 01:15, May 28, 2005 (UTC)

I am aware of the errors which people who are pushing the GiB notation (which has not been accepted in widespread use) have put all over the wikipedia. the article on binary prefix is well written for the most part and uses the unpopular titles for the sake of clarity in discussion. there are some claims in there which should be verified such as the ones you have made in the previous comment. please feel free to come forward with proof that those are the accepted uses in those fields (perhaps from IEEE or such) and we can go about citing sources. I have not changed them to the correct uses because i have not done the same. the abstract (such as this article) are easy enough to back up and that is why i edit it. to state what constitutes common use would take more reseach (which i dont currently have time to do) Cavebear42 17:20, 23 May 2005 (UTC)

That's a very good idea. I'll start collecting references. - Omegatron 17:56, May 23, 2005 (UTC)
And so you did, and a fine job i might add. I added 2 lines below and noted them as mine so as not to confuse the remainder of this work done by Omegatron. I think we are settled in data rate issues pretty firmly as being the standad SI interptitations of the units. What I said in Talk:Gigabyte, however, is still true. the original and (from a computer's standpoint) accurate definitions are still very much the standard definition. We can not abandon all that computers are built on and pretend like transistors have 10 fingers. The mebibyte notation has not gained widespread use and we can not pretend as though it has. The standard rule for such things on the wiki is to use the most common usage as used in the world. Also, the manual of style tells us "For units of measure use SI units, unless there are compelling historical or pragmatic reasons not to do so" and i feel that this is definatly both historical and pragmatic. I am willing join IEEE in conceeding this common usage in the name of clarity and, for lack of other options from IEEE or SI, use the unpopular IEC titles. In order to set a standard to be uniformly implemented across the wiki, I would like to suggest that we create a boilerplate of some sort to explain that we use the less popular titles directing inquisitive readers to this article. I would also like to see us create a rule in the Manual of Style in order to guide editors. I don't know what it takes to create such rules, but I think that this might be a wise move. perhaps it's time to move this discussion (once again) to Wikipedia talk:Manual of Style? Cavebear42 23:36, 31 May 2005 (UTC)
Actually, I think standards trump common usage. See aluminium. The Manual of Style says "For units of measure use SI units, unless there are compelling historical or pragmatic reasons not to do so". So whether their commonness is a "compelling historical reason" to use them or whether these recommendations even count as standards is debatable.
"What I said in Talk:Gigabyte, however, is still true. the original and (from a computer's standpoint) accurate definitions are still very much the standard definition. We can not abandon all that computers are built on and pretend like transistors have 10 fingers."
My biased POV: So the physicists should have their own definition of the mega- prefix so that quoting the speed of light is the more convenient whole number 3 instead of 2.99792458? Or the chemists will come up with their own version of the standard prefixes so that Avogadro's number starts with 6 instead of 6.0221415? The whole point of the SI prefixes is to maintain a consistent set of multipliers so that the various disciplines mesh well together. Laziness on the part of computer engineers shouldn't inconvenience everyone else. - Omegatron 13:29, Jun 1, 2005 (UTC)
Just my 2c since I've been wondering whether to edit this article on the same lines. When measuring bits (or clock cycles), the most common usage appears to be powers of 10 ie. SI. The confusion appears to arise when talking about bytes, whether these should be powers of 10 or powers of 2. Operating systems such as Windows and Linux measure bytes in powers of 2, as do RAM manufacturers. The odd man out seems to be makers of secondary storage devices such as hard disks and USB drives who insist on bytes measured in powers of 10. As far as the manual of style goes, I'd recommend bits are always SI, bytes are always powers of 2. --kudz75 01:04, 1 Jun 2005 (UTC)
hmmm, where to start? clock cycles would be true. they have very little to do with computers as that it's a measure of time. the measure of time existed before computers and stands independently of computing components. the same would be true for computing terms which done requre computing components such as the gigaflop. these are not derived from a binary system comming out of a transistor or similer device. the case of data rates is interesting. a bit is either 1 or 0 and in that sense, its binary. however data rates dont care about what the data is, they care how much is moved. if you move 100 bits in 1 second, it doesnt matter what those bits were and therefore si units work fine. data storage is where the tables turn. if i have 1 bit of memory, i can store 2 values (0, 1). if i have 2 bits, i can store 4 values (00, 01, 10 , 11). if i have 3 i can store 8 values (000, 001, 010, 011, 100, 101, 110, 111). it is easy to see that this is a 2x case where x is the number of bits you have. the fact that we break them into bytes and then count them from there is a historical norm, it was once thought that 8 bits could represent any nessicery number/chariter and therefore was the largest nessicery size for data storage.
as time progressed, we started to make larger devices and used the kilobyte (historical usage) to be 10 bytes of data. this would, of course, store 210 different bytes, not 102 differnt bytes. in hindsight, we should ahve found something other to call it than the kilobyte. back then calling it the kibibyte would have us currently not talking about it. the fact that we were already grouping into 8 bits at that time is the reason this is a bytes discussion. you see, if we never went to the 8 bit theory, we would have declared the kilobit (historical useage) to be 210 bits. this is why it would be a bad idea to keep the powers of 2 called this confusing tytle in bytes and just not bits (which is what we are pretty much doing now). now, hard drive manufacturers can claim whatever reason they want for using powers of 10 and not powers of 2 but the fact remains that the people who make the drives know the difference and that a drive with ~74 GiB of data has a giant 80GB written on the box. if one drive wrote 74 and another wrote 80 at the same price, you can guess which would sell better. in any case, we count in powers of 10 because we ahve 10 fingers, computers count in powers of 2 because they have 2 states. the confution here came when we chose not to give a new name to this new way of counting. I did not make this decition personally, but it was made and saying that it shouldnt have been doesnt change history. the best we can do a tthis point is figure out how to correct it.Cavebear42 18:05, 1 Jun 2005 (UTC)
IEC prefixes are *not* unpopular. They are gaining widespread use in newly developed applications. The OSes is the main inertia that is holding them back. You are mistaken about the lack of "other options" from IEEE and SI, because both have accepted IEC prefixes years ago. IEEE has published a standard with IEC prefixes, this standard has also been accepted by ANSI which re-released it, and at BIPM it has been decided to insert pointer to IEC prefixes in the upcoming new international SI edition, while American localisation of the current SI edition by NIST has referred to IEC prefixes all along. I have been tracking these issues for years now on my IEC prefixes and symbols for binary multiples page with links to all relevant documents and software applications that use them. Delicates 21:04, 3 Jun 2005 (UTC)
perhaps you and i define "popular" diffently, i mean that not only your adverage consumer has never hear of it, i mean that your adverage person who works with electronics and computers on a day to day basis also has never hear of it. that it has been referenced in a document somewhere does not make it popular. I see no point in pointing the finger at OS'es or anything/anyone else. there is a prudent reason why the historical definitions were used. last i heard SI did not have a unit to measure data storage. perhaps they will soon but i havent see that shown. IEEE asked people to not use KB to mean 1024, they did not ask people to use KiB to mean 1024. this is why i said that it was not accepted as their standard. when i get some more time on my hands, ill read your page, it sounds like an interesting thing to track. Cavebear42 21:20, 3 Jun 2005 (UTC)
IEEE Std 1541-2002, IEEE Trial-Use Standard for prefixes for Binary Multiples is a two-year trial standard that would have ended in 2004. Can't find what happened next. - Omegatron 19:36, Jun 6, 2005 (UTC)
It has been accepted into full-use on March 19. Delicates 21:50, 6 Jun 2005 (UTC)

I completely oppose pervasive use of "KiB" instead of "KB" to mean 1024 bytes. This is terrible, terrible. No one but the geekiest nerds knows what this means (this is not a pejorative claim). Is this being discussed anywhere else? --Locarno 18:59, 2 June 2006 (UTC)

Is this being discussed anywhere else?
Standards committees everywhere?
Atlant 19:08, 2 June 2006 (UTC)
No one but the geekiest nerds know what "KB" means, either. — Omegatron 20:05, 2 June 2006 (UTC)

[edit] Various references

[edit] Binary measurements (kilo- = 1024)

[edit] CDs

  • Data capacity of CDs - Data capacity in Mb for a CD-ROM
    • 74 min
    = 333,000 sectors * 2048 bytes / sector
    = 681984000 bytes
    = 650.4 Mb
    • 80 min
    = 360,000 sectors * 2048 bytes / sector
    = 737280000 bytes
    = 703.1 Mb
  • For 74 minute CD-Rs, the capacity is 74*60*44100*2*2*2048/2352 = 681984000 bytes, or 650.390625 binary MiB (exactly, no roundoff error).
  • For 80 minute CD-Rs, the capacity is 80*60*44100*2*2*2048/2352 = 737280000 bytes, or 703.125 binary MiB (again, this figure is exact, not rounded off). [5]

(please note that they meant Megabytes (MB) int his article when they said Mb) Cavebear42 23:36, 31 May 2005 (UTC)

[edit] Memory

  • "As an example, 64 MB of RAM memory always means 64 times 1,048,576 bytes, never 64,000,000." [6]

[edit] Decimal measurements (kilo- = 1000)

[edit] DVDs

  • Understanding DVD - Data capacity in GB for a DVD-R
    • 2,294,922 sectors * 2048 bytes / sector
    = 4,700,000,000 bytes
    = 4.7 GB
  • For DVD+/-R[W] media, the exact capacity is 4697620480 bytes, or just shy of 4.7 decimal GB. The capacity of a DVD-R is certainly nowhere near 4.7 binary GB. [7]

[edit] Data rates

  • "Lending confusion to this mess though, in some areas only decimal values are used such as when the term, "56K modem" works at a maximum speed of 56,000 bits per second, not 57,344." [8]
  • "Just to avoid confusion, 33.6 Kbps = 33600 bps, 28.8 Kbps = 28800 bps (where bps means bits per second), and so on." [9]
  • "Traditionally, Ethernet networks operate at 10 Mega-Bits per Second (10,000,000 Bits per second)" [10]
  • 1.4.48 bit rate (BR): The total number of bits per second transferred to or from the Media Access Control (MAC). For example, 100BASE-T has a bit rate of one hundred million bits per second (108 b/s). IEEE 802.3 standard Cavebear42 23:36, 31 May 2005 (UTC)
Is this true? Why would you have a power of 2 with a decimal unit? Does 128 kb/s really mean 128000 b/s? 128 is 2^7, which implies we're using binary. Can you provide a link to convince me otherwise? Tango 19:31, 24 December 2005 (UTC)
I don't know why the strange combination was chosen, but try it for yourself with your favorite constant bit-rate MP3:
Bitrate: "192 kbps"
Length: 262.3 s
Expected size if "k" = 1000: 192 * 1000 * 262.3 / 8 = 6295200 bytes
Expected size if "k" = 1024: 192 * 1024 * 262.3 / 8 = 6446284 bytes
Actual size:                                          6296347 bytes
– Smyth\talk 13:23, 25 December 2005 (UTC)
128 is a power of two, but the other available rates (32, 40, 48, 56, 64, 80, 96, 112, 128, 160, 192, 224, 256 and 320) are not. I'm sure it has to do with frame sizes or something. — Omegatron 03:58, 19 June 2006 (UTC)
Well, ignoring the fact that some of those rates are powers of 2, what you'll actually find is that all of those rates are multiples of 8, as in 8 bits per byte/octet.
So the BYTE rates would be 4 KB/s, 5 KB/s, 6 KB/s, 7 KB/s, 8 KB/s, 12 KB/s, 14 KB/s, 20 KB/s, 24 KB/s, 28 KB/s, 32 KB/s, and 40 KB/s. You can see a more-rational progression in those MP3 encoding rates, right?
Atlant 16:44, 19 June 2006 (UTC)

[edit] Hard drives

  • "Drive manufacturers, including Hitachi Global Storage Technologies, market their drive capacities in terms of decimal capacity. In decimal 1 kilobyte (KB) is equal to 1,000 bytes, 1 megabyte (MB) is equal to 1,000,000 bytes, and 1 gigabyte (GB) is equal to 1,000,000,000 bytes. Operating systems and some software programs (fdisk, partitioning utilities, system BIOS, etc…) all view the drive capacity in terms of a binary capacity. In binary, 1KB is equal to 1,024 bytes, 1MB is equal to 1,048,576 bytes, and 1GB is equal to 1,073,741,824 bytes." Why does my hard drive report a lower capacity than what is on the drive’s label? (Hitachi)
  • "Note that the Maximum Capacity shows only 3099 MB instead of 3240 MB. This is because some system BIOSs recognize a Megabyte as 1,048,576 bytes (binary). Drive manufacturers recognize a Megabyte as 1,000,000 bytes (decimal)." Hitachi
  • "This has to do with the way nearly every harddrive manufacturer in existance calculates hard drive size. They all define 1 gigabyte = 1,000,000,000 bytes instead of the 1 gigabyte = 1,073,741,824 bytes which it *really* is ... This is standard industry practice" [11]
  • "Hard drive size is given in Gigabytes (GB). A Gigabyte is one billion bytes or one billion characters." [12]
  • "Hard drive manufacturers define 1 gigabyte as exactly 1,000,000,000 bytes. By their definition, a 45BG hard drive is exactly 45,000,000,000 bytes. The true definition of 1 gigabyte is actually 1,073,741,824 bytes" [13]

[edit] Organization recommendations

  • IEC
    • Standard: IEC 60027‐2, Second edition, 2000‐11, Letter symbols to be used in electrical technology — Part 2: Telecommunications and electronics
    • "These prefixes for binary multiples, which were developed by IEC Technical Committee (TC) 25, Quantities and units, and their letter symbols, with the strong support of the International Committee for Weights and Measures (CIPM) and the Institute of Electrical and Electronics Engineers (IEEE), were adopted by the IEC as Amendment 2 to IEC International Standard IEC 60027-2: Letter symbols to be used in electrical technology — Part 2: Telecommunications and electronics. The full content of Amendment 2, which has a publication date of 1999–01, is reflected in the tables below and the suggestion regarding pronunciation." [14]
  • IEEE
    • Standard: IEEE 1541–2002, IEEE Standard for Prefixes for Binary Multiples
      • "1541-2002 (SCC14) IEEE Trial-Use Standard for Prefixes for Binary Multiples [No negative comments received during trial-use period, which is now complete; Sponsor requests elevation of status to full-use.] Recommendation: Elevate status of standard from trial-use to full-use. Editorial staff will be notified to implement the necessary changes. The standard will be due for a maintenance action in 2007." IEEE-SA STANDARDS BOARD STANDARDS REVIEW COMMITTEE (RevCom) MEETING AGENDA 19 March 2005
      • "1541-2002 IEEE Standard for Prefixes for Binary Multiples (Upgraded to full use from trial use)" [15]
    • Information for authors — "Information for IEEE Transactions, Journals, and Letters Authors"
      • TABLE OF UNITS AND QUANTITY SYMBOLS
      • "mega-: SI prefix for 106. The prefix mega shall not be used to mean 220 (that is, 1 048 576)."
      • "kilo-: SI prefix for 103. The prefix kilo shall not be used to mean 210 (that is, 1024)."
    • "Faced with this reality, the IEEE Standards Board decided that IEEE standards will use the conventional, internationally adopted, definitions of the SI prefixes. Mega will mean 1 000 000, except that the base-two definition may be used (if such usage is explicitly pointed out on a case-by-case basis) until such time that prefixes for binary multiples are adopted by an appropriate standards body." [16] (the IEC standard has been published since this note was released and later published by IEEE itself)
  • NIST
    • "The IEC has adopted prefixes for binary multiples in International Standard IEC 60027-2, Second edition, 2000–11, Letter symbols to be used in electrical technology — Part 2: Telecommunications and electronics. ... Although these prefixes are not part of the SI, they should be used in the field of information technology to avoid the incorrect usage of the SI prefixes." NIST Special Publication 330 2001 Edition The International System of Units (SI)
    • "Because the SI prefixes strictly represent powers of 10, they should not be used to represent powers of 2. Thus, one kilobit, or 1 kbit, is 1000 bit and not 210 bit = 1024 bit. To alleviate this ambiguity, prefixes for binary multiples have been adopted by the International Electrotechnical Commission (IEC) for use in information technology." nist.gov
    • "The new prefixes will eliminate the present confusion between powers of 1000 and powers of 1024 since in the field of information technology the SI prefix names and symbols for decimal multiples are now often used to represent binary multiples." News briefs Section 1.9
    • "With significant input from the National Institute of Standards and Technology, the IEC adopted kibi (Ki), mebi (Mi), gibi (Gi), tebi (Ti), pebi (Pi) and exbi (Ei) to represent exponentially increasing binary multiples. A kibibyte, therefore, equals 2 to the 10th power, or 1,024 bytes. Likewise a mebibyte equals 2 to the 20th power, or 1,048,576 bytes. The new prefixes for binary multiples, which parallel the metric prefixes, will increase precision in expressing electronic information." Representative's Report — April 1999
  • SI/BIPM
    • "These SI prefixes refer strictly to powers of 10. They should not be used to indicate powers of 2 (for example, one kilobit represents 1000 bits and not 1024 bits)." [17]
    • "These SI prefixes refer strictly to powers of 10. They should not be used to indicate powers of 2 (for example, one kilobit represents 1000 bits and not 1024 bits). The IEC has adopted prefixes for binary powers in the international standard IEC 60027-2: 2005, third edition, Letter symbols to be used in electrical technology — Part 2: Telecommunications and electronics. The names and symbols for the prefixes corresponding to 210, 220, 230, 240, 250, and 260 are, respectively: kibi, Ki; mebi, Mi; gibi, Gi; tebi, Ti; pebi, Pi; and exbi, Ei. Thus, for example, one kibibyte would be written: 1 KiB = 210 B = 1024 B, where B denotes a byte. Although these prefixes are not part of the SI, they should be used in the field of information technology to avoid the incorrect usage of the SI prefixes." [18]
    • "A decision was made to include a marginal note discussing the binary multiples along the lines of that given on p.14 of the NIST Special Publication 330, 2001 edition". Report of the 15th meeting (17 –18 April 2003) to the International Committee for Weights and Measures
  • ISO?
  • ANSI
  • W3C
    • Units in MathML — Section 5.3.5 -- Prefix, and Appendix B — shows how to incorporate IEC prefixes into mathematical markup.
  • SAE
    • "Thus 1 kbit = 103 bit = 1000 bit and not 210 = 1024 bit, where 1 kbit is one kilobit." [19]

[edit] Comment in the article

Warning: These values are wrong, SI uses 10-based counting, not 2-based. SEC (below) is 2-based. This also seems formatted quite messily (spaces everywhere).

Comments to the article like that belong here. Or fix the article if you think its wrong. --kudz75 06:26, 30 May 2005 (UTC)

Added again by User:66.231.16.111 as a HTML comment - Omegatron 19:28, May 30, 2005 (UTC)

non standard usage? i noted a warning about this being incorrect, and i commented on the spaces used for formatting, but i mreant around the table headings (" Symbol " or " Value ") ... not the numerical seperator used for reading clarity. the original author says SI kilo for bytes is 2^10 = 1024, that's the SEC KiB (noted below). SI kB or KB is 10^3 = 1000 ... hard disk manufactorers say "1 GB = 1 000 000 000 bytes" because they use SI numbering ... or 10 based counting, which is what SI is for, not base 2 counting, which SEC does. I added this as a comment this time so that i don't pollute the document, but I didn't know who to take this to

[edit] new table

I think the new table "Approximate ratios between binary prefixes and their decimal equivalent" should be folded into the preexisting tables. ("> 109 (7.4% error)" and so on) - Omegatron 14:22, Jun 6, 2005 (UTC)

[edit] Nominal 1.44 MB floppies and Windows XP

I've rewritten some text on the 1000*1024 hybrid "megabyte" used e.g. in floppies. This text was quite properly restored by User:Smyth after deletion by an anon. I just checked http://www.cdw.com and as of 2005 every vendor still refers to the standard floppy as nominally 1.44 MB.

Now, as for Windows XP, the situation is curioser and curioser. I was going to put something in the article but changed my mind pending any rational explanation of what Windows XP is doing.

As of the last time I tried, which was five minutes ago: when formatting a 3.5" floppy, Windows XP's formatting utility designates the diskette and the formatting operation as

3.5" 1.44MB 512 bytes/sector

That is, Windows XP still uses 1.44MB as the nominal capacity of a floppy.

But, after formatting, Properties reports the "capacity"

1457664 bytes 1.38 MB

(which is exactly 2847 sectors BTW... and only 1.4235 "hybrid" 1024000 megabytes, not 1.44, so obviously this is the usable capacity after the overhead of the FAT directory is deducted).

Now, 1457664 / 1024 / 1024 = 1.39014 MiB. That is, the second value is NOT consistent with MB meaning MiB, and cannot be explained as roundoff error since the fraction BOTH rounds AND truncates to 1.39 MiB, not 1.38 MiB.

Sounds like some kind of unaccountable sloppiness on Microsoft's part. I can come up with the following wild-ass guess. Suppose there was some point in the code's history in which the code computed 1457664 / 1024 / 1000 = 1.4235 hybrid "megabytes."

Now suppose that for some reason that was arbitrarily truncated to 1.42 MB for display.

Now suppose someone came along and decided that it should be displayed in 1024 * 1024-byte "MB."

Now suppose that instead of fixing the calculation they slapped on a correction.

Now suppose that for some reason they based the correction on 1.42 rather than 1.4235.

1.42 * 1000 / 1024 = 1.3867

Finally, suppose for some utterly unaccountable reason they decided to truncate rather than round... well, I guess you could get 1.38.

Given that all of the intermediate values in the appropriate calculations can be expressed EXACTLY in binary fractions OR decimals OR floating point with a very reasonable number of decimal places, this would seem to suggest sloppiness.

Yes, I remember the days when computers were still occasionally used for computing and programmers were expected to know the rudiments of mathematics and numerical analysis. Just hand me that slide rule, Sonny, and some carbon paper to put in my IBM Selectric. Dpbsmith (talk) 13:18, 21 Jun 2005 (UTC)

Oh, you must have missed the announcement. Computing isn't about math or accuracy anymore. It's now about obfuscation and elitism. - Omegatron 14:01, Jun 21, 2005 (UTC)
Every discipline or area of study suffers from this type of confusion when long standing traditions in a narrow area need to be standardized with other areas. For instance, in electrical engineering j is the unit for the imaginary number when everyone else uses i. Or "magnetic field strength" is not analgous to "electric field strength", the complementary term is "magnetic flux density" because someone else coined magnetic field strength for something else. Computer science is no different when it comes to SI prefixes and information measurement. --kudz75 00:46, 27 Jun 2005 (UTC)
It doesn't matter, but you're missing the point here. This is not a question of binary prefixes and which unit XP decided to use. The issue is that it is difficult to guess what Windows XP is doing because either it's calculating available space in MiB and getting the calculations wrong, or else it is accurately calculating something I don't understand. Which is very, very annoying as there is no good reason at all for the calculation to be inaccurate. Dpbsmith (talk) 01:21, 27 Jun 2005 (UTC)
They're probably calculating it by taking the number of bytes (1457664), dividing by 1024 (1423.5), truncating (1423), then dividing by 1024 (1,389648437) and rounding down... maybe. --Quadduc 01:52, 8 March 2006 (UTC)

[edit] POV discussion

There is a discussion going on at Talk:kilobyte about the POVness of kilo- = 1000, etc. - Omegatron 04:13, Jun 26, 2005 (UTC)

[edit] Which should we use in Wikipedia?

Discussion at Wikipedia:Village pump (policy)#Unit Disagreement.2C MiB vs. MB. - Omegatron July 8, 2005 12:58 (UTC)

A vote has been started on whether Wikipedia should use these prefixes all the time, only in highly technical contexts, or never. - Omegatron 14:49, July 12, 2005 (UTC)

[edit] Popularity contest

User:Pmsyyz wrote (in bold): As of 2005 this naming convention [Kibi...] has not gained widespread use, but its use is growing. It is strongly supported by many standardization bodies, including IEEE and CIPM.

  • Do we have a chart (google results over time maybe?) for this rate of change of usage of kibi-units?
    • Yes, it would be a good idea for someone to try to back this statement up. My guess is that it's true, but I wouldn't have put it in myself without some kind of source citation. Dpbsmith (talk) 12:26, 18 July 2005 (UTC)
  • What would be the use of "many" standards bodies agreeing on something? Surely the whole idea of standards is that there's just one authorative standard for each thing that needs standardising?
    • Standards organizations frequently adopt each others' standards. In the computer field there are many examples of ANSI and ISO standards in which the text of the standard itself is identical. Of course, each organization follows its own rules and has its own committees to make the decision. When something like this does happen it's reasonably significant. Dpbsmith (talk) 12:26, 18 July 2005 (UTC)
  • What does "supported by" mean in reference to standards? Does it mean they've each published a standard saying which style should be used?

Ojw 11:51, 18 July 2005 (UTC)

Have you read the #Organization recommendations section above you? - Omegatron 13:27, July 18, 2005 (UTC)

[edit] Why decimal bytes for HDD?

I think there is a non-neutral POV in this sentence, "Hard disk drive manufacturers state capacity in decimal units, so what is advertised as a "30 GB" hard drive will hold 30 × 10^9 bytes, roughly equal to 28×2^30 bytes (i.e. 28 GiB)."

It implies that HD manufacturers specifically use the decimal designation to inflate the capacity designation. While it may be true at this point in history, it almost certainly was not the original intent of the engineers who created the first hard drives. I propose something like, "Hard disk drive manufacturers state capacity in decimal units. Since most computer operating systems report drive usage and capacity in binary units, the difference causes an apparent loss between the advertised capacity and the formatted, usable capacity."

Second, since the article speculates on the tradition of using decimal units for HDD capacity, I don't think the immediate subsequent statement is accurate, "This usage has a long engineering tradition, and was probably not influenced by marketing. It arose because nothing about the physical structure of the disk drives makes power-of-two capacities natural: the number of platters, tracks and sectors per track are all continuously variable."

I propose, "The decimal unit capacity in hard disk drives follows the method used for serially accessed storage media which predated direct access storage media like hard disk drives. Paper punch cards could only be used in a serial fashion, like the magnetic tapes that followed. When a stream of data is stored, it's more logical to indicate how many thousands, millions, or billions of bytes have been stored versus how many 1024, 1,048,576, or 1,073,741,824 bytes have been. When the first hard disk drives were being developed, the decimal measurement was only natural since the hard disk drive served essentially the same function as punch cards and tapes. Thus today, any device that is addressed or seen as "storage" uses the decimal system to identify capacity."

JJLatWiki 18 July 2005 16:09 (UTC)

I agree it could be expanded or worded differently. It's a wiki, so be bold and change it to the way you like!
Also, can you sign your posts by typing ~~~~ after them? It will show your username and time of posting, like so - Omegatron 16:22, July 18, 2005 (UTC)
I couldn't find how to sign the edits, thanks.
I'm not yet comfortable with modifying an encyclopedia without a review and comment process. But, I'll take your advice...
JJLatWiki 17:19, 18 July 2005 (UTC)
If only vandals had your attitude... :-)
Editing the article is the review and comment process. Other users (like me) have this article on their watchlist, and we see any changes made to it, and make any changes to your edits that we think are appropriate, then you change our edits, and so on. Read through some of the links on the welcome message on your talk page to find out more.
Asking on the talk page first if you are not sure about an edit is helpful, too, of course. Thanks. - Omegatron 18:27, July 18, 2005 (UTC)
For instance, I just made a few small changes to your addition, and others will come along and change them even more.
I assume you got this info from a website? Can you reference the pages by adding a link after each description? Just put the URL in [single brackets] and it will look like this: [20] - Omegatron 18:38, July 18, 2005 (UTC)
Is it correct to say "Since most computer operating systems report drive usage and capacity in binary units, ..."?
I happen to use Windows 2k which gives Properties of my C: drive as:
"Capacity 30,065,098,752 bytes 28.0 GB"
I think Microsoft would do the world a favor if they said 28.0 GiB but they certainly make it clear that this is a 30 billion byte drive. To support the "most computer operating systems" assertion I'd suggest we would have to review the various current utilities and properties on current OS's like XP, various LINUX, Apple X, etc. I suspect XP is not unlike Win2K so given XP has by far the largesst installed base, most will not hold up. BTW, my recollection is that MSDOS in all its flavors reported integer capacity with no prefixes or even commas :-). So most may also not be historically true. I'd like to see less definitive language which puts the source of the confusion at the inconsistent and/or unexplained usage of prefixes by the OS companies. I'm thinking about such language but I thought I'd post this talk before I edited the article. Yr thoughts?Tom94022 01:07, 18 May 2006 (UTC)
It still use binary units rather than decimal units even in your example (so for a user unclear about what a "giga" the only two numbers he can compare will appear inconsistent) and if you look at disk from explorer with "show details", you will only get the binary unit in "size" without the number written in full. Thus, I see no reason to reformulate the text. --Per Abrahamsen 07:25, 18 May 2006 (UTC)
The text is completely correct. If 2k/XP measured in decimal units, it would say;
"Capacity 30,065,098,752 bytes 30.0 GB" — Omegatron 11:27, 18 May 2006 (UTC)
We agree, 30,065,098,752 is decimal units! Windows is inconsistant, using both decimal units to 11 places and decimal units with non-standard binary prefixes. So the current text is incorrect since Windows does provide capacity in decimal units! I don't know much about Apple and I'm told that some Linux have a flag -T for ds that displays capacity with decimal prefixes. So a more correct sentance might be something like:
"Since some operating systems continue to use non-standard prefixes without explanation ..."
BTW, I note someone else has changed the text but IMHO it is still incorrect.Tom94022 20:50, 26 May 2006 (UTC)

[edit] Legal disputes

I added the Legal Disputes section. The information I provided is publically available in many places on the web. Within each case is exactly the kinds of debates going on here.

In my opinion, the difference though is that these people are attempting to take advantage of the debate and claim that corporations are literally charging "per megabyte" and that the corporate megabyte is smaller than the "commonly understood" megabyte and so the consumer is being deceived and cheated. In my opinion, the plaintiffs are filing frivolous lawsuits without merit.

Even though the capacity available to the user is (almost universally) less than the capacity designation, consumers do not pay a dollar amount per mega/gigabyte, therefore they are not paying something for nothing. Likewise, there are no hard drive or Flash drive manufacturers who designate capacity in the binary method and are therefore harmed by the other manufacturers' deception.

EVEN IF the capacity designation were accurate according to the binary method, and a 30GB drive had a formatted capacity of 32,212,254,720 bytes, it is a practical impossibility to store 32,212,254,720 bytes of user data on such a drive. Who is to be sued for THESE losses?

JJLatWiki 18:50, 18 July 2005 (UTC)

You make good points. But I would make a weak case for there being some kind of consumer issue here, however. The issue is this: can the consumer make a fair comparison? I don't think it's terribly farfetched to say that, left to themselves, companies will eventually engage in "specsmanship" in which they use deliberately confused, varying, and obfuscated measurements in order to make comparison difficult. It's no accident that supermarkets shelve all the General Foods cereals together, rather than putting all the different kinds of cornflakes together.
In the 1960s and 1970s stereo components makers were engaged in a kind of "horsepower race" and different companies used differing definitions of a "watt." There really was no easy way for the average consumer to know whether the a "20 watts per channel" system from one vendor was really comparable to a "20 watts per channel" system from another. In the case of stereo systems, where, all things being equal, a 20 watt system really costs quite a bit more than a 15 watt system, if a manufacturer could sell a 15 watt system and represent it as a 20 watt system by using a slightly different definition of "watt," the financial gains were meaningful.
It was very, very difficult to shop for fuel-efficient cars until the EPA tests were introduced, and similarly for household appliances.
In the case of disk drives, however, I don't think that different disk vendors are using different definitions of a gigabyte. As far as I know, a nominal 80 gig Maxstor drive is perfectly comparable to a nominal 80 gig Seagate.
A very analogous case is the tradition of measuring screen sizes by the diagonal. This made perfect sense in the 1950s, when picture tubes were, in fact, circular. People might have been disappointed or confused when they bought a nominal 21 inch set and found that the picture on it was only 16 inches wide, but everybody's 21 inch set was the same size. However, in the 1990s there were quite meaningful differences in usable picture size between different vendor's nominal 15" computer monitors.
There are good reasons for customers to want vendors to use a common, well-defined way of specifying disk capacity. Dpbsmith (talk) 19:10, 18 July 2005 (UTC)
In my opinion, corporations should be left to their own devices UNTIL they use "specsmanship" with an intent to deceive. Back when the speed national speed limit was reduced to 55, I don't think consumers were harmed by Chrysler's "zero to fifty" (or was it 55) acceleration scores. Obviously in that case, the measurement system was built into the rating.
"There are good reasons for customers to want vendors to use a common, well-defined way of specifying disk capacity." I quite agree. In the case of the hard disk drives, I think it was a common and well-defined system. It just didn't agree with what the new consumers expected and worse were being told by their OS vendor. The major Flash drive manufacturers used the same common and well-defined system, but it again didn't jive with consumer expectations that resulted from their chosen OS.
The binary spec is meaningless to consumers with regards to RAM except for direct comparisons to other computers being considered for purchase. With regards to disk storage, the binary spec is only meaningful because the OS vendor arbitrarily chose it. Remember when people actually bought software to add "virtual" RAM. I don't think it was ever licensed per megabyte, but what a glorious time that would have been for a memory warehouse. They could just ship paper licenses for 64KiB, 128KiB, 256KiB, 512KiB, and 1MiB of v-RAM. JJLatWiki 19:33, 18 July 2005 (UTC)

[edit] We against the industry.

I don't understand nor like the "ibibyte" thing.

Aw, c'mon, it's not that hard! The average "filesystems designer", which you claim to be, should have little trouble with this stuff. —Herbee 09:12, August 22, 2005 (UTC)
Just think that all filesystem specifications from the first to the newest use Kilobyte and not Kibibyte, I think, it is a problem ;) Claunia 09:38, 22 August 2005 (UTC)
Granted, but so what? This stuff is still not hard to understand, even if you don't like it. —Herbee 16:06, August 23, 2005 (UTC)

For decades, the industry always used 1000 based for drives capacities while the software used 1024 based multiplies. They always used Kb as 1000000 bytes in drives and Kb as 1048576 bytes in software.

Silly mistake, and that from a "computer science teacher"! The industry have been known to use K to mean either 1000 or 1024, but never a million. —Herbee 09:12, August 22, 2005 (UTC)
You never make a mistake? Simply put Kb when I wanted to say Mb Claunia 09:38, 22 August 2005 (UTC)
Sorry, I just couldn't resist. Yes, I do make mistakes, just as silly as yours, and I do get kicked in the butt for it. It's human nature, I guess. Don't get upset, it's all part of the game. —Herbee 16:06, August 23, 2005 (UTC)

Why should we change the whole world while the people is still using that?

Sigh… —Herbee 09:12, August 22, 2005 (UTC)
You against the world. Wanna be the new Ghandi? Claunia 09:38, 22 August 2005 (UTC)
You lost me there. It's more like innovation alongside conservatism. Not against: the dinosaurs will manage to become extinct all by themselves…;-) —Herbee 16:18, August 23, 2005 (UTC)

Is great that the IEC created new terms to avoid confussion, but it is creating confussion on non technically aware people.

How so? The meaning of K isn't changed: it's just as undefined as ever. Why would people be confused about a new symbol (Ki) with a precise definition? Are they, perhaps, already so full of symbols that they just cannot handle one more? Or is it the case that a certain "systems administrator" doesn't see the difference between K an Ki? —Herbee 09:12, August 22, 2005 (UTC)
THIS is a public encyclopaedia. The intention is the whole public, and not only ones with computer, mathematic and science knowledges. Most people doesn't even known what the unit prefixes mean. Claunia 09:38, 22 August 2005 (UTC)
As you say, the meanings of 'kilobyte' haven't changed; it's still ambiguous. Someone created kibibyte which is well defined, but instead of also creating a kidibyte (or whatever 1000 bytes would be called), they left it as "kilobyte" which is still ambiguous (unless you can convince everyone in the world to stop using kilobyte to mean 1024 bytes, which seems unlikely)
Won't work. Your "kidi" would have exactly the same meaning as "kilo", so what's the point? You don't expect the BIPM to obsolete the "kilo" prefix, do you? Or would you have us use "kilo" to mean 1000, except for bytes, where it's "kidi"? You must by "kiding"! —Herbee 16:32, August 23, 2005 (UTC)
kilobyte = 1024 or 1000 bytes. It's not "exactly equal" to anything. Just look at the size of this talk page, and show me another dictionary definition which is as disputed as this one? Ojw 17:51, 23 August 2005 (UTC)
Billion! :-) Dpbsmith (talk) 19:32, 23 August 2005 (UTC)
Essentially, solving a dispute by telling half the people that they're wrong didn't work, because it would require those people to agree. Creating two new units (kibibyte and kidibyte, for example) would allow an upgrade path from the disputed term to a clearly-defined term in every instance, but that option wasn't chosen. Ojw 17:12, 22 August 2005 (UTC)
But kilo- already means 1000 in every other instance it is used. Computer terminology is the only aberration. — Omegatron 18:29, August 22, 2005 (UTC)
I bet the IEC had many of the same debates we're now seeing on WP. But I assume this particular issue was handled with the question of "what is most technically correct?". That leads to the question, why would we want 2 prefixes that both mean 1000? Having 2 isn't an "upgrade path" because that implies that the intermediate designation will be discontinued at some future date and then we get to have these debates all over again. "But we've been using 'kidi with a D' bytes for years now. Using 'kilobytes' to mean 1000 bytes is just too confusing." But it's wrong to say that this change implies that half the people are wrong. Only the first two people in history to use kilo to mean 1024 were wrong.--JJLatWiki 15:00, 24 August 2005 (UTC)

Try to explain why they are reading gibibytes in wikipedia while their Windows XP says gigabytes. Was enough difficult to explain they the differences between drive manufacturers gigabytes and real gigabytes.

Please try to sign your comments, User:Claunia. You can do this by typing four tildes (~~~~). —Herbee 09:12, August 22, 2005 (UTC)
Still getting in use with wikicode Claunia 09:38, 22 August 2005 (UTC)
Maybe you should stop thinking Microsoft gigabytes as "real" gigabytes. It will be years before OS's start differentiating between giga and gibi. Obviously it's easier to ignore the issue. Maybe someday some sewer dwelling lawyer will decide to sue Microsoft for falsely claiming their leech of a client's hard drive had only 50 terabytes of free space when it actually had 55 terabytes. When that happens, Microsoft might join the larger community. Until then, explain to people that Windows XP has a flaw in how it calculates a gigabyte that makes it seem like their is less space. Or just keep telling them, like we've all been doing for years and years, that there are 2 common meanings for gigabyte and Windows wastes a lot of space on the hard drive so use those measurements as a rough estimate...--JJLatWiki 15:41, 24 August 2005 (UTC)
IMHO, most (if not all) distros of Linux differentiate between GiB and GB. Nippoo 09:32, 13 July 2006 (UTC)

"Is great that the IEC created new terms to avoid confussion, but it is creating confussion on non technically aware people."

Non-technically aware people don't know what a kilobyte is, either. Most assume it means 1000 bytes, like every other usage of the kilo- prefix they have encountered. — Omegatron 13:24, August 22, 2005 (UTC)

It's not "we against the industry." You can't get more "industry" than the IEC and other organizations which have endorsed the standard. If it must be phrased in terms of a conflict, it is more like "engineers versus marketers." Engineers have a vested interest in making measurements clear. Marketers have a vested interest in making them fuzzy, to make it harder for consumers to compare products. And, once their competition adopts a slightly misleading usage that puts them in a better light, it's a marketer's job to make that their own company follows suit.

To the extent that we take sides, Wikipedia should be on the side of "making things clearer." To the extent that imprecise or commercially loaded language is part of ordinary discourse, we should note that fact and explain it.

We should do whatever is needed to make sure that users understand.

In this case, we have a usage which is officially endorsed by a number of standard organizations, is easily understood, is precise, but is less familiar to most readers. Against it, we have a usage which is imprecise, ambiguous, not used consistently by those who use it, but is more familiar to most readers.

To my mind, on balance we should favors the first usage, primarily because the "common" usage is not easily understood. The average person has no idea whether a gigabyte of RAM stores as much as a gigabyte of disk. When someone uses the word "gigabyte," nobody, no matter how experienced in the industry, really knows whether they mean the binary or the decimal usage. Sure, you can guess at the meaning, but we should not put our readers into a situation where they need to guess. Dpbsmith (talk) 15:58, 22 August 2005 (UTC)

Definitely agree. We've already covered the policy of this here — Omegatron 18:29, August 22, 2005 (UTC)

[edit] Incorrect is POV?

Does anyone else interpret the "wrongly...", "even though this is incorrect..." as POV pushing? Don't get me wrong, the MiB/GiB/etc... standards are a welcome change, but it has been the correct usage for decades, and even in the industry, the prefixes are widespread and regarded as correct. StuartH 08:23, 28 October 2005 (UTC)

I don't think so. The meaning of those prefixes is defined by the BIPM, International Bureau of Weights and Standards, and their use in the industry was incorrect according to those definitions. Cans containing 13 ounches coffee are sometimes referred to by their manufacturers as "13-ounce pounds" but I would not expect to be challenged if I were to say "It is incorrect to refer to 13 ounces as a 'pound.'" This is because in the U. S. it is the National Institute of Science and Technology that defines these terms, and Maxwell House is not the NIST.
Furthermore, the confusion is recent because my perception was that during, say, the sixties and seventies, the binary meanings for "kilo" and "mega" were well understood to be sort of a joke. It was only when the PC revolution suddenly expanded the use of computers to a huge nontechnical population that people started to believe that the binary meanings were "real."
What organization would you suggest has the right to define the meanings of the SI prefixes? Is there a serious, widely held viewpoint that says that Microsoft, not the BIPM, is the authority for this?
I am not sure, but I imagine that in countries using the SI, the BIPM is actually the legal authority for the meanings of terms defined in the SI.
Eventually, yes, there is a point when a technical term is misused by marketers and the uninformed to the point where the misuse does become standard, but I don't think this is the case here. Note that as soon as there was perceived to be a problem, the industry's response was to introduce binary prefixes, not to lobby for the BIPM to add the binary meanings for the old terms to the existing standard.
These phrases could be expanded to "even though this is incorrect according to the SI" or "even though this usage is contrary to the international usage that had been in effect for decades," but I think that would be unnecessarily wordy and pedantic. Dpbsmith (talk) 10:42, 28 October 2005 (UTC)
Remeber the ki/Mi/Gi etc prefixes came along after that fact when use of the traditional prefixes in the binary sense was already widespread and you hardy ever seem them used. Plugwash 10:55, 28 October 2005 (UTC)
They came along just about the time there started to be serious confusion about the meaning of the traditional prefixes. Up until the nineties, people who used them in the binary sense understood that that wasn't their correct meaning. The use of the IEC prefixes is indeed not widespread. However, the use of SI prefixes in the binary sense is still, exactly as correct or incorrect as the use of the word "pound" to refer to a 13-ounce can of coffee. Dpbsmith (talk) 12:58, 28 October 2005 (UTC)
The way I see it, though, is that the binary prefixes were "borrowed" from the almost-equivalent SI units because of the need to work in powers of two for addressing and other binary purposes. So they don't really claim to be SI units, just that SI prefixes avoided the need to provide a whole new set of definitions. Maybe they should have used "MiB" and the like from the start, but until now, the context has largely eliminated any ambiguity. But first and foremost, the fact that most of the industry and most consumers still use the old binary prefixes indicates that they are not incorrect to use. StuartH 23:14, 28 October 2005 (UTC)
For decades? I'm sure you meant centuries. Besides the incorrect usage is even a minority in computing industry itself. Most of computing terms use the prefixes correctly. It's the very narrow limited area of terminology that's gone haywire. Delicates 00:39, 29 October 2005 (UTC)
Yes, it is pushing a point of view. The way our language works, someone can coin a new word, or a new prefix. But there is nothing which gives them absolute control in perpetuity over its use. These prefixes aren't "trademarked" or anything like that.
So yes, the originators of the usage can certainly argue that other usage is "incorrect" and contrary to the original intention, but that doesn't really mean that it really fails any linguistic test of correctness.
The BIPM has no legal authority to set any standards; it is just responsible for day-to-day operations. Even the CGPM has no general, plenary legal authority to set general standards on the use of the English language. Anything they say along those lines is more of an advisory nature, not a "legal authority".
Various standards organizations in the computer field did recognize that usage as correct, long before any "kibi-" and the like were ever invented. Gene Nygaard 13:14, 28 October 2005 (UTC)
What's the source of the legal definition of "meter" in the U. S.? Dpbsmith (talk) 14:25, 28 October 2005 (UTC)
Also, which standards organizations were they? Dpbsmith (talk) 15:53, 28 October 2005 (UTC)
I bet if this was about "brontobyte" or "millimicron", no one would object to the use of the word "incorrect", even though they are "widely used", too. — Omegatron 14:18, 28 October 2005 (UTC)
They are? – Smyth\talk 15:51, 28 October 2005 (UTC)
Another reason for using the word "incorrect" is that dictionaries (specifically AHD4 and Merriam-Webster) do not give the binary meanings for kilo-, mega-, giga-, even though the informal binary usage has been current for decades. Dpbsmith (talk) 14:25, 28 October 2005 (UTC)
Then the dictionaries are incomplete and should be updated. :) – Smyth\talk 15:51, 28 October 2005 (UTC)

[edit] Moving away "Specific units of IEC 60027-2 A.2"

In the "See also" section there are some tables specific to IEC 60027-2 A.2 that IMHO should be moved to the article IEC 60027, keeping Binary prefix only for general discussion of the topic. Moreover, in those tables I read that, for SI prefixes, the notations "(or 2x)" and "(or KB)" have been added. Since it is stated that the tables refer to IEC recommendations, and not to "common usage" (regardless my POV), I think that notations should be removed, or moved elsewhere, or the captions should be changed. In the current form, it seems that IEC 60027-2 A.2 allows both decimal and binary meanings for SI prefixes, that AFAIK is not correct. SalvoIsaja 09:36, 6 November 2005 (UTC)

[edit] Decimal-to-Binary Prefixes and Binary-to-Decimal Prefixes Converter

Hi Guys, Okay I added a convertor to the binary prefixes page as an External Link a while ago and noticed it's removed now.

Why would anyone remove it? Is the contribution a bad thing? I fail to understand why it was removed.

It happened here. I think it was an overzealous vandalism revert. Go ahead an re-add it if you want. Also, please sign your comments like this: --~~~~. --P3d0 11:06, 21 June 2006 (UTC)

[edit] Citing of SI prefixes implies "Millibyte" and "Microbyte" and ...

The whole article eschews to mention the other existing prefixes for SI units. All physical units of the SI standard can be scaled up (with kilo, mega, giga...) but ALSO always scaled down (with deci, milli, micro, ...). However for theoretical units like the byte, only the upscaling prefixes are used, because other SI prefixes oh so surprisingly make no sense here.

So when citing it, the article should note that SI may not require, but obviously implies applicability of ALL scaling prefixes. And it should note that therefore the not so widespread use, limited meaningfulness or even non-existence of units like "microbyte" or "decibyte" hint that byte unit prefixes may not be identical in meaning to their SI unit counterparts. (Things like "microbyte" might make sense theoretically for crypthography or computer compression algorithms, but apart from that would be totally idiotic.) --mario

Thank you for your suggestion! When you feel an article needs improvement, please feel free to make whatever changes you feel are needed. Wikipedia is a wiki, so anyone can edit almost any article by simply following the Edit this page link at the top. You don't even need to log in! (Although there are some reasons why you might like to…) The Wikipedia community encourages you to be bold. Don't worry too much about making honest mistakes—they're likely to be found and corrected quickly. If you're not sure how editing works, check out how to edit a page, or use the sandbox to try out your editing skills. New contributors are always welcome. --P3d0 13:08, 7 June 2006 (UTC)
Just wanted to have it discussed first, as it happens to be only my personal nitpick on the topic. (Apart from that I have half a dozen Wikis of my own ;). And I'm rather going to wait for full OpenID support instead of registering yet...
Diminutive prefixes also make sense for data rates. Millibytes per second, and so on. — Omegatron 13:56, 7 June 2006 (UTC)
There are a couple of Google entries for 'mBps' and other such measurement units. While I haven't heard of any of them before (KB/s or MB/s, ok) and I can't rightout imagine any devices where such slow transfer rates could be typically measured, it sounds plausible nevertheless. Only I'd like to differentiate between such compound types (the milli probably stems more from the time aspect of such units) and the "byte" basis type originally discussed in the article. (Though the mB/s argument must then be mentioned as well..) --mario
No one measures in yoctometres or gigafarads, either. All of the units have natural limits to the prefixes that are commonly used with them, but all the prefixes are still valid, for hypothetical examples, etc.
Right, "attometers" or something like that are too small to be measured in practice, and others like an "exajoule" are unlikely to occour in nature. The difference between such physical values and the mathemetical / computer sciences Byte however is, that a millibyte not only cannot be measured, but plainly makes no sense / does not exist. If you think about it, a byte can hold values from 0 to 255, and a bit can old 0 or 1. A supposed "decibyte" however cannot hold information at all (I fail to understand what 80% of a bit could be useful for). OTH I have no reference to back up what is mostly my opionion here. But anyhow the article should note that there is a real discrepancy in applying the same-named SI prefixes to non-physical units like the byte. (This whole articles purpose was to highlight the difference between Base2 and Base10 prefixes/names, not?) --mario
A millibit could easily be a measure of information entropy. Our own article on the topic says English text has an entropy of 1.1 to 1.6 bits per character. I could see someone referring to this as 1100 - 1600 mb. --P3d0 17:11, 16 June 2006 (UTC)
The question is not whether somone could refer to it in this way. Someone could also refer to it as "one trillion, one hundred million to one trillion, six hundred million picobits." Someone could refer to it as 0.0003 to 0.0005 decimal kilodigits per character. Or (I'll leave the math as an exercise for the reader) refer to it in tera-dice-rolls or femto-roulette-wheel-spins. The purpose of this article is not to provide a demonstration of ingenuity, or clever ways to construct hypothetical units that are never used in practice but would have a well-defined meaning if they were. Dpbsmith (talk) 17:30, 16 June 2006 (UTC)
must be applicable for all SI units but make no sense for bits and bytes.
That's not true at all. They make just as much sense as fractional bytes ("4.3 KB/s"). — Omegatron 18:00, 7 June 2006 (UTC)
"4.3 KB/s" is first a compound unit (and not just the prefixed Byte), and second a clear approximation. Of course you can have half a byte (that's a nibble or clear 4 bits). But you seriously can't have 0.3721 of a byte (not with nowadays computers). There needs to be a real example for the existence of fractions of the smallest information container in computer science.. --mario
It takes 20.1 bits to encode a Unicode code point. If you have a random stream of Unicode code points, it will take on average 20.1 bits per character. Information theory frequently takes fractional bits.--Prosfilaes 05:30, 10 June 2006 (UTC)
How could you have millibytes? microbytes? one thousandth of a byte? one millionth of a byte? bit is the smallest unit in computing and a byte consists of 8 bits. one eighth (1/8) of a byte makes sense (= 1 bit). One thousandth of a byte (that is 1/1000 of a byte) doesn't make sense to me. -- McoreD, 2006-06-16T09:03
If I'm giving you 10 apples a day, how many apples do you get in an hour? This isn't difficult math here. — Omegatron 11:33, 16 June 2006 (UTC)
Yes, but nobody but a nerd making a joke would state the answer as "four hundred sixteen and two-thirds milliapples."
Fractional bits do have a meaning and uses. For example, the channel capacity of a noisy channel would likely involve fractional bits. But just because the word "millibits," if it were ever used, would have a well-defined meaning, does not mean that it is a real word in real use. It's rather like "vigintillion." It's in the dictionary, and in theory it has a well-defined meaning, but nobody ever really uses it—except in tables of names of big numbers—because, in the contexts in which it is needed one would just state it in scientific notation.
I am very skeptical that the word "millibit" is really used to any significant extent. I do not think it should be in the article unless someone can cite some good example, say a book on information theory or a research paper on telecommunications or something like that, to show that it is really in reasonably widespread and customary use. Google Books gives five hits on "millibit" but four of them look like scannos to me, and one is apparently an entry in a technical dictionary. Dpbsmith (talk) 14:02, 16 June 2006 (UTC)
Omegatron, I agree there is a mathematical sense to it. But there is no physical sense afaik. 0 or 1 is the smallest unit you can break in machine language. Physically, a millibit would imply a 0 or 1 is made out of 1000 unknown small things. But again, I agree, it mathematically makes sense. :) -- McoreD, 2006-06-20T00:31

Ok, I've totally lost track of what we're arguing here. I think we all agree on the following points:

  • Fractional units of a bit could potentially be used in some fields
  • In practice, such units are never used

Do we agree on this? If so, what's the problem? --P3d0 22:37, 16 June 2006 (UTC)

There isn't one yet. I would have a big problem if, say, someone were to extend the table "Binary prefixes using SI symbols (non-standard, but common)" to include explicit listings of millibits through femtobits. I don't think fractional prefixes need to be mentioned at all. If someone thinks they should be, I would want it limited to a short parenthetical remark like this:
(Quantities of fractions of a bit are encountered in some technical areas, notably information theory. When they are, they are usually expressed numerically using ordinary decimal or scientific notation. Words such as "millibits" are rarely if ever encountered). Dpbsmith (talk) 02:36, 17 June 2006 (UTC)
If there were an actual SI standard it could be argued that the table should list every name endorsed by the standard. However, the SI does not cover units of information, and the use of SI prefixes in this context is, as the table says, "non-standard, but common." I think the article only needs to discuss the words that are in common use. It is significant that the IEC standard does not include any fractional quantities. Dpbsmith (talk) 02:37, 17 June 2006 (UTC)

As I said above, we list yoctometres and gigafarads. They are valid units, even if no one ever uses them. Why should this be any different? If you think we shouldn't list valid units simple because they aren't ever used, then we should be removing them from all unit articles. — Omegatron 20:39, 17 June 2006 (UTC)

This is different, because yoctometres and gigafarads are part of the SI standard. The binary meanings of the SI prefixes are not. They exist only as a matter of customary usage, so only the values which are customarily used are relevant. Dpbsmith (talk) 21:06, 17 June 2006 (UTC)

In case anyone cares, I calculate that one millibit is the quantity of information contained in a statement that you were already 99.93% sure of. For example, the statement "my birthday is not on February 29th" contains approximately one millibit of information. --P3d0 17:51, 19 June 2006 (UTC)

Yes, but which did you actually mean when you wrote the sentence:
  • 1/1000 bit, or
  • 1/1024 bit? Dpbsmith (talk) 18:47, 19 June 2006 (UTC)
Humour value aside, there might be some practical applications of millibits or even microbits (per unit time) on channels with extremely narrow bandwidths or very poor signal-to-noise ratios.
Atlant 22:32, 19 June 2006 (UTC)
The issue is not whether there is any practical application for the concept of fractional bits. There is. Everyone agrees to that. The question is whether or not the term "millibit" is widely and customarily used to measure fractional-bit quantities. It is not. Fractional-bit quantities are customarily referred to by using the word "bit" together with ordinary numeric notation. The proof that it is not a term in common use is that nobody knows whether a millibit customarily refers to 10-3 or 2-10 bits... because it is not customarily used at all.
The reason why a binary understanding of the decimal prefixes arose, and why binary prefixes were coined was was out of a practical need to name quantities of bits which, because of the physical structure of digital circuitry, naturally came in sizes that are powers of 2.
Although fractional bits are useful, I cannot come up with any explanation, however contrived, that would say that multiples of 1/1024 bit occur more often in information theory than multiples of 1/1000 bit. Dpbsmith (talk) 22:44, 19 June 2006 (UTC)
Raisbeck, "Information Theory," MIT Press, 1963, p. 18 refers to "225.7 bits;" on p. 19, "4.76 bits." On p. 25, we find ".72 bit per letter" and ".96 bit per second." On p. 48 he estimates that a chess master playing simultaneous blindfold chess is "taking in .2 bit per second." Dpbsmith (talk) 22:50, 19 June 2006 (UTC)

Let's stay focused here folks. As I said above, we all agree that:

  • Fractional prefixes on bits may be useful in theory
  • In practice, they are never used

So I ask again, what are we arguing? Is the issue whether or not fractional units of a bit should be mentioned in the article? Personally, I don't see any reason we couldn't include the above two points, but I think they are currently much too prominent in the intro, so I'm moving them down to somewhere more suitable. --P3d0 11:05, 20 June 2006 (UTC)

It's also about the fact that we list ridiculous prefixes for other units, that don't even make any physical sense. I like your change, though. — Omegatron 13:33, 20 June 2006 (UTC)
I'd prefer to characterize it this way:
  • Fractional bits are used and are useful in some area such as information and communication theory.
  • It is not clear whether or not SI prefixes are useful, even hypothetically.
  • In practice, they are virtually never used.
  • Even if the term "millibit" had much use, nobody has articulated any reason why one would want to refer to multiples of 1/1024 bit rather than 1/1000 bit.
We are arguing about: what if anything be said about fractional prefixes in the article.
I like P3d0's present wording. I think it's just right. Dpbsmith (talk) 14:03, 20 June 2006 (UTC)

[edit] More Logical?

The section titled "Usage Notes" on hard disk drives notes the following:

“When a stream of data is stored, it's more logical to indicate how many thousands, millions, or billions of bytes have been stored versus how many multiples of 1024, 1,048,576, or 1,073,741,824 bytes have been.”

According to whom? The basic unit on all modern systems is 512 bytes (see LBA). It's more logical to indicate how many blocks (units of 512 bytes) have been stored. If you didn't already know, 512 is half of 1024.—Kbolino 04:12, 5 July 2006 (UTC)

Paper punch cards and magnetic tape have inherent 512 byte block sizes? — Omegatron 16:13, 5 July 2006 (UTC)
I think you mean the common unit of storage in all modern hard drives is 512 bytes. Because CDROM and DVD drives have larger blocks and some tape drives have block sizes that can be set by the user to 32K, 64K, 128K, or even more. All that makes it even less logical to indicate how many blocks have been stored. But that's only assuming we're talking about the actual stored or streamed data, not the capacity. IMO, capacity is best expressed in decimal units and stored data is best expressed in binary units. Maybe the wording should make the distinction between capacity and the total size of the stored data. --JJLatWiki 21:14, 5 July 2006 (UTC)
This serves merely as a reflection of my own ignorance (and, judging by the wording of the last sentence, arrogance). I was evaluating the situation from the limited perspective of hard disk drives. However, I will postulate that it is still more logical to use units of 1024 rather than 1000 because blocks (however they are defined) are multiples of 512. (32K = 32,768 bytes = 512 × 64, for example.). I will similarly disagree with the assertion that stored data and capacity should be represented with different units, as that would be akin to representing the total distance of a journey and the distance already travelled, for example, with different units (i.e., not derived from the same basic definition—miles and kilometers, for example). And finally, punch cards are no longer in use and so should have little bearing on this argument.—Kbolino 07:47, 8 July 2006 (UTC)
The sentence in question is referring to punch cards. To paraphrase, "the decimal measurements for hard drives are used because they were first used for serially accessed storage like punch cards, and there is nothing about a continuous surface of magnetic material that lends itself to a certain block size". — Omegatron 14:08, 8 July 2006 (UTC)
Hmm. How 'bout we forget I said anything and I will work on controlling spontaneous, unjustified outbursts that result from selective reading?—Kbolino 16:54, 8 July 2006 (UTC)
I agree with you that we should represent capacity and stored data size with the same standard, with one caveat: if Microsoft will agree to start reporting drive statistics in decimal terms. As long as the operating system reports drive stats in binary terms, we are stuck with the confusing standards. It's not logical to represent a file that is exactly 1,000,000 bytes in size as 976KB or .95MB, but that is in fact what we have and it's too late to go back. It's unfortunate that consumers trust Microsoft more than the hard drive manufacturers. But unless the hard drive guys rate drive size based on the worst-case-scenario (longest file name permitted by most liberal OS with a drive filled only with files smaller than the largest cluster size option), the consumer will ALWAYS get less storage than the capacity rating. --JJLatWiki 22:57, 10 July 2006 (UTC)
Some facts to perhaps put an historical perspective on this:
  • The first disk drive the IBM 350 (1950's) had 5,000,000 100 6 bit characters organized in 100 character sectors. This predates the SI system.
  • In the 1960's virtually all disk drives used IBM's variable block length format (called, Count Key Data or "CKD"). Any block size could be specified up to the maximum track length. Blocks ("records" in IBM's terminology) of 88, 96, 880 and 960 were often used for obvious reasons. The drive capacity was usually stated in full track record blocking, for example, the 100 Megabye 3330 disk pack only achieved that capacity with a full track block size of 13,030 bytes.
  • CKD continued into the 1990's and perhaps into this day. In the 70's and 1980's most drives were offered with unformatted tracks (the unformatted capacity) with the particular block size and formatted capacity a function of the controller design. For example, the ST412 of IBM PC/XT fame had an unformatted capacity of 12.75 MB (not MiB) and with the Xebec controller and 512 byte sectors it formatted down to 10.0 MB (not MiB). Other controllers supported other block sizes resulting in other formatted capacities.
  • The advent of intelligent interfaces (SCSI and IDE) in the early 1990's took the block size decision into the drive and virtually all chose 512 bytes, for no reason other than that was what IBM had chosen when they picked the Xebec controller for the PC/XT.
So, until relative recently, it was very logical to measure in decimal numbers with decimal prefixes because God gave us 10 fingers and that's the system we learned and used until some bad GUI's took the system reported binary number and mispresented it in the mixed decimal number/binary prefix mess we are in. IMO, this history shows that there is no reason, other than sloppy programming, for the current misuse of SI prefixes and therfore there is no reason to use 512 byte blocks as a measure. As it turns out, this is a particularly bad size for current HDD technology and the industry will likely move to a larger size in the near future. While it is likely to be a binary size, this imminent change will futher make measuring in 512 blocks quite arbitrary.--Tom94022 04:34, 16 July 2006 (UTC)
That should be in the article. — Omegatron 15:54, 16 July 2006 (UTC)

[edit] Distinguish common use and scientific use

Just like nonmetric measures never disappeared despite of any standards (horsepower, calories, inches, feet, gallons, ...) so current common use of binary measures will never disappear nor get more precisely defined on a broader basis. It's another thing in scientific and engineering use - kiB, MiB, GiB, ... are easy, exact notations and hopefully their use will increase. What I'm missing in the discussion is the fact that the terms KB, MB, GB are looked at here as combinations of SI prefix (k, M, G, ...) and information unit (byte). Why not treat them as atomic terms with the common meaning they have in the field for decades? Exactly like B stands for 8 bit, KB stands for 1024 Bytes etc. - Wmk 08:09, 3 August 2006 (UTC)

[edit] Approximately-equal-to

Can someone fix it so that the approximately-equal-to symbols in this section show up better? Like, increase the font size or something? They're not recognizable over on my end (a pretty standard WinXP and Firefox). OzLawyer 17:00, 3 August 2006 (UTC)

Same in mine, but that's not our problem. That's a bug in Firefox and should be reported to them. — Omegatron 17:36, 3 August 2006 (UTC)

[edit] Wear Leveling

People have removed my [citation needed] tag from the section that states that the difference between the binary capacity and the decimal capacity is used for wear-leveling. I've never heard that this is the case, and as such would like to know where people are getting this information since it's (apparently) common knowledge.

Any flash memory invariably has extra cells to keep the yield at a manageable level, so there's really no reason that manufacturers couldn't add extra cells for wear levelling too, other than economics (indeed, some flash memories DO match their binary capacities).

Until somebody shows evidence that wear-levelling is responsible for the "lost" capacity, I'm adding the [citation needed] tag again. 129.128.213.126 20:40, 4 October 2006 (UTC)

I rephrased the entire paragraph to exclude wear-levelling as a reason for describing flash drives in binary multiples. Wear-leveling is NOT the reason or a factor in the "decision" to call them 256MB, but I didn't state that in the article. Flash drives almost universally offer the full decimal megabyte capacity of their designation, but my data is original research that isn't published anywhere, so I left the article more vague. Either way, the [citation needed] tag is much less important, I think. --JJLatWiki 00:56, 5 October 2006 (UTC)

[edit] Legal Disputes, part deux

I just added another lawsuit. And I find it more interesting in that now, 3 of the 4 cases were brought by the same legal team of Gutride and Safier. That's 2 for 3 for Gutride Safier. Half million dollars + $2.4 million. Not a bad deal just for screwing over a few million consumers who will never even know Gutride Safier screwed them. That's just my personal opinion. All of it is personal opinion. Every word was opinion. I was claiming no facts. Please don't sue me, the company I work for, my internet service provider, or Wikipedia for my personal opinion. --JJLatWiki 19:20, 6 November 2006 (UTC)

Why aren't they suing software manufacturers for misrepresenting the size of files? :-) Or maybe the hard drive manufacturers should sue Microsoft for getting them into a lawsuit by reporting the sizes of their drives wrong. — Omegatron 21:00, 6 November 2006 (UTC)
ZACKLY! Microsoft needlessly sucks up more of my drive space than anything else and they're the one giving consumers the wrong capacity. But I would really like to file a class action lawsuit against the the lawyers and the plaintiffs who brought these suits. Because of them, the cost of all hard drives, MP3 players, and flash drives is higher than it would have been. Other than the lawyers getting millions of dollars, and the plaintiffs getting their $1000, no one was helped by any of these lawsuits. (all entries made by me are my own personal opinion, I am not stating any facts or making assertions of any kind) --JJLatWiki 22:17, 6 November 2006 (UTC)
Retrieved from "http://en.wikipedia.org../../../b/i/n/Talk%7EBinary_prefix_c9d9.html"