"4,7 GB = 120 min." - why?

[clokkevi]

On the cover of single-layer DVD-R's / DVD+R's it says "4,7 GB / 120 min"
- why is that?

I know on CD-R's it says "700 MB / 80 min" because that CD-R
has 360,000 sectors, and..:

Data capacity:
360,000 sectors * 2048 bytes/sector data track = 737,280,000 bytes data
(737,280,000 bytes) / (1024 bytes/KB) = 720,000 KB
(720,000 KB) / (1,024 KB/MB) = ca. 703 MB
So it can store ca. 703 MB data.

Audio capacity:
360,000 sectors * 2,352 bytes/sector audio track = 846,720,000 bytes audio
44,100 samples/second * 2 bytes/sample * 2 (stereo) = 176,400 bytes/second
(846,720,000 bytes) / (176,400 bytes/second) = 4,800 seconds
(4,800 seconds) / (60 seconds/minute) = 80 minutes
So it can store exactly 80 minutes of audio.

But when it comes to the amount of sectors on DVD-R's / DVD+R's...
According to OSTA there are some different numbers..
http://www.osta.org/technology/dvdqa/dvdqa6.htm

DVD+R:
DVD+R Specification Version 1.2 - 2,295,104 sectors (4,700,372,992 bytes)

DVD-R:
DVD-R (General) Specification Version 2.0 - (2,294,922 sectors) 4,700,000,000 bytes
Although this is the minimums,
most manufacturers are using 2,298,496 sectors (4,707,319,808 bytes)


The reason for the "4,7 GB" stamp is because they use decimal 10-base notation
- 1 GB = 1,000 KB = 1,000,000 bytes
instead of the usual binary 2-base notation (like Windows use)
- 1 GB = 1,024 KB = 1,048,576 bytes



So.. how does those 2,295,104 / 2,294,922 / 2,298,496 sectors
gets translated to "120 minutes"...?

120 minutes * 60 seconds/minute = 7,200 seconds

(2,295,104 sectors) / (7,200 seconds) = 318,764444444..... sectors/second
(2,294,922 sectors) / (7,200 seconds) = 318,739166666..... sectors/second
(2,298,496 sectors) / (7,200 seconds) = 319,235555555..... sectors/second

Clearly, any rigid "sectors/second" specification
- like the "75 sectors/second" from CD's
- is not used for DVD-R's or DVD+R's.

Some manufaturers, like BenQ, thinks that 120 minutes of digital audio
will take as much as 4,7 GB: :rolleyes:
http://www.benq.com.au/HomeProductList.asp?bc=30
"Offering the storage capacity of 4.7GB of data or 120 minutes of digital audio, BenQ's DVDR media enables you..."
(I don't think I'll ever by any media from BenQ..!)


Let's use the 2,298,496 sectors (4,707,319,808 bytes) DVD size.

4,707,319,808 bytes * 8 bits/byte = 37,658,558,464 bits
37,658,558,464 bits / 7,200 seconds = ca. 5,230,355 bits/second
(ca. 5,230,355 bits/second) / (1,024 bits/Kbits) = ca. 5,108 Kbits/second.

As far as I know, the DVD bitrate is variable.
How can they set "120 min" on the cover?
Could it be some old typing error, that it should say "120 mm" (the diameter of the DVD) instead..? :confused:

[Inventive Software]

The reason for the "4,7 GB" stamp is because they use decimal 10-base notation
- 1 GB = 1,000 KB = 1,000,000 bytes
instead of the usual binary 2-base notation (like Windows use)
- 1 GB = 1,024 KB = 1,048,576 bytes


Unfortunately manufacturers are being this way more and more, and it's very misleading. Hard disk manufacturers do it as well.

4.7 GB is a nicer number than the exact number of bytes or GB that are on the disc. CDs you'll find have maybe 79 mins 57 seconds, which is as near as dammit 80 mins. 703 MB is also as near as dammit 700 MB. The reason is it's very difficult to engineer a CD to exactly the "White Paper" specs. Doing so would also drive the cost up a lot. Making something to within "reasonable limits" is more cost-efficient and also allows for discrepencies within the manufacture process.

120 mins for the DVD recording is the average time you'll get. That particular notation is the standard time you'll get with a DVD recorder (standalone). With different recording modes you can extend the recording time (LP is 240 mins, EP is 360 mins).

In short, it's just a standard that's been adopted, and stuck with out of convenience. Live with it.

[fccHandler]

Originally posted by Inventive Software
Unfortunately manufacturers are being this way more and more, and it's very misleading. Hard disk manufacturers do it as well.
Surprisingly, I recently found out that it isn't their fault. They are actually doing it right, and everyone else (especially Microsoft) has got it wrong.

According to the IEC 60027-2 (http://physics.nist.gov/cuu/Units/binary.html) standard, a gigabyte really is 1,000,000,000 bytes! A gibibyte is 1,073,741,824 bytes.

[Soulhunter]

Seems they gonna use this 120 min. scheme also for future media... :D

http://img274.echo.cx/img274/3733/tdk25da.jpg


Bye

[Mug Funky]

MPEG-2 is VBR, so running time is utterly meaningless.

however, if you assume a constant (or average) bitrate of about 5000 kbps, then you'll get about that long on a DVD-5.

it is all marketing after all, though. i would never put that much video on a DVD-5 - you just can't guarantee it'll look good at that bitrate.

of course, you could fit heaps of video on a DVD-5 if it were in h.264...

[mpucoder]

The time values are for DVR, this is how much time you can record at SP. It's analogous to a T120 VHS tape. (Maybe that's why Beta failed as a consumer medium, people could not equate L750 to 3 hours)

[communist]

Originally posted by fccHandler
Surprisingly, I recently found out that it isn't their fault. They are actually doing it right, and everyone else (especially Microsoft) has got it wrong.

According to the IEC 60027-2 (http://physics.nist.gov/cuu/Units/binary.html) standard, a gigabyte really is 1,000,000,000 bytes! A gibibyte is 1,073,741,824 bytes.
Exactly!
Its always fun to hear people moaning about being cheated about a few 'GB' when they format their brand new hdds ;)

[dragongodz]

well i was not going to bother saying anything but what the hell, may aswell play a little devils advocacy. as such these comments are just meant to show you the other side of the story and should not be taken personally. :)

Surprisingly, I recently found out that it isn't their fault. They are actually doing it right, and everyone else (especially Microsoft) has got it wrong.

According to the IEC 60027-2 standard, a gigabyte really is 1,000,000,000 bytes! A gibibyte is 1,073,741,824 bytes.
that was only introduced in December 1998 to try and help ease the confusion, it failed. computers have been around a LOT LONGER than that and used the terms kilobyte etc first. now because digital A/V programs etc decided to use base 10 and still use the same terms anything else to do with computers should change ? why ? they were first and have a huge amount of both hardware and software out there that simply can not be changed. a better question is why didnt A/V use different terms to start with. so who is really to blame ?

The time values are for DVR, this is how much time you can record at SP.
exactly. which is why dvd media give both size and time, so it is saying it is A/V related and so can use the base 10 to tell you that size. you may say that dvd media is not only used for A/V and you would be right of course. that however doesnt change what the size and time are meant to signify.

Its always fun to hear people moaning about being cheated about a few 'GB' when they format their brand new hdds
also consider that they are giving you what is meant to be the raw size and not formatted. why ? well which formatted size should they list exactly ? FAT, NTFS, EXT, HTFS, OTHER ? they do not produce the exact same size. maybe people would like them to list the size for every possible formatting combination but i doubt they think that should be needed. as for saying HD manufacturers are using base 10 so not at fault, wrong. they have used raw size for a very long time(personal experience with this from the early 80's), well before that IEC doc came out.

as i said at the start, this is just to encourage people to think and see both sides so please dont try and flame me, i wouldnt take that kindly. :devil:

[fccHandler]

Originally posted by dragongodz
that was only introduced in December 1998 to try and help ease the confusion, it failed. computers have been around a LOT LONGER than that and used the terms kilobyte etc first.
True enough. And I'm just as guilty of forever assuming 1KB = 1024 bytes, but I'll play the opposing side just for the sake of argument. ;)

To ease confusion, there has to be some rule that everyone will agree to. The IEC attempted to set a standard in 1998, so you would think Microsoft had enough time to incorporate the new math into Windows XP. Yet they didn't.

The IEC standard is clear and concise (though I confess "gibibyte" doesn't roll off the tongue as fluidly as "gigabyte," LOL). So if/when Microsoft finally adopts it (and perhaps only then), I think it's pretty much guaranteed to become global law. Until that happens, or until a more agreeable standard comes along, Windows users remain mired in confusion.

I have wondered if Microsoft's failure to implement the IEC standard is due to backwards compatibility issues. But I find it difficult to rationalize, considering that many of my old DOS games and programs don't work correctly under XP. As a user, it seems that I gave up "backwards compatibility" when I upgraded from Win98 in the first place...

[movmasty]

Originally posted by clokkevi
"4,7 GB = 120 min." - why?
Well, really at the begin were "135 minutes",

but now also a movie of 90 min. is double layered, usually they fit about 70 min. in one layer...

Anyone up to compare DVD and vhs tapes?

How many vhs240 is a DVD?

[Mug Funky]

interesting... i wonder how much footage in half-d1 could be squeezed on a DVD before the quality falls below that of VHS...

i'm sure you could fit around 8 hours quite comfortably on a DVD-5 if you're only aiming for VHS quality.

[dragongodz]

To ease confusion, there has to be some rule that everyone will agree to. The IEC attempted to set a standard in 1998, so you would think Microsoft had enough time to incorporate the new math into Windows XP. Yet they didn't.
that first statement highlights the problem, how many in the computer related businesses agreed to it ? this includes not only software but hardware.
for example how many ram manufacturers are sticking kibi on their stickers ? none i have seen.
is Mac using kibi (as in on screen, reporting hd and/or ram size etc)? i dont know.
are Linux distros using kibi (again screen)? i know it was added to the kernel several years ago and there was complaints then about it causing confusion. couldnt find anything more about it though.
how about third party software manufacturers(for any platform) ? have they started using kibi in their displays ? cant say i have seen this happening either.

so this is not just a MS thing but a much wider spread non-acceptance.

[movmasty]

Originally posted by Mug Funky
interesting... i wonder how much footage in half-d1 could be squeezed on a DVD before the quality falls below that of VHS...

i'm sure you could fit around 8 hours quite comfortably on a DVD-5 if you're only aiming for VHS quality.
so dvd5=2x vhs240? one $ for 6 $...

Half d1 you mean 352x576/480?
8 hours on a dvd is 159kb/s, this is the mpg1 bitrate, 352x288.

but using lp a svhs can play 8 hours on a tape with better quality than mpeg1

but still using xvid we can put 10 hours of movies on a dvd with better quality respect to the svhs....

[Joe Fenton]

I'm with dragongodz on this one. There isn't a computer user alive (much less a programmer or hardware engineer) that would be CAUGHT DEAD using those crappy IEC terms. That was the stupidest thing ever conceived, and you can bet the IEC didn't consult a single computer expert while making this 'standard'. It's just jealous mathmeticians trying to embarass software programmers making ten times their salary.

I won't ever use those units in anything I write, and neither will anyone I know. They're just plain stupid. Who did they consult when making up those units? The director's three year-old? I have an idea - the base-10 camp can use the following units:

10e3 - kissy
10e6 - missy
10e9 - gissy
10e12 - tissy

Have fun getting magazines to print THOSE units! :p :rolleyes: :cool: :D

[fccHandler]

Well, how about:

Killer-byte
Miller-byte
Godziller-byte
Thriller-byte

In all honesty, I agree the IEC terms sound like baby talk. Big hunky macho programmers don't want to talk like that. I didn't exactly say it in my previous post, but I kinda alluded to a similar misgiving on my part. ;)

But is that the only reason people won't adopt them? Because if it is, don't you think it's kind of a silly reason? I'm sure we'll get over it once everybody is using the terms (if that ever happens).

[Joe Fenton]

You have to give people a compelling reason to change or they won't. So, name one reason to change - the IEC is TELLING you to. Is it compelling? Not in the slightest.

Now name a reason to NOT change - tradition, confusion, the IEC picked stupid sounding terms, and finally, all current computer science materials (books, etc.) would have to be rewritten, republished, and repurchased. Is it compelling? You bet your posterior.

Hell, they couldn't get the US to go metric, and metric terms sound pretty cool. What in God's green Earth makes them think they could push that 'standard' when it makes you sound like an idiot?

[Shinigami-Sama]

Originally posted by Joe Fenton
You have to give people a compelling reason to change or they won't. So, name one reason to change - the IEC is TELLING you to. Is it compelling? Not in the slightest.

Now name a reason to NOT change - tradition, confusion, the IEC picked stupid sounding terms, and finally, all current computer science materials (books, etc.) would have to be rewritten, republished, and repurchased. Is it compelling? You bet your posterior.

Hell, they couldn't get the US to go metric, and metric terms sound pretty cool. What in God's green Earth makes them think they could push that 'standard' when it makes you sound like an idiot?
it's alway annoying have to tell metric <among other things:rolleyes: > ignorant americans what a klick, meter, kilo, mill ect. is
but thats another topic
I agree though, atleast make the stuff sound smart, not lieka 3y/o's baly-rish

[fccHandler]

@Joe Fenton:

The IEC isn't "pushing" it on anyone. They are publishing an international standard (that's their job) which the industry will either accept or reject. IMHO "the terms sound stupid" is not a compelling reason to reject it, but some of your other points are certainly valid.

If all computer science materials will have to be rewritten, well, it's only because they were wrong to begin with. For example, the prefix "Kilo (http://en.wikipedia.org/wiki/Kilo)" (borrowed from the Metric system) is universally understood to mean 1000 units, and it's been that way for 200 years. What has screwed us up now is the fact that the computer industry adopted the same prefix, but redefined it as 1024 units!

The IEC's standard will eliminate this confusion once and for all, and to me that is an overwhelmingly compelling reason to adopt it.

But, like your example of the U.S. going metric, it just ain't gonna stick unless everyone agrees, especially the big players. I think Microsoft could set a tremendous example by adopting the standard in their next OS.

Will we sound like infants when we talk? Sure! But at least we'll agree on what we're talking about.

Remember the term "googol" (from which "Google" gets its name) was invented by a 10-year old child. Do people of today feel stupid saying the word "Google"? Absolutely not!

[Shinigami-Sama]

Originally posted by fccHandler
@Joe Fenton:

The IEC isn't "pushing" it on anyone. They are publishing an international standard (that's their job) which the industry will either accept or reject. IMHO "the terms sound stupid" is not a compelling reason to reject it, but some of your other points are certainly valid.

If all computer science materials will have to be rewritten, well, it's only because they were wrong to begin with. For example, the prefix "Kilo (http://en.wikipedia.org/wiki/Kilo)" (borrowed from the Metric system) is universally understood to mean 1000 units, and it's been that way for 200 years. What has screwed us up now is the fact that the computer industry adopted the same prefix, but redefined it as 1024 units!

The IEC's standard will eliminate this confusion once and for all, and to me that is an overwhelmingly compelling reason to adopt it.

But, like your example of the U.S. going metric, it just ain't gonna stick unless everyone agrees, especially the big players. I think Microsoft could set a tremendous example by adopting the standard in their next OS.

Will we sound like infants when we talk? Sure! But at least we'll agree on what we're talking about.

Remember the term "googol" (from which "Google" gets its name) was invented by a 10-year old child. Do people of today feel stupid saying the word "Google"? Absolutely not!
I"ll agree withteh kilo thing but google not sounding stupid..

anyways, yeah getting this stupid 10/2-base crap sorted out would be nice, much less math in hte long run converting this crap to that crap

[fccHandler]

Originally posted by Shinigami-Sama
getting this stupid 10/2-base crap sorted out would be nice
My point exactly. :)

[dragongodz]

For example, the prefix "Kilo" (borrowed from the Metric system) is universally understood to mean 1000 units, and it's been that way for 200 years. What has screwed us up now is the fact that the computer industry adopted the same prefix, but redefined it as 1024 units!
the thing is people already know that. it is not computers usage that has brought the confusion yet it is being used as the scapegoat.

The IEC's standard will eliminate this confusion once and for all, and to me that is an overwhelmingly compelling reason to adopt it.
this is the problem as i see it. the IEC frelled up and got it around the wrong way. they basically said ok who has the most to change, the computer which has been using this in the consumer marketplace for over 25 years or the A/V market which is a lot more limited market wise and has not been in consumer use for anywhere near as long(remember this is back in 1998) ? the computer has MUCH more to change ? ok lets say they are the ones that need to do it then. sorry but thats not thinking at all. as i said they got it the wrong way around.

[Inventive Software]

Thing is, it makes sense in the end.

Everybody knows 8 bits make a byte, right?

Everybody knows that 1000 bits is exactly 125 bytes, right?

So therefore we can assume that 8000 bits is 1000 bytes, right?

And according to good ol' metric systemation, 1000 bytes would be a Kilobyte, right?

Nope, 'cos since the dawn of computers, they have only worked in 2 base notation. Therefore they see 1 bit as 1 or 0. 2 bits as 00, 01, 10 or 11. (Can you see where this is going?)

Remember from your physics lessons boys and girls, that 2 to the power of n gives the number of possibilities. n is the number of "bits" you have.

2 to the power of 10 gives 1024 possibilities. Hey presto, there's where the crappy 1024 is a kilobyte comes in.

The only way around this is to re-program computer hardware to work in 10-base (decimal to you and me). Believe you me, that's more difficult than it sounds. Binary (2-base) is much simpler for computers to work with, and it's a lot faster than working in any base higher than 2. Work in say base 10 and it's a lot more calculations just to get to the number you are looking for.

We just have to live with it, that's all.

[fccHandler]

Originally posted by dragongodz
this is the problem as i see it. the IEC frelled up and got it around the wrong way. they basically said ok who has the most to change, the computer which has been using this in the consumer marketplace for over 25 years or the A/V market which is a lot more limited market wise and has not been in consumer use for anywhere near as long(remember this is back in 1998) ? the computer has MUCH more to change ? ok lets say they are the ones that need to do it then. sorry but thats not thinking at all. as i said they got it the wrong way around.
I just don't see it that way. Some prefixes are more recent, but for sure the prefix "kilo" has been in use for at least 200 years, and its meaning (until ~25 years ago) was well understood around the world. I remember being taught about the Metric system in elementary school (long before I got my first VIC-20), and there was never any doubt that "kilo" was 1000.

Since computers group bits in powers of 2, it's inappropriate to use "kilo" in computer speak, when you really mean 1024. The Metric prefixes "mega" and "giga" are just as inappropriate. This misuse of the base-10 Metric prefixes is what has caused all the confusion in the first place. It was a Big Historical Mistake, and somebody has got to fix it. We NEEDED new prefixes for binary stuff, so the IEC proposed some. Now it's just up to the big boys to start using them.

BTW, I don't agree that the computer industry has MUCH more to change. Computers change so quickly anyway that people are confused enough just trying to keep up with all the new advances and technologies. The "kibi" thing is minor IMHO.

Fixing Window's source code is simple. CTRL+F: "KB," replace with "KiB." Recompile. :D

It will be less painful for us to adopt the IEC terms for binary counting than to rewrite the base-10 Metric system!

Customer: I asked you for a kilogram.
Salesman: Yep, and I gave you 1000 grams.
Customer: No, you jerk, it should be 1024 grams!
Salesman: (mumbling) Damn computer geek...
**fisticuffs**

[Shinigami-Sama]

Originally posted by fccHandler


Fixing Window's source code is simple. CTRL+F: "KB," replace with "KiB." Recompile. :D

It will be less painful for us to adopt the IEC terms for binary counting than to rewrite the base-10 Metric system!

Customer: I asked you for a kilogram.
Salesman: Yep, and I gave you 1000 grams.
Customer: No, you jerk, it should be 1024 grams!
Salesman: (mumbling) Damn computer geek...
**fisticuffs**
you can fix windows!?! :eek:

[mpucoder]

I agree with fcc - to a point. But first an historical correction. You have to go back much more than 25 years for the first use of kilo in computers. Back to the 50's at least, maybe further. My own personal history goes back only to mid 60's, and the term "kilobyte" was well established. Originally misused by a very small number of computer experts, who understood it was wrong from the start, and had more important things to do than invent new names (the words "bit" and "byte" having exhausted their imaginations). And as long as this small elite group knew it was wrong, there was no need to invent a new set of prefixes. Then, around 28 years ago, thanks to Apple and then Radio Shack, ordinary people could buy a computer. That was the point in time when things should have been renamed, before the masses started using the incorrectly used terms. Now it seems to be almost too late to change. Of course Windows can simply use the new abbreviation, rather than change the calculation, but will that end the confusion, or worsen it?

As for how important it is to change - certainly more important than other changes we've gone through, like cps to Hz, or Angstrom's to whatever they're called now (deci-nanometers?)

btw, the same forward thinking people who thought the misuse of metric prefixes would never be widespread, also thought that no one would need more than 65K - or, apparently, that there was plenty of time to fix the 2 digit year notation.

[dragongodz]

for sure the prefix "kilo" has been in use for at least 200 years, and its meaning (until ~25 years ago) was well understood around the world.
and its meaning for computers is understood now aswell. if you are saying it should have not been used for computers to start with i agree. however it was and nothing was done until 1998, by which point it was too late.

BTW, I don't agree that the computer industry has MUCH more to change.
so all the hardware, all the software, all the related material(including books etc) for computers in general is less than the same things for A/V both now and back in 1998 ? you have to be kidding.

It will be less painful for us to adopt the IEC terms for binary counting than to rewrite the base-10 Metric system!
nobody is saying a rewrite of the metric system. we are saying the use of the same terms for A/V should not have been used and would be the simpler to change now.

[Soulhunter]

Originally posted by Shinigami-Sama

you can fix windows!?! :eek:

Yeah, "format c:\" iirc... :D


Bye

[fccHandler]

Originally posted by dragongodz
and its meaning for computers is understood now aswell.
If that were true, we wouldn't be having this discussion. IIRC it started because there is confusion (even within the computer industry) about what the prefixes mean. I see it when Windows tells me my 4.7 GB DVD will only hold 4.37 GB, or when I buy a new hard drive and Windows reports much less than its advertised capacity.

so all the hardware, all the software, all the related material(including books etc) for computers in general is less than the same things for A/V both now and back in 1998 ? you have to be kidding.
Actually, mpucoder's post highlights something that I didn't fully realize myself... The IEC standard is purely a semantic change. Little (if any) actual code will have to be changed, and I don't see that it will affect hardware devices at all. All the math stays exactly the same, only what we call it changes. And I don't think it's strictly necessary to rewrite existing books. We'll just understand that they were written during the "dark age" of computer history. :)

nobody is saying a rewrite of the metric system. we are saying the use of the same terms for A/V should not have been used and would be the simpler to change now.
You keep referring to an effect on "the A/V market," but I'm sorry I'm not clear on what you mean by that. Can you elaborate?

My view is like this: We have two established camps using the same terminology, which are the Metric system and the computer industry. It is confusing because both use the same prefixes with different meanings. IMHO to end the confusion we should have "kilo" mean exactly one thing, globally. If we make it mean 1000, then we need a new prefix for the base-2 computer industry. If we make it mean 1024, then we have to rewrite the base-10 Metric system. Now, which of these options is less painful?

[Hiro2k]

I'm going to agree with FCC in that we should unify but kibi just sound stupid. If we really wanted to fix the problem we should all agree that when using computer terms a kilo means 1024 and not 1000. That would solve the problems we are seeing and nothing would change. Normal people would now have drive sizes reported correctly without being none the wiser and computer geeks wouldn't have to change the way they speak.

[fccHandler]

Originally posted by Hiro2k
kibi just sound stupid.
Well, see my previous post about this thing called "Google". It only sounds like a stupid word until it becomes a universally accepted nomenclature. Once all your friends are using it, you'll probably feel silly for resisting. I'm fairly confident that people will get over it.

If we really wanted to fix the problem we should all agree that when using computer terms a kilo means 1024 and not 1000. That would solve the problems we are seeing and nothing would change.
That is a possible solution, but I don't think it's better than simply accepting the IEC's proposal. For one thing, manufacturers of blank discs and hard drives will have to change their units. At this point I think it's a rude thing to ask of them, because AFAIK they are the only ones currently doing it right!

Plus there would still be the risk of conflict between us enlightened computer geeks, and those merchants who only deal in the traditional Metric system. (See my other previous post.)

[dragongodz]

You keep referring to an effect on "the A/V market," but I'm sorry I'm not clear on what you mean by that. Can you elaborate?
A/V = Audio and Video.
look exactly what the title of this thread is asking about and the reason given. the size is related to the amount of video. also note that most codecs and encoders etc use base10 while the rest of computer software uses base2.

The IEC standard is purely a semantic change. Little (if any) actual code will have to be changed, and I don't see that it will affect hardware devices at all. All the math stays exactly the same, only what we call it changes.
all markings on hardware whether printing or engraving or labelling would need to be changed. then you would either have to label all old boxes with the new size or have some sort of chart explaining the changes etc etc etc. otherwise people will go to buy something and not know what is what. so a lot of money and resources would need to go in to these "little" changes.

If that were true, we wouldn't be having this discussion. IIRC it started because there is confusion (even within the computer industry) about what the prefixes mean. I see it when Windows tells me my 4.7 GB DVD will only hold 4.37 GB, or when I buy a new hard drive and Windows reports much less than its advertised capacity.
sorry but look at what people are confused about, dvd size , for which the reason has already been given and is only a recent thing with the expansion of dvdr sales. i dont see too many people asking or being confused by ram size etc.
as for hard drives, they were never marked the same as formatted. i know that from personal experience all the way back to the early 80's so it took until 1998 for there to be confusion ? no it didnt, infact its rarely ever asked about except by the occasional newbie.
i already said i agreed it should all have been sorted at the start or atleast when computer popularity started to gain. again this was not in 1998 but back in the 80's.

If we really wanted to fix the problem we should all agree that when using computer terms a kilo means 1024 and not 1000. That would solve the problems we are seeing and nothing would change. Normal people would now have drive sizes reported correctly without being none the wiser and computer geeks wouldn't have to change the way they speak.
and that is the way it currently is for the majority of computer related things. the exceptions being hard drives and A/V related products(software and hardware/media). those are what should be told to change with either correct sizes(HD) and/or clear labelling that size given is A/V related so base10 size. much sinpler than every single thing to do with computers being changed.

since i have had to start repeating myself i wont bother posting again in this thread.

[Joe Fenton]

Hard drive and CD/DVD manufacturers don't do it because they know it's right, they do it because they know it will fool people who ALL understand that the computer units are a different base and give a smaller value. They artificially increased the PERCEIVED size of their products to sell more to the unwary.

Also, the term Google was chosen from the phrase "google eyes" and has NOTHING to do with the math term googolplex (which I might add not a single reputable mathematician has used since it was coined in 1938). Picking a silly name for a BUSINESS is nothing at all like picking silly terms for an ENTIRE BRANCH OF TECHNOLOGY.

Additionally, the terms have always been well understood by people in that branch of technology. I have EXACTLY one GigaByte of memory in my computer, not 1.073741824 GigaBytes. That is COMPLETELY assinine. Base ten units DON'T WORK on computers! Computer scientists and engineers are not going to switch to a system that DOESN'T represent the physical process any more than chemists are going to switch to teaspoons and pecks.

[int 21h]

Death to the metric oppressors!

When I talk to people that aren't so computer literate, rarely does this problem come up, and aside from the odd brawl or two that happens down at the coffee shop between the graduate students studying math and the graduate students studying computer science (yes, there are actually two or three left!), I really don't see the issue here.

If we're looking for a scapegoat, its the marketing executive that got the bright idea to advertise data storage products in base 10 instead of the traditional base 2 ("Hey a bigger number has got to be better.. right!?").

[fccHandler]

Originally posted by Joe Fenton
Also, the term Google was chosen from the phrase "google eyes" and has NOTHING to do with the math term googolplex
Well, wikipedia disagrees with you:
http://en.wikipedia.org/wiki/Google

Base ten units DON'T WORK on computers!
DUH! So why is the computer industry still using base ten Metric prefixes?

[stephanV]

The funny thing is that 'KB' is even wrong if you mean 'kilobyte'. In SI the symbol for the prefix 'kilo' is 'k' and not 'K' (so NOT capitalized). 'K' is used for the unit of temperature: Kelvin.

[Joe Fenton]

Originally posted by fccHandler
Well, wikipedia disagrees with you:
http://en.wikipedia.org/wiki/Google


Okay, I concede that one.

DUH! So why is the computer industry still using base ten Metric prefixes?

Because they give the proper magnitude. No one was violating any standards back when this was decided as there WERE no standards. It was common usage, but that isn't the same thing. The metric system was commonly used for a many years before scientists finally decided to make it the standard. Similarly, the common usage of these terms for computers persisted for many years before someone else decided on their own to make something NOT in common usage the standard. Such a standard will not be accepted.

[fccHandler]

Well, if you truly don't believe that the adoption of the Metric prefixes by the early computer experts was a mistake that needs to be corrected, then let's just agree to disagree.

Anyway, I'm weary of arguing this. Whether the software industry will adopt the IEC standard is certainly questionable. They've had seven years to do so, but nothing much has changed. Oh well.

TTFN

[Joe Fenton]

Yeah, the argument is getting old, but look at it one way - you actually won one of the arguments. I agree that you were completely right about Google. :D

[Mug Funky]

hmm... it could be said there are already alternate terms for kilobyte and megabyte that aren't metric based:

K ("kay")
"meg"
"gig"

these terms could be formalised pretty easily, and sod-all would need to change. they're already in common usage and make sense. it just means that DVDs would have to be re-labelled to say "4.38 gig" instead of "4.7 GB" (which leads to many oversized encodes).

for sure, memory works in increments of powers of two. my machine has 512 (2^9) megs of RAM... before that it had 384 (2^8 + 2^7) then 256 (2^8). also there's the numbering matter:

I have EXACTLY one GigaByte of memory in my computer, not 1.073741824 GigaBytes.

that's a damn good point - it's really really easy to remember powers of two, but not so easy to remember those same numbers * 1.024.