Japanese government finally bids sayonara to the 3.5" floppy disk

Japan is saying sayonara to the floppy disk, which until now was a required medium for submitting some 1,900 official documents to the government. The announcement (Japanese, machine translated) last week from the Ministry of Economy, Trade and Industry brings decades of physical media submission requirements in Japan to an …

[elDog]

But I have a box of hollerith cards with my tax statements on them --- woe is me!

I was planning to get them fed into a card-reader and then transferred to 800BPI tape for subsequent upload to a rotating IBM 360 drum and then finally downloaded to a PC-XT which accepted my 5 1/4 disks.

Guess those tax authorities won't get to see my 1970's returns after all.

(I've had to use all of those technologies and many more ancient and novel. Still the same game: ETL.)

[the spectacularly refined chap]

Re: But I have a box of hollerith cards with my tax statements on them --- woe is me!

It's less than 20 years since what was then the Inland Revenue stopped accepting data on open reel tape. It's probably only slightly over 20 years since they stopped supplying it in that format.

[J.G.Harston]

Re: But I have a box of hollerith cards with my tax statements on them --- woe is me!

I remember in the early 2000s having to collect the electoral register on tape and send it off for processing to be returned on a CD-ROM. ;) I oversaw the project where everything was updated to Express (I think). In the process about 20,000 "ghosts" were cleaned from the system as well.

[TeeCee]

Re: But I have a box of hollerith cards with my tax statements on them --- woe is me!

Sounds like an interesting exorcise.

[Anonymous Coward]

Less "connected" means less likely to be hacked and randsomed.

Especially for tech illiterate civil servants and banks, that's not a bad thing.

[Kevin McMurtrie]

Re: Less "connected" means less likely to be hacked and randsomed.

I'm not sure about that. Floppy disk hacks have been around longer than the Internet. Those device drivers were written long before there were spare resources for luxuries like bounds checking.

[doublelayer]

Re: Less "connected" means less likely to be hacked and randsomed.

What makes you think that the system that took in data on floppy disks had no network connection? Lots of systems had networking and floppy drives. Fewer systems had networking, floppy drives, and an application that was written with security in mind. I'd be more worried about how old the software that was used to process the floppy-provided files was, because if they didn't update the hardware requirements, they may not have changed the software. Keeping in mind that the software was probably written in a time when, even if you did use encryption, it was something that can probably be cracked in seconds nowadays, I don't think my concerns are groundless.

[Anonymous Coward]

Ha! Ha! I'm typing this on a PC which still has a 3.5" floppy drive...

...and the worst bit is, that's not a joke.

Admittedly it's a pretty old computer- it'll be fifteen this year- and even then the floppy drive had already been mostly replaced by pen drives and the Internet.

But the place where I worked still stocked them in 2009, so there must still have been some demand. (*)

I included a floppy when I built that PC because... er, not really sure now. I'm generally quite conservative when it comes to things like that and since they were so dirt cheap by that point that it was probably no big deal to have the option of booting from the few things that didn't work with a pen drive back then, or reading old discs. Or something.

Honestly though, aside from testing it out and possibly booting that old "tomsrtbt" (Linux-on-a-single-1.44MB-floppy distro) a couple of times, I'm not sure I ever used that drive.

But yeah, I'm typing this on a fifteen-year-old PC that still includes a floppy drive, FFS.

(*) And they even made them in black to match modern computers, else I'd have just reused the one from my old beige late-90s PC.

[Roger Lipscombe]

Re: Ha! Ha! I'm typing this on a PC which still has a 3.5" floppy drive...

I've got a USB 3.5" floppy disk drive in the drawer right next to me. I have *no* idea why I purchased it, nor whether I've ever used it.

[katrinab]

Re: Ha! Ha! I'm typing this on a PC which still has a 3.5" floppy drive...

I have a USB floppy drive somewhere in my Big Box of Stuff. I have used it in the past, though not recently.

[Anonymous Coward]

Re: Ha! Ha! I'm typing this on a PC which still has a 3.5" floppy drive...

I keep a "new, old" early 2000's PC explicitly for handling old formats. Besides entertainment value filling in for stuff that doesn't emulate very well; said machine has an Adaptec 39320 SCSI board in, and a 2.88mb floppy. The operational use for the system is mostly in data recovery from esoteric old crap. In my day job, I still run code designed for an IBM 360, and literally every mark of "industrial" computer sold since the late 1970's in Western Europe. I say industrial in quotes; there are systems in wide circulation running DOS and Windows 3.1 in environments that they really should not.

Keeping working hardware around to talk to the old shit you inevitably find in the field has it's, admittedly niche, purposes.

Saner minds might say why not replace the old junk rather than try to keep it ticking over? Modern systems in similar roles are much, much more expensive and short lived. Like certain software vendors; specialised hardware vendors have clocked that corporations with NEED will pay literally anything. Perfectly serviceable hardware goes in the bin because the software functionality dies. "Please buy a new $250,000 instrument". Whereas the 1980's crud sitting on ancient and barely-qualifying as an OS have no such time limits. Keeping the business end of the instrument functioning is more or less the same overhead whether old or new.

We moan about Windows 11 "enforced" upgrades to hardware, but it's a whole lot worse when you're talking about massive and expensive instrumentation systems where the computer component is at most 1% of the end-to-end cost of given system. If this was a sector where cost was not a significant issue (military, Oil & Gas production) they would probably just suck up the bill. But a bunch of other outfits have to control costs in order to turn a profit.

A famous, and recent example are the various 747's that have floppy drives in their hardware stack. The certification needed to put the drive there is vastly more expensive than swapping out the hardware to something a tad more sustainable.

The Japanese resistance to change is in some respects admirable. Consider the rebellion against digital distribution in the form of sales of Vinyl or Tape. I get it. But there is a time and place!

[John Riddoch]

Re: Ha! Ha! I'm typing this on a PC which still has a 3.5" floppy drive...

I stopped having a floppy drive mostly by accident. My PC at the time had a bad habit of taking ages to start up Explorer and I tracked it down to the floppy drive, I think because I hadn't yet reconnected it after some work in the guts and noticed it loaded quicker, so I left it unplugged, but still in the case for when I needed it next. A year later, I realised I hadn't needed it nor missed it and my next PC didn't get a floppy drive installed. If I hadn't had that issue and left it disconnected, I'm sure I'd have probably gotten one installed Just In Case (like you did), but I haven't missed having one over the years.

I still have the drive in a drawer somewhere, although I'm not sure I have a PC I can install it to; I think the floppy drive ports on motherboards were rendered obsolete some time ago.

[Howard Sway]

Persky relies on discoveries of mass quantities of floppy disks by individuals and businesses

Ask the Japanese tax office then - they presumably have millions of the things stored somewhere. Just need to delete the personal data on each one first. Insert disk, click, whirr, buzz, buzz, buzz, click, click, eject disk..........

[theOtherJT]

Re: Persky relies on discoveries of mass quantities of floppy disks by individuals and businesses

I wonder if I still have that degaussing cabinet in the garage...

[43300]

How long before the media starts to deteriorate to a point of being unreliable? Most of these warehouse finds won't have been kept in temperature-controlled environtments.

[AVR]

Humidity's more a problem IIRC. If they were in sealed packages they'd hopefully be safe from that. If not, I expect that many of those stray pallets are remarkably useless.

[navarac]

Still have ...

.... an external Floppy Drive and 1.44 MB Disks (about 20 only now) that are used with a perfectly functioning Mavica Digital Camera. It is fun to get it out and show youngsters who think they know everything and us oldies (77 yo) haven't a clue about technology. When I say my first PC (a 286) had a 20MB Hard Drive, 3.5" and 5.25" Floppy Drive, they think I'm kidding.

We lived through the revolution :-)

[Anonymous Coward]

Re: Still have ...

Not only lived through, but arguably we were the revolution! On the younger end of the scale, I think can't have been 3 or 4 when I picked up a computer for the first time.

I started on the a Sharp MZ-80 which had a built in cassette tape deck. The BIOS could only get you as far as starting the tape loader (or maybe an external device) - no other functionality at all. After a few variants on that theme, we went Amiga 2000, and then (somewhat grudgingly) onto a 486 after Commodore's collapse, which was also the first time we had a HDD.

The bodge nature of the PC's software was very apparent from the get-go, though the raw power was impressive.

[Michael Strorm]

Re: Still have ...

> Still have an external Floppy Drive and 1.44 MB Disks [used] with a perfectly functioning Mavica Digital Camera.

Ah, but did you know that the *original* versions of the Mavica (pre-"Digital Mavica") were actually analogue devices and used analogue floppy disks?

No, I'm not talking about the misguided pedantry of the "all floppies are digital, actually" argument, those truly did- and were intended to- record a fully analogue signal.

Even if the output quality was poor compared to film- being effectively just high-quality standard-definition TV pictures- as someone who was around at the time, I can't begin to imagine how cool it would have been to own one of those cameras back in the late 1980s!

[Alan Mackenzie]

3.5" floppy discs are not analogue.

"Digital" is not the same thing as "online".

[43300]

Re: 3.5" floppy discs are not analogue.

"Digital" is one of the current buzzwords, as in "Digital Transformation" (a phrase which the people using it can rarely define clearly, but seems to mean 'sticking all your data on someone else's computers')

[Bebu]

Re: 3.5" floppy discs are not analogue.

《"Digital Transformation" (a phrase which the people using it can rarely define clearly, but seems to mean 'sticking all your data... someone》

Unfortunately that phrase from my first encounter always evoked images more proctological than technological.

The prexisting canine latin phrase "extractum digitorum" might be to blame. :)

[John Brown (no body)]

Re: 3.5" floppy discs are not analogue.

"Digital Transformation" (a phrase which the people using it can rarely define clearly, but seems to mean 'sticking all your data on someone else's computers'

being a greybeard, I concur. Although it's not always that people can define it clearly as it has come to mean different things at different times, sometimes overlapping. The first time I heard it was at a company who had no computers at all and were buying in their first one. That was a "Digital Transformation" and billed as such to the peons. Other times it meant getting on-line for the first time, often meaning they had email accounts and maybe some web browser access. More recently, as you say, it's transforming to "cloud", other peoples computers :-)

[DuncanLarge]

Re: 3.5" floppy discs are not analogue.

> 3.5" floppy discs are not analogue.

YES they are.

They use frequency modulation and as far as I know a sine wave is not digital.

The C64 datasette (cassette drive) was as close to a digital tape you can get, but it still recorded parts of an analogue waveform into an analogue medium, that of magnetic particles.

It is not possible to move from a 1 to a 0 in zero time, thus it is not digital (the media). It is not possible to record a 1 and a 0 on an analogue medium as it can store any intermediate value, thus th logic 1 you stored earlier may have a absolute value of 500mV (when read by the head) and a logic 0 following it may be at -500mV, but then the media has a bad spot and only alowed a logic one coming after your perfect ones earlier to be at 450mV and the following logic 0 end up being -372mV. On a DIGITAL medium, you will never have that problem.

Digital media must only store discrete digital values. Not even todays flash media can do that, as again, flash media is ANALOGUE. The universe is not digital, such ways of looking at it are artificial. We thus read back our analogue media DIGITALLY by deciding the RANGE at which we have a logic 1 or a 0.

The C64 recorded fast electromagnetic pulses to the tape, but due to the nature of the uiniverse, thost pulses, that square wave, became an analogue waveform. Thus the C64 datasette has a circuite to read that analogue waveform off the analogue medium and "square it up" into the pulse train.

Floppy discs are analogue, as are HDD's and SSD's. We read and write to them in various ways, storing and retreiving analoge values and waveforms, which we then process into a digital bytestream. When a floppy disc drive has pre-amplifiers involved in reading and writing, well, there you go, floppies are analogue.

Optical media however can do that, the marks made on optical media are either there or not there. We dont care if they are deep or shallow as long as they are there or not there. The data is actually not encoded in the marks themselves but in the changes between them.

[Martin an gof]

Re: 3.5" floppy discs are not analogue.

Well-said, that man. Good to hear some basic explanation, however:

Optical media however can do that, the marks made on optical media are either there or not there.

While noting your use of "can", I'd like to put a word in here for Laserdisc, where the video signal is encoded as an FM signal (composite video) reconstructed from those on-off lands and pits, which are used to create a PWM signal. So, in contrast to all the media previously discussed which take an analogue signal from an analogue medium and "process" it to produce a digital signal for the computer, Laserdisc takes a digital signal from an (arguably) digital medium and processes it to produce an analogue signal for the display device!

M.

[Michael Strorm]

Re: 3.5" floppy discs are not analogue.

As I argued below, even that is wrong- Laserdisc video *is* fully analogue (i.e. not converted to digital at any stage) because the PWM pit length isn' t restricted to fixed values.

Even with DVD, the underlying media doesn't force tracks or pit length into fixed, quantised lengths or positions.

What makes it "digital" is how the signal is treated and intended to be processed.

[Michael Strorm]

Re: 3.5" floppy discs are not analogue.

Yes, we know that all already, and most of us weren't under any impression otherwise. I've heard that misguided rationalisation numerous times before.

By your logic, *any medium not reliant on inherently and fundamentally quantised aspects of physics is "analogue", and that includes almost *any* present-day digital technologies, optical media included.

The argument is flawed because What makes a signal or the associated electronics "digital"- in the manner we *mean* when we use the term*- is how they are meant to work when used as intended, how the signal is *treated* and processed.

I would have mentioned Laserdisc even if Martin hadn't got there first, but even his argument is flawed- Laserdisc isn't a digital signal, the PWM pit length isn't constrained to fixed values, so the recorded signal (on analogue Laserdisc) truly *is* analogue all the way through, PCM audio excepted.

You say that we don't care about the depth of the pits on (digital) optical media. My point exactly, we ignore or quntise the analogue aspects that let us treat it as digital. How is that any different to how we use "analogue" floppies as digital?

(And yes, I'm aware that there are fully analogue floppies used for still video photography!)

As I said, what makes it digital is how it's used, not the straw man definition your flawed argument is mis-aimed at.

[NXM]

analogue "digital" storage

My favourite in this field is the 1-bit audio chips you get in cards and toys that play annoying tunes. They have a programmed rom, but instead of holding a bit that's either on or off, each memory cell stores an analogue voltage. The chip samples each cell in turn, copies the voltage via buffer amp, and thus reconstructs the analogue sound waveform. Sounds bloody awful, but it does work.

And I've found many other "digital" chips to be surprisingly analogue, particularly when used outside of their expected parameters. Most recently a Microchip processor that /doesn't/ properly shut down when operated at below its rated supply voltage in a fault situation, and allows its outputs to go into a low-impedance state that it really shouldn't do (according to the datasheet, and you trust those at your peril anyway).

[Martin an gof]

Re: analogue "digital" storage

I've found many other "digital" chips to be surprisingly analogue

A previous boss had the habit of using NAND gates (7400 Four in a DIP) or even inverters (7404, six in a DIP) as op-amps. I suspect this worked better with some technologies than others (LS / HC / plain ol' TTL?) but it did seem to work.

M.

[Michael Strorm]

Don't think it's analogue, but I can imagine that it definitely *is* 1-bit

> 1-bit audio chips you get in cards [..] instead of holding a bit that's either on or off, each memory cell stores an analogue voltage.

Are you sure that's how it works? It's certainly possible to sample and hold discrete analogue values (*), but I'd be exceptionally surprised if that was how a dirt-cheap card did it. Especially as it'd be somewhat meaningless and contradictory to describe the bandwidth/resolution of an analogue sample as "1-bit" (bits inherently implying digital).

My suspicion is that this is half right, but not the bit about the samples being analogue!

My best guess is that the cards are entirely digital and work by taking a stream of 1-bit samples from the ROM, and converting them (almost) directly into on/off pulses for the speaker. And that they're relying on the limitations of the speaker and other parts of the system to filter out the high frequencies and aliasing and "smooth" that out into listenable audio.

This sounds like it shouldn't work- and since it's being done on the cheap I assume that'd be why it *doesn't* work very well!- but it's essentially a very, *very* grungy and low-end version of the same principle underlying the "1-bit" audio encoding used in some very high-end audio systems. (**)

(*) To some extent that's what this and this do.

(**) Pulse density modulation says that- if you do it correctly- then you *can* effectively get a high-quality signal back out if you sample at a high enough rate then low-pass filter it correctly to remove the aliasing. It's the basis of Sony and Philips' Direct Stream Digital audio system, amongst other things.

[John Brown (no body)]

Re: Don't think it's analogue, but I can imagine that it definitely *is* 1-bit

"This sounds like it shouldn't work- and since it's being done on the cheap I assume that'd be why it *doesn't* work very well!- but it's essentially a very, *very* grungy and low-end version of the same principle underlying the "1-bit" audio encoding used in some very high-end audio systems. (**)"

It's also how we got bleeps and bloops, and even (barely intelligible) speech out of TRS-80s and Commodore PETs with minor mods back in the day by simply playing with frequency and duration of the square wave pulses from the cassette port on TRS-80 and user port on PET.. Years later, the IBM PC built that same function into the system board :-)

[anonymous boring coward]

Re: 3.5" floppy discs are not analogue.

Sure, but in that case your internet connection is also analogue. As is your CPU, monitor, etc, etc. Hair splitting about the implementation of digital isn't very helpful.

When we store things as conceptual 1s and 0s it's digital (binary digital, strictly speaking, as you could have more than two states and still be digital).

[Dinanziame]

Old regulations

Roughly six years ago, our business was asked to provide some data to a government body. They did not request floppy disks; sending CSV files by email was fine. However, they still wanted two copies.

[I ain't Spartacus]

Re: Old regulations

I want that report done, and emailed in triplicate, by Friday!

[Gene Cash]

God I feel old....

I remember when 3.5" was TEH NOO KEWL THING on IBM PS/2 Model 30s in our college lab. I remember looking at the first such disk with a friend, and remarking that people will still find a way to staple them to a report.

[I ain't Spartacus]

Re: God I feel old....

Back in the 90s Virgin Mates (Branson's condom brand) did a bit marketing push. They sent cards to journos and politicians extolling the virtues of protected sex - each with a condom stapled into it.

Or perhaps it was more subtle a campaign than I thought at the time? By encouraging those kinds of people to breed, perhaps they were trying to put the rest of us off?

[RAMChYLD]

Re: God I feel old....

I remember my college years. The IT staff were either daft or clueless, USB drives (then still pretty new, 256MB was considered huge and 1GB was something that could last you a lifetime) were banned and everyone had to use a floppy disk. To make sure you suffer they actually disabled USB in BIOS and put hot glue over the USB ports. Really idiotic in hindsight since the floppy disks were unreliable and woe be upon you if your assignment disk decided to develop bad sectors.

[J.G.Harston]

But as long as they are not *required* to use non-disk-based data transfer, but are *allowed* to use non-disk-based data transfer, I see no problem.

[The Central Scrutinizer]

Could someone copy the Internet onto a floppy for me?

For offline reading, you know.

[I ain't Spartacus]

Sorry, we've run out of floppies. Can I print it out for you instead?

[JT_3K]

You joke, but in 2007 a teacher stormed in to my office, threw a 3.5'' floppy so hard it bounced off my desk and hit me in the sternum, then bellowed "I PUT THE INTERNET ON THIS DISK AND WHEN I GOT HOME IT DIDN'T WORK AND I COULDN'T PLAN MY LESSONS AND THE IT HERE IS ALWAYS GETTING IN MY WAY AND I'M BEING OBSERVED AND I DON'T HAVE A PLANNED LESSON"

I sat there a little shell-shocked. They stared at me for ~1-2 seconds and turned heel and walked out. Turns out they'd copied the link to IE on to a floppy, which worked fine in the building but not at home. I had a word with the senior team who chose that it wasn't something they wanted to follow, and had a quiet word with the teacher a few hours later about how little I appreciated their conduct, particularly in front of the students I'd been speaking to before they'd interrupted.

Ironically, the same IT-phobic and highly inept senior lead I'd raised this with, later in that job, called me to his classroom about a VCR that wasn't playing through his projector. Despite me having shown him how to do the input mode on the projector and tested it 20 mins earlier during break. He grandstanded whilst I checked over, citing about how IT never worked. The kids in the front row creased up and loudly repeated when I rather quickly figured the issue and ever so slightly not quietly enough advised him a VCR needs a tape inserting to work. He never did get control over the lesson and needed the on-call team (his senior colleagues) to embarrassingly come and provide additional manpower.

[GraXXoR]

I run my business here in Japan and was very pleasantly surprised to find out last year that I no longer need to use my official “Inkan” registered stamp that I had made for me back in the late 90s and can now rely on my signature or digital versions thereof like everywhere else.

Also moving over to e-tax was a big undertaking about 10 years ago or so and has streamlined one of the most despised business chores that is yearly tax returns.

Speaking of which. Time to get started. Le Sigh.

[An_Old_Dog]

Floppies are Not the Only "Old" Media Still in Use

Magneto-optical discs for medical imaging equipment.

Some years ago at work, I had to spec out a room full of high-end PCs for our new medical imaging software, but they also needed to have M-O drives to exchange data with some older ultrasound scanners. Looking for new PC adaptors and drives which supported the old media/format ... and sourcing blank media, was time-consuming.

[Bebu]

Re: Floppies are Not the Only "Old" Media Still in Use

《....M-O drives to exchange data with some older ultrasound scanners. Looking for new PC adaptors and drives which supported the old media/format ... and sourcing blank media....》

Curious does anyone still make 12" (300mm) MO drives and media?

Many years ago (ca 2012) I had grabbed two MO drives and spare media and added them to my hardware "archive" as new drives weren't being manufactured (by Sony.)

There were hundreds of media recorded with NMR data which I guessed might be required at a later date.

That day arrived. Then I discovered the media was recorded on a SunOS 4 workstation using the native ufs file system with a custom driver to map the 1Kb sector size of the drive to the 512b size required by the SunOS 4 ufs implementation.

I could get hold of a working SunOS box (and my "archive" contained OS install media for most workstation Unixes and hardware) but the custom driver which had apparently shipped on a 3.5" floppy had disappeared into the void leaving only the installation instructions and a quite intimidating licencing agreement (from early 1990s.)

Tried reading the media on a Linux box as the Sun ufs file system was supported but the MO drives sectoring defeated that too. Lession: snaffle the whole working system. Document. Preserve.

Data preservation and archiving is still not formalized in that environment so vast quantities of quite expensively acquired research data are disapearing over the event horizon and in many cases due to now unreadable media.

"..... the answer is blowin' in the wind."

[Anonymous Coward]

Re: Floppies are Not the Only "Old" Media Still in Use

LTO tapes : good for 30 years; however, the drives probably aren't.

The LTO standard defines that a "current" gen drive can read up to two gen's old; and assuming about one gen every 4 years and a drive life of 10.... Yes, the tapes most assuredly are outliving the ability to read them.

I had fun learning to use them in the home lab but ultimately they are up for flipping on retail site of choice.

[anonymous boring coward]

Floppies aren't "digital"?

[Anonymous Coward]

here, have my tax documents on a 100mb ZIP disk!

[John Brown (no body)]

click...click...click...click...click.... :-)

[RAMChYLD]

Well, at least Iomega was kind enough to replace the drives for free even if it was out of warranty. Shame that I lost a lot of stuff that were on zip disks tho.

[Rob 15]

Vinyl storage

I've just recalled those flexi floppy-type records used for distributing games on magazine covers. Perhaps it's time for a comeback. Has anyone got a turntable with Bluetooth?

[Anonymous Coward]

Re: Vinyl storage

Plenty with USB on them. Bung a raspi on the end and voila!

[Chris Miller]

Obligatory!

[Tron]

What you are all celebrating...

... is the erasure of the history of computing. This stuff is very difficult to replicate when mass production ends. You can build a Model T Ford or Spitfire more easily than you can create a working IBM PC from scratch. Our history is being dumped. Keep as much of it as you can.

[doublelayer]

Re: What you are all celebrating...

You are mistaking two things. The first is that we're not celebrating it. I'm not mourning at all, but there is a middle area between sadness and celebration.

The second is more important. No longer using something obsolete is not erasing, dumping, or losing that thing. We have floppy disks and drives in archives, museums, online shops, all over the place really. We also know how they were manufactured such that, if we decided it was worth going into production again, it could be accomplished. It's not happening because there's no point, not because we can't. There is no benefit and some harm requiring people to use something ancient for a historical purpose. It won't prevent companies from no longer manufacturing floppy disks, but it will increase the cost and inconvenience of anyone who had to submit forms that way. It is also a good thing that you're not required to drive a Model T to get your passport, whether you decide to maintain one or not. There are many old things that we don't consider worth our time to maintain, and that is not automatically a bad thing.

[Anonymous Coward]

Re: What you are all celebrating...

UK Car stopping distances in the highway code are still measured in Ford Anglia terms... Thinking times haven't changed (maybe got worse) though stopping distances most assuredly aren't the same.

[anonymous boring coward]

Re: What you are all celebrating...

And, yet, people keep on tailgating. Guess changing the highway code won't make it better.

[ldo]

Trivia Question

I see a lot of people talking about disks having a “1.44MB” capacity. What unit exactly does the “M” stand for, in this case?

[anonymous boring coward]

Re: Trivia Question

Mega.

1.44 x 1024 x 1024

[Michael]

Re: Trivia Question

Mega bytes

1.44 * 1000 * 1000

Mebibytes is 1024.

I think someone made a mistake and used base 10 at some point and it stuck.

The correct size is determined by multiplying the number of tracks, sides,

sectors per track, and 512 bytes per sector, then subtracting the bytes required

to format the disk, and then dividing this figure by 1024. For a "1.44-MB"

3.5-inch floppy disk, there are

80 tracks

18 sectors per track

512 bytes per sector

2 sides

Multiplying the above gives you 1,474,560 bytes. This is the unformatted size.

To determine the number of bytes formatting requires, you need to know how many

bytes are used for the boot sector, file allocation table (FAT), and root

directory.

There is 1 sector used for the boot sector, which is 512 bytes; 18 sectors for

the two FATs (9 sectors each), which is 9216 bytes (512 * 18 = 9216); and 14

sectors for the root directory, which is 7168 bytes.

NOTE: There are two ways to arrive at the 7168 number:

224 entries * 32 bytes per entry = 7168 bytes

-or-

512 bytes per sector (14 * 512 = 7168 bytes)

Adding these figures gives you 16,896 bytes.

Subtracting the amount used for formatting from the total unformatted size gives

you 1,457,664. (1,474,560 - 16,896 = 1,457,664 bytes)

Dividing the above figure by 1024 bytes generates 1440. (1,474,560 / 1024 = 1440

KB)

To convert to megabytes, divide by 1024. (1440 KB / 1024 = 1.406 MB)

[ldo]

Re: Trivia Question

So you end up with 1.4, not 1.44 for the disk capacity in MiB?

[Michael]

Re: Trivia Question

1.4 MiB is correct.

Fun with numbers in computing has been on going for years. Probably why mebibytes were standardised. It happened when I was starting uni around 1998 which led to all sorts of fun in exams and assignments with lecturers using MB to mean either MiB or MB especially when they realised that they had to accept both and update thier questions and marking schemes.

[ldo]

Re: Trivia Question

Whereas in MB, the correct figure is closer to 1.47?

So where does “1.44” come from?

[Dan 55]

Re: Trivia Question

720KiB * 2 = 1440KiB / 1000 = 1.44 "M"

Neither MB nor MiB. Obviously someone in marketing.