Stop getting it wrong!

Stop getting it wrong!

Why did they do this?

Indeed, kebibytes vs kilobytes.
Unfortunately improper use is rampant in reporting.

Because they're dicks.

I ain't writing KiB when I can just write KB.

Why should kilo=1000 in every case except for bytes?

Because bytes aren't base 10.

Because we shouldn't be using base 10 prefixes for base 2 numbers

But why not use one system? Why complicate it?

because the computer counts in 2
why impose a decimal system on it and complicate everything?

Dumb nigger stop using kilo then. Is this the power of the american education system(tm) ?

kB is kilobytes
KiB is Kibibytes

>year 2018
>this shit is still mislabeled

So the base unit is 1.024 bytes? I don't understand.

There is no such thing as base 10 versus base 2 numbers. Numbers are independent of any bases.

There ARE numbers that have a tendency to be round in base 2. For those numbers, *when this roundness is either relevant or naturally present* (and only then), it makes sense to express them in units that can visibly express this roundness. Thus, I have 32GiB of RAM, because RAM naturally comes in units that are round in base 2, and writing that as 34.36 GB would be silly. Yet I have 4TB of disk space storage, because disk spaces do NOT naturally come in base-2 round units, and expressing that as 3.64TiB would be equally silly.

This means there is a use for both prefixes. Naturally, using the same word for the different prefixes would be utter retardation.

More words suck for marketing.

Why does base-2 not make sense for disk space?

Yes, that is true. But that does not change the necessity of having them.

Why did they make computers use two systems?

Computers don't use either system. They just store a number. Base-2 vs Base-10 is just for human readability.

It makes the same amount of sense of base-10 multipliers; that is to say, it's as good a choice as any, and there are no clear preferences. (Disks do not come as powers as 2 as a consequence of the technology.)

Except, of course, that in the lack of any special preferences, it's much more practical to use the same set of multipliers used everywhere else, as the default fallback option.

The only people that have a problem with understanding KiB are the same people who load gallons of fuel into their car and drive 1760 yards to the gun shop to pawn a stolen watch and iphone for a concealed carry permit.

You have a point about the bases but by not being pedantic I was able to get the same point across with far less text.

Fair enough. But I find that a lot of people to not understand this distinction; they seem to think that there is a much more significant distinction between "base 10 numbers" and "base 2 numbers", and that trying to express the latter using base-10 prefixes is a bad idea somehow. So in threads like this, I find it best to be explicit and pedantic about it.

Kekedbytes when

But Microsoft uses KB, not kB.

1KiBe

>Stop getting it wrong!
Stop buying into this propaganda.

Back in the day 1 byte had 8 bits and 1 KB had 1024 bytes. This is still the case.

Marketing people wanted to justify selling 1 TB HDDs that ain't 1 TB so they came up with this shit, oh it's less than 1 TB because it's not 1TiB!!

>Thus, I have 32GiB of RAM
No, you have 32 GB of RAM. That's what's on its box, that's what's listed in the task manager, that's what's listed by diagnostics programs. Your argument is retarded. Also, it's entirely possible to make, say, 2,000,000,000 byte memory modules so I don't even get what the fuck you mean by "they naturally come in base 2"

>Back in the day
Stop being a fucking brainlet
1 kilo of something has always been 1000 of that something
That's just the definition of SI prefixes
>1 kg ≠ 1024 g
>1 km ≠ 1024 m
>1 kB ≠ 1024 B

I hate this shit. It's just disk storage people using this crap for marketing reasons. I'm not going to fucking use these stupid retardo KiB type units when it's always just been KB. Computers work in base 2, so that's obviously what KB is being used with in the context.

I don't though, I've been using the correct nomenclature for over 10 years.
IBM thought they were better than SI.

Computers use a base 2 number system.

1 Byte = 8 Bit
1 Kilobyte = 1,024 Bytes
1 Megabyte = 1,048,576 Bytes
1 Gigabyte = 1,073,741,824 Bytes
1 Terabyte = 1,099,511,627,776 Bytes

This was the case in the 80s and it's still the case now.

And the reason I say it this was the case "back in the day" is that nobody used "KiB" for anything anywhere not that long ago. Mixing in the metric system's kilo and 10 bit numbers really was a marketing stunt.

top - 17:52:19 up 2 days, 16:11, 2 users, load average: 1.41, 1.57, 1.52
Tasks: 467 total, 2 running, 313 sleeping, 0 stopped, 0 zombie
%Cpu(s): 5.0 us, 3.4 sy, 0.0 ni, 91.6 id, 0.1 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 65983560 total, 9305996 free, 31250568 used, 25426996 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 34408484 avail Mem
nah it says KiB.

>No, you have 32 GB of RAM.
I have a bit over 32 GB of RAM. 32 GiB, to be exact.

>That's what's on its box,
Indeed it is (I checked). So the box is inaccurate. Surely marketing terms being inaccurate is not a surprise.

>that's what's listed in the task manager, that's what's listed by diagnostics programs.
Maybe yours is. Windows' tools are well known for displaying numbers in MiB and GiB, and then writing MB and GB. But that just means Windows is reporting incorrect numbers.

>Also, it's entirely possible to make, say, 2,000,000,000 byte memory modules so I don't even get what the fuck you mean by "they naturally come in base 2"
It is certainly possible, but the design of RAM is based around a regular grid of addressable units (which are usually 128 bit wide, in modern RAM). This is much more efficient, hardware wise, if your RAM covers exactly as much address space as can be expressed with a certain number of address lines, for this greatly simplifies the cell selection multiplexing logic. For N address lines, this yields 2^N 128-bit memory cells, or 2^(N+4) bytes. Which makes a 16GiB = 2^34 bytes = 2^30 128-bit cells efficient, and 16GB = a few percent less than that, not efficient.

Indeed they do, which is fine.
Breaking SI prefixes just for computers is not.
That's why KiB/MiB/GiB/TiB/etc exist - as a base2 variant for computing.

Microsoft is the only vendor not on board with this, even IBM - the people who roginally so-opted the SI prefixes, is on board.
With the exception of Apple and HDD makers, everyone just renamed the string from KB/MB/GB/TB/etc to KiB/MiB/GiB/TiB/etc
Apple and HDD manufacturers continue to use KB/MB/GB/TB/etc but in the correct base10 format.

Standards matter and as usual only Microsoft is breaking the standard in silly ways.

>I'm not going to work to the standard for my own arbitrary reasons, but i'll expect everyone else will follow /my/ form and nomenclature, like a standard, becuase I'm better than SI.

>Mixing in the metric system's kilo and 10 bit numbers really was a marketing stunt.
It probably was. But it was a very fortunate marketing stunt, for it is also very necessary.

I have a 4GHz processor. Now you may not believe this, but that means, *and has always meant since the days of the 740kHz Intel 4004*, 4000000000 Hertz.

If my processor can encrypt 1 byte every 10 clock cycles, how much data can it encrypt in a second? If the answer is anything other than "400 MB", something went very fucking wrong. Which means that a megabyte must be 1000000 bytes, not 1048576 bytes.

Likewise, let's say I am measuring the quality of a stretch of road, with a measurement device mounted under my car. The measurements produce 100 kilobytes of data per kilometer of road. That had BETTER be identical to 100 bytes per meter.

Having "kilo" mean either 1000 or 1024 based on context is terrible and leads to awful problems. It is likely true that disgusting marketing is the reason we got into this confusion in the first place; but it is nonetheless something necessary, no matter where it came from. Thus, a megabyte is 1000000 bytes, and if you have a good reason to talk about 1048576 bytes -- which is a totally legitimate thing to do! -- then we have the word "mebibyte" for that purpose. (Yes, I know it's a word that cannot be pronounced without giggling like a 12-year-old. But that doesn't change the situation.)

good luck getting the world to recognize a 10-bit byte

Huh? Who is is talking about 10-bit bytes? Not me. The size of the byte has nothing to do with all this.

The processor deals in 4-byte words for 32 bits of computing. Hard drives use 4096 byte sectors. The Linux kernel uses 4096 byte pages. A char fits in 8 bits. The size of a byte means everything.

Yes. But it has fuck all to do with any of my points in .

i read the entire thread and i still don't know what's the right use

jiggabyte

Is what you are saying that bytes are base ten?

Why isn’t 0x1000 bytes a kilobyte, 0x1000000 a gigabyte, 0x1000000000 a terabyte etc.?

fuck I used the wrong units of magnitude by my point still stands

>I have a bit over 32 GB of RAM
No, 32GiB is 18877906944 bits over 32GB.

kb = 1000 bits
kB = 8000 bits
KiB = 8192 bits
Am I rite here?

>t. HDD manufacturer

yes

To jew people out of getting the size of disk they think they are buying.

So what you're saying is only the "exceptions" of Microsoft and Apple that have 99% market share don't use the KiB bullshit?

Also RAM is still sold in the correct pre-ISO notation, as are CD-Rs.

My bran approves KB = 1000 and kib should go away, for these reasons:
- convenient for calculations
- fits to the metric system
- I have nothing against change and improvements of things

The only reason not to like 1000 is muh marketing tricks. But who cares when it really has benefits at the end?

'murica

Nigger, the only reason the whole kb/kib argument came into existence was that HDD manufacturers cheaped out.

kb equaled 1024 bytes until HDD manufacturers wanted to advertise more storage than they actually put on the disk, so they made up the kib meme and asserted than a kb is 1000 bytes.

This issue can easily be resolved by using the imperial system instead.

underrated post

Double dubs checked.

I hate how windows reports a measurement of GiB in GB. It flares up my autism.

GiB was made up by HDD manufacturers to jew you out of storage space.

A Gigabyte is 1024 megabytes, not 1000.

You don't get jewed if you know how to count.

16(10) == 10000(2)

Git gud and stop blaming my peoples.

>That's just the definition of SI prefixes
Except bits and bytes aren't SI units and the other user is right: "back in the day" the consensus was that prefixes borrowed from SI denote powers of two when used in the context of memory units. There was _zero_ confusion about that, until HDD manufacturers' marketing departments started to pretend otherwise, which lead to to (relatively recent, mind you) invention of those retarded kibi/mebi prefixes.

What do those numbers have to do with anything?
No one uses 5 bit bytes and 16 doesn't relate powers of 10.

Back in the day base-8 was in broader use.

The absolute state of Sup Forums

Because computer addressing doesn't work like that.
Since computers work in base 2, everything winds up as powers of two to make things convenient.

If you break things into chunks of 1000, then you waste 24 addresses.

nvm i'm retarded, it's 02000
More real version is - there were existing SI prefixes and computer guys used them adapted for binary.

>What do those numbers have to do with anything?

He said you just need to know how to count. He followed that statement up with a decimal to binary conversion. The fact that you had no idea what he was doing kind of illustrates his point pretty well.

KB = 3E8 is INconvenient for calculations and should've been never used.

how the fuck did hex enter this conversation?

Marketing.

No it doesn't.
The whole reason we have this 1024 vs 1000 argument is because 1000 doesn't translate into a nice round number in binary.

We use 1000 in our number system (base 10) because it's a nice round number.
In binary that's 01111101000. That is not a nice round number, but in binary 1024 IS a nice round number (10000000000).
Because of this, 1024 makes more sense for addressing, and therefore measurement, than 1000.

Hexadecimal's base is a multiple of binary's (16=2^4), so hex can be used as a shorthand notation for binary numbers.

...

I know how and why hex is used, but it still doesn't bear any relevance to this conversation or to the post that it was replying to.

It was kB when I was a kid so that is what I was taught and I'll keep using it.
Fuck all those Wikipedia progretards. Other than them nobody uses kiB.

Hopefully KiB will die soon, now that everyone is moving to SSDs.

>hopefully imperial will die soon, now that everyone is moving to metric

This is how dumb you sound.

kek

There isn't a single SSD that uses 1000 bytes per kilobyte because that's retarded.
They simply don't work in a way that would allow 1000 bytes to a kilobyte without either becoming significantly more complex or wasting memory.

How is the other an "imperial" system?

And if metric were superior, why aren't you using metric prefixes AND metric units?

As in don't use MiB or GiB units at all on your OS and simply use MB, GB.

O yeah you won't you do that, because you know metric on computers is retarded.

>length units
Should have used weight units

kekibytes?

This diverging nomenclature was solely created to allow manufacturers of Hard Drives and other media storage to charge us more money for less storage.

this is absolutely beautiful

Jesus Christ

>There isn't a single SSD that uses 1000 bytes per kilobyte because that's retarded.
Are you sure about that? See fucking pic related.

I"m actually wondering why RAM manufacturers haven't adapted this yet. They could make 1,000,000,000 byte modules and sell it as 1 GB modules.

Shut up, don't give them any ideas!

Also, that's probably a 512gb drive with 12gb set aside for replacing damaged sectors.

kekistanbytes

KILL ME