I have 40GB in there, but only 830GB data. I suspect that it is deleted test data from Europe-north-1.
Sorry, but I have to disagree there. TiB & co. have surfaced only recently and way back when I learned about computing we only had kB for floppy disks and RAM, and MB for hard disks. They were always power of 2, except for HDDs, so the manufacturers could make them look bigger. With MB the difference was small, so it didn’t matter much, but with TB sized drives it does matter.
RAM is to this day sold in GB and they are definitely the power of 2 variety. I don’t know about SSDs, since they use a technology similar to RAM.
The prefixes used are not unique to data in computer systems. They are used for many different types of units, like kilometer, kilogram, and in all those other cases they have a decimal representation. So… for a consistent system to work, we can’t use the same prefixes to have a binary meaning.
Besides, now we can have fun things like KiM => kibimeter, KiG => kibigram! Who doesn’t want that?
This random user doesn’t disagree with the SI, but predates it. I guess I’m showing my age now. Fascinating reading and it largely says what I said. You are most certainly right that it needs to be consistent. Funny thing is that I have come across xibibytes only very recently, although they have been defined back in '98. I dropped out of the IT industry in 2005 and back then @ IBM nobody used them.
OS’s are unfortunately stubborn in switching to the correct units. It’s confusing, but who knew HDD manufacturers were right after all?
I think the problem is that most stuff predates this standard or grew from something that predated it. Hopefully the world will switch some day to using the correct units. But hey, I’m also still hoping the US will finally switch to metric for real.
i agree with bright that it makes sense to keep with the SI standarts… even if that does require me to learn a few new names for data definitions… i however would also argue that in many aspects of computing it will make sense to use the power of two… ofc today we aren’t really forced to stick to any specific system because we are so many layers removed from direct hardware access…
so making everything just be displayed in SI Bytes makes sense for the simplicity even if the underlying systems will run other sizes… ofc the numbers will all become horrible near infinite decimal numbers… which is kinda a downside…
48GiB/GibiBytes or 51.53961 GB/GigaBytes
so yeah it’s not all sunshine and rainbows moving to using decimal… ofc we could just cut a bit off and just use whole numbers… but it will never be pretty or make good sense when counting in binary on the lower layers…
So yeah… ill just accept that i need to add an i in all the definitions… MiB
PiB GiB … and nobody gets their panties in a twist and the world still makes sense even if the names are kinda stupid
and how you do that exactly when 95% of computation hardware is based on binary…
in binary the multiplication sequence goes 1 2 4 8 16… and in decimal the sequence is 1 10 100 1000 10000…
which is very incompatible… so if we count the bits… then like a 4 bit bus is max value of 15
so a 4 bit bus would be 15 decimal something bus and a 6 bit would then be 63 decimal something bus.
i suppose one could define it in another way… like using like quadratic primes as units both system would share as equal whole numbers …but then …brain explodes
binary simply isn’t compatible with decimal conversion without going into weird numbers schemes and not even sure it would be solve by finding a common large scale denomination.
but maybe…
its a bit like saying you should just write the latin based languages in chinese because there are more chinese so it makes sense to use that language…
however much of it doesn’t translate directly… there will be definitions that both languages doesn’t share, and different definitions that overlap in different ways…
i’m all for having SI but you can’t just say you want it like that and expect the universe to comply… math doesn’t care what we think…
well if you ask me it would be easier just to move on because the mess that is currently with gibibytes and gigabytes is basically impossible to reconcile, and will lead to many issues until all the tech is replaced… you cannot have two different denomination for GB, which there are now…
so it would make sense just to move it along to a different system… or just call it bits… isn’t the whole problem that we denominate a byte as one…
hmmm that might actually work… ofc it would mean that then when people think one symbol is a byte is kinda out the window, but that’s barely even relevant to anyone today…
Nothing needs to be replaced. It would be a tiny update to just display the correct unit. It’s a labeling issue, not a tech issue.
Exactly, we’ve been living in that world since day one. The problem is that existing prefixes with an existing meaning were misused. There are now correct prefixes defined that should be used instead.
All I’m arguing for is that windows simply displays GiB instead of GB. I’m not telling anyone to use the decimal units or calculations, I’m just saying, if you use the binary counting method, use the appropriate unit as well. Coincidentally, this means HDD manufacturers have been the only ones doing it right for years.
