This is up for debate, with computer prefixes now officially aligned with the standard SI prefixes.
You’ll often see a GB meaning 1000MB, and a GiB (gibibyte) meaning 1024MB.
The ISQ (International System of Quantities) and IEC (International Electrotechnical Commission) refer to it that way, and so do many others.
But then again, some keep the more traditional 1024MB is a GB system, and maintain that the SI prefixes shouldn’t count in computing because the base 2 1024 is close enough and it’s the way we traditionally did it. I think Microsoft still does, for example.
In the past, that system was close enough. After all, an additional 24 bytes or kilobytes is a tiny amount. But now that we’re getting into super huge data sizes, the gap is significant. 8 terabytes by the official scale is 8 trillion bytes, but by the “traditional” scale it’s 8.8 trillion bytes, a pretty sizable difference!
In a way, 999 and 1023 are both correct. But 999 is technically the standard, and has been for a while.
Great summarized, especially the “close enough” part. If you think about it this situation never would’ve happened if we would have based the majority of the computer designs on ternary instead of binary.
https://zeta.one/kilobyte-is-1000-bytes/
So, to my limited knowledge, all digital storage is still based on the idea of a switch indicating a 0 or a 1. So, in terms of data storage, you’re using those switches and base 2 is imposed.
You technically cannot build 1000MB of storage because your entire storage system is based 2. Being off by 24 isn’t great, but manageable. However…
Let’s call a KB 1000 bytes, and 1MB 1000 KB: we end up 1MB as 1,000,000 bytes, and 1GB as 1,000,000,000 bytes rather than 1,073,741,824 bytes, ~7.4% off! This error compounds as we go up in units, and quickly leaves one so far from physical hardware as to question one’s sanity!
The real reason for the change is likely to be a little darker - 1.1TB sounds better than 1TB when trying to sell storage (“we give 10% more!”).
You absolutely can build 1000MB storage, literally almost all SSDs to my knowledge use SI units for storage (meaning 1TB = 1000GB). E.g. here, first link I found https://www.crucial.com/ssd/t705/CT2000T705SSD5
I am outraged by your comment and I believe you to be siding with the opressors that are harming us with doublespeak and lies, and ripping us off. I hope this time it doesn’t get flagged by rule 2 , because I am disgusted by this way of thinking and bowing down.
Some of the comment that was deleted because it was too naughty addressed that, it’s not that it’s “made to anger me” it’s to rip people off by misrepresenting the amount they are selling and started much more recently than people think.
However it’s also true that most people who see the units assume it follows the same 1000-based system as literally everything else that uses those SI-prefixes does.
I somewhat doubt that all these international standards organisations are in the pockets of Western Digital and Seagate. It’s far more likely that they think “kilo means 1000, not 1024. Because that’s literally what kilo translates to.” Of course, the end result is still that it benefits storage manufacturers, but I highly doubt that’s what they set out to do.
Regardless of your opinion on the matter, getting that angry at people and dismissing them as bootlickers because they explain the GB vs GiB debate seems over-the-top to me.
In the past, that system was fine. After all, an additional 24 bytes or kilobytes is a tiny amount. But now that we’re getting into super huge data sizes, the gap is significant. 8 terabytes by the official scale is 8 trillion bytes, but by the traditional scale it’s 8.8 trillion bytes, a pretty sizable difference!
Yeah, the 0.8 trillion bytes that the manufacturer is ripping off the customer with.
Not really. The average person sees 8TB and assumes it means 8000GB. And… well… now it does.
Only a very small amount of people (tech nerds) could think it actually means 8192GB, and that each of those GB was actually 1024MB, and so on, and the people who know that also probably won’t be fooled by the misleading thing you say is happening (it isn’t. They’re calling a TB a TB, as they should be).
Personally I’m fine with kilo/mega/giga/tera meaning kilo/mega/giga/tera, as opposed to kilo/mega/giga/tera plus some extra. Nobody is stopping you from measuring things in GiB and getting the numbers you desire.
This is up for debate, with computer prefixes now officially aligned with the standard SI prefixes.
You’ll often see a GB meaning 1000MB, and a GiB (gibibyte) meaning 1024MB.
The ISQ (International System of Quantities) and IEC (International Electrotechnical Commission) refer to it that way, and so do many others.
But then again, some keep the more traditional 1024MB is a GB system, and maintain that the SI prefixes shouldn’t count in computing because the base 2 1024 is close enough and it’s the way we traditionally did it. I think Microsoft still does, for example.
In the past, that system was close enough. After all, an additional 24 bytes or kilobytes is a tiny amount. But now that we’re getting into super huge data sizes, the gap is significant. 8 terabytes by the official scale is 8 trillion bytes, but by the “traditional” scale it’s 8.8 trillion bytes, a pretty sizable difference!
In a way, 999 and 1023 are both correct. But 999 is technically the standard, and has been for a while.
Great summarized, especially the “close enough” part. If you think about it this situation never would’ve happened if we would have based the majority of the computer designs on ternary instead of binary. https://zeta.one/kilobyte-is-1000-bytes/
So, to my limited knowledge, all digital storage is still based on the idea of a switch indicating a 0 or a 1. So, in terms of data storage, you’re using those switches and base 2 is imposed.
You technically cannot build 1000MB of storage because your entire storage system is based 2. Being off by 24 isn’t great, but manageable. However…
Let’s call a KB 1000 bytes, and 1MB 1000 KB: we end up 1MB as 1,000,000 bytes, and 1GB as 1,000,000,000 bytes rather than 1,073,741,824 bytes, ~7.4% off! This error compounds as we go up in units, and quickly leaves one so far from physical hardware as to question one’s sanity!
The real reason for the change is likely to be a little darker - 1.1TB sounds better than 1TB when trying to sell storage (“we give 10% more!”).
You absolutely can build 1000MB storage, literally almost all SSDs to my knowledge use SI units for storage (meaning 1TB = 1000GB). E.g. here, first link I found https://www.crucial.com/ssd/t705/CT2000T705SSD5
under footnote 6. 1 GB = 1 billion bytes
Good point, I was thinking in terms of addressing being base 2 - (so when you call a memory address you’re working in base 16 normally).
Also that rather affirms the idea - selling less while disguising it as more seems a more likely genesis.
I’m old and did not know this. Huh.
Still 1024 in my heart.
Removed by mod
Well, well, aren’t you polite.
yeah , hide behind politeness.
I am outraged by your comment and I believe you to be siding with the opressors that are harming us with doublespeak and lies, and ripping us off. I hope this time it doesn’t get flagged by rule 2 , because I am disgusted by this way of thinking and bowing down.
There, no swear words.
Well ok, I’m sorry you feel that way.
I hope you realise that international standards organisations aren’t setting standardised prefixes specifically to anger you.
Some of the comment that was deleted because it was too naughty addressed that, it’s not that it’s “made to anger me” it’s to rip people off by misrepresenting the amount they are selling and started much more recently than people think.
I can see that argument.
However it’s also true that most people who see the units assume it follows the same 1000-based system as literally everything else that uses those SI-prefixes does.
I somewhat doubt that all these international standards organisations are in the pockets of Western Digital and Seagate. It’s far more likely that they think “kilo means 1000, not 1024. Because that’s literally what kilo translates to.” Of course, the end result is still that it benefits storage manufacturers, but I highly doubt that’s what they set out to do.
Regardless of your opinion on the matter, getting that angry at people and dismissing them as bootlickers because they explain the GB vs GiB debate seems over-the-top to me.
Yeah, the 0.8 trillion bytes that the manufacturer is ripping off the customer with.
Not really. The average person sees 8TB and assumes it means 8000GB. And… well… now it does.
Only a very small amount of people (tech nerds) could think it actually means 8192GB, and that each of those GB was actually 1024MB, and so on, and the people who know that also probably won’t be fooled by the misleading thing you say is happening (it isn’t. They’re calling a TB a TB, as they should be).
Personally I’m fine with kilo/mega/giga/tera meaning kilo/mega/giga/tera, as opposed to kilo/mega/giga/tera plus some extra. Nobody is stopping you from measuring things in GiB and getting the numbers you desire.