Eli5 Why there arent like 20gb USBs instead each usb has twice as much as the previous(8gb,16gb,32gb,64gb,128gb
In: 1
Because of binary. All data storage sizes tend to be powers of 2 because of the binary nature of computing and storage. They are doubling because they’re the next power of 2
Those usb sticks are built by grouping memory chips on a pcb.
Imagine how hard the logistics would be if instead of a standard 6 pack beer we had 7 packs, 5 packs, 11 packs, 13 packs
In thus case a 6 pack is a memory module of 8gb, if you ask the manufacturer to build a 42.069 gb usb they will say “i can do either 32 or 64, im not going to design and manufacture a special design just for you (i can do it, but it will cost you more than making a 64 gb version)”
Edited because 40 is a multiple of 8 lmao
Bits and bytes are exponential starting with 1. It’s the same reason that ram modules generally follow the same pattern. It’s also the reason that 1kB is actually 1024 bytes and not 1000.
There are some instances where it isn’t followed but it’s just easier that way.
Because 2^3 = 8, 2^4 = 16, 2^5 = 32,…
Computers work with only 2 values 1 and 0 called bits.
You can imagine the chips storing these as apartments.
Each door in the apartment must have a number (address).
2^x tells us the maximum number of doors with x being the number of bits needed to write the address for the door.
If you want to write 20 you will find that 2^4 (16) is not enough and you need 2^5, but at that point you might as well use all 32 addresses.
Every storage location needs an address, just like a house needs an address. In order to access that byte of info, you need to basically plug in the correct address. Of course, like all computers, they deal with electricity which is either on or off. We represent that with 1 or 0 respectively.
So let’s say you’ve got a 2-bit address. The first bit has 2 options, and the second bit also has 2 options, so there are 2*2=4 addresses possible. Those are as follows:
00 – 01 – 10 – 11
If I increase the number of bits in the address by 1, to 3 bits, the first two bits can be any of those 4 while the new bit is 0, then repeat the same 4 options while the new bit is 1. So you double the number of locations that can be addressed for each bit you add:
(0)00 – (0)01 – (0)10 – (0)11
(1)00 – (1)01 – (1)10 – (1)11
If I clean that up a bit it looks like this:
000 – 001 – 010 – 011 – 100 – 101 – 110 – 111
8 options for 3 bits. 16 options for 4 bits. And so on… There’s one other thing going on here. There are two prefixes that can be used for data. The normal metric prefixes (kilo-, mega-, giga-…) Are supposed to mean 1000 of the previous. A kilometer is 1000 meters. A megajoule is 1000 kilojoules. In data, they added a separate set of prefixes (kibi-, mebi-, gibi-…) Which are actually an even power of 2 multiple instead of 1000. It works out fairly well because 2¹⁰=1024 which is pretty close to 1000. So a kibibyte is fairly close to a kilobyte. And a megabyte is pretty close to a mebibyte.
I think what happened is that the common community of computer consumers didn’t take to the new prefixes because it’s a bit confusing to have two separate sets that mean different things. And it’s not like you ever need to convert between kilobytes and megajoules so why not just use the regular prefixes anyway. So now everything is called kilobytes or megabytes when in reality, they actually mean kibibyte and mebibytes.
Latest Answers