The great capacity non-conspiracy (TiB vs TB)

[solarisguy]

I am too young :) but those older tried to teach me once that K=1024 and k=1000

 

[anodos]

If you actually looked carefully at the PDP11 peripherals handbook you would see the Massbus RS03 fixed head disk (1970s), which had a capacity of 512K. 64 cylinders with 64 * 128B sectors (base 2). I'm tiring of this discussion though and will check out.

I still think it's an argument best resolved by switching to sexigesimal.

 

[rs225]

Anodos, I did actually look carefully. If you look REALLY carefully, you will see the fixed-head disk with 64 or 128-heads has a primary use as a swap partition. In other words, it is a fantastically expensive hybrid part that can be considered part of the (virtual) memory system. That is the reason that only the fixed head disks use base-2 sizing, while all their other storage is base-10. In other words, it proves my point again.

from the document: "... the RJS03 and RJS04 increase throughput substantially for timesharing applications which involve significant amounts of program swapping."

edit: removed incorrect assumption that these were single track devices.

I still think it's an argument best resolved by switching to sexigesimal.

Which again, is not my point. My point is that hard drive makers do not deserve the blame they get for the overall situation. They have changed nothing, and done nothing wrong.

 

[rs225]

I am too young :) but those older tried to teach me once that K=1024 and k=1000

That actually has some sense to it. Capital-K is Kelvin, lower-k is kilo=1000, and capital-K where it makes no sense to be Kelvin is 1024. :)