Re: Using standardized SI prefixes
On Thu, Jun 21, 2007 at 09:32:09AM +0200, Adam Borowski wrote:
> On Thu, Jun 21, 2007 at 01:11:52PM +1000, Hamish Moffatt wrote:
> > I think Ben's point is that we don't know.
> >
> > You seem to claim that binary units (ie powers of 2) are natural
> > everywhere related to computers, but I disagree. It's natural for
> > memory and structures like it, but not for bitstream quantities like
> > network traffic.
>
> But they don't use powers of 10 any more than they do powers of 10. While
> bps speeds are an oft-quoted case that "always" use powers of 10, the
> connection I got here is guaranteed min=max 1Mbps which as far as I can
> measure it goes right at 1048576 bits per second, rain or sleet.
> And the ISP is one of the most despicable, cheating, greedy ones you can
> imagine -- for example our company pays for that 1Mbps more than in a
> civilised place you would pay for 100Mbps, so if they seen a place to
> overadvertise something, they would.
>
> And as far as I know, usually 1Mbps stands for 1024x1000 bits where network
> speeds are concerned, to be wrong by both the correct and yours
> interpretation :p
The raw network transports (eg Ethernet and SONET) *are* quoted in powers
of 10, and they mean it. Gigabit ethernet is really a billion bits
(10^9) per second. OC-3 is really 155,520,000 bits per second.
Powers of 10 are perfectly natural in this case (imho). They are what we
humans are used to as the default. For computer memory structures where
an N bit address bus means you have 2^N bits of storage, powers of 2
make some sense, but not in the general case.
Hamish
--
Hamish Moffatt VK3SB <hamish@debian.org> <hamish@cloud.net.au>
Reply to: