[arin-ppml] debunking the myth that Moore's law helps

Leo Bicknell bicknell at ufp.org
Wed Dec 16 23:24:35 EST 2009

In a message written on Wed, Dec 16, 2009 at 07:57:58PM -0800, Michel Py wrote:
> - The bandwidth on the Internet is driven by demand.
> - The demand is driven by devices that connect, namely computers and now
> TVs.
> - The {speed | power | memory} of these devices is directly related to
> Moore's law.
> - Michel's law 0.9 says: "A device that is twice as fast or has twice as
> much memory will consume twice as much bandwidth".
> Therefore I say: there is a direct relation between Moore's law and
> Internet bandwidth. Call this simple, simplified, simplistic, whatever.
> And debunk it please.


Computing power seems to double every 1.2 years, roughly.  Note
that this is loosely coupled to moore's law, but there is a

Picking an arbitrary spot in time, in 1991 the first 14.4k modem
was introduced.  That was 18 years ago, or 15 doubling cycles, at
the speed supercomputers have evolved.

14.4kbits * 2^15 is 450 Gigabits.

So if you could afford a 14.4K modem in 1991 which, while expensive
was something the average joe could afford.  It might cost as much
as your computer at that time, you should be able to afford a
450Gigabit home connection today.

Video is a prime example.  Quality has improved, but as much due
to better compression as to more bandwidth.  Turns out we burn a
bunch of CPU cycles to conserve bandwidth, as bandwidth is vastly
harder to increase.  It may not seem like YouTube is saving bandwidth,
but raw 1080i uncompresed video is ~1.5Gbits/sec, while YouTube has
encoded it as ~12Megabits/sec.  All the CPU is going to make up for
the fact that we don't have 1.5Gbits/sec to the home.

So no, twice the speed does not equal twice the bandwidth consumed.

       Leo Bicknell - bicknell at ufp.org - CCIE 3440
        PGP keys at http://www.ufp.org/~bicknell/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 826 bytes
Desc: not available
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20091216/676df279/attachment-0001.sig>

More information about the ARIN-PPML mailing list