Detecting USB 2.0 port(s) on ASUS P4B266 motherboard

[Taavi Burns]

On Wed, Mar 17, 2004 at 05:51:05PM -0500, Henry Spencer wrote:
> On Tue, 16 Mar 2004, Jason Slaughter wrote:
> > > That's 480Mbps.  1394 (aka FireWire400 aka iLink.400) is 400Mbps.  Then
> > > there's the faster FireWire800 which of course runs at 800Mbps.
> > 
> > ...remember that's "480Mbps magical Intel theoretical marketing numbers," 
> > vs "400Mbps reasonable number that could realistically be used as an 
> > indication of actual performance." :)
> 
> John Mashey is fond of saying that such values are "guaranteed not to
> exceed" numbers. 

I'm not as familiar with FireWire, but I have studied the USB spec
(and rewritten the ISR for a USB device device [sic] driver).  They
call the 1.1 12Mbps link 12Mbps because it runs at 12MHz, and can
therefore support 12M bits per second...theoretically.  This
unfortunately includes sync (USB goes in async packets, much as RS-232
goes in async bytes) and coding.  If you feed a packet to the USB bus
that is only ones, you will experience a 14% degredation in speed due
to padding (a string of zeros does not suffer from this; I suppose the
reasoning is that a string of zeros is "normal" and a string of ones
is "less normal").

AFAIK, ethernet works the same way; 10bT runs at 10MHz, 100bT runs at
100MHz.  Hence the speed does not include sync, headers, or
collisions.

NB: USB never suffers from collisions, but it uses a heck of a lot of
CPU, as the host controller is responsible for ASKING each device to
speak; a device can ONLY ever respond to the host controller's
request, never initiate a communication; this is probably where USB
loses out the most vs Firewire.

-- 
taa

    I'd rather listen to Newton than to Mundie. He may have been dead
    for almost three hundred years, but despite that he stinks up the
    room less.
       -- Linus Torvalds
/*eof*/
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml