Max IDE Cable Length?

[Gene Holmerud]

Altho this question was asked relitive to IDE, other
protocols have other considerations.

There is an aspect of "round" cables that is not taken
into account here.  The individual wire pairs (1&2,
and 3&6 in the case of ethernet, other pairing for
other protocols) are twisted.  And each pair is
twisted at a different number of turns per foot. 
Since the electrical signals within each pair are
"differential" (that is, the signal is the voltage
differ
ence between the wires of a pair, but is independent
of the independent of other wires in the cable and
independent of stray voltages), isolation is pretty
good.  No, not as good as individual coaxials, but way
better in usabality, performance & price. The
twisting, and the difference in the number of turns
per foot help cut down the crosstalk.  And most
crosstalk gets induced into other pairs commonly, thus
not effcting the "difference" voltage within a pair.

The cable length restrictions (and the complementary
packet restrictions) in ethernet have to do with
making sure all receivers on a subnet will detect a
collision.

Gene
--- Bryce C <Plug@BryceCo.Net> wrote:
> While they might be pretty and easier to move
> around, I don't reccoment
> using round IDE cables.  With all the wires lumped
> together, the
> interference from one wire can, and does affect
> every other wire near it
> instead of just 2.  This has a negative impact on
> the bus speed as
> compared to ribbons.  I've benchmarked both cables
> using dbench and I
> found the round IDE cable to be WAY slower than the
> ribbon type.  The
> only thing I would use a "round" cable on is the
> floppy drive. There,
> the access speed is slow enough such that you won't
> notice (or bench)
> the difference.
> In short (hehe), stay away from round IDE cables.
> 
> On Tue, 2003-05-06 at 10:25, Fritz wrote:
> > I purchased some of those new-fangled "round" IDE
> > cables the other day.  They're much easier to work
> with
> > and they probably do allow better air flow in the
> cabinet.
> > 
> > These cables are 24" long.  I was under the
> impression
> > that the IDE Spec limited cables to 18".  The
> salesman
> > said "I'm sure that they're OK, otherwise no one
> would
> > be making them and we wouldn't be selling them".
> > Well, with logic like that I promptly opened my
> wallet
> > and ....
> > 
> > Has it got anything to do with the old ATA/33/66
> vs. the
> > new ATA/100/133?  Or the old 40 conductor vs. the
> new
> > 80 conductor?  Anyone have any clues about this?
> > 
> > The system *seems* to work fine.  Perhaps, my data
> is
> > quietly being corrupted, bit by bit ............
> :-(
> > 
> > Fritz Kolberg
> > 
> >
> ---------------------------------------------------
> > PLUG-discuss mailing list -
> PLUG-discuss@lists.plug.phoenix.az.us
> > To subscribe, unsubscribe, or to change  you mail
> settings:
> >
>
http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss
> -- 
> Bryce C <Plug@BryceCo.Net>
> CoBryce Communications
> 

> ATTACHMENT part 2 application/pgp-signature
name=signature.asc



__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com