Hi. I'd not like to argue, but...
>
> [log in to unmask] wrote:
> >
> >         Hi.
> >         Capacitors are for filtering voltages, and specially filtering
> > due to sudden currents inside chips on normal functioning. However,
> > they are not too critical. Their use on timing functions is probably
> > almost null in today's mobos's oscillators.
> >         Don't worry too much (but worry about PB).
> > >
> > > Packard Bell POS P75, Win95a
> > >
> > > An article in "Microtimes" said that cheap motherboards use aluminum
> > > electrolytic capacitors instead of tantalum capacitors.  It said that
> > > the "cheaper capacitors which would degrade over time and no longer
> > > supply correct voltages and correct timing to the CPU."
> > >
>
>
>
> Sorry Javier, but I disagree with you.
>
> Capacitors have several different jobs depending on how they are used
> within the electronic circuit.  One of the jobs a capacitor can do is
> to filter out noise from a digital signal.
>

        On a good electronic design, the only function of capacitors should
provide filtering (lower the ripple) of the supply rails to the individual
chips, specially to the memory chips. The ripple would come from variations
in demanded current, due to normal functioning of the chips. And for that you
need capacitors with a good high frequency response (ceramic).
        One thing a novice electronic designer can do, if seeing a glitch
somewhere, is put a capacitor to kill it. What he should do is try to
improve his design so that the glitch doesn't appear.
        Another thing is noise due to large traces in the PCB near to
high frequency signals. That is not solved with capacitors, but with a
better PCB design.

> As motherboard bus speeds increase, the need for noise filtering
> becomes much more critical.

        In today's circuits, current consumption is more due to
switching logical levels; so faster, more current. So yes, you
need to provide good supply filtering all over the mobo.

>
> As for the article about aluminum vs tantalum capacitors....
> yes there is a quality and performance difference but it is
> very slight -- in most cases.

        No one would use aluminum for filtering, other than at
the connector.

>
> Remember the bottom line.... you get what you pay for.
> If company A sells a product for $2500, and company B sells an identical
> product for $1600, and company C sells it for $1200, that difference in
> price is generally due to the quality of the materials used in the
> product.
>
> Will the higher priced product outlast the lower priced product?
>      almost always
> Will the higher priced product function better than the lower?
>      maybe, but not necessarily
>
> --
> Jim Meagher
> =====

        What should be considered is the use of the equipment. Say you
buy the cheapest case: probably its main switch is as cheap as posible,
so try to switch it as less times as possible. And for the mobo, on
those with the best names, probably all things which should be considered
in design (capacitors, PCB layout, quality control) are as optimal as
posible.

************************************
Javier Vizcaino. Ability Electronics. [log in to unmask]

Starting point:        (-1)^(-1) = -1
Applying logarithms: (-1)*ln(-1) = ln(-1)
Since ln(-1) <> 0, dividing:  -1 = 1
     (ln(-1) is complex, but exists)