At 03:04 24/10/02 +1300, Nathan Legg wrote:

>Yes, you are correct. I mean, (quoting myself:)
>"I would believe this, however, i dont see any regulators specifically
>powering the pci bus. All i see are "supply islands", which are powered
>directly by the psu. Hell, even your keyboard has a direct connection to the
>psu (for a demonstration, spill coffee over it).
>
>The only regulators that I see are for the CPU.

(I'm not picking on you here Nathan, I'm replying to the entire thread)

Umm,

There is no reason why the ADSL card can't include its own regulator on the
card. In fact, I would be very surprised if it didn't. Part of the job of a
(decent) regulator is to filter out any noise and hash that may be getting
introduced onto the power rail by other devices in the computer, and to
prevent your own noise and hash from getting back onto the computers supply
rail.

The "lack of power" argument against internal modems is simply bogus. Has
anybody bothered to measure the current drawn by a typical external modem ?
I'd be surprised if it was more than about 300ma at 12v, which is 3.6
watts. Even if it drew as much as 1 amp (my highest guess) you're still
only talking 12 watts. Anyone with access to a typical external DSL modem
and a multimeter care to measure it for us ? (Cisco 827's don't count as
"typical" here btw :)

The "noisy environment" argument has some merit, but not much. It's true
that if the card was unshielded then interference from the insides of the
computer could be a problem at the frequencies used by DSL, but it is
relatively simple for the manufaturer to put a simple tin shield around
most of the board to eliminate that problem _entirely_.

If they don't, and they get interference due to a lack of shielding, then
it is a DESIGN FAULT of the card, not a fundamental limitation of internal
cards. An external modem should have shielding inside the case too by the
way - to prevent the opposite - the modem radiating uneccesary garbage in
the AM broadcast band.

In case anyone wonders why I keep harping on about this, comming from an
Electronics background (Computers were only ever my secondary hobby,
unfortunately they turned into a job in recent years ;) I hate seeing
pseudo-technical explanations which have little or no factual backing
behind them being bandied around by people who are quite possibly well
skilled in the area of computers but somewhat shaky in the area of
electronics and radio. (and things like EMI and basic principles like
shielding and power supply filtering)

I tend to side with Craig's Winmodem explanation, as the same types of
connection stability problems occured with dialup winmodems vs hardware
controller based external modems. That, and the fact that internal modems
due to their cheaper price tend to have poorer design and quality in general.

But there is no fundamental reason WHY an internal modem must be inferior.
(Eg, someone could, if they wanted, design an internal modem that worked as
well as an external modem, its just nobody has done it yet :)

Regards,
Simon

-- 
This message is part of the NZ ADSL mailing list. 
see http://unixathome.org/adsl/ for archives, FAQ, 
and various documents. 
To unsubscribe: send mail to majordomo@lists.unixathome.org 
with "unsubscribe adsl" in the body of the message