[COLUG] My weak understanding of _Hardware_

David McGlone dlmcglone at columbus.rr.com
Tue Nov 28 11:47:43 EST 2006


On Monday 27 November 2006 13:44, Jim wrote:
> David McGlone wrote:
> > If it was the power supply, most likely you would not get power to the
> > fan.
>
> I've learned over the years to check the power supply rather early on
> when debugging systems with problems. Failing to do so can waste
> much time chasing symptoms caused upstream by the power supply.
> Spend the little bit of time it takes to measure the voltages.
>
> > From your description of the problem the computer is not doing the POST 
> > so it sounds like you have a faulty BIOS chip.
>
> There are many many things that have to work right to have the
> BIOS run. The failure of any one of those many things can
> cause the failure of the whole system. It is too early to
> determine that the _likely_ cause is the "BIOS chip".
> By the way, by itself, the term "BIOS chip" is sloppy.
> It could be the memory chip that hold the BIOS code
> (as in "flashing the BIOS"). It could be the battery
> backed-up RAM that holds the BIOS settings.

I forgot he said he pulled the RAM off the board. With missing ram he should 
have got a beep that continued for as long as the system had power applied to 
it.
 
I guess if you want to be "politically correct" and not so "sloppy" I should 
have referred to it as the CMOS chip or ROM chip.(whichever you want to refer 
to it as) The "battery backed up RAM" you are referring to is NVRAM 
which is sometimes still referred to as the "CMOS chip". The NVRAM stores the 
BIOS  code, but the code on this NVRAM, is what I would say "flashed into the 
CMOS". The reason for this is because if the battery goes dead and all power 
is lost to the NVRAM the actual CMOS resets itself back to the factory 
settings which are stored in the CMOS itself. Furthermore not all computers 
have NVRAM. So if it was what you refer to as "the memory chip that holds the 
BIOS code" which ranks just as bad as or worse than "BIOS chip" the system 
would still boot because the CMOS would reset itself.

Moreover, not all ROM chips were flashable only EPROM, EEPROM or Flash ROM 
(Again whichever you want to refer it as) was flashable. In the early days 
ROM chips had to be replaced if it went bad or someone wanted an upgraded 
version to support new innovations.

Finally, assuming the power supply is good, and there are no bad caps and no 
cracks in the board, and the RAM is there, there are not "many many things 
that have to work before the BIOS runs" unless you are counting each and 
every capacitor on the board that carries electricity to the CMOS or ROM 
(whichever you want to refer to is as) that contains the BIOS code which is 
the first thing to receive power, which in turn runs the POST which in turn 
checks the system RAM, Keyboard, Mouse, HD, CD-ROM etc are working and are 
present.

Conclusion, If he would have got a long beep after removing the RAM the next 
logical step would be to check the CMOS chip.
-- 
David M.


More information about the colug432 mailing list