Is the ATX12V 2x2 power connector installed ? In section 1.6 of
the manual, there is a picture of how a 2x2 should look, when plugged
into the 2x4. Verify your yellow wires (+12V) are oriented the
same way as the picture in the manual. (I'm using the downloadable
manual from the web site.)

You cannot get a POST beep unless the processor is powered.

Also, the PCI Port 80 POST display (two digit display), will rest
at 0xFF or 0x00 at powerup (hardware-wise, it has to have
some known initial value). If the processor is not functional,
the code will stay at that value. So rather than "FF" meaning
Fully Functional, it means Function Frozen :-)

In the distant past, when an couple models of Antec supply,
were connected to a number of different Asus motherboards,
the Vcore regulator would shut down, due to a timing problem.
I didn't get a full explanation at the time, as to what
the exact reason was. But it almost seemed the Vcore was
shutting down on overcurrent, because it was taking so long
for the 12V rail to reach full voltage. So that is another
failure mechanism, only with no feedback other than that a
different brand of ATX supply would work OK.

There is also a small possibility of a bad processor. In
which case, swapping something like a Sempron or other
lightweight substitute, might identify what is broken.

Pulling the RAM, may give the BIOS a chance to start. I think
I've had one motherboard that was silenced, by a stick of
RAM that was bad below 640K. You'd get a beep code
with no RAM present, and the POST display may advance past 0xFF.
That tells you the processor is running.

In the reviews here, someone installed a 9850 and the board
worked OK. Someone else installed an X2 6400+ and the Vcore
regulator went up in smoke (see ref to "PWM"). There was an
article on Anandtech, where they ruined a few AM2 boards before
figuring out the boards were limited to 95W processors. There
is no CPU Support chart that I could find on the Abit site,
so no way to see what the manufacturer claims. The short form
spec on the Abit site suggests the board takes Phenoms, but
whether it takes the top end ones, is something you normally
resolve by looking at their detailed "tested" list.

http://www.newegg.com/Product/ProductReview.aspx?Item=13-127-046&SortField=0&Sum
maryType=0&Pagesize=100&SelectedRating=-1&VideoOnlyMark=False&Page=1

(There is a manual link on this page.)
http://www.abit.com.tw/page/en/motherboard/motherboard_detail.php?pMODEL_NAME=AX
78&fMTYPE=Socket%20AM2

ftp://ftp.abit.com.tw/pub/download//manual/english/ax78_v100.zip

After that nasty incident, Asus responded by boasting about
the newer boards they made, rated at 140W. Since many
manufacturers made substandard boards in the same "wave"
of product releases, I have to conclude AMD told them to
do it (make lightweight Vcore regulators). Later
boards corrected this stupid idea. This is Asus cleaning
up their act.

http://event.asus.com/mb/140w/

When you find an AM2+ board, with 4 phases, the actual phase
configuration is 3+1. On an AM2+, like a Phenom, the core
is powered by three phases, and the memory controller interface is
powered by the remaining phase. It means effectively, that
most of the power is coming from three phases, which is why
these things can fry. (Yes, there is a regulator next to the RAM,
but that powers the RAM sticks. The memory controller on the
processor, has its own supply, and the phase is located
in the Vcore area on the motherboard.)

http://c1.neweggimages.com/NeweggImage/productimage/13-127-046-05.jpg

As far as i know, Abit (UAbit) is stopping motherboard production
at the end of the calendar year. Based on the comments in that
Newegg listing, parts of the company were already broken
(poor tech support, no rebate delivered). So the writing is
on the wall. You could well be looking at the purchase of
another board, if it turns out to be a motherboard problem.
UAbit will still be around, just not making motherboards.
While there is a promise of warranty support for three years,
if you cannot get through to them, that would be hard to arrange
(some companies do stuff like that, "pull a paper bag over their
heads" so they cannot hear from users).

http://channel.hexus.net/content/item.php?item=15225

Good luck,
   Paul

 Meanwhile an engineer who even designed power supplies says
otherwise.  Dave’s only recommendation is to keep replacing things
until something works.  Circuits such as power supply filters and
other specifications mean his mythical power drop cannot (and does
not) happen.  But then Dave has no solution other than to keep
replacing parts on “it could be this or could be that” speculations.
Dave fears what he does not understand such as multimeters.   We
sometimes get a tech who does that same thing.  Either he learns how
to solve problems or he is quickly unemployed (as Dave would be).

 A multimeter is the only tool (other than $thousand equipment) that
can report a power supply - definitively.  How many others only post
speculation or personal attacks rather than a solution?  "It might be
this or might be that" is called shotgunning.  If does not work as
demonstrated by the next sentence.

Normal is for a computer to boot with a defective power supply.  And
normal is for the meter to identify that same supply as defective.
How would I know that?  Our techs routinely demonstrated reality using
a multimeter.  Sometimes a complete system failure is averted by
identifying a slowly getting worse power supply.  Only the meter could
identify the defect.  After all, a computer may still boot with a
defective supply.

 How important is a meter?  It is a first tool that even an auto
mechanic students get and use.  Even car mechanics need a meter to
find and fix simple problems.  Even auto mechanics would quickly be
unemployed if they used Dave’s diagnostic recommendations.

 Until the many components of the power supply ‘system’ are verified
(a minute of labor), then no other labor will be productive – generate
‘definitive’ answers.   Yes, other good parts can act defective if a
power supply system is not first identified.  Ie. Paul’s ‘pull the
ram’ beep test.

 Ignore the insults and claims based only in 'it could be this'
logic.  Get the meter for replies from the minority who actually know
how computers work.

Of course you have to be careful so you don't do any harm.  :-|

You didn't know?

Wow.