Maybe it won't explode...

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View


Well, I decided to do some detailed investigation into the alleged
"Extreme Overclock" video show a chip exploding from overclocking that
appears to be hoax, and to try and find out what would REALLY happen if
one tries to overclock and overvolt a chip that much.

I just dug up this:

Notice this:

"Hosting a Counter-Strike server can be very demanding on the CPU
though. I
was determined to get the K6-2 up to 600 for the LAN party on the
weekend. I found out how to get my motherboard to supply the CPU up to
volts. The motherboard voltage regulators were burning hot at that
and the CPU was practically on fire. It was overheating like mad. I
I had to remove the heat plate to get rid of the cheap MAD goop on the
inside and get the core right on the heatsink. Got a razor blade and my

trusty computer hammer (grins wickedly) and started hacking away at it.
minutes later the heat plate was off, as I thought I may have hit it a
too hard. I put her in and fired her up, POSTed great, but was still
overheating at 3.5v 600 MHz. It would run for about 15 minutes full
before it would overheat and lock up.  Since it lasted that long I
it makes just a little more heat than the heatsink can dissipate.

I rigged up another fan to the side of the heatsink to get more airflow
also cool the motherboard some. I thought "This is is, 600 MHz here I
I fired her up, unknown to me, for the last time. The screen was black,
I got no response from the system. In an instant I lunged for the power

switch. I was too late though, I now knew the acrid smell of a fried
glancing down I saw smoke gently rising from below the heatsink. The
fan I installed had gotten cought on a capacitor next to the CPU socket
it appeared it was seated properly, but it was jsut a hair above the
It took 3.5 volts at 600 MHz with no heatsink to kill her. That's 1.3
over spec and 150 MHz over spec. She was dead before she knew what hit
she felt no pain. Ironically just as I fried it, the Iron Maiden song
the Good Die Young" started playing. I just burst out laughing at the
it took the loss out of the occasion. We ended up using the Athlon 500
our server at the LAN party that weekend. "

That's just half a volt below 4V like in the "APD Duron Vaporizing"
video, and there was no explosion, it just burnt up, even without the
heatsink. So I doubt that even at 4V it would explode. Ie. the video is
definitely faked. Not to mention that the explosion and fire comes from
*beneath* the chip:

and not from the die where you'd expect it to if this was actually
caused by it overheating catastrophically. Interestingly, a second,
smaller chip on the board is seen to blast out (much easier to see if
you step through the video frame by frame), also suggesting an energy
release beneath the board (like from a hidden firecracker or other
explosive device), kicking out components due to the shockwave. I
didn't notice this until just now. In later frames you can see it
resting on the ZIF socket handle, also forced open by the explosion (it
looks like a small "comb-like" thing (the pins) perpendicular to the
top of the handle. I didn't know what that was until now when I
examined it...).
So it seems likely that this was done using an explosive of some sort,
probably a small firecracker.

Also, by the way, what _exactly_ is burning when a CPU is destroyed
with it smoking and making a foul smell? What is there in the die that
could burn (the smoke comes out of the die)? And, what would such a
wrecked chip's die (the actual circuit) look like under an electron

re:Maybe it won't explode...

wow shit :) ;) :)
regards, One_shot_weasel

Site Timeline