Debugging a system freeze?

[William Brown]

> ...even when the hardware is MEANT to be overclocked?

When a cpu is made, they test it's ability to run stable at the max
frequency of that die. If it passes, it gets shipped at say, 3.8Ghz. If
it fails, it gets tested if it runs stable at say 3.4. If it passes,
great, they lock it to 3.4 and ship. If not, they slow it down again, to
say 3.0 ghz and test etc etc etc.

So really, these chips are running at what the manufacturer, intel or
amd, who hire some of the worlds best electronic engineers in the world,
determines to be a stable and fast clock rate. They may be able to
achieve "a bit more" performance, but they would rather produce a high
quality working product, than an unstable broken one.

Now comes along someone who thinks "Yeah man, I can over clock and get
more bang for my buck!". And intel's or amd's marketing managers say
"Yes, we can milk them for more money by letting them 'overclock' our
hardware, all we need to do is flag a few registers on the same die, and
charge more for it!". Such is overclocking in this day and age. What now
do you have to show for your overclocking effort? A damaged cpu die?

What must be pointed out is that anyone who overclocks is assuming they
know more about cpu manufacturing, design and operations that intel's or
amd's best engineers. I beg to differ.

-- 
Sincerely,

William Brown

pgp.mit.edu
http://pgp.mit.edu:11371/pks/lookup?op=vindex&search=0x3C0AC6DAB2F928A2



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 945 bytes
Desc: OpenPGP digital signature
URL: <http://lists.fedoraproject.org/pipermail/users/attachments/20120619/f25457c7/attachment.sig>