Overclocking: sense or nonsense?

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

Threaded View

Why do people overclock their systems when a CPU hardly ever runs at
maximum clock frequency? My E6600's core speed is usually 1600 MHz
instead of the maximum of 2400 MHz. So why bother with overclcoking?
Just wondering.


Re: Overclocking: sense or nonsense?

Quoted text here. Click to load it


Some people do that as a "sport". If the CPU "could" be overclocked, it have
to be overclocked - just for the reason, because it "could" be overclocked.

The next fact is the benchmark point. "My CPU is faster than Yours" -
analogue to car motors.
If somebody gets 200 horsepowers out of his engine, there is another guy
somewhere who tries to get out 205 horsepowers. If the raise of power could
be recognized or not - thats not important. Important is the result of the
test block in the test room to show a buddy the victory of 5 more

In some cases there could be a little sense in overclocking - if You plan to
replace Your system soon and want to get the maximum power for a 3D-game per
If the CPU or the board is damaged, You will replace it with a bigger one.
So no matter if it is broken. I for myself would sell the old system to get
a little amount of money and then buy a bigger one.

But there is not the fun of reaching higher levels of power with the old


Re: Overclocking: sense or nonsense?

Carsten Beckermann wrote:
Quoted text here. Click to load it
I agree with that. The first time I overclocked my computer, it
was a 200 MHz Pentium. When I made it run at 266 MHz, it made a
very noticeable difference in performance for games and video
playback. But that was only part of the reason. The other part
was the satisfaction of making something perform better than the
manufacturer intended.

My last overclock was with a PIII 450 MHz o/c to 550 MHz. By that
time I already had a 1 GHz machine and would have pushed the
older machines even higher if the motherboards had provision to
do so, not really caring if the old CPUs got fried. I did
improvise an additional heatsink out of aluminium sheet for the
onboard voltage regulator of the P1 which heated up much more
than the CPU. It was fun.

Nowadays, having been on the mellower side of 50 for several
years, I no longer play demanding games. Modern CPUs give me more
computing power than I really need and I don't bother with
overclocking. OTOH, I still upgrade now and then just to keep up
with the times. I've just finished assembling an AMD Phenom II x4
920 with 4 gigs of RAM and a 1TB HDD. Go figure. :).

Re: Overclocking: sense or nonsense?

On Fri, 22 May 2009 14:11:17 +0200, Mark

Quoted text here. Click to load it

Hello Mark... I consider it totally nonsense.

I would never break my system (because sooner or later an overclocked
system will break) just because I want a couple of more frames in a

The suggested factory clock frequency keeps your system safe and let
you use the CPU for a long time.

http://ManuelMarino.com (Arts, Music)
http://NewOnlineShopping.net (Clothing, Gifts)
http://TechGamesBlog.com (Technology, Games)

Re: Overclocking: sense or nonsense?

On Fri, 22 May 2009 23:09:59 +0200, Manuel

Quoted text here. Click to load it

Overclocking only "breaks" something when you do it beyond
your capability or ability to recognize a problem and reduce
the speed to an acceptible overclocked level.

Like many things in life, only those who know how should do
more than basic things.   Failures result from insufficent
knowledge of what equipment is capable of long term, and how
to determine a stability level appropriate for the system.
There isn't actually a 100% guarantee a system at stock
speed is 100% stable either, any system should undergo the
stability tests an overclocked one must pass.

Quoted text here. Click to load it

Most of my systems have been overclocked.   The CPUs have
lasted for a long time, in perfect working order when the
systems were retired.  There are a lot of myths about
overclocking that aren't really true, a CPU will tend to
fail from extremely excessive overclocking but more likely
something else  happened like installation error in getting
the heatsink mounted properly, long term use with a failed
fan, or a motherboard's VRM capacitors failing due to user
error in having inadequate case cooling or manufacturer
error of selecting poor quality and/or defective capacitors.

Re: Overclocking: sense or nonsense?

kony wrote:
Quoted text here. Click to load it

Yeah, my old AMD64 system has been running 24 X 7 for years now. Can't
remember how many years but it has been going since an AMD64 was the newest
hottest processor around. The overclock is only around 15% IIRC but it has
been running with increased core and memory voltages since new. The only
actual problem with the system has been the DVD burner and a squeaking
120mm case fan. If one is careful there is no reason a reasonably
overclocked system can't be reliable and long-lived.

John McGaw
[Knoxville, TN, USA]

Re: Overclocking: sense or nonsense?

Somewhere on teh intarwebs Manuel wrote:
Quoted text here. Click to load it

I consider my girlfriend to be the most beautiful woman in the world.

Quoted text here. Click to load it

That is just plain wrong. Why answer a question if you have no knowledge of
the subject at hand? Just to see a reply of yours?

I've been overclocking my systems since I got my Pentium 90MHz to run at
120MHz. Since then *every* system I've owned that could be overclocked has
been overclocked. Some by 50% or more. Guess what? Not *ONE* system "broke".

Then again, overclocking used to be quite complicated, you really needed
your wits about you to get it right so I can understand why you say what you

These days it's simple. With my E7300 (for instance) I just went into BIOS,
changed the FSB from 266MHz to 333MHz and now the CPU runs at 3.3GHz instead
of 2.6GHz. No drama, no voltage or heat increase, it just works at the speed
of a CPU that would have cost me twice as much to buy.

Quoted text here. Click to load it

I bet you never turn your stereo up above 5/10 as well.....

"Build a man a fire, and he`ll be warm for a day. Set a man on fire, and
he`ll be warm for the rest of his life." Terry Pratchett, Jingo.

Re: Overclocking: sense or nonsense?

Quoted text here. Click to load it

Indeed it does - even in summer with no air conditioning.

Thus, for the rest of the year you can safely raise the clock speed and still
not overheat anything.   And if you like to have your AC on in summer . . .

*Careful* overclocking can let you to boost the performance of your PC *without*
risking much.

I've had a Celeron 1.6Ghz running at 1.9GHz for over a year with stock cooling,
and the CPU temps don't go over 55C.


Terry V.

Re: Overclocking: sense or nonsense?

On Fri, 22 May 2009 14:11:17 +0200, Mark

Quoted text here. Click to load it

Consider that unless this is the first PC you've ever owned,
at some point you bought it because it had higher specs,
higher performance, as it is seldom as economical to buy a
new system instead of repair a failed one.

If they had offered you a low end CPU or a midgrade CPU for
the same price, wouldn't you opt for the midgrade?  Wouldn't
the same be true if it were a choice of midgrade or high

Similar could be said for overclocking, while some people do
it for sport, others are simply getting more for less money.

The trick is to know the reasonable limits so you don't
shorten the mainboard lifespan, introduce instability, or
cause enough heat that higher airflow (more noise, more dust
buildup, and more fan wear) is required.

The other factor is that the reason your system seems fast
enough could be that for the moments when it is running at
full speed it makes you wait less.  It's the elimination of
waiting that is important, how fast it can go for the
moments it  needs to rather than the average CPU load level
over it's entire period of use.

Suppose you're decoding (watching) an HD movie.  Suppose the
average CPU utilization were 45%, but during complex high
movement scenes the playback is jerky for a second over and
over again throughout the movie because the CPU wasn't fast
enough.  If it happens, you could replace the CPU, or you
could click a half dozen times in a bios to get a free few
dozen % more performance... making sure to test stability
before relying on it for anything important, personally I
like to find the highest speed (within the parameters
desired) it tests stable, then downclock from that ceiling
speed to ensure a greater stability margin.

Lastly, since an overclocked system is faster, it remains
viable for longer into the future.  Take a system a few
years old running an Athlon XP @ 1.6GHz for example, with an
ample amount of memory and HDD it's still fairly sluggish
running Vista.  Up the CPU to 2.2GHz or so and the
improvement is noticed.

Re: Overclocking: sense or nonsense?

kony wrote:
Quoted text here. Click to load it
... snip ...
Quoted text here. Click to load it

However any overclocking is always giving up safety margins.  When
communicating with some peripheral that originally worked, you may
now occasionally miss a bit, for example.  

As you may conclude, I am opposed to overclocking.

 [mail]: Chuck F (cbfalconer at maineline dot net)
 [page]: <http://cbfalconer.home.att.net
            Try the download section.

Re: Overclocking: sense or nonsense?

CBFalconer wrote:
Quoted text here. Click to load it
One thing I learned a long time ago, that each 6 degree Celcius
temperature increase halves the predicted lifetime of silicon.
So by overclocking a cpu, parts of the chip will have a shorter
live, even when the general temperature stays the same by
better cooling.

Re: Overclocking: sense or nonsense?

On Sun, 24 May 2009 05:58:11 +0200, Sjouke Burry

Quoted text here. Click to load it

Consider an example, one with a CPU close to the era of the
one in a system you have.  Consider a Celeron 800MHz versus
a Pentium 3 1.1GHz.  They are essentially the same CPU,
except the Pentium 3 has double the cache size and at the
factory Intel set it to use a higher bus speed which is only
a matter of which pin connection tells the motherboard the
speed to use.

If you set them both at the same voltage, but set the
Celeron to a 133MHz front side bus like the Pentium 3 uses,
you have a fairly significant performance increase
overclocking it, but may not have any lesser margin with it
than with the Pentium 3 at stock speed, since it is nearly
the same CPU running at the same resultant speed.

It may reduce margins, but the margins may be raised by
increasing voltage.  Personally I agree margins are
important so whenever I find the upper limit, I reduce the
speed below that, and never keep the voltage set to the
absolute lowest it tests stable running with, and do quite a
lot of  testing.  

Even then, overclocking is suited for systems that are not
mission critical, where if there were a rare error, it is
not a big problem.  To some extent I feel the same way about
most PCs today, built with consumer grade parts, not using
ECC memory, running software that is not fault tolerant and
prone to viruses.

Quoted text here. Click to load it

If done in moderation, overclocked CPUs have historically
lasted longer than the viable lifespan of the system,
certainly longer than parts like fans, motherboard or PSU.  

Generally I like to have lowest noise, dust buildup, and fan
wear possible, even if the dust is collecting on intake
filters I'd rather not clean them twice as often if half the
airflow will suffice to keep a system stable and long
lasting.  Towards this end I find the shortest lived parts
are the capacitors.

Even when setting fan speeds low, with use of a good enough
heatsink that the CPU temps always stay under 60C, I have
had no problems with CPU failures from overclocking except
once years ago when voltages were set by jumpers and I had a
slotket with a jumper diagram I read upside down, so I set
far far too high a voltage.

One other time a CPU became thermally distressed (though it
still works fine two years later) because it had low quality
thermal grease on it that the liquid portion had pumped out
of leaving dry solids. The fault was the thermal compound,
as I have not had the problem again after abandoning use of
that batch of compound, but it could be argued that since
the CPU was overclocked and thus producing a little more
heat, it may have pumped out the grease a little faster than
it otherwise would.

A lesson learned from this is an overclocker should do what
anyone seeking to build a quality system would.  Use a high
quality heatsink (need not be expensive if the overclock is
not extreme), high quality heatsink grease, at least medium
quality motherboard and PSU.  If one buys a low end PC Chips
motherboard and $15, 500W PSU, the parts themselves are
simply unfit for oveclocking, though they would be unfit for
running a high-end CPU at stock speed too.

Glancing over at a system I have had running 24/7 for years,
it has an Athlon XP overclocked to 2.2GHz.  It must have
been built in 2004, that is the date of the older folders
created on it's hard drive (though I think I have cloned the
original drive onto a larger, newer one since then).   It
has ran overclocked since then, 5 years already.  It runs at
about 50C under load, 40C idle.  I think it would run cooler
if I had cleaned the dust out of the heatsink within the
past year or two.  :-)

Either way, I have no complaints about it lasting 5 years
already and still working fine.  It's value now is quite
low, the only reason it is desirable that it keep working is
the time it would take to set up a different system to fill
the same role.

Site Timeline