Burn In: What, Why, how?

[HansSvetty]

I read that I should "Burn in" my new Athlon 1000 before overclocking.

How is this done and for how long?

I guess after that, I raise the multiplier and FSB comes last.

Is it ok to raise the multiplier 'til I hit the wall, or do I raise it bit by bit over time?

Thanks,

hans

 

[nealh]

You are going to get alot of opinions..but burn is not real...if you ask the real overclockers like compuwiz1....I have overclocked 3 processors and have seen no burn in effect....

If the processor is going to overclock it will ...some will say after running the chip at higher voltage for period of time I am able to increase spped and/or lower voltage to get better or same result..probably more like the the thermal compound is more eveny distributed allowing better heat transfer and improved performance and stability

let the bashing begin....

 

[HansSvetty]

Good info, thanks!

hans

 

[nealh]

Personally in my limited experience with 2 intel chip..piii700@945 and p450@630 and my AMD TBird..if the chip is unlocked you can adjust on many mobo both the multiplier and fsb..
1.2ghz + are unlocked(? 1.4 based on 200fsb some reports are these are locked)...

I just plug it in..boot the first at default to make everything is working and then try to push to the limit.
Most the 1ghz are locked so unless you unlock it..by connecting the L1 bridges with a defooger kit or conductive pen, lead pencil..you will only be able to adjust the fsb..

So increase the fsb..many of the 1ghz with an axia or avia have hit 1300-1500..you may need to increase the voltage in order to get the chip to run..interestingly many have reported that an increase in voltage did not help and increased problem....it will be safe to go up to 1.85v

You can try at default and see if it posts if not ..increase the voltage if it loads the OS and is stable(you can use Prime95, cpu stablility test to test it)..then try higher..

Note increasing the voltage and or fsb will increase heat and decrease the life of the processor...

 

[HansSvetty]

Nealh,

Some good advice. Sounds like a plan.

Appreciatively,

hans

 

[nealh]

Good luck if you need help keep asking ..the people on these forums are terrific help..you can also pm if you like

 

[kgraeme]

Real burn in is just a test of stability at whatever speed you do it at. The principle is that if you work the processor really hard for a continuous stretch, say 48 hours, and it doesn't crash, then it's not going to crash on you in daily use.

We used to do this on the bench all the time with any component. We basically worked over the CPU, drive, monitor, whatever and if it failed we got a replacement right away. If it passed, then we kept it.

Burning in to "set" the speed of the CPU is a myth. At least as far as I've seen.

 

[pen^2]

the only thing burning-in could do as far as overclocking potential goes, is just the matter of thermal interface material settling more intact with the cpu slug or something to that effect depneding on the type of the chemicals used. largely burning in does nohthing tho

 

[aircooled]

Never done it. Never needed to, Never affected my OC. You don't have to burn in your non-oc system. So why burn in an oc chip? If you buy a Duron 650 or a Duron 800, they are the same chip with different multipliers, so why should I burn it in...

 

[nicowju]

I've never burned in a processor... not my Celeron 300A, not my Celeron 366, not my Celeron 400, not my P3-700's, or my P3-600's, not my 600MHz Duron, or my 1.2GHz T-bird

 

[ErikaeanLogic]

So far, every 1GHz+ T-Bird I've owned has given up an extra 100MHz after 2 weeks of intensive use. Believe me, I try to squeeze out every drop of performance on day 1 with these chips, I give up, I try 2 weeks later and voila!, I get an average of 100 extra MHz. I would say this effect is due to an x-factor I call "burn-in"; I can't scientifically explain it but I do scientifically observe the phenomenon in every one of these cpu's.

ex: 1GHz@1.4-->1.5, 1.2GHz@1.4-->1.5, 1.33GHz@1.5-->1.6

BTW, I use a few different stages of stability testing. I begin by looping the UT fly-by for over 24 hours. That'll usually weed out systems with overextended settings pretty quickly. I use UT because that's what I play and it seems more sensitive to system instability than most other games. Next, I loop 3DMark2001 and look for hops, skips, and locks. After that I'm usually pretty satisfied. If I suspect something's up with the system and UT and 3DM2K1 haven't picked it up, I'll run Prime95. A word about that: Prime95 is not optimized for Athlon cpu's and can make a system fail which would otherwise be stable; I just use it because it'll cause a flaky system to crash within 15-20 minutes.

Just my $0.02.

 

[DieselMan]

Erik, how interesting; I go with the same game/benchmarks for stability UT is what I start with overnight, and then I loop 3dmarks2001 and Prime 95!

 

[ErikaeanLogic]

Well, David, it seems that great minds think alike!

 

[cirrus1]

How do you make the UT fly-by loop forever? Is it a switch or do you use a seperate program?

I also can't seem to find a link to download Prime 95. I've got an old version so I want to download the newest

<edit>NM I found the answer to it myself </edit>

It's the demo that begins the game that loops by default and Prime 95 can be downloaded from http://www.mersenne.org/freesoft.htm

 

[mp3turbo]

hello,

for stability testing, I always use RAR among others (win2000 installation, photoshop6.01 LARGE image manipulations so that memory gets tested too, 3dmark2001, quake3, truck racing, black&amp;white). I'll pack 4-5GB of data to some archive, it takes long time with 100% cpu activity, and then continuously check that archive for CRC errors in night - little .bat will do the trick:

:starttest
rar t name.rar
goto starttest

if overclocked system can survive this Photoshop6.01/RAR torture, I'll declare it stable. My bro's beloved passages are those games

bye, mp3turbo.