Does a CPU "Burn in" work anymore?

ps2cho · Feb 4, 2010

I know some people had luck with it back in K7/K8 AMD days...I did not own an Intel rig back then..

So what about now days? What do you guys think?

Basically run Prime95 for 24hours+ which causes a drop in temperatures.

This is in regards to my i5-520M in my new Sony Vaio laptop. I just replaced the TIM on it and for some reason I am getting 1c higher than with the stock junk. Maybe it needs some time to cure? I would have thought it would have done better right off the bat though

pejsaboy · Feb 4, 2010

I'd be interested in the consensus as well. I wouldn't mind trying to trim some vcore and/or increasing a little more fsb if it seems to be a viable option.

jason4207 · Feb 5, 2010

A lot of times the factory junk is thick, and the mounting mechanism isn't that tight. This works OK together, but if you want to improve the situation you run into a little problem. You replace the thick thermal junk with a nice thin coat of high-quality TIM, but now the mounting mechanism isn't able to cope with the reduced spacing and it doesn't hold it down tight.

So, it could be that the mounting mechanism needs to be modified very slightly to give more pressure.

Bobnova · Feb 5, 2010

Burn-in has been a dubious concept from day 1, personally i think the only reason it result in lower temps is that it was the TIM curing, not the CPU itself doing anything.

I agree with jason about the mounting stuff, i would take it back off and look at your contact pattern (take pics of the bottom of the heatsink and the top of the cpu right when you take it off, and post 'em) to see what it looks like.

reap3r · Feb 5, 2010

Yep, try adding more pressure to the mounting mechanism.. and while some TIMs do need some time to cure, I doubt it would have that big of an effect on your temps.. :shrug:

AEternal1 · Feb 6, 2010

the tim does take some time to fully cure, it took 3 days for my temps to reach their nominal, which was a whopping 3* difference.

ps2cho · Feb 10, 2010

Can you guys see how it works here?

I am not sure how to tackle either of the mounting mechanisms to tighten them. The screws are as tight as they can be...

Any ideas?

Bobnova · Feb 10, 2010

Bend the little metal tabs on the heatsink up slightly, that will make them have to bend more to be flat against the mobo, and hence put greater pressure on things.

jason4207 · Feb 11, 2010

Is that the TIM that you applied? Looks excessive.

EarthDog · Feb 11, 2010

Did it ever work? Only anecdoteally and possibly placebo...ly

. Like Bob mnetioned, in this context you are talking asbout the TIM curing. Burn in is a completely different concept...at least it is to me.

Burn in was, to me, always a bunch of horse hooey as far as lowering voltages on a CPU.

ps2cho · Feb 12, 2010

jason4207 said:
Is that the TIM that you applied? Looks excessive.

That is the stock paste...So you can see why I have the feeling the increased temperatures is due to a now excess gap.

I'm a little confused over bending the metal...It doesn't seem to me like it would do anything other than bend the metal. Once it has tightened down it will be in the same position.

ozzlo · Feb 12, 2010

I've always been very dubious of CPU burn in because Intel does thoroughly test each their chips with a variety of environments such as excessive heat/voltage...
Here is a related video i've found that briefly describes the process... (vid appears to several years old but is still relevant).

http://intelpr.feedroom.com/ModEmbedThisPop_dsp.jsp?nsid=a-d3b433e:126c09dd0c9:-6535

skip to 2:20 where they start describing their testing process.

I've actually seen a much older video which showed several boards capable of holding 96 processors each inside a giant oven type device it said that the chips were "baked" for 48 hours.

jason4207 · Feb 12, 2010

ps2cho said:
That is the stock paste...So you can see why I have the feeling the increased temperatures is due to a now excess gap.

I'm a little confused over bending the metal...It doesn't seem to me like it would do anything other than bend the metal. Once it has tightened down it will be in the same position.

With the tabs bent the screws should be harder to tighten down. That should result in more pressure.

Does a CPU "Burn in" work anymore?

ps2cho

Member

pejsaboy

Member

jason4207

Senior Member

Bobnova

Senior Member

reap3r

Member

AEternal1

Registered

ps2cho

Member

Bobnova

Senior Member

jason4207

Senior Member

EarthDog

Gulper Nozzle Co-Owner

ps2cho

Member

ozzlo

Senior whose title will pier

jason4207

Senior Member

Similar threads