2600K Lapping?

What up guys, Recently I read an article about CPU lapping,
Is this any good?
I guess you will decrease the lifetime of your CPU?
Or if you do this good and safe, it's okay?

You wont decrease the lifetime of your cpu, most likely you will just waste your time and void the warranty.

If you have accurate means of measuring intake ambients , ie calibrated probes capable of 0.1C relative/repeatable accuracy, directly in front of fan for air cooler or rad, and can repeat same load temps within 0.1 on same mount, and about .5C on different mounts (though average multiple mounts back to about 0.2C avg difference) ....then repeat same after lapping, it is unlikely you will see any difference, other than mounting error variation.

Almost all cpus will have a slight convex shape to IHS, and waterblocks the same convex bow, which will flatten upon mounting pressure and give the ideal slightly more pressure in center mount.

I lapped both my i950 and i940. I use dallas 1 wires probes with crystalfonz interface to measure air temp of each fan intake on rad and water temp to 0.1C relative accuracy, and did multiple measurements/mounts before lapping my i950 and i940, and multiple measurements after. The average before and after temps were within 0.1-0.2C on both, ie within error of testing, ie no difference.

That being said, I dont care about intel warranties, but frankly lapping IMO is just a waste of time, unless you have very rare cpu with a deformed/concave IHS.

The cooling solution is going to remove the same amount of heat, granted. That's what you measured, rge. The purpose of lapping is to attempt to reduce the temperature difference between the thing generating the heat (CPU) and the thing removing the heat (heatsink/waterblock).

To clarify this point, it's a bit like the difference between air cooling and water cooling. The heat dissipated by the air cooler or water system will be the same in either case if clockspeed/voltages are kept constant, but the component temps are typically better with a quality water cooling setup.

So the real question isn't did your cooling solution remove the same amount of heat before/after lapping (which is what you measured by taking air and water temps), but did the load temps of your CPUs change measurably/significantly after lapping?

Woah, thanks for the responses guys!
I'm gonna do some research for sure!

My question is WHY? Like was mentioned above it voids the warranty. I wouldnt imagine that you are at a point where every degree C counts. So why bother wasting the time and voiding the warranty for a degree or two...

EarthDog said:

My question is WHY? Like was mentioned above it voids the warranty. I wouldnt imagine that you are at a point where every degree C counts. So why bother wasting the time and voiding the warranty for a degree or two...

Click to expand...

QFT. Unless your overclock is limited by chip temps and you've already exhausted other reasonable options to reduce those temps, lapping shouldn't even be on your radar.

JeremyCT said:

QFT. Unless your overclock is limited by chip temps and you've already exhausted other reasonable options to reduce those temps, lapping shouldn't even be on your radar.

Click to expand...

That's a good point, I thought it would be funny to have some better temps without even replacing my cooling.

Well, it might work. Then again, it might not. If you're aware of the risks going into it, are confident you can do a competent job, and don't mind losing your warranty, have at it and let us know how it goes. I'd advise you to look for a better thermal paste or better heat removal device(s) first though. Better fans, better fan placement, better heatsink, that sort of thing. Maybe just reseat the heatsink on the CPU. There's lots of things you can do to reduce temps that won't void your warranty, and some don't cost much $$s at all.

Thanks sir, not that my temps are bad at all! but you know, some improvement should be cool.

JeremyCT said:

The cooling solution is going to remove the same amount of heat, granted. That's what you measured, rge. The purpose of lapping is to attempt to reduce the temperature difference between the thing generating the heat (CPU) and the thing removing the heat (heatsink/waterblock).

To clarify this point, it's a bit like the difference between air cooling and water cooling. The heat dissipated by the air cooler or water system will be the same in either case if clockspeed/voltages are kept constant, but the component temps are typically better with a quality water cooling setup.

So the real question isn't did your cooling solution remove the same amount of heat before/after lapping (which is what you measured by taking air and water temps), but did the load temps of your CPUs change measurably/significantly after lapping?

Click to expand...

Rereading my post, it wasnt very clear, you had to read into it I was talking about ambients one minute then core temps the next. My point was you can not measure any change in core temps accurately, unless you measure/control for intake ambient temps accurately, not to mention other errors/mounting variance, etc.

What did not change at all before and after lapping was the core temps with exact same load over time, averaged over same time... core temps were within 0.1 to 0.2C difference before and after lapping.

I see people claiming they checked their temps on their present mount 1x, then lapped their cpu and get X better temps and claim lapping works. But they fail to account for few C difference in intake ambients, ie dont measure them directly, dont take into account different mount pressure, tim amount differnce, etc. So left wonder if just better mount, ambient temp variance, etc.

Edit: actually found my post on from 2 years ago on xtreme where I did test, post 3674. I had just finished testing tim mx2 vs mx3, had 8 mounts of mx2 all 71 +/- about 0.5C. then did 2 more mounts got 71C avg both times on all cores. Then lapped cpu as described, and retested with mx2, got literally exactly same temps (again core temps are actual core temps corrected for any 1-2C change in intake ambients, and double checked ambient variance with water temps. since it was second cpu that i got exactly 0C gain with using strict measurements, had ran it only once. But then went back to tim testing as if I never lapped it, and data on both mx3 and mx2 same before and after lapping, so had at least 5 mounts after, and probably 10+ before.

I wont waste my time lapping any more cpus, unless IHS is visibly really screwed up, and would likely just rma if it were that bad.

Back in the day lapping was the thing to do. Now adays your more likely to make temperatures worse by lapping the IHS. Heatsink might help depending but not the IHS.

JeremyCT said:

Well, it might work. Then again, it might not. If you're aware of the risks going into it, are confident you can do a competent job, and don't mind losing your warranty, have at it and let us know how it goes. I'd advise you to look for a better thermal paste or better heat removal device(s) first though. Bettier fans, better fan placement, better heatsink, that sort of thing. Maybe just reseat the heatsink on the CPU. There's lots of things you can do to reduce temps that won't void your warranty, and some don't cost much $$s at all.

Click to expand...

My 2 ¢: sounds like a cost-benefit analysis. The risks of lapping vs a negligi

JeremyCT said:

Well, it might work. Then again, it might not. If you're aware of the risks going into it, are confident you can do a competent job, and don't mind losing your warranty, have at it and let us know how it goes. I'd advise you to look for a better thermal paste or better heat removal device(s) first though. Bettier fans, better fan placement, better heatsink, that sort of thing. Maybe just reseat the heatsink on the CPU. There's lots of things you can do to reduce temps that won't void your warranty, and some don't cost much $$s at all.

Click to expand...

My 2 ¢: sounds like a cost-benefit analysis. The risks of lapping vs a negligible difference. Seems un-worth it.

Lapping is not worth it in your case. On a 95W CPU that's already not running hot nor is it at a really high OC/OV the lapping to get every degree out is rather pointless. It would void out the warranty as well 100%. It's doable, but IMO it's not for you at the moment. BTW no way we have the same setup! 2600K/MIVE/580 What are the odds. Well pretty good but still. Coincidence.

Agreed with the rest here that lapping the CPU is really not worth it.

Hmm, 2600k/MIVE/580(s) club?

rge - thanks for the explanation. I was thinking it was really odd for you to do faulty testing. I had to read your post a few times to make sure I wasn't missing something. Turns out I was of course, but at least it wasn't my fault this time.

Janus67 said:

Agreed with the rest here that lapping the CPU is really not worth it.

Hmm, 2600k/MIVE/580(s) club?

Click to expand...

Sounds cool, hijacking my own thread