Core i7 4960X Ruined During Delidding Attempt

After a number of horribly failed efforts at lapping back in the day, I can sympathise.
 
if it's soldered on.. he just took one for the team. Props to him. heh Now it should give fair warning to the next guy that tries delidding one of those.
 
the info is incorrect, he did not rip the die off, he used heat gun and accidentally de-solder the SMD capacitor, the die is fully functional and survived.
 
Reminds of of that "Now you know" commercial.

奇怪 已經繞了好幾圈了 怎鐵蓋還拆不下來!! ㄇㄉ 熱風槍伺候!!
 
reminds me of when i de-lidded my opteron 165...and it worked! dropped the temps around 8 C if i remember right...under water :)
 
Makes me glad my i5 3570k was a breeze to rip off. The temp change is totally worth it for water setups
 
the info is incorrect, he did not rip the die off, he used heat gun and accidentally de-solder the SMD capacitor, the die is fully functional and survived.

This makes sense; the pic looks like a solder blob.

It looks like the way to remove it is to cut around the edges with tungsten wire, and then use a heat gun.

I'm running 4.7GHZ on my i7-3930; it hits 60C in euler3d on the 6 pipe heatsink I've got on it.

I can't see going to the extra hassle, personally...
 
everyone i ve seen delid cleans the top of paste.... kinda weird....

It's solder, not paste.

He is talking about the paste on top of the heat spreader...

When Sandy went to Ivy and then the new architecture of Haswell, most of the empty space from the shrinkage went to on die GPU. Since Ivy-E has even more dead silicon and empty space compared to Sandy-E, hopefully it will over clock better than either Sandy-E or Ivy.

What do you mean dead silicon and empty space? Sandy-E/Ivy-E dont have the GPU at all. Its simply not there. It should probably overclock pretty well as it's going to be a fairly large die and it appears to be soldered. I wish I could justify a reason to upgrade my 3930k to an Ivy-E.
 
Glad intel wised up and brought the solder back. Sucks this guy was dealing with a pre-release cpu and had to find out the hard way.

On the up side, i doubt he purchased the processor. Maybe Intel will bring back the solder on models of their cpus, They certainly need to.
 
*cringe*

Why do I see intel not sending him samples in the future?
 
First thing I said to myself is, " Hey that looks just like the old un-lidded Thunderbird/Durons I used to play with".

Once or twice I even cracked a core ever so micro an amount. Dead.
 
"Just according to keikaku"

avatar42163_8.gif
 
First thing I said to myself is, " Hey that looks just like the old un-lidded Thunderbird/Durons I used to play with".

Once or twice I even cracked a core ever so micro an amount. Dead.

What are you Talking about? Thunderbirds don't have lids. How can something be delidded if it never had a lid in the first place?
 
This makes sense; the pic looks like a solder blob.

It looks like the way to remove it is to cut around the edges with tungsten wire, and then use a heat gun.

I'm running 4.7GHZ on my i7-3930; it hits 60C in euler3d on the 6 pipe heatsink I've got on it.

I can't see going to the extra hassle, personally...
Sandybridge-E uses solder instead of thermal paste so there would be little advantage to delidding. The regular Ivybridge and Haswell CPUs use thermal paste between the die and the heat spreader, which leads to increased temperatures (sometimes quite a bit so). That's why they were delidding this - to see how Intel joined the heat spreader and the die. It appears to be solder and that is good news.
 
That guy kicks ass. Hope Intel doesn't use solder on that chip shen it is released. Bare die access for cooling pleasure is a gift.
 
That guy kicks ass. Hope Intel doesn't use solder on that chip shen it is released. Bare die access for cooling pleasure is a gift.
I disagree. I'd rather have solder and not have to worry about delidding to get acceptable performance.
 
Back
Top