Arctic MX-5 vs Indium Foil DIY thermal pad

It is, but the retention mechanism does deform the IHS. There is an active thread about this being an issue with LGA 1700.
I haven't seen the new ihs but normally they are a pretty thick piece of metal. Near impossible to bend as they have ridged edges. The silicon itself is also not bending.

Last time around people were complaining about a bending CPU it was just the pcb of the chip. Maybe the design was significantly updated for the worse tho.
 
I haven't seen the new ihs but normally they are a pretty thick piece of metal. Near impossible to bend as they have ridged edges. The silicon itself is also not bending.

Last time around people were complaining about a bending CPU it was just the pcb of the chip. Maybe the design was significantly updated for the worse tho.

It probably varies from generation to generation. I had to think for a bit, but I remembered where I first read about lapping a CPU in situ. Apparently someone e-mailed Noctua a few years ago about the subject, and the rep replied, "lapping the CPU/ IHS outside the socket doesn't help, as it gets deformed once installed inside the socket. If you would like to have the best performance with a lapped heatsink, you would have to flatten the CPU when it is installed inside the socket." I took it to heart because I've lapped several CPUs over the years and sometimes I get the desired effect, and sometimes it just makes things worse. I could never explain why that was, and this seemed like the most plausible explanation. X99 is one of those platforms where everything I lap just gets worse.

https://www.reddit.com/r/hardware/comments/74fnfl/interesting_reply_from_noctua_regarding_lapping/
 
This was a very interesting read. Reminds me of the days I experimented with paste/HS combinations on Pentium III CPUs. Thank you for the effort.

Back in those days, I was the only person I knew who actually used thermal paste. Most coolers came with some kind of shitty thermal pad already stuck on, and everyone I knew just used that. I would buy the Radioshack stuff and gob it on.
 
Back in those days, I was the only person I knew who actually used thermal paste. Most coolers came with some kind of shitty thermal pad already stuck on, and everyone I knew just used that. I would buy the Radioshack stuff and gob it on.
I hated those pads. I took them all out.
 
Back in those days, I was the only person I knew who actually used thermal paste. Most coolers came with some kind of shitty thermal pad already stuck on, and everyone I knew just used that. I would buy the Radioshack stuff and gob it on.
I hated those pads. I took them all out.


That pink gooey shit? I vaguely remember that. I can't remember where from though.

I don't think I ever used it.

I definitely used my fair share of white silicone goop from blister packs though :p

Times have changed quite drastically though. Neither my 286 nor my 486 sx25 shipped with heatsinks at all. They just had bare chips exposed.

All it took my sx25 to be coaxed into running at 50Mhz was adding a small HSF and some silicone goop thermal paste. Mine was probably one of the few sx50's out there :p

It did wonders for my framerate in X-Wing and Tie-Fighter (though I don't think I used the term framerate back then. They just ran smoother)
 
Back in those days, I was the only person I knew who actually used thermal paste. Most coolers came with some kind of shitty thermal pad already stuck on, and everyone I knew just used that. I would buy the Radioshack stuff and gob it on.
oh geez, the "Bubblegum" was the WORST. Removing it to use proper thermal paste was a huge pain
 
It probably varies from generation to generation. I had to think for a bit, but I remembered where I first read about lapping a CPU in situ. Apparently someone e-mailed Noctua a few years ago about the subject, and the rep replied, "lapping the CPU/ IHS outside the socket doesn't help, as it gets deformed once installed inside the socket. If you would like to have the best performance with a lapped heatsink, you would have to flatten the CPU when it is installed inside the socket." I took it to heart because I've lapped several CPUs over the years and sometimes I get the desired effect, and sometimes it just makes things worse. I could never explain why that was, and this seemed like the most plausible explanation. X99 is one of those platforms where everything I lap just gets worse.

https://www.reddit.com/r/hardware/comments/74fnfl/interesting_reply_from_noctua_regarding_lapping/
Ihs arnt straight. They have a designed curve where the middle is abit higher then the edges allowing better contact in that section then the edges when tightened down. If your lapping chips you should lap the heatsink as well, choose thermal paste well, and apply a good amount of pressure to properly mate the surfaces with the paste between. If not done properly it will hurt temps and there is very minimal benefit to doing it correctly with the design of modern cooling solutions.

I'm going to stand by my statement the ihs isnt bending. Its a pretty sturdy piece of material.
 
Ihs arnt straight. They have a designed curve where the middle is abit higher then the edges allowing better contact in that section then the edges when tightened down. If your lapping chips you should lap the heatsink as well, choose thermal paste well, and apply a good amount of pressure to properly mate the surfaces with the paste between. If not done properly it will hurt temps and there is very minimal benefit to doing it correctly with the design of modern cooling solutions.

I'm going to stand by my statement the ihs isnt bending. Its a pretty sturdy piece of material.

Well if I ever get a dead X99 motherboard for a cheap enough price, I'll hack it up and let you know who was right. Until then, rock on.
 
Well if I ever get a dead X99 motherboard for a cheap enough price, I'll hack it up and let you know who was right. Until then, rock on.
Interesting article on ihs shape, one thing I totally neglected to think about was the uneven thermal expansion and fatigue of the material after a thermal cycle. So I guess you have to run the CPU and lap it at the same time?

Or.. If you can get a fairly flat surface use a soft enough metal to make up for the expansion. (Kinda like thermal paste)

One of the cooler pictures from the article depicts exactly how I would want my ihs to look in a ideal world.

13-Intel-HEDT-1024x612.jpg

https://www.igorslab.de/en/cpu-heat...measures-now-at-amd-and-intel-rethink-basics/

I suppose you could try it in a socket to get your answer but I dont think that would be a very scientific method. If you get a chance to delid one try bending the ihs. It really is extremely ridged on the generations ive dealt with.
 
Last edited:
That pink gooey shit? I vaguely remember that. I can't remember where from though.

I don't think I ever used it.

I definitely used my fair share of white silicone goop from blister packs though :p

What a coincidence...

I've been working on refurbishing my Latitude E6540 today and needed a spare tiny screw to put the thing back together.

I have this drawer where I for 20+ years or so have just been throwing spare little baggies of screws and other miscellaneous small computer hardware just in case I ever need it. I was on the hunt for the small screw that holds the screen hinge to the body of the laptop.

I eventually found a screw that would work, but in the process I found these.

Three old yellowed blister packs of white silicone goop thermal paste.

PXL_20220116_012213771.jpg


I wonder how long those have been in there :p

I threw in a baggie of jumpers I found in the picture as well , for good measure.
 
I vaguely remember having some luck with nailpolish remover, but it has been so long that I am not sure anymore.
that tracks, acetone will royally f*ck up most polymers

They were, but the potential OC gains from keeping lower temperatures were also much better back then.
Truth! Those older CPUs were so much more temperature-sensitive in general. Keeping something like a 60W Thunderbird thermally happy could actually be really difficult considering temps over ~60* could cause instability and over 90*ish could cause permanent instability, there was no IHS and no way to monitor hotspots, and the available heatsinks consisted of extruded aluminum with poor retention and 60x10mm fans
 
Back
Top