Yea, I am trying to reproduce it frankly, it was overheating previously (I didn’t manage to measure the ambient at the place previously, it is definitely warmer than the room when I created those graphs), the sensor readouts were hitting 100-105C. And when that happened, my Wifi disconnected and I got dropped out of the game.
@Jieren_Zheng Huh, very interesting. If the CPU goes over 100C, the system should shutdown (I’m pretty but not entirely sure). I wonder what those sensors were for. In any case, I hope you come to a solution!
That’s why I was saying I am not sure that the sensor readouts were real. It could be a momentarily spike and faded out before the system took action though. I notice the system alarms had delays in reporting so I guess the temperature sampling rate isn’t as quick so users don’t get crashes for temporal spikes.
I just got mine from Aliexpress now.
Might get to work on it later tonight and see how much better it runs 
Just reporting back after about 3 weeks with the pad/sheet. It seems to be stable so far, at idle, I’m getting a decent average low temp of just a tad shy of 30c (over 15 minutes averaged):
(Ambient temperature of 22-23c)
…and still able to boost to 63.5 watts when needed.
2 Likes
@Second_Coming After 3 weeks, I’m still impressed with your results! Color me jealous, maybe with some PTM7950. Maybe slather it on.
Just kidding?
Edit: BTW, I haven’t run any extensive tests recently but all still seems well on my end. And I have a feeling I could get slightly better results maybe by spending more time tweaking heatsink contact, but meh (for now).
I am finally done with the repaste (succeeded on my second attempt), the thermal pad is mucchhhh harder than thermal paste.
Interestingly, a 1260p has a larger die than the 1240p (sorry for the lousy camera). Is that the iGPU die? Because that’s the only difference I know between the 1240p and the 1260p.
I actually did not see a huge difference in thermals frankly.
The initial higher temps are because I forgot to prop the laptop up.
So I repeated the test again.
Seems rather close. That being said, I think my room ambient temperatures are slightly warmer as 2 other gaming desktops are running games.
So perhaps a few degrees lower?
I have a empirical comparison on Yuzu running Xenoblade Chronicles before and after, before the system would thermally throttle very quickly and I would get a lot of alerts from temperature alarms. After, it took longer to hit throttling temperatures, and I get much lesser alerts.
So overall, I guess it is better.
I suspect that the iGPU is not very efficient (as seen with old Iris 6100 on a i5-5257U and HD4000 on a i7-3667U). I kinda wish that I can change the panel to a 4:3 1080p screen instead.
1 Like
That’s the primary difference in my case as well. And that’s what I was after, because it means I can have a more performant boost period.
(i.e. That initial thermal transfer is really quick when CPU starts to heat up, way faster than all other thermal paste I’ve tried)
I will probably try one day to hook up an eGPU and get the 3D printed case, add maybe a waterblock or just large heatsinks for passive cooling (probably the latter for much lesser cables).
With an increased cooling capacity, I wonder can I maintain 65W throughout.
Yeah it’s a bit tricky though not too bad after learning how it works. I dug up a Reddit comment showing an application video (skip to 12:15).
Oh, nice catch! I mistakenly thought the smaller die on the bottom was the iGPU. I did some digging, and it turns out it’s the PCH (Platform Controller Hub) die.
Here’s some info for Alder Lake (BGA:
Both the BGA type-3 and BGA type-4 packages are multi-chip modules, with the processor die and the PCH die.
From the image in the article, your P series looks to be BGA Type3:
On the i7-1165G7, it looks like the iGPU is on the bottom of that upper larger die:
So it would make sense if the 1260p has a larger processor die (the one at the top in the pic) than the 1240p due to a larger iGPU.
I’m correcting my original post, and someone please correct me if I’m still incorrect.
Edit: oof, I can’t seem to edit my original post, @moderators can one of y’all please help (again )
If that’s the case I think that’s pretty good – a few degrees delta is a lot in regards to thermal material.
Glad to hear! I’m hoping that this stuff has good longevity, and thanks for the effort/insight.
3 Likes
Yes, your freezer suggestion actually helped! I was so nervous when I started to see condensation when I was applying.
I had the same thoughts too. I also struck by how big it was compared to my i7-3667U and even a delidded Ryzen 5900X (my current desktop CPU).
1 Like
I saw these on Amazon:
https://www.amazon.com/Thermalright-120x120x1-0mm-Conductive-Resistance-Silicone/dp/B09FJW6FX2/
https://www.amazon.com/Thermalright-Conductive-Resistance-Temperature-120x120x2-0mm/dp/B09ZJYNRRD/
With these seems workable except with a much greater thickness?
The thinnest I see is 0.5mm which could increase mounting pressure without warping the heatsink?
Has Framework given all the thermal pad dimensions and thicknesses yet?
Would be handy info to have.
Framework uses Shin-Etsu X-23-8117 thermal paste, not a pad.
I’m not talking about the paste on the CPU/GPU I’m talking about the thermal pads.
Do you mean the ones on the VRMs? For those, I don’t think so.
Framework, can we have the thicknesses please?
From what I’ve read (and general experience), those types of thermal pads tend to have worse performance than thermal paste for CPU lids/bare dies. What makes the PTM7950 work so well is supposedly the phase change capability-magic. So though I saw those Amazon links (or different Thermalwright pads and brands), I personally didn’t buy.
Yeah, though from what I’ve read and after seeing how the PTM7950 pad was after removing the heatsink (a thin layer on the die as well as heatsink), I think the heatsink’s already making pretty good contact as is and an increased thickness may hurt heat transfer rather than help. I suppose someone could stack two layers of PTM7950 to test this theory.
The only way to know for certain is to test it though!
I was thinking that it has increased thermal conductivity (theoretically) to offset the thickness.
For desktops, mounting pressure does have an effect on cooling but too much, it impacts memory channels due to compression of CPU pins and stuff.
For now, I kinda don’t want to repaste again.
Maybe I should have tried it before getting the PTM phase change ones haha