There are technically two issues. One is the liquid metal escaping.
Two, which I think is more important, is the bonding (of the silver shim with the heatsink).
On a CPU only workload (cinebench) I get 43/44W consistently with multiple versions of the paste (my Thermal grease, the original liquid metal, and the Arctic Silver).
When I am playing a game with a GPU+CPU load, the bigger active area (of the GPU+CPU) caused the hotspot temperature to drop.
Furthermore, as @PSierra117 has demonstrated, replacing the shijm and modding the heatsink allows for substantially higher thermal transfer, of 62W in cinebench.
This is near 50% higher than stock or Heatsink Compound or Arctic Silver, and still 10% higher than the seemingly PTM results.
Not surprised. GIven that the fan blows on the separate GPU and CPU heatsink, but only one is doing the work.
This is good for expansion shell users, who would otherwise want to believe they are losing out significantly on the fans’ horsepower. Not really.
I do get higher than 14500 (probably near 15000 territory), 7840HS. 3992MHz, sometimes 4017.
But it’s still only 42Watts.
I kept saying, “don’t worry about score, just check power counter”
Seems like the higher-pitched sound is coming out of the inner slots in thos back vents per this post. I closed the same 4 inner slots on both sides and it also helped me, so I’d recommend rather leaving the outer slots open.
If LM is no longer used, can’t you just remove the shim? When I searched about vapor chamber most results were vapor chamber vs heat pipe, not vapor chamber & heat pipe. The FL16’s heat transfer route is like CPU → shim → vapor chamber → heat pipe. Heat transfer is like electricity conduction, putting resistors in series adds up the total resistance. Is using vapor chamber between the CPU and heat pipe really a good idea?
Yeah I just tested my RMA board that I’ve had for around 3 months and got similar results compared to the numbers I got when I first installed the board. I filled out the PTM 7950 request form, so we’ll see how it goes when those start shipping.
Heat pipes aren’t flat, you loose contact area each time the pipes connect to each other.
The shim/vapor chamber is to make sure the entire die is touching the heatsink assembly.
Without the shim, there is no contact, or you’re relying on thermal material to make up the difference, which isn’t as good as just copper
I came across this thread and tried to perform this test on my Batch 1 FW16 machine. And it seems I got the worst results across the thread Am I missing something, or the situation is so bad?
prerequisites:
Original FW16 180W charger - connected
Battery - 100%
Performance mode is on results:
Cinnebench result - 11858
During the test: CPU package power max - 27.5W
Core 4 reached 100 degrees C
I think that’s the lowest score that’s been reported on this thread so far. I’m sorry to hear that your performance is likely worse than it started after all that work. Thanks for sharing though so the rest of us can learn from it.
At the manufacturing level, sure. DIY? I really doubt it.
It would be nice if Cooler Master would revamp the design of the vapor chamber so it would contact the CPU die without need of an added shim. But as it stands, I doubt you could just remove the shim and still have everything contact properly.
thank you for your reply.
Now the only question I have: will the replacement to PTM7950 help, or it’s something more severe, and I should write straight to the customer support? Could the improper LM thermal application harm the CPU somehow?
Updated my previous support case asking to rma for a ptm board -
“ Thank you for your reply. Currently, the implementation started at the factory in mid-November, but there is still an existing stock of mainboards without PTM pads that need to be utilized first.
My problem is those PTM kits wont be shipping until “early” 2025. And it seems like priority is being given by FW to new orders rather than those with existing issues.
And the “we have existing non PTM boards that need to be used first”? Where? FW are willing to still use these boards with known issues?
Am I over-reacting? Maybe. But I feel like FW should be servicing existing customers with issues first. And can I just request a PTM kit? Sure. (for the record I already had) but should I have to for a manufacturing defect? Not really.
Meanwhile I have a £2k+ laptop that will not perform to its best ability and makes a lot of noise during the simplest tasks. A lot of my colleagues who have to put up with the fans would now never consider a FW.
They should be writing off the LM boards (or at least retrofitting PTM in factory). I know its cost to FW but what use are they if all boards are going to degrade?
Why do “NEW” customers get a special treatment and get new boards? What are we existing customers chopped liver??? We paid the same price as new customers so why are we getting stuck with the defective board? I got my machine in September!!
Have Cooler Master pay for the remaining defective boards!
Right, that will certainly get things going at a faster pace… /s
Entering contract negotiations with CM to have them pay for remaining boards will just result in a ton of lawyers fees and lots of delay.
I think people are blowing up a bit here over what’s at the end of the day, an enthusiast’s tinkering. PSierra’s results are very good, and he’s documented his methods/means very well. If you want similar results, go ahead and make the changes yourself now, no need to wait.
Keep in mind however, that Framework and CM have delivered exactly what’s on the tin. The standard heatsink with LM is capable of holding ~45W sustained. The board and heatsink design work as stated.
What PSierra’s modifications show are when fully optimized, this design can actually deliver much higher, to be competitive with the top of class machines using these specs.
The PTM change from LM still only solves one of the issues found anyway. It does not resolve the shim-heatsink interface. Nothing stops you from buying some PTM online and replacing the LM yourself. If you really want to get into it, go for the PTM-Shim-PTM sandwich method, if soldering is a step too far.
Also, for what it’s worth, every laptop I’ve ever owned with a dGPU and 45W rated CPU chip has made a lot of noise. The framework 16 isn’t an outlier on that front. +2K costs on a windows machine does not mean quiet. If you wanted that, you should be shopping Macs.
Am I missing something, or the situation is so bad?
