I did not mean to ruffle your feathers, Im just saying 'Buyer beware!'mslinn wrote: ↑Mon Jun 24, 2024 6:09 pmAre you suggesting that it would better not to use a thermal pad on the larger, hotter NVMe? If not, what are you proposing? Install a larger NVMe but without any cooling? Never make any changes to this perfectly optimized device?
You also seem to imply that it is better to run the P3S without a thermal pad on the factory-installed NVMe. That seems unlikely to me.
I dispute your intimation that I should live in fear that someone blows up something, and they sue me as a private individual. That would be very unlikely; the suit would have to be brought in Canada, and the cost of the lawsuit would much greater than the expense; even if they won, they would not get their expenses covered.
The worst that should happen is that the system throttles back as it heats up. Doing this repeatedly to the extent that the system ages quickly is unlikely because it would run unacceptably slowly. Without a concrete suggestion, this sounds like fearmongering.
What would be helpful would be publishing the results of running a P3S hard while monitoring the CPU heat via ssh, and putting a temperature probe on various NVMes.
Data and provable facts rule. I welcome any test results that you can provide.
I have never seen where an M.2 share the heatsink of a significant heat source like a processor.
I agree with you, testing is required. Im sure Ableton did this for their SSD solution.
It could be that a thicker thermal pad mitigates heat flow back to the M.2.
Since this is your proposition, you could do the experiments to prove the idea sound.
Maybe install heat sensors on the m.2. and monitor externally.