Did he say his system isn't crashing any longer? He just says his CPU is running cooler now. If It's crashing around the 48-52C mark, that means it isn't because of the temperatures, because the temperature regardless of how heavy the activity isn't what's causing it, purely because anything below 60 degrees is considered a pretty cool temperature overall. It is rather a setting based on the voltage or the RAM that is causing instability.
i was talking about the temps, by fixed i meant he decreased his temps by at least 10c. which is starting somewhere, i have had a similar issue before and it turned out to be the cooler, once i fixed that, it fixed my issue.
So system still crashed after the TIM reapplication. I set everything back to default and then re-ran Cinebench. Passed OK.
I have now ramped up my clock speeds to 4.0GHz @ 1.3v with 1.125v SOC and RAM at 2800Mhz (XMP profile 1) and Cinebench passed at that. This was done incrementally, re-testing after each increment.
So seems stable, BUT... Temps whilst posting this are back up at 50C and idling around 42-45C (with odd spikes up to 53C now and again before it drops back).
Maybe need the NHD15 or the Black Rock mega cooler?
i have had something similar happen to me before and it turned out to be the cooler. this is why i am sure..
i'm going to assume that your rams are on your motherboard's QVL list, otherwise you shouldn't have bought them, so going with that they shouldnt run into any stability issues running @advertised speeds which is 3000mhz.
but you could run a mentest and see what happens, if it still crash try running each stick alone and see what happens