Good argument but remember guru 3d did measure the edges and temp was lowish. Thats the data we have besides the cpu temp and it adds up.
Please pay attention and read the details of my comments. They measured it with a FLIR device pointed at portions of the heatspreader outside of the dies. The portion of the die in question is the portion not covered by the water channel in the AIO cooler represented by increasing levels of bad cooling as follows :
> The portion of the heatspreader above the die which is the area of the copper plate where there are no copper fins where the water flows across doing the most efficient transfer (thermal mass)
> The portion of the heatspreader above the die which is the area of the copper plate where there are raised screw holes (no metal on metal contact)
> The portion of the heatspreader above the die where there simply is no copper plate contact because these contact plates weren't designed for threadripper and their diameter isn't large enough to cover the Heatspreader surface above the dies...
@ub4ty, with all your worries related to heat conduction in metal, you seem to forget about the bottleneck which is located in the boundary layer between metal and air.
I actually didn't forget. Please lets not play games of thermal mechanics when you know exactly what my commentary is centered on and exactly how it significantly contributes to thermal inefficiencies on top of the obvious boundary layer impacting everything.
The bottleneck of these coolers will be radiator size.
The thick copper base plate of the Noctua coolers in particular will be fine. I also trust that Noctua got the shape of the bottom face right (probably needs to be plane) and made the mounting procedure as painless as possible for this socket. Both of these points are important for good thermal contact, and IME they got all this right on other sockets.
The issue with the Noctua is the heatpipes rising off the contact plate way before you get to the edge of the contact plate. Given that they stay in contact all the way to the edge of the contact plate on non threadripper models, I see nothing but thermal inefficiency with them not being in full contact on the TR models. I've contacted Noctua about this and updating the TDP #'s and haven't gotten an answer. Your commentary about the NH9/NH12 are sound. However, you should see the obvious issue that were at thermal limits with these coolers yet people are being shy about critiquing thermal inefficiencies.
I'm done with this debate. My commentary has been backed by people with the actual devices in hand who bothered themselves to do deeper analysis. Do as you will with such information. I've decided what I'm going to do.