not yet, maybe in 3 years :)

If reapplying thermal paste and verifying the heatsink pushes down on the IHS evenly doesn’t help then I suppose what might help is a delid, clean the die, grind down the IHS sides (to reduce the gap between the die and the IHS), apply liquid metal and reseat the IHS

I’m seeing this behavior when performance power plan is selected in windows, it makes all the cores to boost to max turbo 3 frequencies in idle and makes the computer to eat a couple of more watts

Among the classic desktop versions the higher the gen and model, the better power efficiency at low wattages, so an i9-14900K would give you the highest performance at a 95W power limit.

Additionally with undervolting you’d get a bit more performance out of it at the same power consumption

How are you planning to cool it and what are you going to use it for?

i7 or even lower would be quite sufficient for this in a laptop, but I suspect i9 might be more energy efficient and thus easier to cool, but I’m judging this based on the desktop cpus, not laptop. So if the price didn’t matter at all I’d consider researching the laptop i9 models, but if it matters even a little bit then i9 is a too pricy overkill.

Where I live the 13 and 14 gens are the same price, so when considering a 13 gen check the price of the 14 gen model. The performance might not be that greater, but the power efficiency (and thus the ability to cool it) is much better.

I’m willing to try some scenarions you’d wish to know about at 1000rpm fan speed and report the temperature and performance results (eg how long did it take to encode some audio or video etc) as long as I don’t have to pay for the software in question (like games) though😅

with such a low price difference I’d be seriously considering it and comparing them more thoroughly, however where I live the 13900K is still about 50% more expensive than the 14700K and just a couple of EUR cheaper than the 14900K

Yeah this might be a problem since the contact frame doesn’t have a tensioning mechanism other than just tightly fastening the bolts, so the required pressure force is left for the heatsink mounting mechanism - both of which must result in quite uneven pressure on the pins especially with air heatsinks with only 2 mounting points mounted on vertical boards. I plan to use spring washers under the contact frame bolts for this reason

the sanding is necessary if there is a gap between the IHS and the die (otherwise filled with the original indium solder) and reducing this gap might perhaps be the main reason why this works and can sometimes reduce temperatures even by more than 10°

thanks for the new photos! Well it’s strange, it seems to be doing the opposite on my board: the lower the DC loadline, the more realistic the reported power figure is, but the farther away the reported VID is from the actual vcore. At DC loadline=1 I got the closest power reading to the current I measured on the EPS cable (which is still about 20% higher)

yeah I was indeed able to manipulate the reported VID value with the DC loadline. I’m therefore interested in the original intel configuration value and a way of reliably retrieving it without it being manipulated by the DC loadline setting. Knowing an arbitrarily manipulated value isn’t useful for the question I asked.

that says VID can be altered

can you please specify more exactly what did you mean by that? Which sentence in the intel specs says that?