Great post and makes sense to me.
Why would intel post 1.52V if they expected CPU's to die or degrade quickly at that voltage?
And I'm sure that is Intel being conservative so chances are they could go a little higher, but the legal team likely ensured they covered their ass!
The same applies to motherboard manufacturers who's 'auto-overclocking' options tend to run voltage much higher than an experienced manual overclocker would. But they too are covering their asses so there's no way they would allow the BIOS to run unsafe voltages because that comes back on them.
I read so much about "you must not go above 1.35V" and I take this advice with a grain of salt as it comes from old timers who are basing this on older architectures and back to the day when people actually kept PC builds for 5-8 years as a 'normal' life expectancy.
I doubt many enthusiasts/gamers keep their builds more than 2-3 years, and degradation is unlikely during that time if kept within intel specs.
On a personal note, I have a 8086K that will do 5ghz at 1.33V 24/7 on all cores, 5.1ghz at 1.37V on all cores, but seems to really struggle to achieve 5.2ghz without bumping voltages to 1.42-1.43V which isn't a good trade off IMHO for that paltry 100mhz. I use a Corsair 150i Pro AIO which is an 'ok' cooler.
I haven't delidded, as delidding will only help with temps, but won't change the voltage requirements of this chip.
If I could get this thing to 5.3-5.4ghz under 1.5V then I would consider delidding to control temps, but it's a catch 22 situation here as I won't ever know until I delid as the temps would be too high under stress testing without the delid.
Maybe when the warranty is up I'll delid, but most likely will have moved on from Coffee Lake by then.