It was actually pretty great when I worked for a company making things here in england:
“That needs to move 50cm” meant it had to move exactly 500mm
“That needs to move a foot” meant just kick it over about a foot
It was just an unspoken thing that metric meant precise and imperial was just caveman measuring
Assuming a ground clearance of 10cm, and a wheelbase of 2.4m (source: my ass) then you can construct an arc under the wheels
This arc says you could drive such a car on an earth with radius as small as 7.25m. Actually, it could be slightly smaller because of where the wheels would contact, but I’ve lost interest
To make you less anxious:
A friend of mine had issues with his (much older) PC, stuttering in games and similar but it still worked
When I took a look I found it was pegged throttling at 100deg after running for a while. This had been going on for months
Eventually found the AIO pump had completely died, any cooling was due to passive conduction through the materials and water
We replaced the cooler and now it’s been running fine for another 3 years and going
TL:DR: modern CPUs can run hot, and safely boost. As it gets too hot it will start reducing clocks but it’s highly unlikely you damage anything unless you go out of your way to overclock, overvolt or ditch the cooler entirely
I honestly wouldn’t worry about it, especially with the weather as it is at the moment- my PC is getting about the same with a 3900x, nice noctua cooler and loads of airflow It’s been going over 30 in my office and even the best coolers can’t break that physics
Check again when it cools down, might make it easier to see if/where there are issues
It’s not stateless end-to-end, it just means the client needs to keep track and pass the state rather than drivers or hardware
I’m not 100% on the motivation but from an architectural standpoint it does make sense - your software can now do many new and weird things without a hardware change
One example I saw was allowing an arbitrary number of streams to be processed simultaneously, just passing the different context state for each stream
TL:DR The stuff the dedicated module is doing will go inside specific Mediatek chips on specific premium monitors
Really weird it’s taken this long - I remember reading that the modules were expensive and assumed it was just because they were early generations and Nvidia was still working things out