~5W. My 7840HS laptop idles at ~7W.

Isn’t that only because RDNA3 is available with that many more CUs? Afaik the per-unit and per-clock RT performance of RDNA3 is barely ahead of RDNA2.

So now he talks to his audience as if they were kids.

…who do you think their main audience is? It’s entertaining tech content for 12m subscribers. It should be plainly obvious that it isn’t for adult professionals interested in niche workloads and deep reviews.

This video isn’t even meant to be a review.

I’m sure the engineers knew, but yeah they’ll be trying to save cents. Gigabyte made the very same mistake on several of their Waterforce blocks for a couple years. There was a whole slew of users reporting stripped material and corrosion in their loops.

Microsoft goes all in with supporting ARM on Windows

I’m not sure their prior attempts count as going “all in”. Never seemed like there was genuine effort and thought behind it before.

Now’s the first time they’ve had a proper fire under their ass to take it seriously.

Not long ago they said OLED was yrs out

Did they?

Valve said a new performance target was years out. And it still is.

So you don’t have an actual suggestion, you’re just rambling.

I don’t think you understand GN’s methodology and how they ensure results are accurate.

if you want to test cities skylines don’t do it this way.

Stop with your pointless walls of text and suggest a better methodology then.

It needs to be repeatable, represent typical game performance and have lower run-to-run variance than GN’s methodology.

recent

Polaris is 5-7 years old. Why would your recent GPU purchase be something that old anyway?

Unless it’s used, in which case you should expect drawbacks already.