Outside of a few small local businesses that actually care about doing right by people, loyalty hasn’t mattered for decades dude. Companies don’t give a shit about any of us. Why even bother thinking in terms of loyalty, it’s completely misaligned with how they operate.

Lol I can tell you just used Google Lens or some shit and then proceeded to make it sound like you knew what you were talking about by assuming it was Japanese (it’s not).

The weird thing is, I’m not sure any customers actually do care. it genuinely just feels like engineers finding ways to masturbate over how thin they can get something.

Maybe, but Microsoft’s competitors are doing a lot better on the battery life front so they’re leaving a lot on the table for competitors to swoop in by not fixing their sleep and wake issues. It was a big consideration for the company I work at to go with Apple machines because they do lots of field work and need the machines running all day. I can say from experience it’s incredibly frustrating to leave home with my MS Surface on a full charge only for it to have majority of the battery drained by the time I pull it out of my backpack due to waking up when it wasn’t supposed to.

A lot of my leftist friends will still let the bad be the enemy of any sort of good whatsoever it seems. It's exhausting as a leftist when you can never be outraged enough for other leftists.

It will prefetch the instructions and put into the pipeline the branch it thinks is mostly likely. It may do ahead-of-time speculative execution on certain instructions but not always. If it missed the correct branch it will flush the pipeline and start the pipeline over again from the correct branch. Afaik it doesn’t execute or prefetch both branches. The other guy is saying it does but that doesn’t really gel with my own recollection or the Wikipedia article he cited. You can see some further discussion that suggests only one branch gets prefetched here here and here. Reasons cited for only predicting one branch are: 1) Two pipelines with all the associated circuitry to look ahead, decode, and speculatively execute is incredibly expensive in terms of both processing requirements and die real estate. 2) Caching both would thrash your caches with new data constantly. 3) Modern branch prediction is already so accurate, there’s really no need for two pipelines anyways.