I did a fair x86_64 vs ARM comparison benchmarking a real world activity such as compressing a 4K blu-ray movie mux with x265. The Apple M4 Pro uses about 35% less energy than the Ryzen AI MAX 390 for this task. I expected 50% less but 35% is still impressive IMHO. You would still have to burn through a battery charge cycle to compress a single 4K movie with x265 but it's still nice to see that laptops aren't limited to thin client activities anymore. On a side note I'm not sure why people are talking about x86 so much, that obsolete slop was abandoned decades ago. We're on x86_64 now, how is this not common knowledge?
>>107845811you didn't do shit, you copied a few questionable screenshots
The architecture is called x86. x86_64 refers to a specific revision of that architecture that's longer ago now than the 8086 was when x86_64 was. Older revisions are i386, i486, etc. The latest is x86_64_v4.
>>107845811>a fair x86_64 vs ARM comparison >x265You're retarded. Go back.
>>107845836It's all pretty easy to look up. Anyway the focus isn't on x265 FPS, they're all about the same, it's the TDP.>>107845867This is x265 4.1 so it's had tons of ARM SIMD optimizations done to it. If I wanted to be impartial I would have just looked up AI benchmarks.https://x265.readthedocs.io/en/master/releasenotes.html
>>107845811>power has become so expensive in the USA that the wattage became the most important thing for Americans if they want to be able to turn on their computerGrim.
>>107845862A modern x86 CPU is no longer "x86" as it no longer has the 4GB RAM limitation. Using this as an umbrella term for modern intel/AMD processors is like calling modern Americans "Spanish crown invaders".>>107845935Dam, I totally forgot most of the world hooks up their laptops to a car battery they wheel around everywhere while Americans are stuck with 50-70 Wh lithium ones.
>>107845993> A modern x86 CPU is no longer "x86" as it no longer has the 4GB RAM limitation.By that logic, x86 should only refer to the 16bit version of the cpu. But you happily use x86 to refer to the 32bit version. Why are you being inconsistent.
>>107845811Does not the M4 has hardware encoder?
>>107846337Hardware is for fags.
>>107845811I'm happy to see arm doing well, but why the fuck do people keep using this benchmark? there aren't a lot of people out there compressing bluray discs. do a browser benchmark. everyone uses a web browser.
>>107846415Browser benchmarks use hardware acceleration so it doesn't really measure CPU performance.
>>107846490you can disable that. i'm open to other tests, but encoding a bluray is 'real world' but not 'common task'.
