3 ms·
AMD Claims New Laptop Chip Is 30% Faster Than M1 Pro
- senttoschool 4y agoThey used Cinebench. It's a terrible general purpose CPU benchmark. It measure the performance of CPUs for Cinemark which is a niche of a niche piece of software. See why Cinebench sucks as a general purpose CPU benchmark: https://www.reddit.com/r/hardware/comments/pitid6/eli5_why_does_it_seem_like_cinebench_is_now_the/ https://www.reddit.com/r/hardware/comments/pitid6/eli5_why_d... Geekbench is better. SPEC is the best.
- coldtea 4y agoBeing 30% faster is the easy part. Being 30% within the same thermal behavior is the hard part. "AMD chips configured with Ryzen AI are 20% faster in AI tasks than Apple's M2 chip while being 50% more energy efficient, according to the company." What about general CPU performance? That's the AI engine... "To showcase the new chip's performance, AMD compared the performance of a high-end Intel chip, the M1 Pro , and its new Ryzen 9 7940HS processor rendering an object in the popular application Blender." With CPU rendering or GPU rendering? Also note how the M1 Pro is a year old by now, while the new AMD chip is not even on the market yet. Besides M2 Pro is already under production (as is M2 Max). Plus, the M1 flagship is M1 Max, which AMD didn't compare to. Let's say AMD might be close, but it's nowhere a 30% win over Apple's chip state of the art in general as the title might mislead someone to beliebe..
- gorjusborg 4y ago> Being 30% faster is the easy part. Being 30% within the same thermal behavior is the hard part. Yeah, but I'd argue it isn't heat, but power consumption that matters (understanding that heat/power use are directly related). The Apple M processors aren't about top performance, they are about top performance/watt. I'd love to see intel/amd pull off the magic required to bring x86 into comparable perf/watt with apple silicon, but I wont hold my breath. This smacks of marketing to try to redefine the metric of competition back to raw performance. Reminds me of the Hz wars.