Hacker News Reader

Comment by ricardobeat | Hacker News Reader

Multiple rounds of 20% improvement will still not give you the performance jump brought by M1. It takes tasks that were simply unthinkable on a light laptop and not only makes them possible but manages to beat something like a 10-core i9 [1], in a machine without a fan.

[1] https://arstechnica.com/gadgets/2020/11/mac-mini-and-apple-s...

View on HN

Multiple rounds of 20% improvement like what we've been getting with iPhones each year?

How quickly we forget... the 10-core i9 was being embarrassed by AMD's Zen 2 mobile processors this spring.

https://www.notebookcheck.net/10-core-Intel-Core-i9-10900F-d...

Looks like single core Zen 3 scores about 2100 on Geekbench 5. M1 scores about 1700.

https://browser.geekbench.com/v5/cpu/singlecore https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q...

I'm confused why we feel like AMD Zen 3 has to "catch up" when it it's 29% ahead.

Yes I get it, per watt, battery life. The M1 is awesome for those metrics. But AMD chips have more "absolute" power. So you pick the one that is better for your needs - very powerful portability, or simply ultimate total performance (or something in between).

Those 2000+ scores are from hobbyist enthusiasts overclocking the Zen 3 CPUs, with elaborate liquid nitrogen cooling setups at the highest scores.

Stock Zen3 single core scores average around 1650: https://browser.geekbench.com/processor-benchmarks

There also appears to be an awful lot of them running MacOS which doesn’t really make sense (overclockers with bleeding edge Ryzen systems don’t have a lot of overlap with the Hackintosh crowd). Seems like it might be at least partially a data error.

What I'm trying to say is you'll not see a laptop that resembles a Macbook - silent, slim, light, with day-long battery life and top performance, with Zen CPUs anytime soon. It's simply outside the performance envelope. 35W vs 10W is a huge difference, even before you account for the slow cores / fast cores architecture vs frequency boosting.

https://www.anandtech.com/show/16252/mac-mini-apple-m1-teste...

> We noted that although Apple doesn’t really publish any TDP figure, we estimate that the M1 here in the Mac mini behaves like a 20-24W TDP chip

It also looks like the 15W AMD Ryzen 7 4800U trades blows with the 20W Apple Silicon M1.

On the other hand, I agree - OEMs are just barely starting to warm up to AMD (so "Macbook-like" laptops are rare, the Huawei Matebook is a blatant copy, but it's not the original) and the M1 is more power efficient, enabling amazing battery life in laptops. I didn't contest that idea.

I just think there's a classic reality distortion field happening when comparing the Apple Silicon M1 to "all other CPUs" that, for many, veers rather far from the truth.

Reply on news.ycombinator.com