←back to thread

Apple Silicon M1 chip in MacBook Air outperforms high-end 16-inch MacBook Pro

(www.macrumors.com)
1080 points antipaul | 1 comments | 12 Nov 20 00:56 UTC | HN request time: 0s | source
Show context
nichch ◴[12 Nov 20 02:15 UTC] No.25065659[source]▶
How am I supposed to interpret this? A MacBook Air surpasses my i7-8700k in single and (almost) multi core performance?
replies(1): >>25065673 #
minxomat ◴[12 Nov 20 02:17 UTC] No.25065673[source]▶
Yes, in fact, the A14 (iPhone 12) already surpassed most Intel chips: https://images.anandtech.com/doci/16226/perf-trajectory_575p...

Intel is now #3

replies(1): >>25065717 #
FartyMcFarter ◴[12 Nov 20 02:24 UTC] No.25065717[source]▶
A modern mobile CPU with a TDP of 6 watts is beating a modern desktop CPU with a TDP of 125 watts? Is it just me or this seems too good to be true?
replies(4): >>25065751 #>>25065766 #>>25065784 #>>25065798 #
minxomat ◴[12 Nov 20 02:29 UTC] No.25065751[source]▶
It's a mobile CPU with many silicon advantages (widest decoder in industry, memory closer, deepest re-order buffer of any CPU and much more) plus a sane ISA and optimized OS. So yeah, you're seeing the benefit of Apples integration. That's why even the Anandtech page calls that graph "absurd", because it seems unreal, but it's real.
replies(1): >>25065796 #
refulgentis ◴[12 Nov 20 02:36 UTC] No.25065796[source]▶
Gotta be frank because it's not getting through: you're jumping way ahead here. Every time one of these threads has happened, there's an ever-increasing # of people who vaguely remember reading a story about Apple GeekBench numbers, so therefore this one is credible too - I used to be one of those people. This has been going regularly for 3-4 years now, and your interlocutor as well as other comments on this article are correct - comparing X86 versus ARM on GeekBench is nonsensical due to the way GeekBench eliminates thermal concerns and the impact of sustained load. Your iPhone can't magically do video editing or compile code faster than an i5.
replies(2): >>25065827 #>>25066030 #
1. minxomat ◴[12 Nov 20 02:42 UTC] No.25065827{5}[source]▶
My comment, and this specific thread, isn't even related to GeekBench. The graph I linked used SPEC instead of GB5. The gigantic architectural deep dive over on Anandtech even includes a discussion on the strengths and limits of both benchmarks, and how they make sense based on further micro-architecture testing.

The reason that graph doesn't include the A14 Firestorm -> M1 jump was simply timing. We know the thermal envelopes of the M1 and the cooling designs. We now have clock info thanks to GB5. So yes, the data is pretty solid. No one's saying that the iPhone beats the Mac (or a PC) at performance when you consider the whole system. Just that the CPU architecture can and will deliver higher performance given the M1 clock, thermals and cooling. Remember that The A14/M1 CPUs are faster at lower clock speeds.