This bodes well for the rumored entry level MacBook with an A-series chip inside. If they can get the price on those down to $500-$600 it’ll run circles around everything else at the price point.
Wouldn't it be so cool if you could actually do anything with it? Imagine building a Bugatti and installing an unremovable speed limiter that forced you to top out at 75 MPH, because of course you're never going to take it to a track.
It was only intended for highway travel at best.
But hey, you can scroll at 120hz now, right? Think different!
All this tells me is that Intel and AMD are the only manufacturers making leading chips that you can do anything with.
It’s a really a shame to put such a good cpu in such an artificially limited phone. Apple already has all the elements required for effective convergence but would rather have us lug gimped devices out of greed. The current situation is so sad for actual innovation.
I don’t want a MacBook with an A19, I want to use the A19 I already have connected to a screen and a keyboard with a proper software stack.
PassMark and Geekbench are closed source. I don't know why I should trust them to e.g. treat fundamentally different kinds of devices in a sane and fair way. They have vastly different cooling mechanisms, for one. It matters if they can sustain a certain load for 5 minutes or for 5 hours.
This is great news for the entire ARM ecosystem. The fact that ARM is now exceeding the best x86 CPUs marks a historical turning point, and other manufacturers are sure to follow suit.
This is cool but it’s just one benchmark. It’s not clear to me what exactly PassMark even measures. That doesn’t mean the result is “wrong” but take it with a grain of salt
PassMark single threaded test is a very simple synthetic benchmark according to their docs:
> The single threaded test is an aggregate of the floating point, string sorting and data compression tests.
It’s an impressive score, but there’s more to performance in real apps than these three simple tests. This is a reflection of the high burst clock speed of the chip and the great job they did keeping it fed, but PassMark single threaded is about the simplest measurement you can do.
Modern CPUs aren’t really optimized for peak floating point throughput because any GPU will do a much better job at that. It’s also not clear if they actually use SIMD features of the chips, where desktop and server class parts should be able to walk away from the Apple phone chip due to their higher power limit alone.
Another reason why Apple should allow the UTM project to use the JIT API in their iOS app. I want to virtualize ARM64 Linux VMs on the fastest CPU in the world. Apple could put Android's Termux to shame, but they don't want to do that, and therefore I have no use for them. What great things are iPhone users going to do with a fast locked down phone? Play Roblox at 120 FPS? Doom scroll at 2x speed? Not know how to use their phone (but faster)?
18 comments
[ 4.1 ms ] story [ 38.7 ms ] threadI don’t know if that will be the M5 or M6 series but with active cooling and wall/big battery power I bet it will be impressive.
And a pro/ultra variant with lots of cores might post some very nice multi core numbers.
Way to go Apple. That’s a great accomplishment.
It was only intended for highway travel at best.
But hey, you can scroll at 120hz now, right? Think different!
All this tells me is that Intel and AMD are the only manufacturers making leading chips that you can do anything with.
Edit: Fixed last sentence.
Not sure why the A18 passmark scores are quite a bit lower than A17
I don’t want a MacBook with an A19, I want to use the A19 I already have connected to a screen and a keyboard with a proper software stack.
PassMark and Geekbench are closed source. I don't know why I should trust them to e.g. treat fundamentally different kinds of devices in a sane and fair way. They have vastly different cooling mechanisms, for one. It matters if they can sustain a certain load for 5 minutes or for 5 hours.
> The single threaded test is an aggregate of the floating point, string sorting and data compression tests.
It’s an impressive score, but there’s more to performance in real apps than these three simple tests. This is a reflection of the high burst clock speed of the chip and the great job they did keeping it fed, but PassMark single threaded is about the simplest measurement you can do.
Modern CPUs aren’t really optimized for peak floating point throughput because any GPU will do a much better job at that. It’s also not clear if they actually use SIMD features of the chips, where desktop and server class parts should be able to walk away from the Apple phone chip due to their higher power limit alone.