So what? People don't generally buy Apple devices to run different OS's on them. They buy them for macOS or iOS/iPadOS. Having a CPU that uses very little power and at least matches the x86 equivalents is a complete no brainer to me.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Apple M1 CPU
- Thread starter benparker
- Start date
Remains to be seen. Apples <> Oranges
*SpecINT and Geekbench don't count as real workloads
Don't forget it's the Intel U series the M1 is mainly replacing, which itself doesn't perform that great compared to other mobile chips. Having a look over the presentation again it doesn't seem like they're comparing to the high end CPUs at all. A fat chunk of the "98% laptops sold in the last year" will use low end CPUs, and will be significantly lower than the Air's $999 starting price.
The 2020 models had a mix of U and G7 CPUs, but they all still had pretty low base clocks. The G7 variants might still be able to compete will against the M1 in some workloads however.
https://www.tomshardware.com/news/apple-arm-mac-announcement-M1
Somewhat unsurprisingly a M1 Quad core (as it's Big.Little 4+4) is 3.5x faster than a unspecified Ice Lake processor?Apple Macbook Air with M1 said:
What's the betting that it's the 1.1 GHz dual-core Intel Core i3 1000NG4 available in the current Air?
Faster than ever isn't really quantifiable though?MacBook Pro (13-Inch) with M1 said:
Even in the 13" Pro it still only has to be faster than the current 2.0Ghz quad core i5, which is known to be thermally limited?
@RavenXXX2 SpecInt2006 isn't representative of anything though
Yes Apple's performance may have doubled almost year on year, but there is a much more significant performance uplift between the 4Core/4 Thread 4790K and the 10Core/20Thread 10900K than that graph suggests
The below is more representative of the gains between a 4790K/10900K an also shows the performance gains (or lack thereof of a fairly recent laptop chip)
Yes Apple's performance may have doubled almost year on year, but there is a much more significant performance uplift between the 4Core/4 Thread 4790K and the 10Core/20Thread 10900K than that graph suggests
The below is more representative of the gains between a 4790K/10900K an also shows the performance gains (or lack thereof of a fairly recent laptop chip)
Associate
- Joined
- 29 Jun 2016
- Posts
- 543
I've got no doubt these M1 devices are rather good, I mean they're heavily leaning on the (very good) ARM ecosystem.
The whole argument that ARM is somehow miraculously 3.5x faster is bogus though. It's not like they're magically able to get more from transistors than other processors, nor is the M1 directly comparable to x86 offerings as the M1 is a true SoC and x86 are not.
The whole argument that ARM is somehow miraculously 3.5x faster is bogus though. It's not like they're magically able to get more from transistors than other processors, nor is the M1 directly comparable to x86 offerings as the M1 is a true SoC and x86 are not.
Associate
- Joined
- 27 Apr 2007
- Posts
- 966
They are on 5nm so they can actually use more transistors right now until everyone plays catch-up.
Apple have a full ARM architecture license so their designs are very different from the vanilla off the shelf ARM ones, hence the large performance gains.
It's irrelevant to end users on where you choose to draw the line on the semantic debate on what a SoC is. Benchmarks beat semantics.
These chips are being used in the same chassis so they are a direct replacement for the Intel parts.
Zen 3, the King is dead!
They are on 5nm so they can actually use more transistors right now until everyone plays catch-up.
Apple have a full ARM architecture license so their designs are very different from the vanilla off the shelf ARM ones, hence the large performance gains.
It's irrelevant to end users on where you choose to draw the line on the semantic debate on what a SoC is. Benchmarks beat semantics.
These chips are being used in the same chassis so they are a direct replacement for the Intel parts.
Zen 3, the King is dead!
I would like to see these independently benchmarked against Intel's offerings let alone AMD's, Apple are claiming to have the world's fastest chip, Apple also want me to believe the 10900K is only 19% faster than the 7700K, they are positioning the A14 28% above the 7700K. In which case its not the fastest CPU in the world, its not even close to as fast as the outgoing EOL Ryzen 4800H Laptop chip let alone the out going soon to be EOL 3990X.
Apples marketing is literally "just lie, make it up"
Last edited:
Source?
It's essentially a single threaded benchmark. However there's no details as to clock speeds, compilation parameters etc or anything that allows the comparison to be fair.
In terms of relevance, it's imo irrelevant as nothing is single threaded anymore, and multithreaded performance scaling can vary depending on a huge number of factors
@humbug
The screenshot is from AnandTech. And you're calling Apple liars? I'm not entirely sure what your point is whatsoever.
The screenshot is from AnandTech. And you're calling Apple liars? I'm not entirely sure what your point is whatsoever.
