Apple's new exciting M1 CPU was just announced and promises significant performance gains over the companies previous systems equipped with Intel core processors. Unfortunately, in Apple's presentation, the performance numbers shown were at best vague and non-conclusive. That leaves us in the dark as to where exactly the M1 will actually land in terms of CPU performance. That changes today with a tweet from @andysomerfield, who has real performance numbers for the Apple M1 chip in the Affinity Photo benchmark. The M1 is put against one of Apple's older 2019 iMacs with an 8th generation Intel Core i5 6-core CPU that features a boost frequency of up to 4.1GHz.
Affinity Photo is a photo editing service with a built-in benchmark that measures vector and rasterization performance. Apple's M1 processor scored the following:
- 504 points in Single-Core Vector
- 2032 points in Multi-Core Vector
- 538 points in Multi-Core Raster
- 6966 points for GPU Raster
- Combined 532 Single Core
- Combined 7907 Single GPU
For the Core i5 and Radeon Pro 580X combo in the 2019 iMac, here are the results:
- 310 Single-Core Vector
- 1515 Multi-Core Vector
- 393 Multi-Core Raster
- 8133 GPU Raster
- 407 Single Core
- 5568 Single GPU
In the CPU tests, the M1 chip wins hands down, being on average 25% faster than the Core i5 CPU. The M1 is also not much slower than the RX 580X in the GPU scores. This is exciting for Apple's M1 chip, which clearly demonstrates its ARM based architecture can go toe to toe with previous-gen Intel x86 chips in performance. In the past, ARM was a great for attaining great power efficiency and long battery life. Now Apple is demonstrating we can have the best of both worlds: high performance and long battery life.
There are of course caveats. This is only one set of benchmarks, which could be specifically optimized for the M1 chip. This is also against a base model 8th Gen Core i5. We need to see a lot more data to get a handle on how the M1 truly compares against other processors across a variety of tasks. Either way, it'll be very interesting to see how the M1 stacks up against Intel's and AMD's latest offerings. Those have more cores and higher frequencies, which the initial M1 isn't likely to match, but it should still prove great for laptop battery life.
Stay on the Cutting Edge
Join the experts who read Tom's Hardware for the inside track on enthusiast PC tech news — and have for over 25 years. We'll send breaking news and in-depth reviews of CPUs, GPUs, AI, maker hardware and more straight to your inbox.
Aaron Klotz
Freelance News Writer
Aaron Klotz is a freelance writer for Tom’s Hardware US, covering news topics related to computer hardware such as CPUs, and graphics cards.
More about macbooks
Latest
16 CommentsComment from the forums
Joseph_138 Being faster than a chip that is almost three generations old and no longer in production is a meaningless comparison. 8th generation i5's also didn't have hyperthreading, yet, so comparing it to a processor that presumably does use some form of multi-threading is not a balanced test. The testing also targets a mid range i5, not a top of the line i9, making it even more meaningless because power users, whether they be using a computer for business or for gaming, will gravitate closer to the higher end of the CPU stack. I would like to see this chip tested against a 10th gen i9, then it will show how weak this processor really is.
Reply
jwcdis Like the first commenter, let's not compare this to a 2 generation old CPU. This should be compared to 10th/11th generation. We are hardware enthusiasts here not apple fanboys. Also let's point out the obvious, every new piece hardware is going to demolish its predecessor; that's a given
Reply
nofanneeded I have the feeling that Apple is cheating using hardware codes inside their chip that can fool specific benchmarking tools..
Also , there is no way to to test their chip with neutral software yet ...
Reply
Math Geek the comparison is against a 2 gen old cpu but there's a lot to be learned there.
how has the intel chip grown in performance since gen 8? if the m1 is 25% faster and we know how much faster the new chips are to the 8th gen intel chip..... follow the bread crumbs.........
Reply
D e n i9-10900, 32GB RAM, GeForce RTX 2070 Super
vector single cpu 301
vector multi cpu 1878
raster multi cpu 775
raster single gpu 8950more info here - affinity photo benchmark results page
Reply
spongiemaster D e n said:
i9-10900, 32GB RAM, GeForce RTX 2070 Super
vector single cpu 301
vector multi cpu 1878
raster multi cpu 775
raster single gpu 8950more info here - affinity photo benchmark results page
According to the article, an i5-8500 with a single core boost of 4.1Ghz scored 310 in the Vector Single CPU test. Your post shows a 10900 that has a single core boost of 5.2Ghz scored a 301 in the same test which is 3% slower than the 8500. i5 -8500 is a 6 core/6 thread 3.9Ghz all core turbo (1515 multi cpu score). 10900 is a 10 core/ 20 thread 4.6Ghz all core boost (1878 multi cpu score). Either the 8500 tested too fast, or the 10900 tested slow, but they can't both be right.
Reply
mariusmotea And there is Ryzen 5000 series available. Can we see some scores?
Reply
MaikelMox Comparing what is most likely a dedicated coprocessor inside their new silicon (heavily optimized for one particular task) to a general purpose CPU is not exactly an apples to apples comparison, not even if it is Apple vs Apple :p
This reminds me a lot to the hype around "GPU accelerated" software which AMD promoted in the past around their APUs. There was a real potential for superior performance but it was totally linked to software being written specifically for it, which ultimately nobody did. As soon as they have been able, AMD has proudly gone back to promoting a high performance CPU design.
In this case, however, Apple controls the SW development on iOS so I admit they could be able to generate some superior tools combining their HW & SW skills. Be sure that they will command a price premium for that, maybe this time they are entitled to it.
Apple marketing department at its best. It seems that the reality distortion field has been activated again.
Reply
TerryLaze MaikelMox said:
Comparing what is most likely a dedicated coprocessor inside their new silicon (heavily optimized for one particular task) to a general purpose CPU is not exactly an apples to apples comparison, not even if it is Apple vs Apple :pAnd it doesn't need to be.
Photo Editing is a major market for apple sales and how fast a new apple product is in this specialized field will determine how desirable it will be for that market,they don't care how the computer does it as long as it does it, if this is a software suit that people use on mac that is.Reply
MaikelMox True, it is only the sensationalistic title of the article which pisses me off a bit.
I agree that the product development strategy likely makes sense.Reply
