Now that Microsoft’s Windows on ARM emulation layer can also run 64-bit x86 code, there will inevitably be questions about how it compares to the Apple M1. Qualcomm’s 8cx platform and minimal refresh earlier this fall were never known for speed – these chips have been sold on the basis of excellent battery life, not pure performance.
Even with that said, the M1 eats Windows on ARM-powered laptops for breakfast. That’s the conclusion PCWorld reached after comparing the Snapdragon 8cx to the M1, with a Core i5 (4C / 8T, 1GHz base, 3.6GHz boost) machine deployed for good measure. Moving to 64-bit apps on Windows doesn’t seem to do anything for the 8cx’s overall performance.
It must be said that this is not particularly surprising, but it highlights the critical weakness in the nascent Windows on ARM ecosystem. Qualcomm essentially stopped updating its SoC this year, and while there was technically an overhaul for devices like the Surface Pro X, the performance is practically identical. There has been no architectural update.
Early benchmarks should, as always, be taken with grains of salt, but the Surface Pro X is a fairly well-known amount at the moment and its performance is already known to be inferior to x86. And one of the problems that Windows struggles with on ARM, despite having been around for a number of years at this point, is that it has fewer native applications. Cinebench R23, for example, Apple M1 has native code paths, but nothing equivalent to the Surface Pro X. That means we are effectively comparing native performance on M1 to emulated x86 performance over Windows on ARM.
Image by PCWorld.
Both the Core i5 and SQ1 lose badly here, but the Core i5 achieves 39 percent of the M1 performance in multi-threaded, while the SQ1 achieves ~ 23 percent of the M1 in both. Things don’t improve in Handbrake:
Image by PC World.
There are results for the M1 in Handbrake 1.4.0 with video toolbox and H.265. The x86 systems have been tested with the latest version for the platform, 1.3.3. The M1 is 2.87x faster than the Intel Core i5, but the i5 is more than 2x faster than the SQ1.
I’ve mentioned before that Windows may become more important on ARM if ARM chips prove to have lasting advantages over x86, but these results show just how hollow the gap is. It is true that the SQ1 is running emulated code here, while the M1 is not, but it hardly matters. If you look at the SQ1 and mentally improve its scores 1.5x, it’s nowhere near the performance of M1 or x86.
In short, Microsoft needs an ARM hardware developer who is willing to invest in regularly improving its platform with real performance improvements. Besides Qualcomm, Nvidia, AMD and Samsung are said to be the top three potential players. Samsung has discontinued its own CPU efforts so they seem to be out of the running. AMD has built ARM chips relatively recently, but it has shown no interest in pulling focus away from Ryzen and competing with its own x86 CPUs. That leaves Qualcomm and Nvidia. Qualcomm could launch a true successor to the 8cx this year and probably improve single-thread performance by 1.4x or more as the CPU would be at least two generations more advanced.
Right now Qualcomm seems like the best option for a competing product, but I’m not convinced we should count the idea of a future Windows on ARM system with Nvidia hardware. I have absolutely no backup of any inside information, but Nvidia is a competitive company that once had some real plans for itself on the desktop. The company’s name is literally derived from the Roman goddess of envy, Invidia. If ARM CPUs lead the way in a way that Intel and AMD can’t answer – and yes, that’s a really big “if” – I can see Nvidia throwing its hat in the metaphorical ring.
Regardless, PCWorld’s tests make it clear that we are a long way from any Windows on ARM laptop that can compete with the Apple M1.
Read now: