Real-World Test: Photoshop for Apple Silicon is Really, Really Fast
Earlier today, Adobe finally released the full-blown non-Beta version of Photoshop for Apple Silicon, and we got a chance to test it out. I won’t beat around the bush: the benchmarks are really impressive, matching or outperforming Intel-based computers that cost two to three times as much. We’ve been expecting this release for some time, […]
Earlier today, Adobe finally released the full-blown non-Beta version of Photoshop for Apple Silicon, and we got a chance to test it out. I won’t beat around the bush: the benchmarks are really impressive, matching or outperforming Intel-based computers that cost two to three times as much.
We’ve been expecting this release for some time, playing around with the Beta version of Photoshop for Apple Silicon and crossing our fingers that the final version would take full advantage of the M1’s impressive performance, unlike Lightroom CC. Now we have our answer, and it’s even better for photographers and photo editing than we could have hoped.
How We Test Performance
When testing Photoshop performance in our reviews here on PetaPixel, we use Puget Systems‘ PugetBench Beta v0.8 benchmark. Why the older Beta version? Because it’s the last version that included the Photo Merge test: a data point that is uniquely important to photographers but too much of a pain for Puget to include in the latest versions of the benchmark. As you’ll soon see, it’s a good thing we kept this older benchmark for testing, because Photo Merge happens to be the M1’s superpower.
In the charts below, you’ll see four computers listed: M1 Mac mini (Apple Silicon), M1 Mac mini (Rosetta 2), 13-inch MacBook Pro, and Dell XPS 17. The idea was to show how x86 Photoshop runs on Intel hardware (13-inch MBP and XPS 17), via Rosetta 2 emulation on the M1 Mac mini, and then compare those three scores against the Apple Silicon-optimized version running on the same Mac mini.
Keep in mind that both the 13-inch MacBook Pro and the Dell XPS 17 boast a full 32GB of RAM to the Mac mini’s 16GB. The XPS 17 is also running a 10th Gen, 8-core Intel Core i9-10875H alongside a GeForce RTX 2060 Max-Q GPU with 6GB of VRAM. Finally, both the 13-inch Intel MacBook Pro ($3,000) and the Dell XPS 17 ($3,000) that we tested cost a whole lot more than the fully-loaded M1 Mac mini ($1,700) used for this comparison.
Results: Apple Silicon vs Rosetta 2 vs Intel
Unsurprisingly, the M1 Mac mini loses to the competition in raw GPU performance, more-or-less matching the onboard graphics of the quad-core Core i7 that’s in the 13-inch MacBook Pro (full review here). But even with this score working against it, the Mac mini running Apple Silicon-optimized Photoshop managed to get the second highest Overall score we’ve ever seen out of PugetBench.
What’s more, none of the computers we’ve reviewed, not even the most expensive 16-inch MacBook Pro you can buy or the Razer Blade Studio Edition, has ever broken the 100 mark on the PugetBench Photo Merge test. Running optimized Photoshop, the M1 Mac mini hit 130+ in run after run after run.
What does that mean in real terms? An Intel-based 13-inch MacBook Pro with a 2.3GHz quad-core i7 and 32GB of RAM takes about 2 minutes and 45 seconds to merge a 6-photo panorama shot on the Nikon D850 (full res .NEF Raw files). The M1 Mac mini running optimized Photoshop does this same task in 1 minute and 14 seconds. Even a $6,700 fully-loaded 16-inch MacBook Pro with an 8-core Intel Core i9 and discrete GPU takes 1 minute and 52 seconds.
All of this from a chip that sips power so slowly that we were able to get almost 16 hours of 4K video playback out of the M1 MacBook Pro we reviewed in December.
Final Thoughts: What Does This Mean for Creatives?
After the lackluster improvements that we saw when we compared Apple Silicon-optimized Lightroom against the x86 version running via Rosetta 2, we didn’t come into this test with high hopes. To see the scores jump this much, when Rosetta 2 was already doing such a great job with the x86 version of Photoshop, was frankly mind-blowing. I thought the first run was a mistake; by the 6th I was forced to start believing my eyes.
As a reviewer, you’re always worried that you’ve over-sold something that really impressed you. I said that the M1 MacBook Pro “changes everything” and called the M1 Mac mini “the best Mac for most photographers,” and while I was pretty confident, snarky comments have a way of making you doubt your own sanity. Benchmarks like this help to prove that neither of these statements are as hyperbolic as they might initially seem to be.
When companies like Adobe take full advantage of what Apple has created in the M1, the results are undeniable.
What remains to be seen is how Intel, AMD, and the PC market in general will choose to respond. Will ARM-based PCs take off now that more and more software companies are taking the time to optimize their apps for Apple Silicon, or are AMD’s latest Ryzen processors efficient-yet-powerful enough to keep creators from jumping ship? And how can Intel possibly fight a war on two fronts, with AMD killing it on x86 and Apple blowing minds on ARM?
Whether you love or hate Apple is irrelevant. This isn’t about Apple Silicon vs Intel vs AMD and whether you’re a Mac or a PC. What’s happening here is a re-igniting of competition. And when companies compete, the customers win.