NewsPhotography

Performance Test: Lightroom Classic is Way Faster on Apple Silicon

This morning, Adobe unveiled the long-awaited version of Lightroom Classic that is fully optimized for Apple Silicon devices, and we had a chance to test it out before release. Our hopes were high for a program so famously sluggish. Could Apple’s M1 processor deliver a hefty performance boost for photographers? In short: yes. The latest version of Lightroom Classic was up to 25 percent faster on the M1 than on our more expensive Intel-based Mac.

Our experience with Apple Silicon-optimized apps from Adobe has been 50/50 so far. When Lightroom for M1 was released in December 2020, we found meager performance gains above and beyond what M1 could already do via Rosetta 2 emulation of the Intel version. Photoshop, on the other hand, experienced a massive performance uplift when it was optimized for Apple Silicon, blowing us away with its GPU, Filter, and especially its Photo Merge scores in Puget Systems’ PugetBench benchmark.

For Adobe, the stakes are particularly high with Lightroom Classic. Most enthusiast and professional photographers prefer Lightroom Classic to its cloud-based sibling Lightroom CC, and Capture One 21 — Lightroom Classic’s biggest rival — showed significant performance gains in almost every category with the release of Version 14.2 optimized for Apple Silicon.

In other words: this had better be good.

The Test

For this comparison, we timed the import and export of 110 Sony a7R IV photos and 150 PhaseOne XF images on three versions of the software and two different machines1:

  • ARM-optimized Lightroom running on an M1 iMac with 16GB or RAM
  • Intel-optimized Lightroom running on the same M1 iMac via Rosetta 2
  • Intel-optimized Lightroom running on a 13-inch Intel MacBook Pro with 32GB of RAM

The imports were configured with 1:1 previews, all other boxes unchecked, and exports were performed in two flavors: sRGB JPEGs at 100 percent, and AdobeRGB TIFFs at 16-bit with no compression. Note that these tests are a little different than previous benchmarks we’ve run, where we used Standard previews at import. Switching over to 1:1 previews allows us to ignore changes in screen resolution/default preview size, and since 1:1 previews take longer to create they give us a better basis for comparison.

The results are presented as the average of at least three consecutive runs to compensate for inconsistencies, although we didn’t see a major change from run to run in any of our tests.

The Results

In the import test, the ARM-optimized version was about 13.5 percent faster than Rosetta 2 emulation for both Sony and PhaseOne files, and 20 to 24 percent faster than the Intel version running on an Intel chip. A pretty good start for Adobe:

And the good news kept on rolling across every test we ran. Exporting 61-megapixel Sony a7R IV files and 100MP PhaseOne XF files as 100 percent JPEGs was 10 to 12 percent faster than Rosetta 2 emulation, and about 24 percent faster than the Intel Mac.

The larger 16-bit TIFF exports saw an improvement of nine to 13 percent compared to Rosetta 2, and a huge 22 to 25 percent improvement compared to the Intel version running on the Intel Mac.

Here are all of the exports side-by-side, so you can compare them on the same scale:

We’ll be honest: we did not expect this much of a bump. After the lackluster performance of Lightroom CC for Apple Silicon, we thought we might see a five to 10 percent improvement at the most. But these results paint a much more impressive story. Sure, Lightroom Classic for Apple Silicon isn’t two or three times faster than its Intel counterpart running on similar hardware, but ask yourself this: when was the last time Adobe released a new version of Lightroom Classic was 25 percent faster in any category?

To the best of our knowledge, the closest we ever got was a meager 11 percent jump in 2018, and that benefit was only available if you had a relatively powerful multi-core processor and at least 12GB of RAM.

By comparison, today’s update applies to every Apple Silicon Mac, including the extremely affordable M1 Mac mini that we dubbed the best Mac for most photographers. And when we finally get to test this on the rumored 14-inch or 16-inch MacBook Pro with a 2nd generation Apple Silicon processor, we expect these numbers to get even better.

When Companies Compete, Photographers Win

We weren’t sure what to expect when we fired up the ARM-optimized version on our M1 iMac, but these results leave no doubt in our minds: Adobe owes Apple a thank you note.

If you’re already using an M1 Mac, you’ll see a 10 to 12 percent improvement across the board compared to the Intel version running via Rosetta 2 emulation. Already good news. But if you’re sporting an Intel machine like our quad-core 13-inch MacBook Pro (which, by the way, has twice the RAM1 of our M1 iMac and costs $700 more than an equivalent M1 MacBook Pro) you can expect even bigger gains — a whopping 20 to 25 percent improvement in import and export times.

That’s nothing to scoff at, and Adobe’s own tests show an even bigger jump in performance when the two machines are more evenly matched. When comparing an M1 MacBook Pro with 16GB RAM against an Intel MacBook Pro with 16GB of RAM and a Core i5 processor, the company found a 54 percent increase in performance.

The release of the M1 SOC has allowed Adobe to show the biggest performance gains in Photoshop and Lightroom that we’ve seen in several years and paves the way for similar gains (we hope) for ARM-based Windows computers.

This is good news for Adobe, great news for Apple, and bad news for Intel… but that’s not why we’re excited by updates like this. We’re excited because major performance gains from any corner of the industry inspire innovation and competition across the whole. And that, dear reader, is great news for photographers.


Footnotes:

113-inch MacBook Pro: 2.3GHz Quad-Core Intel Core i7, Intel Iris Plus Graphics, and 32GB of LPDDR4X RAM.
24-inch iMac: Apple M1 SOC with 8-core CPU, 8-Core GPU, and 16GB unified memory.


Author: DL Cade
Source: Petapixel

Related posts
AI & RoboticsNews

H2O.ai improves AI agent accuracy with predictive models

AI & RoboticsNews

Microsoft’s AI agents: 4 insights that could reshape the enterprise landscape

AI & RoboticsNews

Nvidia accelerates Google quantum AI design with quantum physics simulation

DefenseNews

Marine Corps F-35C notches first overseas combat strike

Sign up for our Newsletter and
stay informed!