I've been waiting for BareFeats to get their hands on one of these things to see what their testing shows. First set of benchmarks:
It's only compared to the i7 13" MBP currently, but the FCPP and Lightroom Classic results are absolutely mind-blowing. These aren't synthetic benchmarks--these are minutes-long, processing-intensive, real-world operations, and the M1 bested the i7-1068NG7 by well over two minutes... in an operation that took it three minutes.
The M1 comes in at 3.5x faster in a lengthy Lightroom Classic operation, and it's running under Rosetta. FCPX transcode is well over 5x faster. The GPU gaming benchmarks are not as ridiculous, but still over double the performance of Intel's previous-gen GPU in two of three tests.
I've been as impressed as anybody with the synthetic benchmark performance of the M1, but these results are just ridiculous. One assumes part of the reason for this is that it's not just CPU--these kind of image-manipulation actions presumably combine CPU, GPU, specialized processing units, and advantages of the memory architecture, but this shows that in situations where the chip can really bring all of its architectural advantages to bear the potential benefits are something else entirely.
M1 MacBook Pro versus Intel MacBook Pro
real world speed test results for performance minded Macintosh users
barefeats.com
It's only compared to the i7 13" MBP currently, but the FCPP and Lightroom Classic results are absolutely mind-blowing. These aren't synthetic benchmarks--these are minutes-long, processing-intensive, real-world operations, and the M1 bested the i7-1068NG7 by well over two minutes... in an operation that took it three minutes.
The M1 comes in at 3.5x faster in a lengthy Lightroom Classic operation, and it's running under Rosetta. FCPX transcode is well over 5x faster. The GPU gaming benchmarks are not as ridiculous, but still over double the performance of Intel's previous-gen GPU in two of three tests.
I've been as impressed as anybody with the synthetic benchmark performance of the M1, but these results are just ridiculous. One assumes part of the reason for this is that it's not just CPU--these kind of image-manipulation actions presumably combine CPU, GPU, specialized processing units, and advantages of the memory architecture, but this shows that in situations where the chip can really bring all of its architectural advantages to bear the potential benefits are something else entirely.
