16-inch MBP 2x slower than M1 MacBook Air in a real-world Rust compile

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by codeulike on Tuesday November 17, 2020 @02:25PM (#25128425)

Note that the comparison is also compiling to ARM vs compiling to Intel, so its partly to do with the simplicity of the compilation process between the two architectures.

https://twitter.com/rikarends/status/1328762958118346753

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by rualca on Tuesday November 17, 2020 @02:36PM (#25128564)

I'm also curious to learn where was the bottleneck in both tests.

I've read stuff about how the M1-packing MacBook Air shipped with a SSD whose burst write speed was far higher than the one shipped in older MacBooks. The bottleneck on build jobs tends to lie on disk access, specially with projects comprised of a significant number of small files whose build also outputs a bunch of small files.

This is one of the reasons behind doing builds on RAM drives.

If that's the case then these weird speedups might not be due to magic properties on Apple's M1 professor but due to the fact that the processor doesn't idle as much while waiting for all those reads and writes to finish.

If anyone has any data on this, please do share.

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by bydo on Tuesday November 17, 2020 @03:21PM (#25129248)

The SSDs are faster, but only by ~20%: https://techcrunch.com/wp-content/uploads/2020/11/BlackMagic...

From the TechCrunch review, which is pretty breathless but also contains a lot of good data: https://techcrunch.com/2020/11/17/yeah-apples-m1-macbook-pro...

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by spronkey on Thursday November 19, 2020 @05:39AM (#25148134)

Sequential performance is only one facet of overall SSD performance. Some older Apple SSDs weren't the greatest at random reads and writes - it's possible these new ones are quite a bit faster in that department, FWIW.

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by stingraycharles on Tuesday November 17, 2020 @03:36PM (#25129455)

Can’t you just instruct the compiler to compile for Intel on ARM and vice versa? We do this all the time, build ARM artifacts on an x86 CPU with GCC.

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by codeulike on Tuesday November 17, 2020 @03:56PM (#25129714)

Yes, but that isn't what was done in this case.

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by an_opabinia on Tuesday November 17, 2020 @02:51PM (#25128746)

Well that pretty much settles the matter, doesn't it.

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by ChrisRR on Wednesday November 18, 2020 @03:55AM (#25134553)

This should be a higher up. A fairer comparison would be compiling for the same target

There could be all manner of processor optimisations that are taking time to process. Like for like would be much more indicative

[ Parent | on HN ]

Re:16-inch MBP 2x slower than M1 MacBook Air in a real-world... by urmish on Tuesday November 17, 2020 @05:47PM (#25130961)

You can get better optimizations with CISC. If anything, compiling to a RISC ISA and expecting the same optimizations is actually tougher.

[ Parent | on HN ]