Over 6 years, 2x the performance in single-core and 4x (for 2.5x the cores, depending how you count) seems surprisingly low compared to stuff I've read (not experienced yet) with the M1.
I think a lot of the perceivable speedup in M1 Macs is less to do with these benchmarks and more to do with optimizing the OS and Apple software for the M1 and the various accelerators on the SoC (video encoding and ML).
Of course M1's are perceived as being faster because:
- video camera encoding/decoding is offloaded to a dedicated processor;
- disk encryption is offloaded to a dedicated processor;
- rendering is offloaded to dedicated GPU's;
- matrix multiplication is offloaded to dedicated matrix multiplication processors (neural engines);
- no wasteful memory transfers (even though the memory bus is 2048 bit wide anyway) across dedicated processors.
Kind of like channel processors running channel programs in IBM mainframes, brought to a mainstream ultraportable laptop.
And, on top of that, the screen that drops the refresh rate by a factor of 6 when there is no need to refresh the picture too often.
Ironically, the CPU is left with pretty much one job to do: to compute. It is possible to get fast compilation times whilst doing all of the above at once. Or have an app that does all of the above at once, too (no, it won't save Microsoft Teams). And be power efficient, too; hence a long battery life and an Apple marketing team brouhaha: «We can do this all day». They are right – they can actually do all of that, and they can brag about it now because they have a full list from the above fulfilled.
System performance is a holistic matter, not just single/multicore comparisons oftetimes taken out of the context.
We already had dedicated GPUs for all the graphics/video stuff and the T2 chip for the disk encryption in the Intel Macs. That leaves the unified memory architecture as the only innovation.
Since Apple has been relying on Intel processors for many years now, I'd bet they've spent more time optimizing MacOS for Intel at this point. On the other hand, iOS has given them a quick way to transfer optimization knowledge.
> I'd bet they've spent more time optimizing MacOS for Intel at this point.
Keep in mind a lot of the core libraries on macOS and iOS have been shared since the inception of iOS. The kernel, CoreFoundation, and FoundationKit are on both platforms. There's also a bunch of other shared frameworks even before Catalyst. So much of macOS's foundational code has been running and optimized for ARM almost as long as Apple has been using Intel.
Also remember that a lot of apple customers stick with absolutely ancient laptops and are then amazed when a modern one is much faster e.g. I had to explain what an NVME drive is to someone who is a great developer but just not a hardware guy.
Don’t really understand this argument - Apple has had 13 years to optimise Mac OS for Intel and only a couple of years for M1 / can’t see how a video encoding accelerator affects responsiveness.
That’s a fair point - I think that it works both ways too - they know a lot about optimising CPU design for iOS / MacOS especially in low power environments.
I think you are on the right track, but I think the performance gain really comes from the RAM being on the chip itself, which raw number crunching won't make use (but actual usage will make great use of)
I waited for these to come out for a long time, and now "should I upgrade" isn't even a question. It's a "shut up and take my money" situation at this point, but they aren't even taking preorders over here yet. I mean, it could probably run a browser, several electron apps, AND android studio at the same time, how cool is that?!
Summary: single core performance gain is 2x whereas multi-core performance gain is 4x.