A Notebookcheck analysis of the performance and efficiency of the new Apple M3 SoC, Apple's first 3nm chip, in comparison with AMD, Intel and Qualcomm.
4-Cores (Performance) Top = 4.056 GHz (or ~3.6 GHz when all cores are loaded)
4-Cores (Efficiency) Top = 2.748 GHz
10-Core GPU:
Up to 60% faster than M1
Up to 20% faster than M2
Dynamic caching, where cache and memory are allocated dynamically based on the actual requirements of applications.
Hardware-based ray tracing, mesh shading and AV1 decoding.
However, only two displays can be used at a time.
The other interesting summary figure is more-or-less maintaining power efficiency as before on lower power with comparable performance to other chips at higher power usage.
=
Either way it is spun (pun intended), it’s very impressive performance boost and fast cadence of release by Apple. notebook.check is a handy website, thanks for posting OP.
I imagine their development pipeline for the M* chips is very long. M4 is already being masked and M5 is being finalized with M6 development progressing and perhaps the M3 design team being ported to begin work on M7. Still very impressive but M3 has likely been prototyped since M1 was announced. I wouldn’t be surprised at all if they go big on a design and iterate down to what’s feasible with current manufacturing and yields.
One major difference between apple and intel is that apple generally has a much shorter time between spec and mass production. If you make assumptions based on knowing that Intel has six consecutive server chips in the pipeline, you might not be quite correct applying that to apple.
True but they also had a rather nice advantage up until basically today.
Qualcomm announced their Elite and Microsoft needs to get their shit right if they want to start transitioning some of their customers towards an architecture different than AMD64, at least on the laptop category.
Apple could just relax a little their schedule if they believe the competitors won’t fill the gap. After all, being faster than yourself doesn’t give you more money. You can keep milking the existing market up until your competitors start to get a little closer.
Well, if you blow away the competition such that it migrates more people to your platform and that migration increases people purchasing your add-ones and services, it could make a lot more money.
Context of that generational leap makes it far less impressive, and its clear that Apple wont be able to pull this off again in the next 2 years unless there are major changes.
Apple used to have M silicon trail A silicon architecture by a generation, this is the first year Apple has put them on the same level, its essentially 2 generations of architecture leap this year.
M2 was on N5P, M3 is on N3B, that node difference alone can account for half of the uplift.
I’d actually argue that M3 is underwhelming for what went into it, it also wasnt ‘one year’, but 16 months.
8-Core CPU:
10-Core GPU:
The other interesting summary figure is more-or-less maintaining power efficiency as before on lower power with comparable performance to other chips at higher power usage.
=
Either way it is spun (pun intended), it’s very impressive performance boost and fast cadence of release by Apple. notebook.check is a handy website, thanks for posting OP.
I imagine their development pipeline for the M* chips is very long. M4 is already being masked and M5 is being finalized with M6 development progressing and perhaps the M3 design team being ported to begin work on M7. Still very impressive but M3 has likely been prototyped since M1 was announced. I wouldn’t be surprised at all if they go big on a design and iterate down to what’s feasible with current manufacturing and yields.
One major difference between apple and intel is that apple generally has a much shorter time between spec and mass production. If you make assumptions based on knowing that Intel has six consecutive server chips in the pipeline, you might not be quite correct applying that to apple.
True but they also had a rather nice advantage up until basically today.
Qualcomm announced their Elite and Microsoft needs to get their shit right if they want to start transitioning some of their customers towards an architecture different than AMD64, at least on the laptop category.
Apple could just relax a little their schedule if they believe the competitors won’t fill the gap. After all, being faster than yourself doesn’t give you more money. You can keep milking the existing market up until your competitors start to get a little closer.
Well, if you blow away the competition such that it migrates more people to your platform and that migration increases people purchasing your add-ones and services, it could make a lot more money.
20% jump in one generation is massive. If they continue this the M series chips will be (even more) crazy fast in no time.
Context of that generational leap makes it far less impressive, and its clear that Apple wont be able to pull this off again in the next 2 years unless there are major changes.
Apple used to have M silicon trail A silicon architecture by a generation, this is the first year Apple has put them on the same level, its essentially 2 generations of architecture leap this year.
M2 was on N5P, M3 is on N3B, that node difference alone can account for half of the uplift.
I’d actually argue that M3 is underwhelming for what went into it, it also wasnt ‘one year’, but 16 months.
I am wary of a stat that says “up to” rather than “on average”