AnandTech took a deep dive into the new M2 chip announced yesterday, focusing in particular on the M1 versus M2 chip performance. These chips are available in the all-new MacBook Airand in an updated version of Entry-level 13-inch MacBook Pro.
The site says that while Apple has been as timid as ever when it comes to chip specifics (e.g. clock speed), we already know quite a bit…
Chip M1 vs. M2: Similar to A14 vs. A15
At a high level, there have been a limited number of changes with the M2 – or at least as many as Apple wants to disclose at this time – with a focus on a few critical areas, compared to the windfall that was the original M1 SoC. . While this is all preliminary before further disclosures from Apple or getting to grips with the hardware itself, the M2 looks very much like a derivative of the A15 SoC, similar to how the M1 was derived from the A14 . As a result, at first glance, the M1 to M2 upgrade looks quite similar to the A14 to A15 upgrade.
Apple said the M2 chip is based on “a second-generation 5nm process,” which almost certainly means TSMC’s 5NP process. This improves performance without increasing transistor density, and this is supported by the main illustration showing that the M2 chip is physically larger than the M1 (image above).
M1 versus M2 in numbers
Smith has helpfully collated all of the data shared by Apple into a single comparison chart.
Efficiency cores are more powerful
AnandTech suggests that a key difference between the M1 and M2 chips is that the latter has more powerful efficiency cores.
The M2 retains Apple’s base 4 performance plus 4 efficiency configuration. Apple doesn’t disclose what generation of CPU cores they are using here, but based on performance expectations and timing, there’s every reason to believe these are the Avalanche and Blizzard cores that were introduced for the first time on the A15. […]
Avalanche offers only modest performance gains, as Apple has invested most of its upgrades in improving overall power efficiency […]
Based on what we’ve already seen with the A15, this larger update of this generation is on the efficiency side. Blizzard CPU cores behave more and more like not-so-small cores, offering relatively high performance and a much larger backend design than what we see with other Arm efficiency cores.
35% faster GPU, but at a power cost
Overall, the M2 GPU seems to be more power efficient, but if you want to use all the power, it will use more power than the M1.
Apple touts two performances for the M2’s GPU. At iso power (~12W), the M2 should deliver 25% faster GPU performance than the M1. However, the M2’s GPU can, for better or worse, also consume more power than the M1’s GPU. At its full power state of 15 watts, Apple says, it can deliver 35% more performance.
It’s important to note that the base M2 chip supports Apple’s ProRes and ProRes RAW codecs, which are only available on the Pro, Max, and Ultra versions of the M1.
50% more memory bandwidth
Another thing the M2 base borrows from the more powerful versions of the M1 chip is LPDDR5 memory support, compared to LPDDR4 for the M1 base. This gives the chip about 50% more memory bandwidth, which will definitely help GPU performance.
No I/O enhancements
Smith says Apple’s specs for USB and other I/O protocols appear unchanged from the M1. Apple hasn’t said anything about the M2’s PCIe capabilities, so that too is likely unchanged.
As always with AnandTech rooms, worth reading the whole thing.
FTC: We use revenue-generating automatic affiliate links. After.
#chip #chip #Heres