This month Apple introduced the M2 chip – the second generation dedicated silicon chip for Mac computers. After the M1 chip revolutionized the Mac starting in November 2020, how much better did the M2 really get?
Apple says the M2 chip takes the performance per watt of the M1 even further with an 18 percent faster CPU, 35 percent more powerful GPUs, and a 40 percent faster Neural Engine. There are also other important improvements such as increased memory bandwidth and support for up to 24GB of unified memory.
As Apple’s second major iteration of silicon, it may not be clear how significant the M2’s improvement is over the M1, especially as it sticks to the 5nm manufacturing process, contrary to some rumors that have suggested an even bigger leap in performance and efficiency. In store. The M2 chip debuted in the 13-inch MacBook Pro and the redesigned MacBook Air, so how far the M2 is better than its predecessor may be an important consideration when buying a new Mac. Read on to learn more about the differences between the two slides.
- Manufactured using TSMC’s 5nm (N5) process.
- 16 billion transistors
- 4 high-performance “firestorm” cores
- 4 “Icestorm” energy-saving cores
- 3.2GHz CPU clock speed
- CPU cores debuted in the A14 Bionic chip of the iPhone 12 lineup
- Octa-core graphics processor
- Support 8GB or 16GB unified memory
- Memory bandwidth 68.25 GB/sec
- nervous motor
- H.264 and HEVC hardware acceleration media engine
- Video decoding engine
- video encoding engine
- Image Signal Processor (ISP)
- Made with TSMC’s Enhanced 5nm Process (N5P)
- 20 billion transistors
- 4 high-performance “Avalanche” cores
- 4 energy-saving Blizzard cores
- 3.49GHz CPU Clock Speed
- CPU cores debuted in the A15 Bionic chip of the iPhone 13 lineup
- 10 GPU cores
- Standard memory support 8GB, 16GB or 24GB
- 100 GB/sec memory bandwidth
- 40% faster neural engine
- H.264, HEVC, ProRes and ProRes RAW hardware acceleration media engine
- High bandwidth video decoding engine
- video encoding engine
- ProRes Engine Encryption and Decryption
- The “new” image signal processor (ISP)
Manufacturing process and transistors
Like the A14 Bionic, the M1 chip was built using TSMC’s first-generation 5nm manufacturing process. On the other hand, the M2 uses TSMC’s second generation 5nm process like the A15 Bionic chip. The M2 adds four billion additional transistors for a total of 20 billion – 25 percent more than the M1. An improved 5nm manufacturing process is at the heart of many of the M2’s performance and efficiency improvements.
The M1 and M2 have four high-performance cores and four energy-saving cores, but while the M1 features the “Firestorm” and “Icestorm” cores of the A14 Bionic chip, the M2 offers the “Avalanche” and “Blizzard” of the A15 Bionic chips. According to Apple, this results in 18 percent greater multithreaded performance than the M1.
In early Geekbench benchmarks, the M2, which runs at 3.49GHz compared to 3.2GHz for the M1, received a single-core score of 1,919, which is roughly 12 percent faster than the single-core score of 1,707 for the 13-inch MacBook. forefront. The M2 has a multi-core score of 8,928, an increase of about 20 percent from the 7419 score for the M1 model. This is completely in line with Apple’s claim that the M2 chip is up to 18 percent faster than the M1.
Both chips have high-performance cores with 192KB of L1 instruction cache and 128KB of L1 data cache. Energy-saving cores contain a 128KB L1 instruction cache, a 64KB L1 data cache, and a 4MB L2 shared cache. The only difference here is that the shared L2 cache is larger on the M2 chip — 16MB instead of 12MB on the M1.
The M2 features two more GPU cores on the M1, which results in a moderate increase in graphics performance. Apple says the M2 has up to 25 percent higher graphics performance than the M1 at the same power level, and up to 35 percent better performance at its maximum power. In early Geekbench Metal benchmarks, the M2 chip scored 30.627, a marked improvement over the 21001 score obtained by the M1.
The M1 and M2 have dedicated video encoding and decoding engines for H.264 and HEVC hardware acceleration, but the M2 video engines are also capable of ProRes and ProRes RAW acceleration to enable playback of multiple streams of 4K and 8K video. Additionally, the M2 media engine includes a higher bandwidth video decoder, supporting 8K H.264 and HEVC video.
The M1 and M2 come in configurations with either 8GB or 16GB of unified memory, but the M2 adds an additional top-tier configuration with 24GB. The M2 memory controller can also deliver 100GB/s of combined memory bandwidth, a significant improvement over the M1’s 68.25GB/s memory bandwidth.
The M2 features improvements to many of Apple’s custom silicon technologies. For example, the Neural Engine can process up to 15.8 trillion operations per second — more than 40 percent of M1. The M2 also has Apple’s latest Safe Zone and a new Image Signal Processor (ISP) for better image noise reduction.
Overall, the M2 chip offers moderate improvements over the M1, even if most M1 users are unlikely to notice significant improvements when upgrading to the M2. The M2’s developments are generally worthwhile, if not transformative, and the chip is sure to provide a more modern experience with Apple’s low-spec silicon devices — especially those that come from an Intel-based device.
While the M2 offers improvements across the board thanks to an improved 5nm process, updated cores, and additional GPU cores, the major upgrades come for users who need to work with video, as well as those with memory-intensive workflows. The M2’s high-bandwidth video decoder and dedicated ProRes and ProRes RAW video engine deliver meaningful improvements to video editors, while the unified 24GB memory layer and 100GB/s memory bandwidth significantly enhance the capability of the M2 hardware. To handle memory-hungry apps and intense multitasking.