Wen: the aurora
Source: Python Technology [public ID: pythonall]
This year’s Singles’ Day, when everyone is busy shopping, Apple inc. released a Mac product equipped with self-developed M1 chip, marking apple officially entered a new era of self-developed chip. The M1 chip adopts the classic large and small core design of the ARM architecture, together with the neural engine and powerful GPU, to provide a more dramatic performance improvement for the Mac.
How strong the M1 chip is
The M1 chip is Apple’s first designed specifically for the Mac and the first ARM based chip for the Mac platform. The M1, like the A14, uses the same 5nm process and integrates 16 billion transistors, 35% more than the A14. The CPU adopts a 4+4 core design (four high performance cores and four high performance cores), which is also stronger than the A14’s 6 core design. The GPU also integrates eight cores and supports 2.6TeraFlops, making it the highest performance integrated graphics card in the industry. The memory adopts the “unified memory” design, so that the CPU and GPU share the same memory pool, which not only makes data sharing more efficient, reduces latency, but also saves power. The M1 processor also includes a 16-core neural network engine, including the Secure Enclave security module, which can perform 11 trillion calculations per second, increasing machine learning speed by a factor of 15.
From CPU, GPU, neural network engine to Apple T2 chip, the overall integration is highly integrated, in short, it is no longer a simple CPU. M1 also supports a number of other technologies, such as the latest ISP (image signal processor), the latest hardware security partition, storage controller, low power and high power media encoding and decoding engine, etc. And for consumers, the battery life is also a big highlight. Compared to the latest PC processors, the performance at 10W power consumption is twice that of the “Friend”, and the power consumption at the same performance is only a quarter of the Intel chip.
Know before you buy
Apple has been trying to unify the ecosystem, to solve the pain point of the ecological split between x86 Mac and ARM iPhone/iPad, to unify the underlying development architecture of all products, so that the software of iPhone/iPad can also be used normally on Mac computers, so that there will be a large space for integration. Of course, integration is not an overnight thing, may take several years to complete, integration process will also appear a series of problems.
-
After integration, the software was switched to THE ARM architecture. Although the software developed in the future need not be adapted to the platform, the software on the Mac x86 platform in the past cannot be switched to the ARM platform immediately. During this period, Running the old software on the M1 Mac required Rosetta2 to simulate the x86 environment, meaning there was a process of instruction translation, so it was definitely different from running on real x86.
-
The M1 MacBook supports up to 16 gigabytes of memory, while the x86 MacBook supports up to 32 gigabytes.
-
The hard disk supports a maximum of 2TB, and the x86 version supports a maximum of 4TB.
-
Can’t upgrade to 10 gigabit network port, can’t connect to external unique display, only with thunder 3 port, but not the latest thunder 4.
-
For those who prefer to buy a Mac and install Windows, note that it is not available on the latest M1 Mac.
conclusion
At present, there are only three products equipped with M1 chips: MacBook Air, MacBook Pro 13.3 and Mac Mini. M1 is only the first generation product, and it is the same as the A-series chips. The general initial generation of products are more problems, need to go through continuous optimization. So if that’s something you care about, you can wait. If you like to try new things and don’t care about the details, you can jump right in. OK, that’s all for this time, if you like, check back.
Welcome to follow the wechat official account: Python technology, here we have personally written 100 days of actual practice, there are all kinds of interesting programming practice, there are all kinds of learning materials, and a large group of lovely friends to discuss with each other.
