Since its introduction in 1965, Moore’s Law has been the norm for the global semiconductor industry for decades. But gradually seeing a bit of fatigue in the past few years, there is a possibility that it may end. If this is true, Intel will follow Moore’s law for decades. ) It is bound to need a new strategy to continue the trend. In addition, it is thought that the end of Moore's Law is good news for NVIDIA. However, the CPU will not disappear or be eliminated. It will still exist. Not so stunning. According to Forbes, the SeekingAlpha columnist Bruce Pile pointed out that in the current global semiconductor industry, Intel is facing increasing challenges and gradually not seen as unbreakable. In the past PC era CPU It plays a major role in computing, so the clock rate and crystal density are very important. Since then, with the emergence of more meta-computing requirements, companies such as NVIDIA have invented a new type of processor with different architectures such as graphics processing chips (GPUs). , With thousands of parallel cores, it can engage in a greater number of simple tasks with less transistor density, making it suitable for use in games and other applications Since then, the next shift is underway. GPUs are considered as computing devices. Nowadays they are also suitable for emerging technologies such as artificial intelligence (AI). Graphics cards are also really suitable for a large number of parallel digital operations, and these cores can increase linearly. Computing power, bypassing the limitations of today's Moore's Law, differs from the case where the performance of a CPU in a CPU doubles by only 10%. Therefore, as NVIDIA CEO Huang Renxun previously stated, Moore's Law is dead, and GPUs will replace Intel and others. The CPU produced by the chip maker does not believe that the CPU will disappear in the future, and that the PC and other computing functions will only provide the computing needs in the future. However, these PC type functions will not become an amazing part of the industry. The most amazing part of the industry is that if GPUs provide more computing needs in emerging technologies, it can be seen that development APIs convert computing tasks to parallel graphics operations performed by the GPU, or are visible in the data center. It is possible to offload more computing tasks from the CPU to provide operations on the GPU. Even this would help NVIDIA GPU vendors continue to be in emerging technologies Benefits, such as AI, data centers, automotive electronics, medical platforms, etc. However, Pile believes that Intel's current size is still important, such as Intel's market value is still nearly 1 times more than NVIDIA, sales performance than any other publicly listed chip makers Still more, it still dominates the global CPU market. In addition, even if the 28- and 40-nanometer wafer process can produce useful chip products, the Intel 14-nanometer wafer fab is still a competitive advantage for any 28- and 40-nanometer process production. If the chip is produced on a 14-nanometer process, it will help with smaller size, lower cost, faster speed, higher performance, and lower power consumption. In this case, if there is a design that cannot afford the 14-nanometer process, only 28 and A 40-nanometer chip company is likely to become a potential acquisition target for Intel. In this case, Pile believes that with NVIDIA’s current market value of US$137 billion, Intel will have difficulty buying NVIDIA. Currently, NVIDIA and Samsung Electronics Samsung's 14-nanometer manufacturing process produces NVIDIA chips, but it's worth looking at whether Intel can produce a lower-priced, faster product than NVIDIA GPUs in the future. And it's still going on in NVIDIA. Under long, estimated a few years NVIDIA will not be large-scale acquisition of Intel.
