The difference of computing power between CPU and GPU
Main differences between CPU and GPU:
1. CPU is the central processing unit of computer
GPU is the graphics processor of computer CPU is a very large-scale integrated circuit, which includes Alu arithmetic logic unit, cache memory and bus CPU is the core of a computer's control and operation, its main function is to interpret the instructions issued by the computer and process the big data in the computer software GPU is the abbreviation of image processor. It is a kind of microprocessor which is specially used for PC or embedded device to perform image operation The work of GPU is similar to that of CPU mentioned above, but not entirely. It is designed to perform complex mathematical and geometric calculations. This game has high requirements in this respect, so many game players also have deep feelings for GPU 
so CPU and GPU are two completely different things, they just sound the same
extended data:
CPU and GPU are different in design because they are used to deal with different tasks at first, and some tasks are similar to the problems that GPU is used to solve at first, so we use GPU to calculate. The operation speed of GPU depends on how many pupils are employed, and the operation speed of CPU depends on how powerful professors are employed, The professor's ability to deal with complex tasks is very good, but for less complex tasks, it still can't hold many people
of course, today's GPU can also do some slightly complicated work, which is equivalent to upgrading to the level of junior high school students and senior high school students, but it still needs CPU to feed data to the mouth before it can start to work, which is still managed by CPU
CPU: central processing unit, is a very large scale integrated circuit, is a computer operation core (core) and control unit (control unit). Its main function is to interpret computer instructions and process data in computer software
GPU: graphics processor, also known as display core, visual processor and display chip, is a kind of microprocessor which is specialized in image operation on personal computers, workstations, game machines and some mobile devices (such as tablet computers, smart phones, etc.)CPU and GPU are designed for two different application scenarios respectively
1. CPU needs strong versatility to deal with different data types, and at the same time, it also needs logical judgment, which will introce a lot of branch jump and interrupt processing. All these make the internal structure of CPU extremely complex
GPU is faced with highly unified, interdependent large-scale data and pure computing environment without interruption
extended data
application direction of CPU and GPU
GPU is suitable for high predictability and a large number of similar operations, as well as high latency and high throughput architecture operationsGPU English full name graphic processing unit, Chinese translation for "graphics processor". GPU is a concept relative to CPU. As graphics processing becomes more and more important in modern computers (especially in home systems and game enthusiasts), a special graphics core processor is needed<
the role of GPU
GPU is the "brain" of the display card, which determines the grade and most of the performance of the card, and is also the basis for the difference between 2D display card and 3D display card. 2D display chips mainly rely on the processing power of CPU when processing 3D images and special effects, which is called "soft acceleration". 3D display chip is the so-called "hardware acceleration" function, which integrates 3D image and special effect processing functions into the display chip. The display chip is usually the largest chip (with the most pins) on the display card. Now most of the graphics cards on the market use NVIDIA and ATI graphics processing chips< When NVIDIA released geforce 256 graphics processing chip in 1999, it first proposed the concept of GPU. GPU makes the graphics card less dependent on CPU, and does part of the original CPU work, especially in 3D graphics processing. The core technologies of GPU include hardware T & L, cubic environment material mapping and vertex blending, texture compression and bump mapping, al texture four pixel 256 bit rendering engine, etc
in short, GPU is a display chip that can support T & L (transform and lighting, polygon conversion and light processing) in hardware, because T & L is an important part of 3D rendering, which is used to calculate the 3D position of polygons and process dynamic light effects, also known as "geometric processing". A good T & L unit can provide detailed 3D objects and advanced light effects; However, in most PCs, most of the T & L operations are handled by the CPU (that is, the so-called software T & L). Because the CPU has many tasks, in addition to T & L, it also does non 3D graphics processing work such as memory management and input response. Therefore, in the actual operation, the performance will be greatly reced, and the graphics card often waits for CPU data, Its computing speed is far behind the requirements of today's complex 3D games. Even if the working frequency of CPU exceeds 1GHz or higher, it will not help much. Because this is a problem caused by the design of PC itself, it has nothing to do with the speed of CPU<
about CPU and GPU
first question:
the competition of GPU is much more fierce than that of CPU. The CPU of general PC is only Intel and AMD. In terms of GPU, N and a are the two leading manufacturers, but Intel, 3S and other manufacturers can proce low-end procts. Although their procts are not as good as the first two, they can also meet the needs of users in many applications. Therefore, N and a will not die until they run hard. CPU manufacturers did not adopt the advanced technology of GPU, because CPU manufacturers have their own investment in proction lines. It is impossible to eliminate the original proction lines and replace them with new ones. In that way, it may be difficult to recover the original investment. GPU manufacturers, for various reasons, usually design their own procts for others, such as TSMC. In order to receive business, OEM manufacturers have to upgrade their proction equipment continuously to survive. So the above reasons
the second problem
as you said, CPU not only processes AI, plot and other data of the game, but also completes some image aspects. Every time Microsoft releases a new DX, not every GPU can support the new features of DX, so some image tasks have to be completed by CPU. There are also some features, such as the gravity feature, which used to be completed by the CPU. Now some GPUs can also support these tasks. These tasks are completed by the GPU<
the third problem
GPU is equivalent to the CPU dedicated to image processing. Because it is specialized, it is strong. When processing images, its work efficiency is much higher than that of CPU. However, CPU is a general data processor, which is strong in processing numerical calculation. It can complete tasks that GPU can't replace, so GPU can't replace CPU< In addition,
now amd has acquired the design manufacturer of a-record graphics card chip, amd sees that in the future, CPU and GPU can take the lead in the competition only by taking a road of integration. How CPU and GPU work together to maximize efficiency is the problem AMD is considering, but also Intel's problem
the fourth problem
one of the remarkable features of Microsoft's release of Windows 7 is to combine the powerful power of GPU and CPU to enhance the value of GPU in hardware use. In Windows 7, CPU and GPU form a collaborative processing environment. CPU computes very complex sequence code, while GPU runs massively parallel applications. Microsoft uses DirectX compute to take GPU as one of the core components of the operating system. DirectX Compute It enables developers to take advantage of GPU's massively parallel computing power to create compelling consumer and professional computing applications. In short, DirectX compute is the GPGPU general computing interface developed by Microsoft. It wants to unify the GPU general computing standard. That is to say, after Windows 7, the hardware status of GPU will be second only to CPU and play a greater role.
CPU and GPU are very different because of their different design goals. They are designed for two different application scenarios. CPU needs a strong versatility to deal with various data types, while GPU is faced with highly unified, interdependent large-scale data and pure computing environment without interruption
the difference between them lies in the structure difference of cache system and digital logic operation unit: Although CPU has multi-core, the total number is not more than two digits, each core has enough cache and enough digital and logic operation units, and it also has a lot of hardware to speed up branch judgment or even more complex logic judgment; The number of cores of GPU is far more than that of CPU, so it is called many cores (NVIDIA Fermi has 512 cores). The cache size of each core is relatively small, and the number of logical operation units is small and simple (GPU is always weaker than CPU in floating-point calculation at the beginning). As a result, CPU is good at dealing with complex computing steps and complex data dependent computing tasks, such as distributed computing, data compression, artificial intelligence, physical simulation, and many other computing tasks. Due to historical reasons, GPU is proced for video games (so far its main driving force is still the growing video game market). In 3D games, a kind of operation often appears is to perform the same operation on massive data, such as: to perform the same coordinate transformation on each vertex, and to calculate the color value of each vertex according to the same lighting model. The multi-core architecture of GPU is very suitable for sending the same instruction stream to the multi-core in parallel and executing with different input data. From 2003 to 2004, experts outside graphics began to notice the unique computing power of GPU, and began to try to use GPU for general purpose computing (GPGPU). After NVIDIA released CUDA, AMD and Apple also released OpenCL, GPU began to be widely used in the field of general computing, including: numerical analysis, massive data processing (sorting, map rec, etc.), financial analysis, etc
in short, when programmers write programs for CPUs, they tend to use complex logical structures to optimize algorithms, so as to rece the running time of computing tasks, namely latency. When programmers write programs for GPU, they use the advantage of GPU to deal with massive data and improve the total data throughput to cover up the latency. At present, the difference between CPU and GPU is graally narrowing, because GPU has made great progress in dealing with irregular tasks and inter thread communication. In addition, the problem of power consumption is more serious for GPU than for CPU
generally speaking, the difference between GPU and CPU is a big topic. It can even take 32 class hours and more than ten lectures in a semester. Therefore, if the questioner has more specific questions, he can further raise them. I will try to answer within my knowledge.
the difference between them lies in the structure difference of cache system and digital logic operation unit in the chip: Although the CPU has multiple cores, the total number is not more than two digits, each core has enough cache and enough digital and logic operation units, and there are many hardware to accelerate branch judgment or even more complex logic judgment; The number of cores of GPU is far more than that of CPU, so it is called many cores (NVIDIA Fermi has 512 cores). The cache size of each core is relatively small, and the number of logical operation units is small and simple (GPU is always weaker than CPU in floating-point calculation at the beginning). As a result, CPU is good at dealing with complex computing steps and complex data dependent computing tasks, such as distributed computing, data compression, artificial intelligence, physical simulation, and many other computing tasks. Due to historical reasons, GPU is proced for video games (so far its main driving force is still the growing video game market). In 3D games, a kind of operation often appears is to perform the same operation on massive data, such as: to perform the same coordinate transformation on each vertex, and to calculate the color value of each vertex according to the same lighting model. The multi-core architecture of GPU is very suitable for sending the same instruction stream to the multi-core in parallel and executing with different input data. From 2003 to 2004, experts outside graphics began to notice the unique computing power of GPU, and began to try to use GPU for general purpose computing (GPGPU). After NVIDIA released CUDA, AMD and Apple also released OpenCL, GPU began to be widely used in the field of general computing, including: numerical analysis, massive data processing (sorting, map rec, etc.), financial analysis, etc
in short, when programmers write programs for CPUs, they tend to use complex logical structures to optimize algorithms, so as to rece the running time of computing tasks, namely latency. When programmers write programs for GPU, they use the advantage of GPU to deal with massive data and improve the total data throughput to cover up the latency. At present, the difference between CPU and GPU is graally narrowing, because GPU has made great progress in dealing with irregular tasks and inter thread communication. In addition, the problem of power consumption is more serious for GPU than for CPU.
First, we need to explain what CPU (central processing unit) and GPU (graphics processing unit) stand for. CPU is the central processing unit, GPU is the graphics processor. Secondly, to explain the differences between the two, we must first understand the similarities between them: both have bus and external connections, have their own cache system, and digital and logical operation units. In a word, both of them are designed to complete the calculation task
let's intuitively draw the last diagram:
{rrrrrrr}
from the diagram, we can see that CPU and GPU have their own storage (orange part, the actual storage system is more complex than the diagram), control logic (yellow part) and operation unit (green part), but the difference is that the control logic of CPU is more complex, while the operation unit of GPU is small but numerous, GPU can also provide more registers and program controllable multi-level storage resources
the difference between them lies in the structure difference of cache system and digital logic operation unit in the chip: Although CPU has multi-core, the total number does not exceed two digits, each core has enough cache and enough digital and logic operation units, and there are many hardware to accelerate branch judgment or even more complex logic judgment; The number of cores of GPU is far more than that of CPU, so it is called many cores (NVIDIA Fermi has 512 cores). The cache size of each core is relatively small, and the number of logical operation units is small and simple (GPU is always weaker than CPU in floating-point calculation at the beginning). As a result, CPU is good at dealing with complex computing steps and complex data dependent computing tasks, such as distributed computing, data compression, artificial intelligence, physical simulation, and many other computing tasks
GPU is proced for video games e to historical reasons (its main driving force is still the growing video game market so far). In 3D games, one kind of operations that often appear is to perform the same operation on massive data, such as: to perform the same coordinate transformation on each vertex, and to calculate the color value of each vertex according to the same lighting model. The multi-core architecture of GPU is very suitable for sending the same instruction stream to the multi-core in parallel and executing with different input data. From 2003 to 2004, experts outside graphics began to notice the unique computing power of GPU, and began to try to use GPU for general purpose computing (GPGPU). After NVIDIA released CUDA, AMD and Apple also released OpenCL, GPU began to be widely used in the field of general computing, including: numerical analysis, massive data processing (sorting, map rec, etc.), financial analysis, etc
in short, when programmers write programs for CPU, they tend to use complex logic structure to optimize the algorithm, so as to rece the running time of computing tasks, namely latency. When programmers write programs for GPU, they use the advantage of GPU to deal with massive data and improve the total data throughput to cover up the latency. At present, the difference between CPU and GPU is graally narrowing, because GPU has made great progress in dealing with irregular tasks and inter thread communication. In addition, the problem of power consumption is more serious for GPU than for CPU
GPGPU, GPU with CPU processing power. It's mainly the work of GPU. The ability of GPU can help the CPU to calculate. (general graphics processing) is beyond the ability range of GPU, and even has general data processing. Whether it can be a CPU depends on whether it is placed on the motherboard
GPU, a chip for graphics processing GPU is also a kind of CPU. Compared with graphics card, there was no special GPU for graphics processing in early graphics card, which did not support 3D
CPU, a chip for data processing (graphics can also be regarded as data). Central processing unit, a relative concept. Imperceptibly identified as the thing on the motherboard, it controls the GPU by sending instructions to the GPU. It's actually a microprocessor
amd-apu, speed up processor. Integrated GPU core CPU, and fusion. Different from Intel i-series intelligent (GPU CPU works independently)