Why bitcoin hurts graphics card
2. Modern multi-core CPU aims at instruction set parallelism (ILP) and task parallelism (TLP), while GPU is data parallelism (DLP)
3. GPU often has a larger bandwidth of memory, that is, the so-called video memory, so it will also have good performance in high throughput applications
4. CPU is a general-purpose operation. In short, it means brainless calculation and violent calculation. Whether it's 100 children sharing apples or simulating the earth, it can be decomposed into the most basic calculation of 1 + 1 = 2. GPU is specially designed to deal with high-order mathematical algorithms, such as calculating the relative positions of light source, object, viewpoint and shadow, which requires trigonometric functions to be piled up. The bitcoin digger uses SHA-256, a secure hash function invented by the US National Security Agency, which is generally used for password encryption and decryption. This algorithm can perform a large number of 32-bit integer right-rotate operations, which is very suitable for GPU which is good at large-scale concurrent computing and password cracking.
Why use graphics card to mine
compared with CPU, GPU is used for general purpose calculation. So it can stack hundreds of stream processors awesome. Each stream processor is like a small CPU. Although its ability to run complex programs is far from the power of CPU, it can not support many processors, so the actual performance, especially the single precision floating point performance is much stronger than that of CPU. p> However, as we have mentioned before, a card is very suitable for general computing such as virtual currency because it has many stream processors, and the speed of obtaining specific solutions is much higher than that of N card. Therefore, many miners rush to buy a card to calculate virtual currency for profit
of course, with NVIDIA's Pascal graphics card, n card's ability in mining virtual currency is not as different as a card before. So in the current situation that a card is generally out of stock, those mid-range N cards have become the second choice for miners
What are the disadvantages of mine cardfinally, I want to talk about why we can't choose mine card when we are shopping for second-hand cards. Because the miners are mining bitcoin, the most important thing is to seek profits. Therefore, the graphics card they use is running and calculating all the time, and its strength is far greater than that of daily game applications. In other words, running such a calculation for one day is equivalent to running a graphics card for one month
in addition, the places where virtual currency is mined are mostly the places where multiple graphics cards gather. Thousands of graphics cards are constantly calculating in a room, and their heat is also quite huge. Miners can not choose the chassis with good heat dissipation conditions like ordinary consumers, so that the heat can be released smoothly. In this way, the graphics cards can still operate for 7 * 24 hours under high temperature conditions, It also deepens its aging degree
the result of doing so is that the graphics cards used for mining are lost prematurely. If you buy these graphics cards, the result will be unstable operation, restart and other things will continue to happen
moreover, at present, manufacturers also have strict requirements on mine cards, such as banning second-hand sales and shortening the shelf life. If users buy these mine card graphics cards, future maintenance will also be a big challenge
Starting from the essence of bitcoin, the essence of bitcoin is actually a special solution generated by a bunch of complex algorithms. A special solution is one of the infinite (in fact, bitcoin is finite) solutions that can be obtained from the equations. Every particular solution can solve the equation and is unique
why use graphics card to mine
compared with the complex computing of CPU, GPU used by graphics card is used for general computing. So it can stack hundreds of stream processors awesome. Each stream processor is like a small CPU. Although its ability to run complex programs is far from the power of CPU, it can not support many processors, so the actual performance, especially the single precision floating point performance is much stronger than that of CPU. p>
the actions in bitcoin can be considered as the use of graphics card to do continuous general computing, and the computational complexity is relatively low and quite repetitive. The stream processor used by graphics card is just suitable for such an algorithm. It's like constantly solving equations. The graphics card is 2000 middle school students, while the CPU is 8 math doctors. Although the mathematical knowledge of the doctor of mathematics is far more than that of the middle school students, the speed of solving medium-sized equations such as virtual currency is certainly not as fast as that of 2000 middle school students
so we used graphics card to mine virtual currency at the beginning. After that, someone developed a specific algorithm to make the stream processor of the graphics card greatly strengthen the virtual currency solution, which is the origin of the mining machine
1. CPU is mainly optimized for serial instructions, while GPU is optimized for massively parallel operations
2. Modern multi-core CPU aims at instruction set parallelism (ILP) and task parallelism (TLP), while GPU is data parallelism (DLP)
3. GPU often has a larger bandwidth of memory, that is, the so-called video memory, so it will also have good performance in high throughput applications
the CPU needs strong generality to deal with various data types, and at the same time, it also needs logical judgment, and it 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. GPU uses a large number of computing units and ultra long pipeline, but it only has very simple control logic and saves cache. The CPU not only occupies a lot of space by cache, but also has complex control logic and many optimization circuits. So unlike CPU, GPU is good at large-scale concurrent computing, which is suitable for password cracking
in short, CPU is a general-purpose operation. In short, it means brainless calculation and violent calculation. Whether it's 100 children sharing apples or simulating the earth, it can be decomposed into the most basic calculation of 1 + 1 = 2. GPU is specially designed to deal with high-order mathematical algorithms, such as calculating the relative positions of light source, object, viewpoint and shadow, which requires trigonometric functions to be piled up. The bitcoin digger uses SHA-256, a secure hash function invented by the US National Security Agency, which is generally used for password encryption and decryption. This algorithm can perform a large number of 32-bit integer right-rotate operations, which is very suitable for GPU which is good at large-scale concurrent computing and password cracking.