The difference between a card and N card of Ethereum mining mach
in terms of eth, a card generally needs to brush the BIOS to achieve the highest computing power, and it needs to install a special mining driver
the N card is easy to operate, and the driver can be installed at will. However, the price performance of a card will be higher than that of N card after digging eth
however, in terms of mining life, if eth can't dig in the future, a card may cost a lot. For the time being, a card can only dig Eth and Zec. Zec has proced professional mining machines. In the future, it may not be cost-effective to dig video cards. There are really few currencies that a card can dig. N card is different. Although eth is slightly lower than a card, n card supports many algorithms. There are more than 100 counterfeit coins that can be g. One day, Eth and Zec collapse, and you can continue to live by changing any currency. Aiminer supports both a-card and n-card mining.
1: a card is good at drawing, and N card is good at playing
2: a card drive is bad, n card drive is good
3: a card has good color and N card has fast speed
4: the power consumption of a card is low, and the power consumption of N card is large
5: the price of a card is better than that of N card.
A card originally refers to ATI's graphics card, but after ATI was acquired by AMD, AMD's graphics card is also called a card. N card refers to NVIDIA's graphics card. A card and N card are developed and proced by two different graphics card manufacturers
The architecture of N card seems perfect, but its disadvantages are also obvious. Because each stream processor corresponds to independent instruction transmitter and control unit, the volume is huge, and the control unit accounts for a large proportion of the consumption of transistors. In the case of the same number of transistors, n card can do a lot less computing units. When the number of stream processors is relatively small, the performance will be insufficient when processing 4D instructions (because it takes four stream processors to process one instruction), so the frequency of stream processor of N card will be more than twice as high as the core frequency to make up for the defect in number. Another disadvantage is the huge power consumption2. The n-card architecture is highly efficient, flexible and easy to perform in practical applications. But the power consumption is difficult to control, and less processing units limit its theoretical computing ability
In the aspect of a card, although the general 1D stream processor is also used as the execution unit, the instruction level parallel architecture is adopted. Every five stream processors are a group, and each group can receive up to one 5D instruction at a time (while the N card receives 1D instruction), and all the instructions are packaged into 5D instructions (while the N card is split into 1D instructions), So the architecture of a card is also called 5D architecture. This design can achieve high instruction throughput, can make a huge computing unit with fewer control units, and has less transistor consumption, so the stream processor of card a is generally 4-5 times of that of card n, the theoretical computing power is far stronger than that of card n, the power consumption is relatively lower, and the chip area with the same performance is relatively small The disadvantages of a card architecture are also obvious. Although the total computing performance is strong in theory, once mixed instructions or conditional instructions are encountered, it is difficult for the front-end to realize complete 5D packaging, which often turns into 3D, 2D and 1D, resulting in only 3, 2 or even 1 stream processor in each group working, and almost half of the units are wasted. In order to optimize this kind of architecture, it is necessary to rece the appearance of mixed and conditional instructions (which requires a lot of effort of programmers), or eliminate them (which is impossible). So in the software optimization, a card is at a disadvantage, often unable to play its e performance
4. The advantage of a card architecture lies in its theoretical computing ability, but its execution efficiency is not high, and its adaptability to complex and changeable tasks is not strong. Without the support of software, it is often unable to play its e performance. Therefore, in addition to the support of game manufacturers, a card often releases driver patches optimized for a certain game
5. The performance of a card and N card in the game has their own advantages and disadvantages. In most game tests, they have their own advantages and disadvantages, which can be said to be equal. Although the architecture of N card and a card is different, in order to be compatible with all kinds of software and hardware, they are designed according to certain standards, so the implementation of performance is the same. The launch of each single game is a landmark publicity effect, and the support tendency of single game manufacturers has become a place for the two companies to compete. Therefore, we often see the two brands in turn. And this phenomenon leads to a considerable number of beginners into a misunderstanding: who supports more games, whose graphics card is good. In fact, the fact is not so simple. In the eyes of every game vendor, there will always be only a number of players, and they will not be stupid enough to give up one for the sake of the other. Therefore, even if the game is claimed to be optimized for one graphics card, it will leave a certain degree of backwardness for the other. Therefore, in most game tests, even if two graphics cards win or lose each other, the gap is not big
The differences between the two are as follows:
1. Different manufacturers
1. N card: refers to the graphics card procts developed by NVIDIA company
2. A card: refers to the graphics card procts developed by AMD company
Second, the architecture is different
1. N card: the control unit accounts for a large proportion of the consumption of transistors. In the case of the same number of transistors, n card can do a lot less computing units
2. A card: it adopts a general 1D stream processor as the execution unit, but it adopts an instruction level parallel architecture, with every five stream processors as a group
Three, different advantages1, n card: n card architecture is highly efficient, flexible, and easy to play its e performance in practical application. But the power consumption is difficult to control
2. Card a: the stream processor of card a is generally 4-5 times of that of card n, the theoretical computing power is also far stronger than that of card n, and the power consumption is relatively lower
The structure of general graphics card is as follows:capacitance: capacitance is a very important component of graphics card, because the quality of display image mainly depends on the quality of capacitance, which directly affects the quality of graphics circuit
Display memory: it is responsible for storing all kinds of data that the display chip needs to process, and its capacity and performance directly affect the display effect of the computer. The new video cards all use DDR3 / ddr5 video memory, and the mainstream video memory capacity is generally 1GB ~ 2GB GPU and fan: GPU is the chip of graphics card, which is responsible for most of the computing work of graphics card, rather dry the role of CPU in the computer. The function of GPU fan is to heat the GPUgraphics card interface: it is usually called golden finger, and can be divided into PCI, AGP and PCI Express. PCI and AGP display long interface are basically eliminated, and PCI Express is the mainstream graphics card in the market
peripheral interface: the long peripheral interface is responsible for the output task of the graphics card. The new graphics card includes a traditional VGA analog interface and one or more digital interfaces (DVI, HDMI and DP)
bridge interface: the medium and high-end graphics card can support multiple blocks working at the same time, and the bridge interface is connected between them through the bridge
n card is characterized by attention to detail, and it has more advantages under the standard image quality
a card of the same level is not as fast as n card in most games. N card is optimized, and its anti aliasing and large resolution are better than a card. In terms of cost performance, a card of the same level is higher.
a card of the same price, n card, a card is better
with the same ability, a card and N card are about 200 yuan more expensive than a card.
1. A card is suitable for playing games, and N card is suitable for design
2. Generally speaking, n-card is more high-end than a-card. It's also fun to buy the right n-card, and it can do pictures, film and television design. If you only play games, buy a 256 video memory will not be eliminated for a while
3. The technical emphasis of N card and a card is that n card pays attention to 3D performance and speed, and a card pays attention to 2D image quality
4. The advantage of N card is that the frequency of the graphics card can reach nearly 100%. The disadvantage is that the power consumption is high. The advantage of a card is that the power consumption is low. The disadvantage is that the software optimization is not enough
5. The n-card architecture is highly efficient, flexible and easy to perform in practical applications, but its power consumption is difficult to control and its theoretical computing power is limited by fewer processing units
6. The advantage of a card architecture lies in its theoretical computing ability, but its execution efficiency is not high, and its adaptability to complex and changeable tasks is not strong, so it needs software support to play its e performance.