Does Ethereum use n card or a card

1. Hello, the mining of bitcoin is actually doing distributed computing, and the strongest point of graphics card is floating-point computing. A card and N card are more suitable for floating-point operation than a card. N card should spend more energy on rendering, and a card has more Alu units than n card. At this point, a card is 2-3 times faster, so mining speed with a card is faster

2. The two can't be compared, it depends on which aspect you are in, for example,
after the general game is launched, n-card users can play it one month earlier than a-card users
the default image quality of a card is the highest, and that of N card is the lowest
there are also a-card users who like to watch DVD and n-card users who like to play 3dsmax
a card is better at running D3D, n card is better at running OGL
there are also instructions when the driver is updated:
n card: the driver improves the quality of AF, and enables you to have * *% performance improvement in... Games. It is recommended to download
a card: the driver improves the quality of DVD playback and corrects the bug in the game. It is highly recommended to install
in addition, shrinking n-cards usually have no official name, while shrinking a-cards usually have official names
in addition, the n-card driver likes to conflict with small games, and the a-card driver likes to conflict with big games
in addition, the n-card driver likes to play digital games, as if the driver developers change a few lines of code every day, and then hand it over to the hardware manufacturer for release, which will be officially released three months later. The monthly number of a card increases by 0.1 on time, even if the development is not completed

3. The difference between the early n-card and a-card is
a-card design
n-card playing games
in terms of cost performance, n-card is better than a-card. It is recommended that you buy n-card, and if the budget is sufficient
it is recommended that you buy seven rainbow flame warlord gtx660, and one is gtx660ti, but it costs several hundred more ti
the cost performance is not as good as gtx660!

4. At present, there is no way to sell a card cheaply, because it is a pricing strategy to compete for the market, which does not mean amd does not want to sell at a high price. Now NVIDIA's new architecture saves power, has high performance, and does well in advertising. AMD's card consumes a lot of power, and some online game watches are not as good as NVIDIA's. If you sell them for the same price, few people will buy them. The performance of Au depends on the core and frequency. The price is power consumption, and some games are not as good as IU, Now AMD's problem is the same as Intel and NVIDIA's in the past. The core of the problem is to compete with high power consumption. You can see where the price of a card comes from when mining, so it's not worth the price. Things that are easy to sell are not cheap, and the profits are also large. You can't sell things that are not easy to sell except to rece the price, or you will die

5. At the same price, consider the color effect better, choose a card, consider the game support more choose n card. A card has advantages for EA games, such as the best flying car and battlefield series.

6. A card with the same price is certainly better, because ATI's graphics card is based on cost performance, just like HD3850. The performance of this card is similar to geforce
9600GT, but the price is about 100 yuan lower than 9600GT.

7. This question mainly depends on what price you want to buy. In fact, it doesn't matter what the low-end graphics card is. The performance difference between the low-end procts of the two brands is very small, almost negligible. The mid end proct amd has a slight advantage in the Chinese market. The main reason is that the price of AMD's procts is really cheap. Although there are many disadvantages, the price advantage is unmatched by NVIDIA, High end procts, I personally think NVIDIA's graphics card is better. There are too many advantages, such as energy saving, convenience, stability, low temperature, physical acceleration, sufficient supply, and so on. Moreover, the performance is not inferior to AMD's. although the price is 100 yuan to 200 yuan higher than AMD's, these advantages are enough to make up for this difference
if it's me, under GTX250, I will mostly consider starting with a card, which is also 599 yuan. Why don't I choose a card with higher performance? However, when the GTX250 is above, I don't think about a card at all, because anyway, more than 1000 yuan has been spent. I don't care whether it is 1100 or 1200. The performance is almost the same. Moreover, n card also has more physical acceleration. I won't give myself a lot of trouble for just 100 yuan< br /> ×××××××××××××××××× Powerful dividing line ××××××××××××××××××
it's really hard to say the last question added by the landlord, because havok was acquired by Intel. Although Intel let havok operate independently, it seems that it is the default that havok cooperates with AMD, but don't forget that the real owner of havok is Intel. Intel's default cooperation with AMD is to contain NV and get something from AMD. AMD and Intel are old rivals, People all over the world know that AMD's doing this can be said to be competing with a tiger. If it's not careful, it will be swallowed by Intel. However, if it doesn't do this, it won't work. NV has steadfastly defeated ATI with PhysX on high-end graphics cards. Why NV's PhysX has been out for nearly two years and AMD hasn't supported havok up to now is the reason. This shows that AMD has not bowed its head like Intel, However, it's hard to say when this situation will be supported. Now Intel's new generation of core
I3 processor integrates GPU. When Intel can completely plug the graphics card into the CPU, both AMD and NV will be in danger. At that time, the a card and N card will all become Intel's
as for
aersaisi1's saying that there are no more than three games that support physics acceleration, it is absolutely a-fan's words. Now there are more than 150 games that perfectly support physics acceleration, and there are a lot of them if you look for them. It is suggested that
aersaisi1 go to the phone first and then answer the questions
in addition, is PhysX really dispensable? The following is a video comparison with or without PhysX. Which a card can achieve this effect at this stage?

8. The performance is similar. A card on high score screen, high voltage has advantages. This price is not explained by MSI. The performance is a little lower than that of ice dragon, and the price is much cheaper.

9. Objective evaluation: if you don't care about the price, and often play the latest large stand-alone games, it is recommended that you buy n card, which has better support for the game than a card. If you pay attention to cost performance and have no requirements for the latest large single game, please buy a card. A card has a small advantage in color, and the latest proct supports Dx11. It mainly depends on your purpose.

10.

It's better to choose card a for graphics card. Card a has strong single precision floating-point operation ability and is suitable for drawing. Card n has better support for games. Users who often play games can choose card n

if the drawing requirements are high, it is more appropriate to choose professional graphics card

Single precision number is a way to express approximate value of real number by computer. In VB, the single variable is stored in the form of IEEE 32-bit (4-byte) floating-point value. Its range is from - 3.402823e38 to - 1.401298e-45 when the number is negative, and from 1.401298e-45 to 3.402823e38 when the number is positive

floating point operation is real number operation, because the computer can only store integers, so real numbers are divisors, so floating point operation is very slow and there will be errors. Now most machines are 32-bit, that is to say, if 32 bits are used to represent integers, then it is 0 to 2 ^ 32-1 for unsigned integers and - 2 ^ 31 to 2 ^ 31-1 for signed integers