NVIDIA's flagship card, the RTX 4090 graphics card, uses TSMC's 5nm technology, which is very good and powerful, but the price is 12,999 yuan. Jen-hsun Huang, CEO of NVIDIA, said that this is because of the high price of contract chip manufacturing, and they can't do anything about it.
This is the inevitable problem for chip companies to use advanced technology, and in the past two years, the production capacity of the big price increases, TSMC 5nm cost.
NVIDIA can also rely on performance and branding to drive up the final price of its products to make up for the cost increase. What about AMD? It's not feasible to raise prices as much as NVIDIA, but this time around with the RX 7000 series, AMD is surprising.
Not only did the price of the graphics card not go up,The RX 7900 XTX 7999 is 5,000 dollars cheaper than the RTX 4090.
How does AMD do it when it's so cheap and still makes money? In the RDNA3 architecture analysis released tonight, AMD explains how the RX 7000 graphics card uses small chip design to solve the cost problem.
However, it is a little strange that AMD explains the cost advantage of GPU small chips with the data of Ryzen PPT content instead, probably because there is no difference between CPU and GPU in this respect, I am too lazy to re-make PPT.
According to AMD, 7nm 16-core processors would be expensive if the idea of a native big core were used,It is 2.1 times higher than the cost of two 8-core CCD+12nm IOD cores, which is a big difference.
Not only will larger 7nm chips have to be made, but production will also be reduced, further raising costs.
Between the lines, that is to use the idea of small chip design RX 7000 series graphics card, the cost has a lot of advantages, this is RX 7900 XTX graphics card has the courage to be cheaper than friends 5000 yuan of confidence.