AMD

Radeon Pro VII-AMD's bold response to Nvidia

AMD invited Gamereactor to an online press conference that was supposed to be special, and did not disappoint. It was the revelation of a wonderful new graphics card that, although not directly oriented to gaming, can mess with the business of graphics cards.
The card in question is the new Radeon VII, which meets all the necessary requirements: it is fast, has a huge capacity, and most of all, it is much cheaper with the equivalent of Nvidia. This is relevant because this market sector, from graphics cards facing large productions and high-performance computing has been dominated by Nvidia, and Monopoly is never good.
This type of Nvidia cards reach around 9 thousand dollars, which may seem absurd to a player, but it has been typical for something like the Nvidia Quadro GV100, indicated for large companies. How much does this Radeon VII cost then? The target price at this time is 1900 dollars, and although the capacity of a single board is not equivalent, it is possible to combine two for 3798 dollars and ensure superior performance than the GV100 frame... $ 5,200 less.
This puts AMD on a direct route to the "throne" in terms of dual-precision computing, allowing more complex simulations, design validations, and the use of 64-bit data types. We are talking about advanced works that require graphics cards much higher than the gaming card type, and are used to calculate the type of heat that an airplane can withstand, design cars, and simulate the operation of machines, without actually having to build all of that. It's a huge market, which requires engineering-assisted computing capable of calculating incredible amounts of data, and that's not usually cheap. Projections indicate that this market will yield at least $ 19 billion in five years.
Now, comparing two Radeon VII to a CP100 frame, which costs 7769 dollars, you gain more speed, since it supports PCIe of fourth generation and has six outputs with active IFL, against the ports 4/4 + 1DVI-d DL. In terms of computing power, the AMD dual card guarantees a massive 13.1 TFLOPs, which more than doubles the power available on a GP100. Even the GV100 falls far short, with only 7.4 TFLOP. This in addition to using faster memory than Nvidia'S GDDR6.

Unlike other boards, Radeon VII also includes error correction and remains below 73 degrees, while an RTX 5000 frame exceeds 80 degrees.

This is all very interesting, but what brings of really relevant to the players? Well, this card from AMD will also be used for the production of content at 8K. we are talking about 33,277,600 pixels on the screen, which in the future will be the new standard. AMD has not specifically talked about gaming, but this power will certainly be adapted to gaming cards, which at this time rarely manage to reach 4K and 60 frames per second with all the top graphics capabilities. If we put ray-tracing on top of that, it's even harder to get that target. Several rumors suggest that following the development of these Radeon VII, AMD may be creating gaming cards capable of rivaling or even surpassing NVIDIA'S RTX cards. A RTX 2080 Super, for example, costs more than 800 or even 900 euros in Portugal, but this great pressure from AMD could force Nvidia to review prices to stay competitive.
That is, although Radeon VII is not connected to gaming, its development can greatly impact the gaming graphics card market in the medium term.

Kim Olsen
© Gamereactor