The red devil, AMD has released a flagship GPU recently which they are calling the “Radeon VII”. The ‘VII’ in the represents the 7nm process used in the graphics card which is the main highlight but it also represents the second generation of the Vega architecture. From the start, AMD has been very clear that this graphics card is targeted more towards data centers and is not originally intended for gaming.
AMD has decided not to compete in the higher end gaming market as they have been in a tough spot since the release of Nvidia’s RTX series graphics card featuring ray tracing and AI, they are targeting the mid-range market where they get more sales.
The Radeon VII will start at $700 for which you can also get Nvidia RTX 2080 with ray tracing and DLSS features. So why should you buy a graphics card based on a 2 year old Vega architecture? This is exactly what we are going to find out today. In this article, we have mentioned the features, specs and the benchmark results to help you decide whether it justifies the $700 price tag.
The core spec of the Radeon VII is that the process has been shrunk down from 14nm to a 7nm process which has enabled to fit in more memory. The smaller transistors run faster thus provide more performance and also consumes less power. It will come loaded with twice the memory compared to the last gen, 16 Gigs of superfast HBM2 memory which is really going to boost up the performance in professional work.
The HBM2 is the fastest GPU memory available in the industry and provides nearly twice the bandwidth of the GDDR6 found in Nvidia RTX series graphics cards. This will overkill most of the 4K AAA titles and what’s surprising is that it isn’t just on par with Nvidia’s deal but in some games, it even performs better but of course, it does not have those ray tracing features.
Moving on to the architecture, the new Vega 20 is about 33% smaller than its predecessor but the overall GPU is still a tad bigger as they have used the space to fit in more memory and transistors. The amount of graphics silicon is the same which is impressive considering they have shrunk down the process.
The Radeon VII features 3,840 GCN cores and 60 compute units (CUs) which is slightly less than the RX Vega 64 but thanks to the higher clock speed you get better performance. The improved clock speeds are also the result of the 7nm die shrink which has enabled the developers to fit in more transistors and also benefits in thermal and power.
It will have a base clock of 1,400 MHz and a boost clock 1,750 MHz, though AMD claims that it can reach up to 1,800 MHz at ‘peak engine speed’. All in all with these specs data-intensive frame shuffle especially at high resolutions won’t be a problem. Also, some games can really take advantage of the extra HBM2 memory and Vega exclusive features.
The specs on paper might look really great but what really matters is real-world performance. Without a doubt, I can say that AMD has developed the fastest consumer graphics card that even beats the $1000 Nvidia RTX 2080 Ti when it comes to designing, editing, and other professional work. However, when it comes to gaming it does not compete well enough.
Nvidia RTX 2080 leads in most of the popular gaming titles and even provide better and more realistic reflection and lighting effects. The Turing architecture will, of course, perform better as it was recently released and is also dedicated towards gaming. AMD does make some considerable wins in titles which require more memory like Far Cry 5 but overall Nvidia is the clear winner here.
This does not mean that the Radeon VII is a bad deal, it really blows out Nvidia when it comes to other graphics-intensive tasks. Also, the 7 nm transistors can deliver up to 20% higher performance or 40% power savings at the same performance versus its 16 nm node.
Here we have also shown some benchmark results to give you a better idea of the performance.
The AMD Radeon VII is a really impressive piece of tech which shows that how much Vega architecture is capable of. This graphics card is of course not for gamers but if you want to get some serious work done then this GPU is definitely worth considering. It features the fastest available HBM2 memory which gives the top of the line performance at a justified price tag in my opinion. But if you are looking for a gaming intended GPU by AMD then you will have to wait for their upcoming Navi architecture graphics cards.