Graphics cards to function for power use the PCI Express port, or whatever interface they use to communicate with the rest of the hardware. Therefore, the architecture of the different GPUs is not only designed to achieve the highest possible graphics performance, but also the highest possible in a given consumption. This is why there has been an evolution in the energy efficiency of NVIDIA and AMD GPUs over time.
Evolution of energy efficiency in NVIDIA and AMD
Using information pulled from the TechPowerUp database, a graph was created that tells us about the evolution of AMD and NVIDIA gaming GPUs in terms of power and energy efficiency. The origin of the graphics is ten years ago with the launch of the ATI Radeon HD 5000 based on the AMD Terascale 2 architecture and the NVIDIA GTX 400.
The biggest change came with the launch of the NVIDIA GTX 900, in which NVIDIA first adopted an architecture based on tile rasterization in its GPUs. A technology that AMD promised to apply under the name DSBR in its AMD Vega, but did not do until the launch of the first generation of RDNA known as the RX 5000 series to evolve with the RX 6000. L AMD’s effort in what power consumption placed it in this setting above the current RTX 30.
Another point to highlight is the question of price, as it is known, AMD generally sells its graphics cards at a lower price than NVIDIA. Which, although this can be understood as a benefit to the user, was also the product of an AMD which for generations lost its competitiveness against NVIDIA.
Now that AMD has managed to regain some ground in performance and power consumption over NVIDIA, one shouldn’t expect Lisa Su’s company to remain the cheap option for long. Because at equivalent performance, AMD can capitalize on the same. However, this comparison does not take into account things such as NVIDIA’s DLSS.
AMD has the advantage of more advanced manufacturing nodes
Over the past two generations of PC GPUs, we’ve seen how AMD used TSMC’s 7nm node, while NVIDIA used TSMC’s 12nm node for its RTX 20, then Samsung’s 8nm node for its RTX. 30. Which means that the transistor The density of AMD GPUs was higher than that of NVIDIA and as a result, larger chips were fitted to GeForce branded graphics cards compared to AMD.
Using a more advanced node is also an advantage in terms of power efficiency, so the comparison between AMD and NVIDIA architectures in this regard is not entirely fair when using different manufacturing nodes. If the rumors that Lovelace uses TSMC’s 5nm node hold true, then we will be able to compare AMD and NVIDIA architectures in terms of power efficiency in a much fairer way.