AMD Radeon RX Vega 5 vs NVIDIA GeForce RTX 4070

AMD Radeon RX Vega 5

The AMD Radeon RX Vega 5 is an integrated GPU for notebooks. It is used for the Ryzen 3 4000 APUs, which were launched in early 2020. The GPU is based on the Vega architecture (5th generation GCN) and has 5 CUs (= 320 shaders) clocked at up to 1400 MHz. The performance depends on the configured TDP (12-25 W at launch), the clocks, the cooling, and the memory configuration). The GPU should benefit from fast dual-channel DDR4-3200 RAM (contrary to DDR4-2133 single-channel, which is also possible).

The Vega architecture offers some improvements over the Polaris generation and now supports DirectX 12 Feature Level 12_1. More information is available in our dedicated article about the Raven Ridge architecture.

The performance should be slightly below or on par with the older Radeon RX Vega 6 (6 CUs but lower clock speeds). Therefore, only a few modern games like Fifa 20 or F1 2019 should be able to run in lowest detail settings.

Thanks to the 7nm process and clever power-saving features, the power consumption is comparatively low (according to AMD), so the graphics card can also be used for slim and light notebooks.

NVIDIA GeForce RTX 4070

The Nvidia GeForce RTX 4070 is a high-end desktop graphics card based on the AD104 chip using the Ada Lovelace architecture. The 4070 offers 5,888 cores and 12 GB GDDR6X graphics memory with a 192 bit memory bus and clocked at 21 Gbits. The card offers a good raytracing performance thanks to the 46 dedicated raytracing cores. For AI calculations (and DLSS 3 in games), the RTX 4070 has 184 Tensor cores.

The gaming performance is well suited for 2k/QHD gaming in most demanding games, and should be able to handle 4k gaming for lesser demanding games (or with reduced settings / DLSS activated). The raw gaming performance is comparable to an older RTX 3080 or a AMD Radeon 6800 XT. Compared to the similar named mobile GeForce RTX 4070 Laptop GPU, the desktop variant is much faster.

The AD104 chip is manufactured in the modern 5nm process at TSMC (4N with EUV) and specified with a TGP (Total Graphics Power) of 200 Watt (compared to the 450 Watt of the RTX 4090).