NVIDIA GeForce RTX 4080 Super vs NVIDIA GeForce RTX 5070 Ti

The Nvidia GeForce RTX 5070 Ti is a powerful desktop graphics card that utilizes the GA203 chip, based on the Blackwell architecture. The model offers 8960 CUDA cores, and 16 GB GDDR7 graphics memory with a 256-bit memory bus (896 GB/s). The base clock of the GPU (reference model) is specified at 2.30 GHz, while the boost clock can reach 2.45 GHz.

Architecture

Nvidia GeForce RTX 5070 Ti features new Tensor (5th Gen) and RT-cores (4th Gen), with 1406 AI TOPS and 133 TFLOPS performance, respectively. Blackwell architecture is the first to feature the new DLSS 4 Multi Frame generation, which, thanks to enhanced AI capabilities, can generate more frames per second while using less memory and delivering better picture quality. Nvidia also managed to improve the performance and output quality of DLSS, Frame Generation, and Ray Reconstruction features (also available for the previous RTX 4000 series). Additionally, the new models now support Reflex 2 low-latency mode and Frame Warp.

Performance

Thanks to the new Tensor and RT-cores, as well as some bump in CUDA cores, the card is promised (by Nvidia) to offer as much as 2x better performance compared to the previous RTX 4070 Ti (thanks to the new DLSS 4 MFG feature). According to Nvidia's presentation, even the lower-end RTX 5070 can reach the performance level of the RTX 4090. However, this comparison was conducted using DLSS4 with 4x Multi-Frame Generation.

According to our own tests, the GeForce RTX 5070 Ti is some 5% slower than the RTX 4080 Super and Radeon RX 7900 XTX in the benchmarks, but the situation is different in games. With its 4K resolution, the 5070 Ti is 2% faster than the RX 7900 XTX and at least 3% slower than the 4080 Super. Using the 1080p/2K resolution yields slightly different results: the 7900 XTX and 5070 Ti are on the same level, while the RTX 4080 Super is 10-11% faster.

Our tests also show that the RTX 5070 Ti is 18% slower than the RTX 5080, a significant 66% behind the top RTX 5090, but also ~19% faster compared to the standard RTX 5070.

Power consumption

According to Nvidia, the GB203 chip is manufactured using the modern 4N FinFET process at TSMC and is specified with a TGP (Total Graphics Power) of 300 watts (15 W higher than RTX 4070 Ti). Nvidia now also specifies maximum GPU temperature, which is stated at 88 ℃ for RTX 5070 Ti. The model also requires at least a 750-Watt PSU.