Benchmarking Notes, The NVIDIA GeForce GTX 1070 Ti 3DMark Results
Our graphics benchmarking test bed has the following specifications :
We used the GeForce driver 388.31 for the NVIDIA graphics cards, and Radeon Software 17.9.3 for the AMD graphics cards.
3DMark DirectX 12 Benchmark (2560 x 1440)
3DMark Time Spy is the DirectX 12 benchmark in 3DMark. It supports new API features like asynchronous compute, explicit multi-adapter, and multi-threading.
The NVIDIA GeForce GTX 1070 Ti Founders Edition (Price Check) did very well in this DirectX 12 benchmark. It came in between the Radeon RX Vega 64 (Price Check) and the Vega 56 (Price Check), and was 17% faster than the GeForce GTX 1070.
3DMark (1920 x 1080)
For Direct 11 performance, we started testing the graphics cards using 3DMark at the entry-level gaming resolution – 1920 x 1080.
Due to the relatively low resolution, this is a CPU-limited test for many high-end graphics cards. The GeForce GTX 1070 Ti and the GeForce GTX 1070 were both roughly between the Radeon RX Vega 64 (Price Check) and the Radeon Vega 56 Price Check) in performance.
3DMark (2560 x 1440)
We then took 3DMark up a notch to the resolution of 2560 x 1440. Let’s take a look at the results!
The GeForce GTX 1070 Ti (Price Check) pulled away with the higher resolution. At 1440p, it was 11% faster than the GeForce GTX 1070, matching the Radeon RX Vega 56 in performance. It was just 12% slower than the Radeon RX Vega 64 (Price Check).
3DMark (3840 x 2160)
This is torture, even for flagship graphics cards.
At this resolution, the NVIDIA GeForce GTX 1070 Ti (Price Check) was 11% faster than the GeForce GTX 1070, matching the Radeon RX Vega 56 in performance. It was now 13% slower than the Radeon RX Vega 64 (Price Check).