AMD Radeon R9 370X vs NVIDIA GeForce GTX 1070 Ti Comparison

The AMD Radeon R9 370X and the NVIDIA GeForce GTX 1070 Ti occupy different performance tiers, despite both being end-of-life products. The recorded data shows a decisive overall victory for the NVIDIA card, which wins all three shared benchmark tests with substantial margins. The R9 370X, a 2015 part built on GCN 1.0, trails in raw compute, memory capacity, and API support, while the GTX 1070 Ti, a 2017 Pascal-based card, delivers multiples of the performance in every measured category. The database positions the R9 370X in the 58th percentile of all GPUs, while the GTX 1070 Ti sits at the 56th percentile, a counterintuitive result driven by the different benchmark suites each card was tested with, as the head-to-head results tell a much clearer story.

Where Each One Wins

The use-case split is stark. The NVIDIA GeForce GTX 1070 Ti wins every single head-to-head benchmark in the database, with no recorded wins for the AMD Radeon R9 370X. In the three shared tests, the NVIDIA card dominates across different API workloads. In Geekbench Metal, the GTX 1070 Ti scores 22180 versus 12676 for the R9 370X, a 42.8% advantage. In Geekbench OpenCL, the gap widens to 47.4%, with scores of 49219 and 25893 respectively. The most extreme difference appears in Geekbench Vulkan, where the GTX 1070 Ti achieves 61006 against a mere 9018 for the R9 370X, an 85.2% deficit for the AMD part. These results indicate that the GTX 1070 Ti is not just faster, it is categorically superior in modern API workloads, particularly Vulkan, where its Pascal architecture shows a massive scaling advantage.

The R9 370X does have a narrow edge in one metric from the broader database: its average benchmark score of 15862 is slightly higher than the GTX 1070 Ti's 14277. This is a quirk of the benchmark selection. The R9 370X was tested with Geekbench Metal, OpenCL, and Vulkan, while the GTX 1070 Ti was subjected to a much wider array of tests, including several Passmark DirectX variants and 3DMark Steel Nomad DX12, which drag its average down. For the R9 370X, its nearest rivals in the database include the AMD Radeon RX 7700 (0.1% faster), the AMD Radeon RX 9060 (1% slower), and the NVIDIA GeForce RTX 3060 Ti (1.7% faster), all of which are far newer parts. The GTX 1070 Ti's nearest rivals are the Intel Iris Xe MAX Graphics (0.3% slower), the AMD Radeon Vega 11 (0.5% slower), and the NVIDIA GeForce GTX TITAN (0.7% slower), a group that reflects its lower average score.

For actual gaming and compute workloads, the GTX 1070 Ti is the clear winner. The R9 370X only makes sense in legacy DirectX 11 scenarios or as a basic display adapter, but even there, the data does not support a win. The GTX 1070 Ti also shows much higher Passmark scores in DirectX 9 (207 vs. no comparable test for the R9 370X), DirectX 11 (107), and G3D (14674), all of which are absent from the AMD card's benchmark list, further cementing its superiority in traditional rasterization.

FAQ

Q: Which card is faster in Vulkan?

A: The NVIDIA GeForce GTX 1070 Ti is dramatically faster. In Geekbench Vulkan, it scores 61006 against 9018 for the AMD Radeon R9 370X, a delta of 85.2% in favor of the NVIDIA card.

Q: How much memory do these cards have?

A: The AMD Radeon R9 370X has 2 GB of GDDR5 memory on a 256-bit bus, while the NVIDIA GeForce GTX 1070 Ti has 8 GB of GDDR5 memory on a 256-bit bus. The GTX 1070 Ti also has higher memory bandwidth at 256.3 GB/s versus 179.2 GB/s.

Q: What are the power requirements?

A: Both cards have a TDP of 180 W and a suggested PSU of 450 W. The AMD card uses two 6-pin power connectors, while the NVIDIA card uses a single 8-pin connector.

Q: Which card has better OpenCL performance?

A: The NVIDIA GeForce GTX 1070 Ti wins in Geekbench OpenCL with a score of 49219, which is 47.4% higher than the AMD Radeon R9 370X's 25893.

Q: What is the release date difference?

A: The AMD Radeon R9 370X was released on August 26, 2015, while the NVIDIA GeForce GTX 1070 Ti was released on November 1, 2017. The GTX 1070 Ti is over two years newer.

Q: Do both cards support DirectX 12?

A: Yes, but with different feature levels. The AMD Radeon R9 370X supports DirectX 12 (11_1), while the NVIDIA GeForce GTX 1070 Ti supports DirectX 12 (12_1). The NVIDIA card also supports Vulkan 1.4, while the AMD card supports Vulkan 1.2.170.

Head-to-Head Benchmarks

The head-to-head data is unambiguous: the NVIDIA GeForce GTX 1070 Ti wins all three recorded comparisons, and each win is substantial. The smallest margin is in Geekbench Metal, where the GTX 1070 Ti scores 22180 against the R9 370X's 12676. That is a 42.8% advantage, meaning the NVIDIA card delivers nearly one and a half times the performance of the AMD part in this Apple-focused API. For users running Metal workloads, the choice is obvious.

The Geekbench OpenCL test shows an even larger gap. The GTX 1070 Ti scores 49219, while the R9 370X manages 25893. The delta is 47.4%, which translates to the NVIDIA card being roughly 90% faster in raw compute throughput. OpenCL is a common cross-platform compute API, and the results indicate that the Pascal architecture, with its 2432 shading units and 8.186 TFLOPS of FP32 performance, simply outclasses the GCN 1.0 design, which has only 1280 shading units and 2.637 TFLOPS.

The most lopsided result comes from Geekbench Vulkan. The GTX 1070 Ti posts a score of 61006, while the R9 370X only manages 9018. The delta is 85.2%, meaning the AMD card is just 14.8% of the NVIDIA card's performance in this test. Vulkan is a low-overhead API that benefits from modern hardware features, and the Pascal GPU's support for Vulkan 1.4, compared to the older Vulkan 1.2.170 on the R9 370X, contributes to this massive disparity. The GTX 1070 Ti also has a much higher texture rate (255.8 GTexel/s) and pixel rate (107.7 GPixel/s), which directly impacts Vulkan rendering efficiency.

Beyond the head-to-head tests, the GTX 1070 Ti's other recorded benchmarks reinforce its dominance. In Passmark G3D, it scores 14674, a figure that would likely dwarf any comparable test for the R9 370X, though no direct comparison exists. The GTX 1070 Ti also shows strong compute performance with a Passmark GPU Compute score of 7165, and it handles DirectX 9 with a score of 207, DirectX 11 with 107, and DirectX 12 with 51, all of which are absent from the R9 370X's benchmark list. The data consistently shows the GTX 1070 Ti as the superior card in every measurable aspect.

Specification Differences

The two cards differ fundamentally in their core specifications. The AMD Radeon R9 370X has 1280 shading units, 80 texture mapping units, and 32 render output units. The NVIDIA GeForce GTX 1070 Ti has 2432 shading units, 152 TMUs, and 64 ROPs. These are roughly double the AMD card's counts, explaining the massive performance gap. Clock speeds also favor NVIDIA: the GTX 1070 Ti runs at a base of 1607 MHz and a boost of 1683 MHz, while the R9 370X operates at 980 MHz base and 1030 MHz boost. The NVIDIA card's memory runs at 2002 MHz (8 Gbps effective) versus 1400 MHz (5.6 Gbps effective) on the AMD card.

Memory capacity is another major differentiator. The R9 370X has 2 GB of GDDR5, while the GTX 1070 Ti has 8 GB of GDDR5. Both use a 256-bit bus, but the higher memory clock on the NVIDIA card yields a bandwidth of 256.3 GB/s versus 179.2 GB/s. This 77 GB/s difference is critical for high-resolution textures and compute workloads. The GTX 1070 Ti also has a higher transistor count (7,200 million on a 314 mm² die) compared to the R9 370X (2,800 million on a 212 mm² die), though the AMD card uses a much older 28 nm process while the NVIDIA card uses 16 nm.

Physical dimensions differ as well. The R9 370X is 221 mm long (8.7 inches) and 111 mm tall (4.4 inches), while the GTX 1070 Ti is 267 mm long (10.5 inches), 112 mm tall (4.4 inches), and 40 mm wide (1.6 inches). Both are dual-slot cards, but the NVIDIA card is longer and has a single 8-pin power connector instead of the AMD card's two 6-pin connectors. Display outputs also vary: the R9 370X has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the GTX 1070 Ti has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a. The newer NVIDIA card supports higher-bandwidth display standards.

Architecture Differences

The architectural gap is a generation apart. The AMD Radeon R9 370X is built on GCN 1.0, using the Trinidad chip, and fabricated on a 28 nm process at TSMC. This is the first iteration of AMD's Graphics Core Next architecture, which debuted years before the R9 300 series. The chip contains 2,800 million transistors on a 212 mm² die, with a transistor density of 13.2 million per mm². GCN 1.0 lacks modern features like hardware-level asynchronous compute optimizations found in later GCN revisions, and its DirectX 12 support is limited to feature level 11_1.

The NVIDIA GeForce GTX 1070 Ti uses the Pascal architecture, specifically the GP104 chip, fabricated on a 16 nm process at TSMC. This process shrink allows for 7,200 million transistors on a 314 mm² die, resulting in a transistor density of 22.9 million per mm², nearly double that of the AMD card. Pascal is a fully modern architecture with DirectX 12 (12_1) support and Vulkan 1.4, which explains its superior performance in the Geekbench Vulkan test. The GTX 1070 Ti also has a higher FP32 throughput at 8.186 TFLOPS, and it supports FP16 compute at 127.9 GFLOPS (1:64 ratio), though this is a minor feature.

The R9 370X's GCN 1.0 architecture has a pixel rate of 32.96 GPixel/s and a texture rate of 82.40 GTexel/s, while the GTX 1070 Ti achieves 107.7 GPixel/s and 255.8 GTexel/s. These differences are directly tied to the ROP and TMU counts. The NVIDIA card's render output units are more efficient per clock, and its higher clock speeds amplify the advantage. The AMD card has no ray tracing cores or tensor cores, and neither does the NVIDIA card, as they predate the RTX series. However, the GTX 1070 Ti's Pascal architecture was designed for higher fill rates and better geometry handling, which is evident in its benchmark results.

The Verdict

The data is conclusive. The NVIDIA GeForce GTX 1070 Ti is the superior card in every head-to-head benchmark, with wins of 42.8% in Metal, 47.4% in OpenCL, and 85.2% in Vulkan. The R9 370X cannot compete in any modern API workload, and its only statistical advantage is a higher average benchmark score (15862 vs. 14277), which is an artifact of the different test suites. Users should choose the GTX 1070 Ti for any serious gaming, compute, or workstation task. Its 8 GB of memory, higher bandwidth, double the shading units, and newer architecture provide a comfortable margin of performance that the R9 370X cannot match. The R9 370X may suffice for legacy DirectX 9 or 11 titles at low resolutions, but even then, the GTX 1070 Ti's Passmark scores (207 in DirectX 9, 107 in DirectX 11) indicate it would outperform the AMD card. In short, the GTX 1070 Ti is the only rational choice between the two, based on the recorded measurements.