AMD Radeon R9 370X vs AMD Radeon RX 570X Comparison

The benchmark data presents a clear picture: the AMD Radeon RX 570X is the significantly faster GPU, though the R9 370X has its own strengths in specific metrics. The recorded measurements show an overall win count of 1 to 0 in favor of the RX 570X, driven by a decisive margin in the shared compute benchmark.

Head-to-Head Benchmarks

The single direct comparison available in the database is the Geekbench OpenCL test. Here, the AMD Radeon RX 570X delivers a score of 38,939, while the AMD Radeon R9 370X manages 25,893. This translates to the RX 570X outperforming the R9 370X by 33.5%. This is a substantial generational leap, reflecting the RX 570X’s newer architecture and higher specifications.

This 33.5% advantage is the largest single-metric gap between the two cards in the recorded data. For context, the R9 370X’s average benchmark score of 15,862 places it within 0.1% of the AMD Radeon RX 7700 (score 15,852) and 1% behind the AMD Radeon RX 9060 (score 16,014). The RX 570X, with an average score of 13,871, sits 0.4% ahead of both the NVIDIA RTX A2000 Mobile (score 13,821) and the AMD Radeon 660M (score 13,812). While the RX 570X wins the head-to-head, its average score is actually lower than the R9 370X’s, a quirk of the different benchmark suites recorded for each card. The Geekbench OpenCL result is the only test where both cards have comparable data, and it is the one that defines the performance hierarchy.

The data indicates that the RX 570X’s victory is not marginal; it is a clear and consistent lead in raw compute throughput. The R9 370X, despite being from an older generation, still holds a higher average score across its recorded benchmarks, which is largely due to its Geekbench Vulkan score of 9,018 and Geekbench Metal score of 12,676. However, these tests have no direct counterpart in the RX 570X’s recorded results, so a direct comparison on those workloads is not possible from the database.

Where Each One Wins

The AMD Radeon RX 570X wins decisively in the only directly comparable workload: OpenCL compute. This is indicative of its superior raw processing power, which is backed by its hardware specifications. The RX 570X’s FP32 throughput of 5.095 TFLOPS is nearly double the R9 370X’s 2.637 TFLOPS. Its texture rate of 159.2 GTexel/s is also significantly higher than the R9 370X’s 82.40 GTexel/s. For tasks that rely on these capabilities, such as general-purpose GPU computing or heavy shader workloads, the RX 570X is the clear choice.

The AMD Radeon R9 370X, however, has a stronger showing in its own recorded benchmark suite. Its Geekbench Metal score of 12,676 and Geekbench Vulkan score of 9,018 indicate that it remains competent in those specific API environments. Since the RX 570X lacks recorded results for these tests, the database cannot confirm whether the RX 570X would also win there, but the R9 370X’s scores are the only data points available. Therefore, in applications that are primarily bound by Metal or Vulkan performance, the R9 370X has a proven track record within this dataset, while the RX 570X’s performance in those areas is unmeasured.

The RX 570X also wins on memory capacity. It has 8 GB of GDDR5 memory compared to the R9 370X’s 2 GB. This is a major advantage for modern games and applications that require large texture pools or datasets. The RX 570X’s memory bandwidth of 224.0 GB/s also exceeds the R9 370X’s 179.2 GB/s, further reinforcing its lead in memory-intensive scenarios. The R9 370X has no comparable advantage in the recorded specifications that would give it a win in a specific use case other than its compatibility with older API versions, which is a niche benefit.

The Verdict

The data points to a single, clear verdict: the AMD Radeon RX 570X is the superior graphics card for compute and modern workloads. The 33.5% lead in Geekbench OpenCL is a decisive indicator of its performance class. For users prioritizing raw processing power, higher memory capacity, and newer feature support, the RX 570X is the only logical choice based on these measurements.

The AMD Radeon R9 370X, however, is not without merit. Its higher average benchmark score of 15,862 versus 13,871 for the RX 570X is notable, though this is influenced by the different benchmark suites. Its recorded strengths in Metal and Vulkan suggest it can still handle specific legacy workloads adequately. The R9 370X may be suitable for a system where those specific API tests are the primary concern, or where its smaller 2 GB memory footprint is not a limiting factor. It also has a lower transistor density and a larger process node, indicating it is a simpler, older design.

Ultimately, the RX 570X is the more capable and future-proof option. Its 8 GB memory buffer alone is a massive advantage over the 2 GB on the R9 370X, as modern applications increasingly exceed 2 GB VRAM usage. For any new build or upgrade, the RX 570X is the recommended card. The R9 370X should only be considered for very specific, low-demand scenarios where its proven Metal and Vulkan scores are sufficient and its limited memory is not a bottleneck.

FAQ

Q: Which card is faster in OpenCL compute?

A: The AMD Radeon RX 570X is significantly faster, scoring 38,939 in Geekbench OpenCL compared to the R9 370X’s 25,893, a 33.5% advantage.

Q: How much memory does each card have?

A: The AMD Radeon RX 570X has 8 GB of GDDR5 memory, while the AMD Radeon R9 370X has 2 GB of GDDR5 memory.

Q: What is the average benchmark score for each card?

A: The AMD Radeon R9 370X has a higher average benchmark score of 15,862, while the AMD Radeon RX 570X scores 13,871. This is based on different benchmark suites recorded for each card.

Q: Which card has a higher pixel rate?

A: The AMD Radeon RX 570X has a higher pixel rate of 39.81 GPixel/s, compared to the R9 370X’s 32.96 GPixel/s.

Q: What are the process nodes for these GPUs?

A: The R9 370X is built on a 28 nm process at TSMC, while the RX 570X uses a 14 nm process at GlobalFoundries.

Q: Does the RX 570X support a newer version of DirectX?

A: Yes, the RX 570X supports DirectX 12 (12_0), while the R9 370X supports DirectX 12 (11_1).

Architecture Differences

The two cards represent different generations of AMD’s Graphics Core Next (GCN) architecture. The R9 370X is based on the Trinidad chip and uses GCN 1.0, a design from the Pirate Islands generation. The RX 570X is based on the Polaris 20 chip and uses GCN 4.0, belonging to the Polaris generation. This architectural leap is the primary reason for the performance difference.

The manufacturing process differs significantly. The R9 370X uses a 28 nm process at TSMC, while the RX 570X uses a 14 nm process at GlobalFoundries. This smaller node allows the RX 570X to pack more transistors. The R9 370X has 2,800 million transistors on a 212 mm² die, while the RX 570X has 5,700 million transistors on a 232 mm² die. This results in a much higher transistor density for the RX 570X: 24.6M per mm² versus 13.2M per mm² for the R9 370X.

Core configuration also differs. The RX 570X has 2048 shading units and 128 texture mapping units, while the R9 370X has 1280 shading units and 80 texture mapping units. Both cards have 32 render output units. The RX 570X also supports FP16 compute at a 1:1 ratio with FP32, offering 5.095 TFLOPS for both, while the R9 370X has no recorded FP16 capability. The RX 570X’s API support is newer, including DirectX 12 (12_0) and Vulkan 1.3, whereas the R9 370X supports DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6.

Specification Differences

The AMD Radeon RX 570X has a higher base clock of 1168 MHz and a boost clock of 1244 MHz, compared to the R9 370X’s 980 MHz base and 1030 MHz boost. The RX 570X also has faster memory, with a 1750 MHz memory clock and 7 Gbps effective speed, versus the R9 370X’s 1400 MHz memory clock and 5.6 Gbps effective speed.

Memory configuration is a key differentiator. The RX 570X has 8 GB of GDDR5 memory, while the R9 370X has 2 GB. Both use a 256-bit bus, but the RX 570X’s higher clock speed gives it 224.0 GB/s of bandwidth versus 179.2 GB/s for the R9 370X.

Power requirements differ, with the RX 570X having a lower TDP of 150 W compared to the R9 370X’s 180 W. The RX 570X also requires a single 6-pin power connector, while the R9 370X needs two 6-pin connectors. Both cards have a suggested PSU of 450 W and a dual-slot width.

The RX 570X is longer at 241 mm (9.5 inches) compared to the R9 370X’s 221 mm (8.7 inches). The R9 370X has a recorded height of 111 mm (4.4 inches), while the RX 570X does not have a height listed. Display outputs also differ: the RX 570X offers 1x DVI, 1x HDMI 2.0b, and 3x DisplayPort 1.4a, while the R9 370X has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2.

The release dates are different, with the R9 370X launching on August 26, 2015, and the RX 570X on April 10, 2018. The R9 370X had a launch MSRP of 199 USD, while the RX 570X has no recorded launch MSRP. Both cards are end-of-life products.