AMD Radeon Pro 5700 XT vs AMD Radeon R9 370X Comparison
AMD Radeon Pro 5700 XT
Radeon R9 370X
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon Pro 5700 XT vs AMD Radeon R9 370X
Where Each One Wins
The AMD Radeon Pro 5700 XT wins every recorded head-to-head benchmark against the AMD Radeon R9 370X. The database shows three direct comparisons, and the Pro 5700 XT takes all three: Geekbench Metal, Geekbench OpenCL, and Geekbench Vulkan. The R9 370X has no winning benchmark in this matchup.
The largest gap appears in Vulkan compute. The Pro 5700 XT scores 56593 in Geekbench Vulkan, while the R9 370X manages only 9018, a delta of 527.6%. That is not a marginal lead; it is a generational chasm. Metal tells a similar story: 51272 versus 12676, a 304.5% advantage for the newer card. OpenCL is comparatively closer but still decisive, with the Pro 5700 XT at 59467 against 25893, a 129.7% lead.
For use-case planning, this means the Pro 5700 XT is the clear choice for any workload that stresses modern graphics APIs, GPU compute, or Apple ecosystem applications that rely on Metal. The R9 370X, while end-of-life and from an older architecture generation, retains some relevance only in legacy DirectX scenarios, but the database does not record any DirectX head-to-head results for this pair. The Pro 5700 XT also sits at the 63rd percentile among all GPUs, while the R9 370X sits at the 58th percentile. Its average benchmark score of 18685 places it slightly above the R9 370X average of 15862, a difference of roughly 17.8% in aggregate performance.
Architecture Differences
The two cards come from different architectural eras. The Pro 5700 XT uses the Navi 10 chip built on RDNA 1.0, fabricated by TSMC on a 7 nm process. The R9 370X uses the Trinidad chip built on GCN 1.0, also from TSMC but on a 28 nm process. That process gap is enormous: 7 nm versus 28 nm. Transistor counts reflect the difference. The Pro 5700 XT packs 10,300 million transistors on a 251 mm² die, giving a density of 41.0 million transistors per square millimeter. The R9 370X has 2,800 million transistors on a 212 mm² die, a density of 13.2 million per square millimeter. The newer part fits roughly 3.7 times more transistors into only 18% more die area.
The architectural feature set also diverges. RDNA 1.0 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. GCN 1.0 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The Pro 5700 XT therefore has a higher DirectX feature level and a newer Vulkan version. Neither card has ray tracing cores or tensor cores, so that is not a differentiator.
Power and physical design differ sharply. The Pro 5700 XT has a TDP of 130 W, is configured as an integrated graphics package (IGP) with no power connectors, and requires a suggested PSU of 300 W. The R9 370X has a TDP of 180 W, is a dual-slot card, uses two 6-pin power connectors, and needs a 450 W PSU. The Pro 5700 XT delivers far more compute while drawing less power on paper. The R9 370X also has physical dimensions recorded: 221 mm length and 111 mm height. The Pro 5700 XT has no recorded dimensions because it is an IGP with no outputs.
Head-to-Head Benchmarks
The recorded data shows three direct comparisons, all won by the Pro 5700 XT.
Geekbench Metal: Pro 5700 XT scores 51272, R9 370X scores 12676. The Pro 5700 XT is 304.5% faster. This is the second-largest delta in the set and reflects the Pro card's intended role in Mac-oriented Metal workloads. The R9 370X predates Metal's widespread adoption and its GCN 1.0 architecture is simply outmatched here.
Geekbench OpenCL: Pro 5700 XT scores 59467, R9 370X scores 25893. The delta is 129.7%. This is the closest of the three comparisons, but still a decisive win. OpenCL is the most portable of the three APIs, so the R9 370X shows its best relative performance here. Even so, the Pro 5700 XT's FP32 throughput of 7.675 TFLOPS versus 2.637 TFLOPS for the R9 370X explains the gap. The Pro card also has 2560 shading units against 1280, double the shader count, and 160 texture mapping units against 80.
Geekbench Vulkan: Pro 5700 XT scores 56593, R9 370X scores 9018. The delta is 527.6%, the largest margin in the entire comparison. Vulkan support on the R9 370X is limited to version 1.2.170, while the Pro 5700 XT supports Vulkan 1.4. The combination of newer architecture, more shaders, and faster memory makes this a rout. No other benchmark in the database pits these two directly.
Wins tally: 3 for the Pro 5700 XT, 0 for the R9 370X. The data does not show a single scenario where the older card pulls ahead.
Specification Differences
The database records several fields where the two cards differ. These are the key ones:
Process node: 7 nm for the Pro 5700 XT, 28 nm for the R9 370X.
Transistors: 10,300 million versus 2,800 million.
Die size: 251 mm² versus 212 mm².
Transistor density: 41.0 million per mm² versus 13.2 million per mm².
Base clock: 1243 MHz versus 980 MHz.
Boost clock: 1499 MHz versus 1030 MHz.
Memory clock: 1500 MHz (12 Gbps effective) versus 1400 MHz (5.6 Gbps effective).
Memory size: 16 GB GDDR6 versus 2 GB GDDR5.
Memory bandwidth: 384.0 GB/s versus 179.2 GB/s.
Shading units: 2560 versus 1280.
Texture mapping units: 160 versus 80.
ROPs: 64 versus 32.
Pixel rate: 95.94 GPixel/s versus 32.96 GPixel/s.
Texture rate: 239.8 GTexel/s versus 82.40 GTexel/s.
FP32 performance: 7.675 TFLOPS versus 2.637 TFLOPS.
FP16 performance: 15.35 TFLOPS (2:1) for the Pro 5700 XT; the R9 370X has no recorded FP16 value.
TDP: 130 W versus 180 W.
Slot width: IGP versus dual-slot.
Power connectors: None versus 2x 6-pin.
Suggested PSU: 300 W versus 450 W.
Bus interface: PCIe 4.0 x16 versus PCIe 3.0 x16.
Display outputs: No outputs for the Pro 5700 XT; 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 for the R9 370X.
DirectX support: 12 (12_1) versus 12 (11_1).
Vulkan support: 1.4 versus 1.2.170.
Release date: 2020-08-03 for the Pro 5700 XT, 2015-08-26 for the R9 370X.
Predecessor/successor: The R9 370X lists its predecessor as Volcanic Islands and successor as Arctic Islands. The Pro 5700 XT lists neither.
The R9 370X has a launch MSRP of 199 USD. The Pro 5700 XT has no recorded launch MSRP.
FAQ
Q: Which card is faster in Geekbench Metal?
A: The AMD Radeon Pro 5700 XT scores 51272 in Geekbench Metal, while the Radeon R9 370X scores 12676. The Pro 5700 XT leads by 304.5%.
Q: Is the R9 370X better at any benchmark in this comparison?
A: No. The database records three head-to-head benchmarks (Geekbench Metal, OpenCL, and Vulkan), and the R9 370X loses all three. The win count is 3 for the Pro 5700 XT and 0 for the R9 370X.
Q: How large is the Vulkan performance gap?
A: The Pro 5700 XT scores 56593 in Geekbench Vulkan, the R9 370X scores 9018, a delta of 527.6%. This is the largest margin of any benchmark in the comparison.
Q: What memory configurations do these cards use?
A: The Pro 5700 XT has 16 GB of GDDR6 on a 256-bit bus with 384.0 GB/s bandwidth. The R9 370X has 2 GB of GDDR5 on a 256-bit bus with 179.2 GB/s bandwidth.
Q: Do these cards support the same DirectX version?
A: No. The Pro 5700 XT supports DirectX 12 (12_1), while the R9 370X supports DirectX 12 (11_1). The Pro card also supports Vulkan 1.4 versus Vulkan 1.2.170 for the R9 370X.
Q: Which card requires more power from the PSU?
A: The R9 370X has a TDP of 180 W and a suggested PSU of 450 W. The Pro 5700 XT has a TDP of 130 W and a suggested PSU of 300 W. The older card draws more power despite being slower.
The Verdict
The data is unambiguous. The AMD Radeon Pro 5700 XT wins every recorded benchmark against the AMD Radeon R9 370X, with deltas ranging from 129.7% to 527.6%. Anyone choosing between these two for a compute or graphics workload should pick the Pro 5700 XT unless specific constraints apply.
The Pro 5700 XT is the correct choice for Metal-centric environments, Vulkan workloads, and OpenCL compute. Its 16 GB GDDR6 memory and 384.0 GB/s bandwidth make it far better suited to large datasets. Its 2560 shading units and 7.675 TFLOPS FP32 performance are roughly triple the R9 370X's 1280 shaders and 2.637 TFLOPS. It also supports a newer DirectX feature level and a newer Vulkan version.
The R9 370X has one practical advantage: it is a physical dual-slot card with display outputs (2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2), while the Pro 5700 XT is an IGP with no outputs. If a system requires a discrete card with video output and cannot use an integrated package, the R9 370X is the only one of the two that physically fits that role. Its 221 mm length and 111 mm height are recorded, and it uses two 6-pin connectors.
But for raw performance, the verdict is simple. The Pro 5700 XT is 304.5% ahead in Metal, 129.7% ahead in OpenCL, and 527.6% ahead in Vulkan. It has more than three times the transistor count, more than double the shader count, double the ROPs, and more than double the memory bandwidth. It does all of this at a lower TDP (130 W versus 180 W) and with a lower suggested PSU (300 W versus 450 W). The R9 370X is end-of-life, and the benchmark data shows why.