AMD Radeon R9 370X vs NVIDIA T400 Comparison

FAQ

Q: Which GPU wins in OpenCL compute performance?

A: The AMD Radeon R9 370X dominates the OpenCL test, scoring 25,893 versus the NVIDIA T400’s 17,039. That is a 34.2% margin in favor of the AMD card.

Q: How do the two compare in Vulkan performance?

A: The NVIDIA T400 takes a decisive lead in Vulkan, scoring 15,976 against the R9 370X’s 9,018. The T400 is 77.2% faster in this API, a massive reversal from the OpenCL result.

Q: Which card has the higher average benchmark score?

A: The NVIDIA T400 edges ahead with an average benchmark score of 16,508, while the AMD Radeon R9 370X averages 15,862. The T400 also sits at the 60th percentile among all GPUs, versus the R9 370X’s 58th percentile.

Q: What are the power requirements for each card?

A: The NVIDIA T400 has a 30 W TDP and requires no power connectors, with a suggested PSU of 200 W. The AMD Radeon R9 370X has a 180 W TDP, needs two 6-pin power connectors, and suggests a 450 W PSU.

Q: Which card offers higher memory bandwidth?

A: The AMD Radeon R9 370X provides 179.2 GB/s of bandwidth over a 256-bit bus with GDDR5 memory. The NVIDIA T400 delivers 80.00 GB/s over a 64-bit bus with GDDR6 memory.

Q: When were these cards released?

A: The AMD Radeon R9 370X launched on August 26, 2015, while the NVIDIA T400 arrived on May 5, 2021. Both are now end-of-life products.

Architecture Differences

The NVIDIA T400 is built on the TU117 chip using the Turing architecture, fabricated on a 12 nm process at TSMC. It packs 4,700 million transistors into a 200 mm² die, yielding a transistor density of 23.5M per mm². The AMD Radeon R9 370X uses the Trinidad chip based on GCN 1.0 architecture, manufactured on a 28 nm process also at TSMC. It contains 2,800 million transistors across a 212 mm² die, giving a density of 13.2M per mm².

The T400’s newer process node allows it to fit far more transistors per area, but the R9 370X compensates with a wider memory interface and more raw compute resources. The AMD card features 1,280 shading units, 80 texture mapping units, and 32 ROPs. The NVIDIA card has just 384 shading units, 24 TMUs, and 16 ROPs. Despite fewer execution units, the T400’s Turing architecture brings modern feature support, including DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The R9 370X supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170.

Memory configurations diverge sharply. The T400 uses 2 GB of GDDR6 on a 64-bit bus, achieving 80.00 GB/s. The R9 370X uses 2 GB of GDDR5 on a 256-bit bus, achieving 179.2 GB/s — more than double the bandwidth. The T400’s memory clocks at 1250 MHz with 10 Gbps effective speed, while the R9 370X runs at 1400 MHz with 5.6 Gbps effective.

The T400’s base clock is a modest 420 MHz, but it boosts to 1425 MHz. The R9 370X runs at a higher base of 980 MHz and boosts to 1030 MHz. The T400 achieves a pixel rate of 22.80 GPixel/s and a texture rate of 34.20 GTexel/s. The R9 370X posts 32.96 GPixel/s and 82.40 GTexel/s. In FP32 compute, the T400 delivers 1,094.4 GFLOPS, while the R9 370X reaches 2.637 TFLOPS. The T400 also supports FP16 at 2.189 TFLOPS (2:1), a feature the R9 370X lacks entirely.

Physical and power characteristics differ substantially. The T400 is a single-slot card with no power connectors and a 30 W TDP. The R9 370X is dual-slot, requires two 6-pin connectors, and draws 180 W. The R9 370X measures 221 mm in length and 111 mm in height, while the T400’s dimensions are not listed. Display outputs also differ: the T400 offers three mini-DisplayPort 1.4a connectors, while the R9 370X provides two DVI, one HDMI 1.4a, and one DisplayPort 1.2.

Head-to-Head Benchmarks

The two cards split their direct benchmark comparisons exactly one win apiece. In Geekbench OpenCL, the AMD Radeon R9 370X is the clear victor, scoring 25,893 against the NVIDIA T400’s 17,039. That is a 34.2% deficit for the T400, a significant gap that reflects the R9 370X’s higher shading unit count and memory bandwidth in compute-heavy workloads.

The Geekbench Vulkan test flips the script entirely. The NVIDIA T400 scores 15,976, while the R9 370X manages only 9,018. The T400’s 77.2% advantage is even larger than the R9 370X’s OpenCL lead, suggesting that the newer Turing architecture handles modern graphics APIs far more efficiently. The R9 370X’s older GCN 1.0 design and lower Vulkan version (1.2.170 versus 1.4) likely contribute to this poor showing.

Looking at average benchmark scores across all tests, the T400 holds a narrow edge. Its 16,508 average surpasses the R9 370X’s 15,862 by roughly 4%. The T400’s percentile ranking of 60 versus the R9 370X’s 58 reinforces this slight overall superiority, even though the AMD card wins the single compute test by a wide margin.

The R9 370X’s nearest rivals in the database include the AMD Radeon RX 7700 (15,852 average, 0.1% behind), the AMD Radeon RX 9060 (16,014 average, 1% ahead), and the AMD Radeon Pro W5500 (15,679 average, 1.2% behind). The T400’s nearest rivals include the NVIDIA GeForce RTX 5090 D V2 (16,504 average, 0% difference), the AMD Radeon PRO W7500 (16,415 average, 0.6% behind), and the NVIDIA RTX PRO 6000 Blackwell (16,408 average, 0.6% behind). These clusters show both cards sitting in a similar performance tier, despite their architectural differences.

Specification Differences

| Specification | NVIDIA T400 | AMD Radeon R9 370X |

|---|---|---|

| Architecture | Turing | GCN 1.0 |

| Process Node | 12 nm | 28 nm |

| Transistors | 4,700 million | 2,800 million |

| Die Size | 200 mm² | 212 mm² |

| Transistor Density | 23.5M / mm² | 13.2M / mm² |

| Base Clock | 420 MHz | 980 MHz |

| Boost Clock | 1425 MHz | 1030 MHz |

| Memory Type | GDDR6 | GDDR5 |

| Memory Bus Width | 64 bit | 256 bit |

| Memory Bandwidth | 80.00 GB/s | 179.2 GB/s |

| Shading Units | 384 | 1,280 |

| TMUs | 24 | 80 |

| ROPs | 16 | 32 |

| Pixel Rate | 22.80 GPixel/s | 32.96 GPixel/s |

| Texture Rate | 34.20 GTexel/s | 82.40 GTexel/s |

| FP32 Performance | 1,094.4 GFLOPS | 2.637 TFLOPS |

| FP16 Performance | 2.189 TFLOPS (2:1) | None |

| TDP | 30 W | 180 W |

| Slot Width | Single-slot | Dual-slot |

| Power Connectors | None | 2x 6-pin |

| Suggested PSU | 200 W | 450 W |

| Display Outputs | 3x mini-DisplayPort 1.4a | 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 |

| DirectX Support | 12 (12_1) | 12 (11_1) |

| Vulkan Support | 1.4 | 1.2.170 |

| Release Date | May 5, 2021 | August 26, 2015 |

| Launch MSRP | Not listed | 199 USD |

| Dimensions | Not listed | 221 mm x 111 mm |

The Verdict

The data paints a clear picture of two GPUs built for different eras and different workloads. The AMD Radeon R9 370X is a brute-force compute card from 2015, with 1,280 shading units, 80 TMUs, and 179.2 GB/s of bandwidth. It wins the OpenCL benchmark by 34.2% and offers more than double the FP32 throughput (2.637 TFLOPS versus 1,094.4 GFLOPS). For raw compute tasks that scale across many shaders and high memory bandwidth, the R9 370X is the better choice.

The NVIDIA T400, released in 2021, is a modern workstation card that prioritizes efficiency and API support. Its 30 W TDP, single-slot design, and lack of power connectors make it dramatically easier to integrate into compact systems. It dominates in Vulkan by 77.2%, a result that reflects Turing’s superior graphics pipeline and newer driver optimizations. The T400 also supports FP16 compute, which the R9 370X cannot match.

The average benchmark scores favor the T400 (16,508 versus 15,862), and its 60th percentile ranking versus the R9 370X’s 58th shows a modest but consistent overall edge. However, the R9 370X’s nearest rival list includes the RTX 3060 Ti (16,129 average, 1.7% ahead), indicating it trades blows with much newer hardware in aggregate terms.

For a user choosing between these two, the decision hinges on workload. If the priority is OpenCL compute performance and maximum memory bandwidth, the R9 370X delivers a 34.2% advantage in that specific test. If the priority is modern API performance, lower power draw, and a smaller physical footprint, the T400 is the clear winner — its Vulkan score is 77.2% higher, and its 30 W TDP is six times lower than the R9 370X’s 180 W.

Neither card supports ray tracing or tensor cores, so those features are absent from both. The T400’s newer process node (12 nm versus 28 nm) and higher transistor density (23.5M per mm² versus 13.2M per mm²) explain its efficiency advantage. The R9 370X’s larger die (212 mm² versus 200 mm²) houses fewer transistors but more execution units, reflecting a fundamentally different design philosophy.

The R9 370X’s launch MSRP was 199 USD, though the T400 has no listed launch price. Both are end-of-life products, so availability and driver support may vary. The R9 370X’s predecessor was Volcanic Islands and its successor was Arctic Islands; the T400’s predecessor was Quadro Volta and its successor was Workstation Ampere.

In summary, the AMD Radeon R9 370X is the compute specialist, the NVIDIA T400 is the modern efficiency pick. The benchmark data shows one win each in head-to-head tests, with the T400’s average score and percentile ranking giving it the overall edge. Users who need OpenCL throughput should lean AMD; users who need Vulkan performance or low power consumption should lean NVIDIA.