AMD Radeon PRO V710 vs NVIDIA GeForce RTX 5090 D Comparison

FAQ

Q: How does the NVIDIA GeForce RTX 5090 D compare to the AMD Radeon PRO V710 in raw compute performance?

A: The RTX 5090 D delivers 104.8 TFLOPS FP32, while the Radeon PRO V710 delivers 27.65 TFLOPS FP32. In the 3DMark Steel Nomad DX12 test, the RTX 5090 D scores 14326 versus 853 for the Radeon PRO V710, a 1579.5% difference.

Q: Which card has more memory and bandwidth?

A: The RTX 5090 D has 32 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth. The Radeon PRO V710 has 28 GB of GDDR6 on a 224-bit bus with 504.0 GB/s bandwidth.

Q: What are the power requirements for each card?

A: The RTX 5090 D has a TDP of 575 W with a suggested PSU of 950 W and a single 16-pin connector. The Radeon PRO V710 has a TDP of 158 W with a suggested PSU of 450 W and a single 8-pin connector.

Q: Do both cards support the same APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: Which card has a higher percentile ranking among all GPUs?

A: The RTX 5090 D ranks in the 92nd percentile, while the Radeon PRO V710 ranks in the 88th percentile.

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

A: The RTX 5090 D has an average benchmark score of 77712, while the Radeon PRO V710 has an average benchmark score of 58657.

Architecture Differences

The NVIDIA GeForce RTX 5090 D is built on the Blackwell 2.0 architecture using the GB202 chip, fabricated on TSMC's 5 nm process. It packs 92,200 million transistors on a 750 mm² die, resulting in a transistor density of 122.9M per mm². The AMD Radeon PRO V710 uses the RDNA 3.0 architecture with the Navi 32 chip, also on TSMC's 5 nm process, but with 28,100 million transistors on a 346 mm² die, for a density of 81.2M per mm².

The RTX 5090 D features 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The Radeon PRO V710 has 3,456 shading units, 216 TMUs, 96 ROPs, and 54 RT cores, with no tensor cores listed. This is a fundamental architectural difference: the NVIDIA card is designed with dedicated tensor hardware, while the AMD card relies on its RDNA 3.0 compute units.

Memory architecture differs significantly. The RTX 5090 D uses 32 GB of GDDR7 on a 512-bit bus, achieving 1.79 TB/s bandwidth. The Radeon PRO V710 uses 28 GB of GDDR6 on a 224-bit bus, achieving 504.0 GB/s bandwidth. The NVIDIA card's memory subsystem delivers more than three times the bandwidth.

Clock behavior differs as well. The RTX 5090 D has a base clock of 2017 MHz and a boost clock of 2407 MHz, with memory running at 1750 MHz (28 Gbps effective). The Radeon PRO V710 has a base clock of 1900 MHz and a boost clock of 2000 MHz, with memory at 2250 MHz (18 Gbps effective). Despite lower clocks, the AMD card achieves lower power consumption due to its smaller die and fewer compute units.

The physical and electrical designs diverge substantially. The RTX 5090 D is a dual-slot card measuring 304 mm in length, 137 mm in height, and 48 mm in width, with a 575 W TDP. The Radeon PRO V710 is a single-slot card with no listed dimensions, a 158 W TDP, and no display outputs. The RTX 5090 D uses PCIe 5.0 x16, while the Radeon PRO V710 uses PCIe 4.0 x16.

Where Each One Wins

The NVIDIA GeForce RTX 5090 D wins decisively in every recorded benchmark category. In the 3DMark Steel Nomad DX12 test, it scores 14326 against 853 for the AMD card. In Geekbench OpenCL, it scores 310674 against 116460. The head-to-head benchmark table shows 2 wins for the RTX 5090 D and 0 for the Radeon PRO V710.

For gaming and high-performance compute workloads, the RTX 5090 D is the clear choice. Its 104.8 TFLOPS FP32 and 104.8 TFLOPS FP16 (1:1) performance, combined with 170 RT cores and 680 tensor cores, make it suitable for demanding DirectX 12 Ultimate workloads, ray tracing, and AI-accelerated tasks. The 32 GB GDDR7 memory with 1.79 TB/s bandwidth supports large datasets and high-resolution textures.

The Radeon PRO V710 wins in efficiency-focused scenarios. Its 158 W TDP and single-slot design make it suitable for dense server installations or systems with limited power and space. The lack of display outputs indicates it is intended for compute-only deployments, where its 28 GB of GDDR6 memory and 504.0 GB/s bandwidth can handle moderate workloads. Its 27.65 TFLOPS FP32 and FP16 performance is modest but usable for specific tasks.

The RTX 5090 D also offers display outputs, with 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the Radeon PRO V710 has no outputs at all. For any workload requiring visual output, the NVIDIA card is the only option between the two.

Specification Differences

| Specification | NVIDIA GeForce RTX 5090 D | AMD Radeon PRO V710 |

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

| Architecture | Blackwell 2.0 | RDNA 3.0 |

| Chip | GB202 | Navi 32 |

| Process Node | 5 nm | 5 nm |

| Transistors | 92,200 million | 28,100 million |

| Die Size | 750 mm² | 346 mm² |

| Transistor Density | 122.9M / mm² | 81.2M / mm² |

| Base Clock | 2017 MHz | 1900 MHz |

| Boost Clock | 2407 MHz | 2000 MHz |

| Memory Clock | 1750 MHz (28 Gbps effective) | 2250 MHz (18 Gbps effective) |

| Memory Size | 32 GB | 28 GB |

| Memory Type | GDDR7 | GDDR6 |

| Memory Bus | 512 bit | 224 bit |

| Memory Bandwidth | 1.79 TB/s | 504.0 GB/s |

| Shading Units | 21760 | 3456 |

| TMUs | 680 | 216 |

| ROPs | 176 | 96 |

| RT Cores | 170 | 54 |

| Tensor Cores | 680 | null |

| Pixel Rate | 423.6 GPixel/s | 192.0 GPixel/s |

| Texture Rate | 1,636.8 GTexel/s | 432.0 GTexel/s |

| FP32 | 104.8 TFLOPS | 27.65 TFLOPS |

| FP16 | 104.8 TFLOPS (1:1) | 27.65 TFLOPS (1:1) |

| TDP | 575 W | 158 W |

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

| Power Connectors | 1x 16-pin | 1x 8-pin |

| Suggested PSU | 950 W | 450 W |

| Bus Interface | PCIe 5.0 x16 | PCIe 4.0 x16 |

| Display Outputs | 1x HDMI 2.1b, 3x DisplayPort 2.1b | No outputs |

| Dimensions | 304 mm x 137 mm x 48 mm | Not listed |

| Release Date | 2025-01-29 | 2024-10-02 |

| Launch MSRP | 2,299 USD | Not listed |

The RTX 5090 D is a larger, more powerful card in every measurable specification except power draw and physical footprint. The Radeon PRO V710 consumes 158 W versus 575 W, and fits in a single slot versus dual slots.

Head-to-Head Benchmarks

The 3DMark Steel Nomad DX12 test shows the largest gap. The RTX 5090 D scores 14326, while the Radeon PRO V710 scores 853. This is a 1579.5% delta, meaning the NVIDIA card outperforms the AMD card by nearly 17 times in this DirectX 12 workload. The RTX 5090 D's 170 RT cores and 21,760 shading units provide the raw throughput needed for this demanding test.

The Geekbench OpenCL benchmark shows a closer but still one-sided result. The RTX 5090 D scores 310674, while the Radeon PRO V710 scores 116460. The delta here is 166.8%. This test exercises general compute performance, where the NVIDIA card's 104.8 TFLOPS FP32 versus 27.65 TFLOPS FP32 gives it a 3.8x theoretical advantage, reflected in the actual score difference.

The RTX 5090 D's nearest rivals in the database include the AMD Radeon RX 6650M XT with an average score of 76904 and a delta of 1.1%, the AMD Radeon RX 6850M XT at 78940 with a delta of -1.6%, and NVIDIA Tesla P100 cards at 79396 and 79605 with deltas of -2.1% and -2.4%. This places the RTX 5090 D at the top of its immediate competitive cluster, with an average benchmark score of 77712.

The Radeon PRO V710's nearest rivals are clustered much lower. The NVIDIA P102-100 scores 58528 with a delta of 0.2%, the AMD Radeon RX 6950 XT scores 58392 with a delta of 0.5%, the Intel Arc A570M scores 58239 with a delta of 0.7%, and the AMD Radeon RX 5600 OEM scores 58085 with a delta of 1%. The Radeon PRO V710's average benchmark score of 58657 places it in the 88th percentile, just below the RTX 5090 D's 92nd percentile.

The Verdict

The data is unambiguous. The NVIDIA GeForce RTX 5090 D outperforms the AMD Radeon PRO V710 in every recorded benchmark by substantial margins. The 1579.5% lead in 3DMark Steel Nomad DX12 and 166.8% lead in Geekbench OpenCL establish a hierarchy that no other metric contradicts.

The RTX 5090 D is the choice for workloads that demand maximum compute throughput. Its 104.8 TFLOPS FP32, 170 RT cores, 680 tensor cores, 32 GB GDDR7 memory with 1.79 TB/s bandwidth, and dual-slot design with display outputs make it suited for high-end gaming, ray tracing, AI inference, and content creation. The 92nd percentile ranking among all GPUs and an average benchmark score of 77712 confirm its position near the top of the performance stack.

The Radeon PRO V710 serves a different purpose entirely. Its 158 W TDP, single-slot design, and lack of display outputs point to server or workstation deployments where power density and physical space are priorities. Its 28 GB memory and 504.0 GB/s bandwidth provide adequate capacity for moderate compute tasks, and its 88th percentile ranking with an average score of 58657 shows it is not a weak card in absolute terms. It simply operates in a different performance class.

For a builder assembling a high-performance system with no power or space constraints, the RTX 5090 D is the only rational choice. For a density-optimized compute node where 575 W is unacceptable and display output is unnecessary, the Radeon PRO V710 fits that niche. The launch MSRP of the RTX 5090 D is 2,299 USD, though its performance lead over the Radeon PRO V710 is so large that direct comparison is almost academic. These are not competing products; they are solutions for different problems.