AMD Radeon RX 590 vs NVIDIA RTX PRO 2000 Blackwell Comparison

The NVIDIA RTX PRO 2000 Blackwell and AMD Radeon RX 590 represent two distinct eras of GPU design, separated by nearly seven years of architectural progress. The benchmark data shows a decisive victory for the newer NVIDIA part, though the comparison reveals interesting nuances about how each card handles different types of workloads. The most striking single datapoint comes from the 3DMark Steel Nomad DX12 test, where the RTX PRO 2000 scores 2374.5 against the RX 590’s 1081, a massive 119.7% advantage. However, the full picture includes compute, API-specific, and legacy DirectX tests that tell a more complex story.

Head-to-Head Benchmarks

The sole directly comparable benchmark in the data is 3DMark Steel Nomad DX12, and it is a landslide. The NVIDIA RTX PRO 2000 Blackwell delivers 2374.5 points, more than double the RX 590’s 1081 points. That 119.7% delta represents a generational leap in raw DX12 rasterization performance, reflecting the Blackwell architecture’s ability to extract far more work per clock from its modern shader design.

Beyond that single head-to-head test, the broader benchmark suites reveal a consistent pattern of dominance. In Geekbench OpenCL, the RTX PRO 2000 scores 106087, while the RX 590 has no comparable OpenCL result in the data. The Vulkan gap is similarly lopsided: the NVIDIA card hits 113865 in Geekbench Vulkan, whereas the RX 590’s only Geekbench listing is a Metal score of 48407—a different API entirely. Passmark results reinforce the trend, with the RTX PRO 2000 posting 20049 in G3D and 8396 in GPU Compute, versus the RX 590’s absence from those specific Passmark tests.

The one area where the RX 590 shows a comparative strength is in the Passmark legacy DirectX subtests, though the numbers still favor NVIDIA. The RTX PRO 2000 scores 241 in DirectX 9, 174 in DirectX 11, 122 in DirectX 10, and 80 in DirectX 12 (Passmark’s version). The RX 590 has no corresponding Passmark DirectX 9/10/11/12 results in the data, so direct comparison is impossible. However, the RX 590’s average benchmark score of 24744 sits just 2.1% below the RTX PRO 2000’s 25269 average, which suggests that when accounting for all available tests—including the RX 590’s Metal result—the two cards are closer in aggregate than the Steel Nomad blowout implies.

Architecture Differences

The architectural chasm between these two GPUs is enormous. The RTX PRO 2000 uses the GB206 chip on a 5 nm TSMC process, packing 21,900 million transistors into a 181 mm² die for a density of 121.0M transistors per mm². The RX 590 uses the Polaris 30 chip on a 12 nm GlobalFoundries process, with just 5,700 million transistors spread across a 232 mm² die—a density of only 24.6M per mm². That means the NVIDIA chip crams nearly five times more transistors into a smaller physical area, a direct result of the 7 nm process-node gap.

Compute resources follow the same pattern. The RTX PRO 2000 fields 4352 shading units, 136 TMUs, and 48 ROPs, alongside 34 RT cores and 136 tensor cores. The RX 590 counters with 2304 shading units, 144 TMUs, and 32 ROPs, with no dedicated RT or tensor hardware. The NVIDIA card’s FP32 throughput is 17.03 TFLOPS, more than double the RX 590’s 7.119 TFLOPS, and its FP16 performance matches at 17.03 TFLOPS (1:1 ratio), while the RX 590’s FP16 is also 7.119 TFLOPS (1:1). Pixel and texture rates tell a similar story: 93.94 GPixel/s and 266.2 GTexel/s for NVIDIA versus 49.44 GPixel/s and 222.5 GTexel/s for AMD.

Memory architecture diverges sharply. The RTX PRO 2000 uses 16 GB of GDDR7 on a 128-bit bus, delivering 288.0 GB/s bandwidth. The RX 590 uses 8 GB of GDDR5 on a 256-bit bus, for 256.0 GB/s. Despite the wider bus on the older card, the newer GDDR7 memory runs at a higher effective speed—18 Gbps effective versus 8 Gbps effective—which allows the narrower 128-bit interface to outpace the 256-bit GDDR5 solution. The NVIDIA card also supports the modern PCIe 5.0 x8 interface, while the RX 590 is limited to PCIe 3.0 x16.

Feature support is another generational divider. The RTX PRO 2000 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, plus hardware ray tracing and tensor acceleration. The RX 590 tops out at DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3, with no ray tracing or AI acceleration hardware. Display output also differs: NVIDIA offers 4x mini-DisplayPort 2.1b, while AMD provides 1x HDMI 2.0b and 3x DisplayPort 1.4a.

Where Each One Wins

The RTX PRO 2000 wins in virtually every modern workload category. Its 119.7% advantage in 3DMark Steel Nomad DX12 makes it the clear choice for contemporary DirectX 12 games and applications that leverage the latest rendering features. The 16 GB of GDDR7 memory provides double the capacity of the RX 590, which matters for large textures, high-resolution assets, and compute workloads that exceed 8 GB. The presence of RT and tensor cores gives it a decisive edge in ray-traced content and AI-accelerated tasks, neither of which the RX 590 can handle natively.

The RX 590’s wins are more circumstantial. Its 256-bit memory bus provides higher theoretical bandwidth per byte of data transferred, which can benefit certain bandwidth-bound legacy workloads. Its 144 TMUs outnumber the RTX PRO 2000’s 136, giving it a slight edge in texture-heavy operations that scale with TMU count rather than shader throughput. The 1x 8-pin power connector and 175 W TDP suggest it was designed for mainstream PSUs, though the NVIDIA card’s 70 W TDP and no power connectors make it far more power-efficient—a major practical win. In the aggregate, the RX 590’s average benchmark score of 24744 is only 2.1% below the RTX PRO 2000’s 25269, meaning that in mixed or legacy workloads, the older card remains surprisingly competitive relative to its age.

Specification Differences

| Specification | NVIDIA RTX PRO 2000 Blackwell | AMD Radeon RX 590 |

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

| Process Node | 5 nm (TSMC) | 12 nm (GlobalFoundries) |

| Transistors | 21,900 million | 5,700 million |

| Die Size | 181 mm² | 232 mm² |

| Base Clock | 982 MHz | 1469 MHz |

| Boost Clock | 1957 MHz | 1545 MHz |

| Memory Size | 16 GB | 8 GB |

| Memory Type | GDDR7 | GDDR5 |

| Memory Bus | 128 bit | 256 bit |

| Memory Bandwidth | 288.0 GB/s | 256.0 GB/s |

| Shading Units | 4352 | 2304 |

| TMUs | 136 | 144 |

| ROPs | 48 | 32 |

| RT Cores | 34 | None |

| Tensor Cores | 136 | None |

| FP32 Performance | 17.03 TFLOPS | 7.119 TFLOPS |

| Pixel Rate | 93.94 GPixel/s | 49.44 GPixel/s |

| Texture Rate | 266.2 GTexel/s | 222.5 GTexel/s |

| TDP | 70 W | 175 W |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | 250 W | 450 W |

| Bus Interface | PCIe 5.0 x8 | PCIe 3.0 x16 |

| DirectX Support | 12 Ultimate (12_2) | 12 (12_0) |

| Vulkan Support | 1.4 | 1.3 |

| Production Status | Active | End-of-life |

| Release Date | 2025-08-10 | 2018-11-14 |

| Launch MSRP | N/A | 279 USD |

FAQ

Q: How much faster is the RTX PRO 2000 in 3DMark Steel Nomad DX12?

A: The NVIDIA card scores 2374.5 versus 1081 for the RX 590, a 119.7% advantage.

Q: Which GPU has more memory and bandwidth?

A: The RTX PRO 2000 has 16 GB of GDDR7 on a 128-bit bus delivering 288.0 GB/s, while the RX 590 has 8 GB of GDDR5 on a 256-bit bus for 256.0 GB/s.

Q: Does the RX 590 support ray tracing?

A: No. The RX 590 has no RT cores, whereas the RTX PRO 2000 includes 34 RT cores.

Q: What is the power consumption difference?

A: The RTX PRO 2000 has a 70 W TDP with no power connectors and a 250 W suggested PSU, versus the RX 590’s 175 W TDP with a 1x 8-pin connector and 450 W suggested PSU.

Q: How do their average benchmark scores compare?

A: The RTX PRO 2000 averages 25269 across all its benchmarks, while the RX 590 averages 24744, a difference of 2.1%.

Q: Which card has better API support?

A: The RTX PRO 2000 supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the RX 590 supports DirectX 12 (12_0) and Vulkan 1.3; both have OpenGL 4.6.