AMD Radeon RX 590 vs NVIDIA RTX PRO 4000 Blackwell Comparison
AMD Radeon RX 590
RTX PRO 4000 Blackwell
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon RX 590 vs NVIDIA RTX PRO 4000 Blackwell
The NVIDIA RTX PRO 4000 Blackwell and AMD Radeon RX 590 occupy different eras of GPU design, and the benchmark data reflects a generational chasm. The RTX PRO 4000 Blackwell is built on a 5 nm process with the Blackwell 2.0 architecture, while the RX 590 uses a 12 nm process with the GCN 4.0 architecture. The data shows a single head-to-head benchmark result, and the NVIDIA card dominates it completely, but the story extends beyond that one test when examining their respective positions in the broader performance landscape.
Head-to-Head Benchmarks
The only direct benchmark comparison available is the 3DMark Steel Nomad DX12 test, and the result is stark. The NVIDIA RTX PRO 4000 Blackwell scores 4648, while the AMD Radeon RX 590 scores 1081. This produces a delta of 330% in favor of the NVIDIA card. In practical terms, the RTX PRO 4000 Blackwell delivers more than four times the performance of the RX 590 in this modern DirectX 12 workload. This is not a marginal improvement; it is a complete generational reset. The RX 590 was released in 2018, and its GCN architecture lacks the dedicated hardware features that the Blackwell GPU brings to the table, such as 70 RT cores and 280 tensor cores, which are entirely absent from the RX 590.
The gap is also evident in their average benchmark scores across all tests. The RTX PRO 4000 Blackwell has an average benchmark score of 27135, while the RX 590 sits at 24744. This is a difference of roughly 9.7%, which is far smaller than the Steel Nomad delta because the averages include a mix of older and newer workloads. The RTX PRO 4000 Blackwell’s percentile rank among all GPUs is 72, while the RX 590’s is 70. Both cards sit in a similar tier of overall performance when averaged across many tests, but the nature of that performance is very different. The RX 590’s score is buoyed by its strength in legacy APIs — it supports DirectX 12 (12_0) while the RTX PRO 4000 Blackwell supports DirectX 12 Ultimate (12_2) — but it falls apart in modern, feature-rich workloads like Steel Nomad.
The nearest rival data for each card further contextualizes their standings. The RTX PRO 4000 Blackwell’s closest competitors are the AMD Radeon RX 6700 XT (average score 27425, delta -1.1%), the NVIDIA GeForce RTX 4070 Mobile (27435, -1.1%), and the NVIDIA GeForce RTX 3090 (27565, -1.6%). This indicates that the RTX PRO 4000 Blackwell performs at the level of a high-end desktop GPU from the previous generation, trading blows with the RTX 3090. The RX 590, by contrast, sits alongside the NVIDIA RTX A5000 Mobile (24763, -0.1%), Intel Arc A350M (24647, +0.4%), and AMD Radeon RX 6600 XT (24442, +1.2%). The RX 590’s rivals are all lower-tier mobile or entry-level desktop parts, showing that its performance class is far below the RTX PRO 4000 Blackwell’s.
Looking at the specific benchmark scores for the RTX PRO 4000 Blackwell, the data shows a wide range of capabilities. In Passmark G3D, it scores 28427, and in Geekbench Vulkan it reaches 194168. Its Passmark GPU Compute score is 14805. These numbers are not directly comparable to the RX 590 because the RX 590 only has two benchmark entries in the data: the Steel Nomad score and a Geekbench Metal score of 48407. The lack of overlapping benchmarks beyond Steel Nomad makes direct comparisons difficult, but the Steel Nomad result is decisive. The RTX PRO 4000 Blackwell’s FP32 compute throughput is 36.83 TFLOPS, compared to the RX 590’s 7.119 TFLOPS — a fivefold advantage in raw shader math. This is reflected in the texture rate (575.4 GTexel/s vs 222.5 GTexel/s) and pixel rate (197.3 GPixel/s vs 49.44 GPixel/s), where the NVIDIA card leads by margins of 2.6x and 4x respectively.
FAQ
Q: How much faster is the NVIDIA RTX PRO 4000 Blackwell than the AMD Radeon RX 590 in 3DMark Steel Nomad?
A: The RTX PRO 4000 Blackwell scores 4648, while the RX 590 scores 1081, resulting in a 330% performance advantage for the NVIDIA card.
Q: Which card has a higher average benchmark score?
A: The RTX PRO 4000 Blackwell has an average benchmark score of 27135, compared to 24744 for the RX 590, a difference of approximately 9.7%.
Q: Does the AMD Radeon RX 590 support ray tracing or tensor cores?
A: No. The RX 590 has no RT cores and no tensor cores listed in the data. The RTX PRO 4000 Blackwell includes 70 RT cores and 280 tensor cores.
Q: What is the memory configuration difference?
A: The RTX PRO 4000 Blackwell has 24 GB of GDDR7 memory on a 192-bit bus with 672.0 GB/s bandwidth. The RX 590 has 8 GB of GDDR5 memory on a 256-bit bus with 256.0 GB/s bandwidth.
Q: Which card has a higher percentile rank among all GPUs?
A: The RTX PRO 4000 Blackwell sits at the 72nd percentile, while the RX 590 is at the 70th percentile, making them closely matched in overall standing despite the massive gap in modern workloads.
Q: How do their nearest rivals compare in average score?
A: The RTX PRO 4000 Blackwell’s nearest rival is the AMD Radeon RX 6700 XT with a delta of -1.1%. The RX 590’s nearest rival is the NVIDIA RTX A5000 Mobile with a delta of -0.1%.
The Verdict
The data paints a clear picture for different use cases. The NVIDIA RTX PRO 4000 Blackwell is the pick for anyone running modern DirectX 12 Ultimate workloads, ray-traced content, or AI-accelerated tasks. Its 330% lead in Steel Nomad DX12, combined with its 24 GB GDDR7 memory and 36.83 TFLOPS FP32 throughput, makes it a workstation-class card that sits alongside the NVIDIA GeForce RTX 3090 in average performance. The RX 590, with its 8 GB GDDR5 memory and 7.119 TFLOPS, is a legacy part that is end-of-life. It still holds its own in average benchmark scores due to its strong showing in older APIs, but it lacks the hardware features — no RT cores, no tensor cores — required for contemporary graphics workloads.
For a user upgrading from an RX 590-era system, the RTX PRO 4000 Blackwell is the obvious choice for future-proofing. The RX 590’s nearest rivals are all lower-tier parts like the Intel Arc A350M and the GeForce GTX 1630, which confirms its position as a budget-oriented card from a previous generation. The RTX PRO 4000 Blackwell’s nearest rivals include the RTX 3090, which places it in a much higher performance bracket. The RX 590’s only advantage is its lower power draw relative to its age — it has a 175 W TDP versus the RTX PRO 4000 Blackwell’s 140 W TDP — but that is a nominal difference given the NVIDIA card’s massive performance lead. The RTX PRO 4000 Blackwell also supports PCIe 5.0 x16, while the RX 590 is limited to PCIe 3.0 x16.
Specification Differences
The core specification differences between the two cards are extensive. The RTX PRO 4000 Blackwell uses the GB203 chip on a 5 nm process at TSMC, while the RX 590 uses the Polaris 30 chip on a 12 nm process at GlobalFoundries. The transistor count tells the story: 45,600 million for the NVIDIA card versus 5,700 million for the AMD card, with die sizes of 378 mm² and 232 mm² respectively. This results in a transistor density of 120.6M per mm² for the RTX PRO 4000 Blackwell versus 24.6M per mm² for the RX 590.
Clock speeds differ significantly. The RTX PRO 4000 Blackwell has a base clock of 1230 MHz and a boost clock of 2055 MHz, while the RX 590 has a base clock of 1469 MHz and a boost clock of 1545 MHz. The AMD card has a higher base clock, but the NVIDIA card has a much higher boost clock. Memory is another major differentiator: the RTX PRO 4000 Blackwell has 24 GB of GDDR7 at 1750 MHz (28 Gbps effective) on a 192-bit bus, while the RX 590 has 8 GB of GDDR5 at 2000 MHz (8 Gbps effective) on a 256-bit bus. The bandwidth is 672.0 GB/s for the NVIDIA card versus 256.0 GB/s for the AMD card.
The compute units also diverge sharply. The RTX PRO 4000 Blackwell has 8960 shading units, 280 TMUs, and 96 ROPs, while the RX 590 has 2304 shading units, 144 TMUs, and 32 ROPs. The NVIDIA card’s FP32 performance is 36.83 TFLOPS, compared to 7.119 TFLOPS for the AMD card. The power and physical specifications also differ: the RTX PRO 4000 Blackwell has a 140 W TDP, is single-slot, and uses a 1x 16-pin power connector with a suggested 300 W PSU. The RX 590 has a 175 W TDP, is dual-slot, uses a 1x 8-pin connector, and requires a 450 W PSU. Both cards are 241 mm long, but the RTX PRO 4000 Blackwell is 111 mm tall and 20 mm wide, while the RX 590’s height and width are not listed.
Architecture Differences
The architectural gulf between the two GPUs is fundamental. The RTX PRO 4000 Blackwell is built on the Blackwell 2.0 architecture, which is NVIDIA’s latest design specifically for professional workstation use. It includes 70 RT cores for hardware-accelerated ray tracing and 280 tensor cores for AI and machine learning workloads. The RX 590, by contrast, uses the GCN 4.0 architecture, which predates ray tracing and tensor core hardware entirely — the data lists no RT cores and no tensor cores for the AMD card. This means the RX 590 cannot accelerate ray-traced effects or AI-based features in software, while the RTX PRO 4000 Blackwell is designed to handle both.
The process technology also reflects the generational leap. The RTX PRO 4000 Blackwell is fabricated on a 5 nm node at TSMC, while the RX 590 uses a 12 nm node at GlobalFoundries. This is a two-generation jump in process technology, enabling the NVIDIA card to pack 45,600 million transistors into a 378 mm² die, versus 5,700 million transistors in a 232 mm² die for the AMD part. The Blackwell architecture supports DirectX 12 Ultimate (12_2), while the GCN 4.0 architecture only supports DirectX 12 (12_0). For Vulkan, the RTX PRO 4000 Blackwell supports version 1.4, while the RX 590 supports 1.3. Both cards support OpenGL 4.6.
Memory technology is another architectural differentiator. The RTX PRO 4000 Blackwell uses GDDR7, which is the latest memory standard, while the RX 590 uses GDDR5. This, combined with the wider 192-bit bus on the NVIDIA card versus the 256-bit bus on the AMD card, results in the RTX PRO 4000 Blackwell achieving 672.0 GB/s of bandwidth — more than 2.6 times the RX 590’s 256.0 GB/s. The display outputs also differ: the RTX PRO 4000 Blackwell has 4x DisplayPort 2.1b, while the RX 590 has 1x HDMI 2.0b and 3x DisplayPort 1.4a. The NVIDIA card also supports PCIe 5.0 x16, while the AMD card is limited to PCIe 3.0 x16. The production status further confirms the generational split: the RTX PRO 4000 Blackwell is listed as Active, while the RX 590 is End-of-life, with its successor listed as Vega.