AMD Radeon RX 560X vs NVIDIA GeForce RTX 5090 D V2 Comparison

The Verdict

The recorded data places the AMD Radeon RX 560X and the NVIDIA GeForce RTX 5090 D V2 in entirely separate performance tiers, with no direct head-to-head benchmark overlap. The RX 560X, an end-of-life Polaris part from the RX 500X generation, posts an average benchmark score of 18,626, placing it at the 62nd percentile of all GPUs in the database. The RTX 5090 D V2, a current GeForce 50-series flagship on the Blackwell 2.0 architecture, records an average score of 16,504 on its single 3DMark Steel Nomad DX12 test, sitting at the 59th percentile. Because the two cards are measured under different benchmark suites, a direct comparison of their average scores is not meaningful; instead, the data should be read as an indication of each card's position within its own testing context.

For users selecting between these two, the choice hinges on workload and system constraints. The RX 560X, with a 75 W TDP, no power connectors, and a 170 mm length, is suited for compact, low-power builds where the absence of auxiliary power is a requirement. Its 4 GB GDDR5 memory and 112.0 GB/s bandwidth serve older or less demanding titles and compute tasks. The RTX 5090 D V2, in contrast, demands a 575 W TDP, a 16-pin power connector, a suggested 950 W PSU, and a 304 mm length. It provides 24 GB of GDDR7 memory with 1.34 TB/s bandwidth, 170 ray tracing cores, and 680 tensor cores, making it the clear choice for modern high-end workloads such as ray-traced gaming, AI inference, and large-scale rendering. The data shows no scenario where the RX 560X outperforms the RTX 5090 D V2 in raw capability; the decision is purely about matching the card to the system's power and physical envelope.

Architecture Differences

The two GPUs are built on fundamentally different architectures, nodes, and process technologies. The RX 560X uses the Polaris 21 chip, based on GCN 4.0 architecture, fabricated on a 14 nm process at GlobalFoundries. It integrates 3,000 million transistors on a 123 mm² die, yielding a transistor density of 24.4 million per mm². The RTX 5090 D V2 employs the GB202 chip, built on Blackwell 2.0 architecture, manufactured on a 5 nm process at TSMC. This chip packs 92,200 million transistors onto a 750 mm² die, achieving a density of 122.9 million per mm², a 5x density advantage over the older part.

Memory subsystems diverge sharply. The RX 560X offers 4 GB of GDDR5 on a 128-bit bus, with 1750 MHz memory clock (7 Gbps effective) and 112.0 GB/s bandwidth. The RTX 5090 D V2 provides 24 GB of GDDR7 on a 384-bit bus, with 1750 MHz memory clock (28 Gbps effective) and 1.34 TB/s bandwidth, a twelvefold increase in capacity and a twelvefold increase in bandwidth. The bus interface also differs: the RX 560X uses PCIe 3.0 x8, while the RTX 5090 D V2 uses PCIe 5.0 x16, offering substantially more host bandwidth for data transfers.

Compute resources are orders of magnitude apart. The RX 560X has 1,024 shading units, 64 texture mapping units, and 16 ROPs, producing a pixel rate of 20.40 GPixel/s and a texture rate of 81.60 GTexel/s. Its FP32 throughput is 2.611 TFLOPS, and FP16 is identical at 2.611 TFLOPS (1:1). The RTX 5090 D V2 has 21,760 shading units, 680 TMUs, and 176 ROPs, with a pixel rate of 423.6 GPixel/s and a texture rate of 1,636.8 GTexel/s. Its FP32 and FP16 performance both reach 104.8 TFLOPS (1:1). The RTX 5090 D V2 also includes 170 ray tracing cores and 680 tensor cores, hardware features entirely absent from the RX 560X.

Feature support differs in API coverage. The RX 560X supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. The RTX 5090 D V2 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The newer card's DirectX 12 Ultimate tier includes advanced ray tracing and mesh shader capabilities that the GCN 4.0 part cannot expose. Physical dimensions also separate the two: the RX 560X measures 170 mm (6.7 inches) in length, while the RTX 5090 D V2 is 304 mm (12 inches) long, 137 mm (5.4 inches) tall, and 48 mm (1.9 inches) wide. Both cards are dual-slot, but the power requirements could not be more different: the RX 560X draws no auxiliary power, while the RTX 5090 D V2 requires a single 16-pin connector.

Where Each One Wins

The RX 560X wins exclusively in scenarios defined by low power draw and compact physical footprint. Its 75 W TDP and lack of power connectors mean it can be installed in systems without additional power cables, a trait that suits small-form-factor or legacy machines. Its 170 mm length fits into short chassis, and its end-of-life status implies it may be available in lower-cost used markets, though the database does not record pricing. For light gaming, older DirectX 12 titles, or OpenCL compute tasks, the RX 560X's 17,020 OpenCL score and 20,231 Vulkan score in the database show it holds its own within its class.

The RTX 5090 D V2 wins everywhere else. Its 24 GB frame buffer and 1.34 TB/s bandwidth are decisive for 4K gaming, high-resolution texture packs, and large dataset workloads. The 170 ray tracing cores and 680 tensor cores enable hardware-accelerated ray tracing and AI-based features such as DLSS, which the RX 560X cannot perform. Its 104.8 TFLOPS FP32 throughput is 40 times that of the RX 560X's 2.611 TFLOPS, a gap that dominates any compute-heavy application. The 3DMark Steel Nomad DX12 score of 16,504, despite being from a single test, indicates a high level of sustained performance in modern DirectX 12 workloads. The RTX 5090 D V2's active production status and 2025 release date mean it represents current-generation capability, whereas the RX 560X is firmly legacy hardware.

FAQ

Q: Which GPU has higher average benchmark scores?

A: The RX 560X has an average benchmark score of 18,626, while the RTX 5090 D V2 has an average score of 16,504. However, these scores come from different benchmark suites (Geekbench OpenCL/Vulkan for the RX 560X, 3DMark Steel Nomad DX12 for the RTX 5090 D V2), so they are not directly comparable.

Q: What are the memory capacities and types?

A: The RX 560X has 4 GB of GDDR5 memory on a 128-bit bus with 112.0 GB/s bandwidth. The RTX 5090 D V2 has 24 GB of GDDR7 memory on a 384-bit bus with 1.34 TB/s bandwidth.

Q: Do both cards support ray tracing?

A: No. The RTX 5090 D V2 includes 170 ray tracing cores and supports DirectX 12 Ultimate (12_2). The RX 560X has no ray tracing cores and only supports DirectX 12 (12_0).

Q: What are the power requirements?

A: The RX 560X has a 75 W TDP and requires no power connectors, with a suggested PSU of 250 W. The RTX 5090 D V2 has a 575 W TDP, requires a 16-pin power connector, and suggests a 950 W PSU.

Q: Which card is physically larger?

A: The RTX 5090 D V2 measures 304 mm (12 inches) in length, 137 mm (5.4 inches) in height, and 48 mm (1.9 inches) in width. The RX 560X is 170 mm (6.7 inches) long. Both are dual-slot cards.

Q: What is the production status of each card?

A: The RX 560X is end-of-life, released on 2018-04-10, with a predecessor of Polaris and successor of Vega. The RTX 5090 D V2 is active, released on 2025-08-14, with a predecessor of GeForce 40 and successor of GeForce 60.

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark results between the RX 560X and the RTX 5090 D V2, and neither card has any wins in their respective test sets against each other. This absence is expected given their different testing methodologies. The RX 560X is evaluated under Geekbench OpenCL and Geekbench Vulkan, producing scores of 17,020 and 20,231 respectively. The RTX 5090 D V2 is evaluated under 3DMark Steel Nomad DX12, yielding a score of 16,504. Without a shared benchmark, any numerical comparison between the two is speculative and not supported by the recorded data.

What the data does show is each card's position relative to its own nearest rivals. The RX 560X's average score of 18,626 places it within 1.3% of several peers. It trails the NVIDIA Tesla K80 by 1.3%, which scores 18,866, and the NVIDIA GeForce RTX 2070 by 0.9%, which scores 18,789. It matches the AMD Radeon Pro 5700 XT, which scores 18,685, a delta of -0.3%, and beats the AMD FirePro D500, which scores 18,533, by 0.5%. These tight deltas indicate that the RX 560X sits in a dense performance cluster where a few percentage points separate it from similarly positioned cards. Its 62nd percentile ranking among all GPUs confirms it as a mid-pack performer in the database's broader distribution.

The RTX 5090 D V2's single 3DMark score of 16,504 places it in a different cluster. Its nearest rival, the NVIDIA T400, scores 16,508, a delta of 0%, meaning the two cards are statistically indistinguishable in this test. The AMD Radeon PRO W7500 scores 16,415, a delta of 0.5% behind, and the NVIDIA RTX PRO 6000 Blackwell scores 16,408, a delta of 0.6% behind. The AMD Radeon RX 5700 XT scores 16,361, sitting 0.9% behind the RTX 5090 D V2. The 59th percentile ranking for the RTX 5090 D V2 is notable: despite its massive compute and memory resources, the Steel Nomad DX12 test places it in the lower half of the database's distribution, suggesting that this particular benchmark does not fully exercise the card's capabilities or that the test's scoring scale compresses high-end results. The data therefore advises caution: the RTX 5090 D V2's raw specifications are extraordinary, but its recorded benchmark score only reflects a single workload, and no direct comparison to the RX 560X is possible from the database.