AMD Radeon RX 560

The AMD Radeon RX 560 is an entry-level graphics card built on the Polaris 21 chip using GCN 4.0 architecture, fabricated on a 14 nm process at GlobalFoundries with 3,000 million transistors on a 123 mm² die. It launched on April 17, 2017, with a launch MSRP of 99 USD, and is now end-of-life, succeeding the Arctic Islands generation and preceding Vega. The card measures 170 mm (6.7 inches) in length and occupies a dual-slot footprint. Its specifications place it in the lower tier of the GPU landscape, as reflected by its 28th percentile ranking among all GPUs.

Power and Cooling

The RX 560 has a TDP of 75 W, a figure that allows it to operate without any external power connectors; the specification sheet lists no power connectors at all. AMD recommends a 250 W power supply for systems using this GPU, which is modest by modern standards but entirely consistent with the card's low thermal envelope. The 14 nm process from GlobalFoundries, with a transistor density of 24.4 million per mm², contributes to this efficiency. The die size of 123 mm² and the 3,000 million transistor count indicate a compact, power-conscious design. The card's dual-slot cooler and 170 mm length suggest it can fit into small form factor cases, though the data does not provide height or width dimensions. Power is drawn entirely through the PCIe 3.0 x8 interface, which is the sole electrical connection. This combination of low TDP, no auxiliary power, and a modest PSU recommendation makes the RX 560 a straightforward installation for systems with limited power headroom.

Ray Tracing and Feature Set

The RX 560 does not include dedicated ray tracing cores or tensor cores; the specification sheet lists none for either category. Instead, it relies on the GCN 4.0 architecture, which provides 1024 shading units, 64 texture mapping units, and 16 raster output units. The pixel fill rate is 20.40 GPixel/s, and the texture fill rate is 81.60 GTexel/s. FP32 compute is rated at 2.611 TFLOPS, with FP16 performance identical at 2.611 TFLOPS (1:1 ratio). These numbers indicate a card designed for traditional rasterization workloads rather than hardware-accelerated ray tracing or AI-based features. For API support, the card is compatible with DirectX 12 (feature level 12_0), OpenGL 4.6, and Vulkan 1.3. The absence of RT and tensor cores means any ray tracing or machine learning tasks would have to be handled through software, which is not addressed in the provided data. The GCN 4.0 architecture, while older, offers broad API coverage that remains relevant for many current titles.

Memory Subsystem

The RX 560 comes with 4 GB of GDDR5 memory on a 128-bit bus, yielding a memory bandwidth of 112.0 GB/s. The memory clock runs at 1750 MHz, which translates to 7 Gbps effective. This configuration is typical for entry-level cards of its generation. For high-resolution gaming, the 4 GB capacity may become a limiting factor for texture-heavy titles, and the 112 GB/s bandwidth is modest compared to higher-end cards, though the benchmark data does not directly measure resolution scaling. The 128-bit bus width is half that of many mid-range cards, which can constrain performance at 1440p or above. However, for 1080p gaming, the memory subsystem is likely sufficient for moderate settings, as the card's average benchmark score of 5060 suggests. The effective memory speed of 7 Gbps is a standard figure for GDDR5 of that era, and the 112 GB/s bandwidth aligns with the card's overall performance tier.

How It Compares

The RX 560's average benchmark score is 5060, placing it in a tight cluster of comparable GPUs. Against the AMD Radeon R7 M340, the RX 560 is 0.2% slower, with the M340 scoring 5071. This difference is negligible in real-world terms, and the two cards are effectively tied in average performance. The Intel Iris Pro Graphics 5200 scores 5042, putting the RX 560 0.4% ahead. Again, the gap is tiny, and the RX 560's advantage is within the margin of measurement noise. The AMD Radeon HD 8670M scores 5012, and the RX 560 leads by 1%. This is still a small margin, but it indicates that the RX 560 holds a slight edge over this older mobile part. The NVIDIA Quadro 4000 scores 5000, with the RX 560 ahead by 1.2%. The Quadro 4000 is a workstation-oriented GPU, yet its average score is nearly identical to the RX 560's. All four rivals are within 1.2% of the RX 560, indicating that the card's performance is essentially on par with these older or lower-tier parts. The percentile rank of 28 means the RX 560 outperforms 28% of all GPUs in the database, a modest standing that reflects its entry-level positioning.

Benchmark Performance

The benchmark suite reveals a mixed profile across different API workloads. In Geekbench Metal, the RX 560 scores 23597, while in Geekbench OpenCL it scores 16234. The Metal score is notably higher, which may reflect the card's compatibility with Apple's API, though the data does not explain the discrepancy. Passmark results show a wide spread: DirectX 9 scores 57, DirectX 10 scores 16, DirectX 11 scores 25, and DirectX 12 scores 21. The DirectX 9 score is more than double any other DirectX test, suggesting that older API workloads run relatively better on this architecture. The Passmark G2D score is 485, and the G3D score is 3671. Compute performance is 1437 in Passmark GPU compute. The average benchmark score across all tests is 5060, which aligns with the rival comparisons. The 28th percentile ranking confirms that the RX 560 sits in the lower quarter of all GPUs. The delta percentages against rivals are all under 1.2%, so the card is statistically indistinguishable from its nearest competitors in average performance. This tight clustering suggests that the RX 560's performance is well understood and that any of these cards would deliver similar frame rates in real-world scenarios.

FAQ

Q: What is the TDP of the AMD Radeon RX 560?

A: The TDP is 75 W, and the suggested PSU is 250 W.

Q: Does the RX 560 require any external power connectors?

A: No, it has no power connectors; it draws power solely from the PCIe slot.

Q: What is the memory configuration?

A: It has 4 GB of GDDR5 memory on a 128-bit bus with 112.0 GB/s bandwidth, running at 1750 MHz (7 Gbps effective).

Q: Which APIs are supported?

A: DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3.

Q: How many shading units does it have?

A: It has 1024 shading units, along with 64 TMUs and 16 ROPs.

Q: What is the FP32 performance?

A: FP32 compute is rated at 2.611 TFLOPS, with FP16 at the same 2.611 TFLOPS (1:1).

Who Should Consider It

The RX 560 is positioned for entry-level gaming and light compute tasks. Its 28th percentile ranking means it outperforms 28% of all GPUs, and its average score of 5060 places it alongside the Radeon R7 M340, Intel Iris Pro 5200, Radeon HD 8670M, and Quadro 4000. With 4 GB of GDDR5 and 112 GB/s bandwidth, it can handle 1080p gaming at moderate settings in older or less demanding titles, though the data does not specify resolution-specific performance. The low 75 W TDP and absence of power connectors make it an easy drop-in for systems with a 250 W PSU, including small form factor builds. Its end-of-life status means it is no longer in production, but for users seeking a low-power, legacy card with Vulkan 1.3 and DirectX 12 support, the RX 560 remains a viable option based on benchmark results. The card's DirectX 9 performance is particularly strong, which could appeal to users running older games that rely on that API. However, for modern titles that demand higher VRAM or bandwidth, the RX 560's 4 GB capacity and 112 GB/s throughput may prove limiting. Ultimately, the data shows a card that is best suited for budget-conscious builds prioritizing low power consumption over peak performance.