AMD Radeon RX 560 Mobile

The AMD Radeon RX 560 Mobile is an end-of-life mobile graphics processor built on the GCN 4.0 architecture, using the Baffin chip fabricated by GlobalFoundries on a 14 nm process. It occupies the 50th percentile of all GPUs in the database, with no discrete benchmark scores recorded, so its position rests on its specification sheet rather than measured workloads. Released on January 4, 2017, it is part of the Polaris Mobile generation under the RX M500 series, and it ships as an MXM module rather than a soldered laptop GPU.

Benchmark Performance

The database records no benchmark scores for this part — the benchmarks array is empty and the average score is 0 — so the only quantitative performance anchor is the 50th percentile ranking against all GPUs. That percentile places it exactly at the median of the database: half of all recorded GPUs are faster and half are slower. For a 2017 mobile part, that is a reasonable mid-pack position, but it is not a top-tier result.

The raw compute figures support that reading. The chip delivers 1.887 TFLOPS of FP32 throughput, and the same 1.887 TFLOPS for FP16 because the architecture processes half-precision at a 1:1 ratio with full precision — there is no dedicated FP16 acceleration path. Texture fill is 58.97 GTexel/s from 56 texture mapping units, and pixel fill is 16.85 GPixel/s from 16 ROPs. These numbers define the card's throughput ceiling, and they are consistent with a 50th-percentile part.

Clocks are modest: a 1000 MHz base and a 1053 MHz boost. The gap between base and boost is only 53 MHz, which suggests limited thermal headroom in the MXM form factor. The 896 shading units, 56 TMUs, and 16 ROPs are the shader and fixed-function resources that scale with those clocks, and the pixel and texture rates above are derived from them. Because the boost clock sits so close to the base clock, sustained-load performance will not differ dramatically from short-burst performance — the card is unlikely to drop far below its rated throughput under sustained load.

The 14 nm process node from GlobalFoundries, with 3,000 million transistors on a 123 mm² die, gives a transistor density of 24.4 million transistors per square millimeter. That density is typical of mid-range 14 nm parts and explains why the TDP stays at 55 W while still delivering the listed compute rates. In short, the benchmark picture is one of a balanced, mid-range mobile GPU: no standout peaks, no embarrassing valleys, and a steady 50th-percentile profile.

How It Compares

The nearestRivals field in the database is empty, so there are no direct rival entries with scores or delta percentages to cite. This is not unusual for an end-of-life mobile part — many laptop GPUs are never benchmarked head-to-head in the database. The comparison must therefore be made against the aggregate database rather than specific named products.

At the 50th percentile, the RX 560 Mobile sits exactly at the median of all GPUs ever recorded. That means it is faster than the bottom half of the database and slower than the top half. In the context of its generation — Polaris Mobile, in the RX M500 series — this places it as a mainstream offering rather than a flagship. Its predecessor is listed as "Gem System" and its successor as "Navi Mobile," which frames it as a transitional Polaris-era part between older GCN-derived mobile GPUs and the later Navi generation. Without rival benchmark scores, the percentile is the single most useful comparison metric available.

The absence of rival data also means the card cannot be positioned against specific competitors like NVIDIA mobile parts or other AMD Polaris variants. What can be said is that its 50th-percentile rank, combined with 4 GB of GDDR5 memory and a 128-bit bus, is a coherent mid-range profile. It is not a part that will challenge high-end desktop GPUs, nor is it a low-end part that struggles with basic 3D workloads. The data simply does not support any more granular comparison.

Ray Tracing and Feature Set

The specification lists no ray-tracing cores and no tensor cores. This is a GCN 4.0 architecture part, and GCN 4.0 predates the dedicated RT and tensor hardware that appears in later architectures. Consequently, any ray-traced effects would have to run on the general-purpose shading units — the 896 shaders — which is not a practical path for real-time ray tracing at meaningful frame rates.

API support is solid for the era: DirectX 12 at feature level 12_0, OpenGL 4.6, and Vulkan 1.3. The DirectX 12 feature level 12_0 is the baseline tier of DX12, covering core rendering features but not the higher feature levels that later GPUs expose. Vulkan 1.3 is a modern API revision, which means the card can run contemporary Vulkan titles even though the hardware is older. OpenGL 4.6 is the latest OpenGL revision, so legacy OpenGL applications are fully covered.

The display outputs are listed as "Portable Device Dependent," which is expected for an MXM module — the actual ports are determined by the laptop chassis, not the GPU card itself. The bus interface is MXM-B (3.0), a mobile PCIe-style slot. The architecture is GCN 4.0, built on 14 nm at GlobalFoundries, with 3,000 million transistors. The feature set is therefore defined by the GCN 4.0 feature level: no hardware ray tracing, no AI acceleration, but complete coverage of the traditional graphics APIs.

Who Should Consider It

This is an end-of-life part, released on January 4, 2017, and it is no longer in production. The realistic use cases are therefore limited to existing laptops that already have the MXM module, or to MXM-based systems where a drop-in replacement is needed. The 50th-percentile ranking and the 1.887 TFLOPS FP32 throughput suggest a card that can handle mainstream 3D workloads from its era, but not modern high-end gaming.

The 4 GB GDDR5 memory and 96.00 GB/s bandwidth are the limiting factors for modern titles. At higher resolutions and with high-detail textures, 4 GB fills quickly, and 96 GB/s of bandwidth is modest by current standards. For 1080p-class gaming with medium detail settings, the card is likely adequate, but the data does not include resolution-specific benchmarks, so this is an inference from the memory size and bandwidth rather than a measured result. For users running older games, indie titles, or light productivity workloads that use the GPU for acceleration, the RX 560 Mobile remains serviceable.

The card's 16 ROPs and 56 TMUs also cap its fill-rate performance. The 16.85 GPixel/s pixel rate is enough for moderate resolutions, but high-resolution, high-refresh-rate workloads will exceed what the ROPs can deliver. The 50th-percentile position means it is neither a bargain nor a bottleneck in the abstract — it is simply a mid-range mobile GPU that has aged out of the high-end segment. Anyone considering it today should do so with expectations set by its 2017 release date and its mid-pack percentile.

Memory Subsystem

The memory subsystem is straightforward: 4 GB of GDDR5 on a 128-bit bus, running at 1500 MHz with a 6 Gbps effective data rate, yielding 96.00 GB/s of bandwidth. The 128-bit bus is the defining constraint here. A wider bus would deliver more bandwidth at the same memory clock, but 128 bits is typical for a mid-range part of this class, and 96 GB/s is the consequence.

For high resolutions, 96 GB/s is the practical ceiling. Textures, framebuffers, and geometry all compete for that bandwidth, and modern games with high-resolution textures will pressure it. The 4 GB capacity is the second constraint. At 1080p, 4 GB is workable for many titles, but at 1440p and above, the combination of 4 GB capacity and 96 GB/s bandwidth will cause stutter or texture pop-in when the working set exceeds available VRAM. The 6 Gbps effective memory clock is not high by modern standards, but it is appropriate for the 1500 MHz base clock of the memory.

The 1:1 FP16 ratio also affects the memory subsystem indirectly: because FP16 and FP32 run at the same rate, there is no bandwidth-saving path for half-precision workloads. Everything the GPU does, whether full- or half-precision, must pass through the same 96 GB/s pipe. That makes the memory bandwidth the single most likely bottleneck in demanding workloads, ahead of the compute throughput.

Power and Cooling

The RX 560 Mobile is rated at a 55 W TDP, which is low enough to be handled by the MXM module's cooling solution in most laptops. The card is an MXM Module with an MXM-B (3.0) bus interface, and it lists no power connectors — power is drawn entirely through the MXM slot. This is a key advantage of the mobile form factor: no external PCIe power cables are required, and the 55 W envelope is modest enough that a laptop's existing cooling design should be able to dissipate it.

The database lists no suggested PSU for this part, which is consistent with its mobile nature — the laptop's power supply is fixed and not user-selectable. The absence of power connectors means installation is a matter of slotting the module in and securing it. The 14 nm process and the 123 mm² die with 3,000 million transistors help keep the TDP at 55 W; a larger or less efficient chip would push that figure higher. Cooling is "portable device dependent" in the same way display outputs are: the laptop chassis determines how much airflow the module receives. For a 55 W part, a single heatpipe and small fan are typically sufficient, but the data does not specify any cooling solution, so this is an inference from the TDP rather than a listed spec.

The boost clock of 1053 MHz, only 53 MHz above the 1000 MHz base, suggests the card is thermally limited in practice. A part with more cooling headroom would typically boost further above its base clock. The 55 W TDP, combined with the narrow boost range, paints a picture of a GPU that is tuned for consistent, sustained performance in a constrained thermal envelope rather than bursty high clocks.