AMD Radeon Pro 560 vs AMD Radeon Pro 5600M Comparison

The AMD Radeon Pro 560 and AMD Radeon Pro 5600M are both end-of-life mobile workstation GPUs, but they represent two very different eras of AMD's architecture. The data shows a clear generational gap: the 5600M wins all three head-to-head benchmark comparisons decisively, while the older 560 sits near the 61st percentile of all GPUs. However, the 5600M’s average benchmark score of 16351 places it at the 59th percentile, slightly below the 560's 61st percentile despite its raw performance advantage. This paradox is explained by the benchmark pool: the 5600M’s average is dragged down by its low Passmark DirectX scores, while the 560’s three Geekbench tests are all relatively strong. For a builder, this means the 5600M is the faster part in modern compute and graphics APIs, but the 560 is no slouch in its own right for legacy workloads.

Head-to-Head Benchmarks

The most lopsided result comes in Geekbench Metal, where the 5600M scores 55030 against the 560's 20918. That is a delta of -62% for the 560, meaning the 5600M is roughly 2.6 times faster in Apple's Metal API. This is the single largest gap in the comparison, and it underscores how much RDNA 1.0 improved over GCN 4.0 for compute-heavy tasks. The 5600M’s lead is equally stark in Geekbench OpenCL, scoring 47783 versus 15504 — a -67.6% delta. In practical terms, any OpenCL workload, from video encoding to scientific simulation, will finish in a fraction of the time on the 5600M.

Geekbench Vulkan shows a similar story, with the 5600M scoring 46425 against the 560's 16232, a -65% delta. The 5600M wins all three head-to-head tests, giving it a clean 3-0 sweep. The 560 never comes within striking distance in any metric; its best relative showing is in Metal, where it still trails by more than 60%. For the 560, the only consolation is that its average benchmark score of 17551 is higher than the 5600M’s 16351, but that figure is misleading because the 5600M has seven extra Passmark tests (DirectX 9/10/11/12, G2D, G3D, and GPU compute) that score very low — its Passmark DirectX 12 score is just 39, while its Passmark G3D is 9279. These low scores pull its average down, but they do not reflect real-world modern gaming or compute performance.

Architecture Differences

The 560 is built on the Polaris 21 chip using GCN 4.0 architecture, fabricated on a 14 nm process at GlobalFoundries. It packs 3,000 million transistors into a 123 mm² die, yielding a transistor density of 24.4M per mm². In contrast, the 5600M uses the Navi 12 chip with RDNA 1.0 architecture, made on TSMC's 7 nm process. The fact pack does not list transistor count or die size for the 5600M, so a direct density comparison is impossible, but the node shrink is obvious: 7 nm versus 14 nm is a massive leap in efficiency potential.

The memory subsystems are radically different. The 560 has 4 GB of GDDR5 on a 128-bit bus, delivering 81.28 GB/s of bandwidth. The 5600M has 8 GB of HBM2 on a 2048-bit bus, delivering 394.2 GB/s — nearly five times the bandwidth. This makes the 5600M vastly superior for bandwidth-hungry tasks like 4K texture streaming or large dataset manipulation. The 5600M also doubles the memory capacity, which is critical for modern workloads that exceed 4 GB.

Compute resources differ sharply. The 560 has 1024 shading units, 64 TMUs, and 16 ROPs. The 5600M has 2560 shading units, 160 TMUs, and 64 ROPs. That is 2.5 times the shaders, 2.5 times the texture units, and 4 times the raster units. Pixel rate jumps from 14.51 GPixel/s to 73.22 GPixel/s, and texture rate from 58.05 GTexel/s to 183.0 GTexel/s. FP32 throughput goes from 1.858 TFLOPS to 5.857 TFLOPS — a 3.15x improvement. The 5600M also supports FP16 at 11.71 TFLOPS with a 2:1 ratio, while the 560 does FP16 at 1.858 TFLOPS (1:1), meaning the 5600M can double its throughput on FP16 workloads.

The 5600M runs at lower clock speeds — 822 MHz base and 1144 MHz boost versus the 560's memory clock of 1270 MHz (the 560's core clocks are not listed). Despite lower clocks, the 5600M’s massive shader count and memory bandwidth win outright. The 5600M has a lower TDP of 50 W versus the 560's 75 W, which is remarkable given its higher performance. Both are IGP (integrated) parts with no power connectors and portable-device-dependent outputs. The 560 uses PCIe 3.0 x8, while the 5600M uses PCIe 4.0 x16, doubling the bus width and generation.

Where Each One Wins

The 5600M wins every benchmark where both are tested — Metal, OpenCL, and Vulkan — so it is the clear choice for any macOS or Linux workload that leverages these APIs. Its 8 GB HBM2 memory and 394.2 GB/s bandwidth make it superior for large frame buffers, high-resolution textures, and compute tasks that require frequent memory access. The 5600M’s FP16 capability (11.71 TFLOPS) is a major advantage for machine learning inference and certain graphics effects that use half-precision arithmetic.

The 560 has no benchmark wins in the head-to-head data, but it does have a higher average benchmark score (17551 vs 16351) and a higher percentile ranking (61st vs 59th). This suggests that in legacy DirectX workloads (which the 5600M scores poorly on in Passmark — DirectX 9: 118, DirectX 10: 62, DirectX 11: 59, DirectX 12: 39), the 560 may be more consistent. The 560’s 75 W TDP, while higher than the 5600M’s 50 W, is still low enough for thin-and-light machines, and its 14 nm process is more mature. For older software that does not support RDNA 1.0 well, the 560 could be the safer bet, but the data does not include any direct DirectX comparisons between the two.

The 5600M wins on raw compute power, memory bandwidth, API support (DirectX 12_1 versus 12_0), and efficiency (lower TDP). The 560 wins on nothing in direct comparisons, but its higher average score and percentile suggest it has a niche in legacy benchmarks. In practice, the 5600M is the better part for almost any modern workload, while the 560 might be preferable only for those who need maximum compatibility with older GCN-optimized software.

FAQ

Q: Which GPU is faster in Geekbench Metal?

A: The AMD Radeon Pro 5600M scores 55030 in Metal, while the AMD Radeon Pro 560 scores 20918. The 5600M is 62% faster in this test.

Q: How do the memory bandwidths compare?

A: The 5600M has 394.2 GB/s of bandwidth from 8 GB of HBM2 on a 2048-bit bus. The 560 has 81.28 GB/s from 4 GB of GDDR5 on a 128-bit bus. The 5600M offers nearly five times the bandwidth.

Q: Which GPU has a higher average benchmark score?

A: The 560 has an average benchmark score of 17551, while the 5600M averages 16351. However, the 5600M's average is lowered by its Passmark DirectX scores (e.g., 39 in DirectX 12), which are not directly comparable to the 560's Geekbench results.

Q: Do both GPUs support DirectX 12?

A: Yes, both support DirectX 12, but the 560 supports 12_0 while the 5600M supports 12_1, meaning the 5600M supports newer DirectX 12 features like DXR and variable rate shading.

Q: What is the difference in FP32 compute performance?

A: The 5600M delivers 5.857 TFLOPS of FP32 performance, while the 560 delivers 1.858 TFLOPS. The 5600M is approximately 3.15 times faster in single-precision compute.

Q: Which GPU has a lower TDP?

A: The 5600M has a TDP of 50 W, which is lower than the 560's 75 W. This makes the 5600M more power-efficient despite being significantly faster.

The Verdict

The data is unambiguous: the AMD Radeon Pro 5600M is the superior GPU for any modern workload. It wins all three head-to-head benchmarks by margins of 62% to 67.6%, offers 2.5 times the shading units, 4 times the ROPs, and nearly 5 times the memory bandwidth. Its 8 GB of HBM2 memory is double the 560's 4 GB of GDDR5, and its FP16 performance (11.71 TFLOPS) is in a different league. The 5600M also has a lower TDP (50 W vs 75 W), making it more suitable for thin laptops without sacrificing performance.

If you are choosing between these two for a new build or upgrade, the 5600M is the only rational pick for tasks like 3D rendering, video editing, or compute acceleration. Its DirectX 12_1 support ensures future compatibility, and its PCIe 4.0 x16 interface provides more bandwidth to the CPU. The 560 is not without merit — its higher average benchmark score and 61st percentile ranking suggest it handles legacy DirectX workloads better — but those scores come from a different test suite. In direct comparisons, the 560 never wins. Choose the 5600M unless you have a specific need for a part with a 14 nm GCN 4.0 architecture and are willing to accept 62-67% lower performance in Metal, OpenCL, and Vulkan.

Specification Differences

| Field | AMD Radeon Pro 560 | AMD Radeon Pro 5600M |

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

| Chip | Polaris 21 | Navi 12 |

| Architecture | GCN 4.0 | RDNA 1.0 |

| Process Node | 14 nm (GlobalFoundries) | 7 nm (TSMC) |

| Transistors | 3,000 million | Not listed |

| Die Size | 123 mm² | Not listed |

| Transistor Density | 24.4M / mm² | Not listed |

| Base Clock | Not listed | 822 MHz |

| Boost Clock | Not listed | 1144 MHz |

| Memory Clock | 1270 MHz / 5.1 Gbps effective | 770 MHz / 1540 Mbps effective |

| Memory Size | 4 GB | 8 GB |

| Memory Type | GDDR5 | HBM2 |

| Memory Bus Width | 128 bit | 2048 bit |

| Memory Bandwidth | 81.28 GB/s | 394.2 GB/s |

| Shading Units | 1024 | 2560 |

| TMUs | 64 | 160 |

| ROPs | 16 | 64 |

| Pixel Rate | 14.51 GPixel/s | 73.22 GPixel/s |

| Texture Rate | 58.05 GTexel/s | 183.0 GTexel/s |

| FP32 Performance | 1.858 TFLOPS | 5.857 TFLOPS |

| FP16 Performance | 1.858 TFLOPS (1:1) | 11.71 TFLOPS (2:1) |

| TDP | 75 W | 50 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 4.0 x16 |

| DirectX Support | 12 (12_0) | 12 (12_1) |

| Release Date | 2017-04-17 | 2020-06-14 |