AMD Radeon 680M vs AMD Radeon Pro 560 Comparison

Where Each One Wins

The recorded benchmark data splits cleanly between these two AMD parts, with the AMD Radeon 680M taking every head-to-head win. The Radeon Pro 560 does not win a single direct comparison in the database, while the 680M wins both available tests. This is not a close contest in aggregate terms, but the nature of the wins matters for real-world use.

The Radeon 680M dominates in compute-oriented workloads. In Geekbench OpenCL, it posts a score of 23,468 against the Pro 560's 15,504, a 33.9% advantage. That is a massive gap for integrated graphics versus a discrete mobile part. The 680M also wins the Vulkan test with 21,965 points versus 16,232, a 26.1% lead. These are not marginal victories; they represent a fundamental performance tier difference.

The Radeon Pro 560, despite losing both tests, still holds a respectable position in the broader database. Its average benchmark score of 17,551 places it in the 61st percentile of all GPUs. The 680M, conversely, has a lower average score of 15,270 and sits in the 57th percentile. This discrepancy between head-to-head results and overall percentile is explained by the different benchmark mixes. The Pro 560 has three recorded tests (Metal, OpenCL, Vulkan), while the 680M has three different tests (3DMark Steel Nomad DX12, OpenCL, Vulkan). The 3DMark result for the 680M is notably low at 378, dragging down its average.

For use-case selection, the data suggests the 680M is the clear choice for OpenCL and Vulkan workloads. The Pro 560's Metal score of 20,918 is not directly comparable since the 680M lacks a Metal result, but the 680M's OpenCL and Vulkan wins are decisive. Anyone needing raw compute throughput in cross-platform APIs should choose the 680M. The Pro 560's edge is only in the context of its specific benchmark set, where it achieves a higher average score and percentile ranking due to the absence of a low 3DMark result.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon Pro 560 has an average benchmark score of 17,551, which is higher than the AMD Radeon 680M's 15,270. The Pro 560 also ranks higher in overall percentile at 61% versus 57% for the 680M.

Q: How much faster is the Radeon 680M in OpenCL?

A: The Radeon 680M scores 23,468 in Geekbench OpenCL, which is 33.9% higher than the Radeon Pro 560's 15,504. The delta percentage is recorded as -33.9% from the Pro 560's perspective, meaning the 680M leads by that margin.

Q: Does the Radeon Pro 560 win any direct benchmark comparison?

A: No. In the head-to-head results, the Radeon Pro 560 has zero wins. The Radeon 680M wins both recorded tests: Geekbench OpenCL and Geekbench Vulkan.

Q: What is the Vulkan performance difference?

A: The Radeon 680M scores 21,965 in Geekbench Vulkan, while the Radeon Pro 560 scores 16,232. The 680M leads by 26.1%.

Q: Which GPU has a higher transistor density?

A: The Radeon 680M has a transistor density of 63.0 million transistors per square millimeter, compared to 24.4 million per square millimeter for the Radeon Pro 560. The 680M also uses a smaller process node (6 nm versus 14 nm).

Q: Are both GPUs integrated graphics?

A: Yes, both are listed as IGP (integrated graphics processor) with a slot width of "IGP" and no power connectors. However, the Radeon Pro 560 is part of the Radeon Pro Mac (500 Series) generation, while the 680M is from the Navi II IGP (Rembrandt Mobile) generation.

Head-to-Head Benchmarks

The database records two direct comparisons between these GPUs, and both favor the AMD Radeon 680M decisively.

Starting with Geekbench OpenCL, the 680M produces a score of 23,468. The Radeon Pro 560 manages 15,504. The delta is 33.9% in favor of the 680M. This is the largest margin in either test. OpenCL is a general-purpose compute API, so this result indicates the 680M has substantially higher raw compute throughput. The Pro 560's 1.858 TFLOPS FP32 rating versus the 680M's 3.379 TFLOPS aligns with this outcome, though the architecture differences also play a role.

In Geekbench Vulkan, the 680M again takes the lead with 21,965 points. The Pro 560 scores 16,232, a 26.1% deficit. Vulkan is a low-level graphics and compute API, so this result reflects not only compute power but also driver efficiency and hardware feature support. The 680M's RDNA 2.0 architecture with 12 ray tracing cores likely contributes to its Vulkan advantage.

The 680M's wins are consistent across both APIs. There is no test where the Pro 560 pulls ahead. The narrowest margin is 26.1% in Vulkan, which is still a substantial gap. For context, the Pro 560's nearest rivals in the database include the AMD Radeon 780M (average score 17,588, delta -0.2%) and the NVIDIA GeForce RTX 4060 (17,639, delta -0.5%). The 680M's nearest rivals include the NVIDIA GeForce GTX 580 (15,283, delta -0.1%) and the RTX 2060 (15,290, delta -0.1%). These rival positions show that the Pro 560 sits in a higher overall performance tier by average score, but the 680M's specific benchmark wins are what matter for direct comparison.

Specification Differences

The two GPUs differ in nearly every major specification category. The most striking differences are in memory architecture. The Radeon Pro 560 has 4 GB of dedicated GDDR5 memory on a 128-bit bus, delivering 81.28 GB/s of bandwidth. The Radeon 680M uses system shared memory with a bus width and bandwidth listed as "System Shared" and "System Dependent" respectively. This means the 680M's memory performance is not fixed; it depends on the host system's RAM configuration.

Clock speeds also differ significantly. The Pro 560's boost clock is not listed, but its memory clock is 1270 MHz with 5.1 Gbps effective. The 680M has a base clock of 2000 MHz and a boost clock of 2200 MHz, with memory speed dependent on the system. The 680M's higher clocks contribute to its compute advantage, despite having fewer shading units.

The shading unit count favors the Pro 560: 1024 shading units versus 768 for the 680M. The Pro 560 also has more texture mapping units (64 versus 48). However, the 680M has double the raster operations pipelines (32 versus 16) and adds 12 ray tracing cores, which the Pro 560 lacks entirely. The pixel rate tells the story: the 680M achieves 70.40 GPixel/s versus 14.51 GPixel/s for the Pro 560. Texture rate similarly favors the 680M at 105.6 GTexel/s versus 58.05 GTexel/s.

Power draw differs notably. The Pro 560 has a TDP of 75 W, while the 680M is rated at 50 W. Both are integrated with no power connectors. The bus interface also differs: the Pro 560 uses PCIe 3.0 x8, while the 680M uses PCIe 4.0 x8.

Architecture Differences

These are fundamentally different architectures from different eras. The Radeon Pro 560 uses the Polaris 21 chip built on GCN 4.0 architecture, manufactured on a 14 nm process at GlobalFoundries. It belongs to the Radeon Pro Mac (500 Series) generation. The chip contains 3,000 million transistors on a 123 mm² die, yielding a density of 24.4 million transistors per square millimeter.

The Radeon 680M uses the Rembrandt+ chip on RDNA 2.0 architecture, manufactured on a 6 nm process at TSMC. It belongs to the Navi II IGP (Rembrandt Mobile) generation. The chip contains 13,100 million transistors on a 208 mm² die, yielding a density of 63.0 million transistors per square millimeter. This is nearly three times the transistor density of the Pro 560, reflecting both the smaller process node and the more complex architecture.

The 680M's RDNA 2.0 architecture brings features the GCN 4.0 Pro 560 cannot offer. The 680M includes 12 ray tracing cores, supports DirectX 12 Ultimate (12_2), and Vulkan 1.4. The Pro 560 supports DirectX 12 (12_0) and Vulkan 1.3. Both support OpenGL 4.6. The 680M also has a 2:1 FP16 ratio (6.758 TFLOPS versus 3.379 TFLOPS FP32), while the Pro 560 has a 1:1 ratio (1.858 TFLOPS for both FP16 and FP32). This means the 680M can double its throughput on FP16 workloads.

The 680M is listed as Active production status with a release date of January 2023, while the Pro 560 is End-of-life with an April 2017 release date. The 680M has a predecessor (Vega II IGP) and a successor (Navi III IGP), while the Pro 560 lists neither. The process node difference (6 nm versus 14 nm) is a full generation apart, explaining the efficiency and performance advantages of the 680M.

The Verdict

The benchmark data is unambiguous: the AMD Radeon 680M is the superior GPU for OpenCL and Vulkan workloads. It wins both head-to-head tests with margins of 33.9% and 26.1% respectively. Its higher clock speeds, ray tracing support, newer architecture, and doubled FP16 throughput make it the more capable part in every measured direct comparison.

The Radeon Pro 560 should be chosen only in specific circumstances. Its higher average benchmark score (17,551 versus 15,270) and better overall percentile (61% versus 57%) come from a different benchmark mix that includes a strong Metal result (20,918) and no low 3DMark score. If the workload is Metal-based on macOS, the Pro 560's recorded data shows it performs well, though no direct Metal comparison exists against the 680M.

For users prioritizing OpenCL or Vulkan performance, the 680M is the clear pick. Its 23,468 OpenCL score and 21,965 Vulkan score outpaced the Pro 560 by substantial margins. The 680M also offers modern features like ray tracing and DirectX 12 Ultimate support, which the Pro 560 lacks entirely.

The 680M achieves this performance at a lower TDP (50 W versus 75 W), making it more power-efficient despite being a newer, denser chip. Its system shared memory model means performance will vary with host RAM, but its raw compute capabilities are not in question.

The Pro 560 remains relevant only for legacy macOS systems or scenarios where its 4 GB of dedicated GDDR5 memory is preferable to shared memory. The data shows no scenario where the Pro 560 wins a direct benchmark against the 680M. For any modern workload measured in OpenCL or Vulkan, the Radeon 680M is the definitive choice.