AMD Radeon 680M vs AMD Ryzen Z2 A GPU Comparison

Where Each One Wins

The AMD Radeon 680M is the clear performance leader across every recorded benchmark in the database. Its three benchmark scores are all substantial, with the strongest result coming from Geekbench OpenCL at 23,468 points and a Vulkan score of 21,965 points. The 3DMark Steel Nomad DX12 result of 378 points rounds out the picture. The Ryzen Z2 A GPU has no recorded benchmark scores in the database, so all measured performance comparisons default to the 680M.

The 680M also holds a higher overall percentile ranking. It sits at the 57th percentile among all GPUs, while the Z2 A GPU sits at the 50th percentile. The average benchmark score for the 680M is 15,270 points, derived from its three tests. The Z2 A GPU has an average benchmark score of 0, reflecting the absence of any recorded tests. The database shows zero wins for each side in head-to-head comparisons, but that is because no direct comparative benchmarks exist for this pair; the 680M still delivers all of the measured compute output.

In practical terms, the 680M wins outright in any task that relies on raw shading throughput, pixel fill, or texture processing. Its 768 shading units, 48 texture mapping units, and 32 ROPs all exceed the Z2 A GPU's 512 shading units, 32 TMUs, and 16 ROPs. The 680M also produces a pixel rate of 70.40 GPixel/s versus 25.60 GPixel/s for the Z2 A GPU, and a texture rate of 105.6 GTexel/s versus 51.20 GTexel/s. These are not marginal gaps; the 680M more than doubles the pixel fill and doubles the texture rate.

The Z2 A GPU's advantage is confined to power efficiency and portability context. Its 15 W TDP is one third of the 680M's 50 W TDP. For thermally constrained, low-power console-class designs, that difference matters. The Z2 A GPU also offers a dedicated memory configuration with 16 GB of LPDDR5 on a 128-bit bus, yielding 102.4 GB/s of bandwidth. The 680M uses system-shared memory with bandwidth described as system dependent, so in a fixed memory setup the Z2 A GPU could have a more predictable memory subsystem. But in raw compute terms, the 680M wins every measurable category.

Architecture Differences

Both GPUs are built on AMD's RDNA 2.0 architecture, but they diverge sharply in implementation. The 680M uses the Rembrandt+ chip, which is part of the Navi II IGP generation for Rembrandt Mobile. The Z2 A GPU uses the Van Gogh chip, classified as a Console GPU generation. The 680M is fabricated on a 6 nm process at TSMC, while the Z2 A GPU uses a 7 nm process, also at TSMC. The smaller node gives the 680M a density advantage: 63.0 million transistors per square millimeter versus 14.7 million for the Z2 A GPU.

The transistor counts reflect the scale difference. The 680M packs 13,100 million transistors on a 208 mm² die. The Z2 A GPU has 2,400 million transistors on a 163 mm² die. That is a substantial gap in complexity, and it shows up in the execution resources. The 680M carries 768 shading units, 48 TMUs, 32 ROPs, and 12 ray tracing cores. The Z2 A GPU has 512 shading units, 32 TMUs, 16 ROPs, and 8 ray tracing cores. Every execution block is larger on the 680M, which directly explains its higher pixel rate, texture rate, and FP32 throughput.

Clock speeds also favor the 680M. Its base clock is 2000 MHz with a boost of 2200 MHz. The Z2 A GPU runs at a 1000 MHz base and 1600 MHz boost. The 680M's higher clocks compound its wider execution resources. The FP32 output is 3.379 TFLOPS for the 680M versus 1.638 TFLOPS for the Z2 A GPU. FP16 performance follows the same pattern: 6.758 TFLOPS (2:1) for the 680M versus 3.277 TFLOPS (2:1) for the Z2 A GPU.

The memory systems are different in kind. The 680M relies on system-shared memory with no dedicated VRAM, a bus width described as system shared, and bandwidth that depends on the host platform. The Z2 A GPU has a fixed 16 GB LPDDR5 pool on a 128-bit bus with 102.4 GB/s of bandwidth. Its memory clock is listed as 800 MHz with 6.4 Gbps effective. The Z2 A GPU's memory is not shared; it is a dedicated allocation. The 680M's integrated design means its memory performance varies with the system it is installed into.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The 680M uses a PCIe 4.0 x8 bus interface, while the Z2 A GPU has no bus interface listed. The 680M's display outputs are described as portable device dependent, while the Z2 A GPU lists a single USB Type-C output. The 680M is an IGP with no power connectors, while the Z2 A GPU has no slot width or power connector data recorded.

The release dates differ by nearly two years. The 680M launched on January 2, 2023, and its predecessor is the Vega II IGP with a successor in the Navi III IGP. The Z2 A GPU launched on December 31, 2024, with no predecessor or successor listed. Both are marked as Active in production status.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries for these two GPUs. All comparative analysis must come from the individual benchmark suites recorded for the 680M and the architectural specifications of both parts. The 680M's three recorded benchmarks provide concrete numbers. The Z2 A GPU has no recorded scores, so there are no rival deltas to cite.

The 680M's strongest single result is its Geekbench OpenCL score of 23,468. That figure represents the GPU's compute throughput in a general-purpose workload. Its Vulkan score of 21,965 is close, indicating consistent performance across different API paths. The 3DMark Steel Nomad DX12 score of 378 is a lower absolute number, but that test is a demanding next-generation workload that scales with raw rasterization and ray tracing resources. The 680M's 12 ray tracing cores and 32 ROPs feed that test.

For context, the 680M's nearest rivals in the database are desktop discrete GPUs. The NVIDIA GeForce GTX 580 has an average score of 15,283, a delta of -0.1% versus the 680M's 15,270 average. The NVIDIA GeForce RTX 2060 averages 15,290, also -0.1%. The NVIDIA GeForce RTX 3050 OEM averages 15,199, which is +0.5% relative to the 680M. The AMD Radeon RX 7600 averages 15,171, a +0.7% delta. These deltas are all within one percentage point, meaning the 680M's average benchmark score lands in the same performance band as those discrete cards despite being an integrated part.

The Z2 A GPU has no nearest rivals and no average score, so it cannot be placed in that same field. Its 50th percentile ranking is a default position rather than a measured result. The 680M's 57th percentile is a measured outcome based on its three benchmark entries.

The FP32 and FP16 comparisons are the clearest numerical story. The 680M delivers 3.379 TFLOPS of FP32, which is more than double the Z2 A GPU's 1.638 TFLOPS. The FP16 numbers are 6.758 versus 3.277 TFLOPS, again roughly a 2.06x advantage. Pixel rate is 70.40 GPixel/s versus 25.60 GPixel/s, a 2.75x gap. Texture rate is 105.6 GTexel/s versus 51.20 GTexel/s, a 2.06x gap. These ratios hold consistently across all throughput metrics because the 680M has both more execution units and higher clocks.

The Z2 A GPU does have one clear specification win: memory bandwidth. Its 102.4 GB/s is fixed and dedicated, whereas the 680M's bandwidth is system dependent and cannot be stated as a fixed number. In a platform where the 680M shares system memory bandwidth with the CPU, the Z2 A GPU's dedicated pool could avoid contention. But that advantage does not translate into any recorded benchmark score for the Z2 A GPU.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon 680M has an average benchmark score of 15,270 points based on three tests. The AMD Ryzen Z2 A GPU has an average benchmark score of 0 because no benchmark results are recorded in the database.

Q: How much faster is the 680M in FP32 compute?

A: The 680M delivers 3.379 TFLOPS of FP32 throughput, while the Z2 A GPU delivers 1.638 TFLOPS. The 680M is more than double the Z2 A GPU in this metric.

Q: What memory configuration does each GPU use?

A: The 680M uses system-shared memory with system-dependent bandwidth and no dedicated VRAM. The Z2 A GPU has 16 GB of LPDDR5 memory on a 128-bit bus with a fixed bandwidth of 102.4 GB/s.

Q: Do both GPUs support the same graphics APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What are the process nodes for each chip?

A: The 680M uses a 6 nm TSMC process, while the Z2 A GPU uses a 7 nm TSMC process. The 680M also has a higher transistor density at 63.0 million transistors per square millimeter versus 14.7 million.

Q: Which GPU has more ray tracing cores?

A: The 680M has 12 ray tracing cores, while the Z2 A GPU has 8 ray tracing cores.

Specification Differences

| Specification | AMD Radeon 680M | AMD Ryzen Z2 A GPU |

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

| Chip | Rembrandt+ | Van Gogh |

| Generation | Navi II IGP (Rembrandt Mobile) | Console GPU (AMD) |

| Process Node | 6 nm | 7 nm |

| Foundry | TSMC | TSMC |

| Transistors | 13,100 million | 2,400 million |

| Die Size | 208 mm² | 163 mm² |

| Transistor Density | 63.0M / mm² | 14.7M / mm² |

| Base Clock | 2000 MHz | 1000 MHz |

| Boost Clock | 2200 MHz | 1600 MHz |

| Memory Size | System Shared | 16 GB |

| Memory Type | System Shared | LPDDR5 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 102.4 GB/s |

| Shading Units | 768 | 512 |

| Texture Mapping Units | 48 | 32 |

| ROPs | 32 | 16 |

| Ray Tracing Cores | 12 | 8 |

| Pixel Rate | 70.40 GPixel/s | 25.60 GPixel/s |

| Texture Rate | 105.6 GTexel/s | 51.20 GTexel/s |

| FP32 Performance | 3.379 TFLOPS | 1.638 TFLOPS |

| FP16 Performance | 6.758 TFLOPS (2:1) | 3.277 TFLOPS (2:1) |

| TDP | 50 W | 15 W |

| Slot Width | IGP | Not listed |

| Power Connectors | None | Not listed |

| Bus Interface | PCIe 4.0 x8 | Not listed |

| Display Outputs | Portable Device Dependent | 1x USB Type-C |

| Release Date | 2023-01-02 | 2024-12-31 |

| Predecessor | Vega II IGP | None listed |

| Successor | Navi III IGP | None listed |

| Percentile vs All GPUs | 57 | 50 |

| Avg Benchmark Score | 15270 | 0 |