AMD Radeon 680M vs AMD Ryzen Z2 GPU Comparison

# AMD Radeon 680M vs AMD Ryzen Z2 GPU

The AMD Radeon 680M and the AMD Ryzen Z2 GPU occupy different positions in the integrated graphics landscape. The 680M is an RDNA 2.0 IGP built on TSMC's 6 nm process, while the Ryzen Z2 GPU is an RDNA 3.0 part on TSMC's 4 nm node. The 680M carries an average benchmark score of 15,270 across recorded tests, placing it in the 57th percentile of all GPUs. The Ryzen Z2 GPU, by contrast, has no recorded benchmark scores in the database and sits in the 50th percentile. This makes the comparison heavily reliant on architectural specifications and theoretical throughput figures rather than direct measured performance.

Where Each One Wins

The AMD Radeon 680M wins in the only category where direct measurements exist: actual benchmark results. The database records three scores for the 680M. In 3DMark Steel Nomad DX12, it posts a score of 378. Its Geekbench OpenCL result reaches 23,468, and its Geekbench Vulkan score is 21,965. These figures give it an average benchmark score of 15,270, which places it just ahead of the NVIDIA GeForce GTX 580 (15,283, a delta of -0.1%), the NVIDIA GeForce RTX 2060 (15,290, a delta of -0.1%), and the AMD Radeon RX 7600 (15,171, a delta of 0.7%). The Ryzen Z2 GPU has zero recorded benchmarks, so the 680M wins all measured comparisons by default.

The Ryzen Z2 GPU wins on raw compute specifications. Its FP32 throughput is 8.294 TFLOPS, which is more than double the 680M's 3.379 TFLOPS. Its FP16 output is also 8.294 TFLOPS with a 1:1 ratio, whereas the 680M delivers 6.758 TFLOPS with a 2:1 ratio. The Z2 GPU also has a higher pixel rate at 86.40 GPixel/s versus 70.40 GPixel/s, and a higher texture rate at 129.6 GTexel/s versus 105.6 GTexel/s. The Z2 GPU operates with a boost clock of 2700 MHz compared to the 680M's 2200 MHz. These specification advantages indicate the Z2 GPU is designed for higher sustained throughput, but without benchmark data, the actual performance gap cannot be quantified from the database.

Architecture Differences

The two GPUs come from different architectural generations. The Radeon 680M uses RDNA 2.0, built on the Rembrandt+ chip, and belongs to the Navi II IGP generation. The Ryzen Z2 GPU uses RDNA 3.0, built on the Hawk Point chip, and is classified as a Console GPU. The process nodes differ significantly: the 680M uses TSMC's 6 nm process, while the Z2 GPU uses TSMC's 4 nm process. This node difference contributes to the Z2 GPU's higher transistor density of 142.6 million transistors per square millimeter, compared to 63.0 million per square millimeter for the 680M.

The transistor counts also diverge sharply. The 680M integrates 13,100 million transistors on a die size of 208 mm². The Z2 GPU packs 25,390 million transistors onto a smaller die of 178 mm². The Z2 GPU's die is 30 mm² smaller while carrying nearly double the transistor count, which reflects the denser 4 nm manufacturing process.

Both GPUs share identical core counts in several areas. Each has 768 shading units, 48 texture mapping units, 32 render output units, and 12 ray tracing cores. Neither has tensor cores. The shared core configuration means the architectural differences manifest primarily in clock speeds, memory implementation, and compute efficiency rather than raw unit counts.

Memory access is a major differentiator. The 680M uses system-shared memory with a bus width and bandwidth that are system dependent. The Z2 GPU, however, comes with 16 GB of dedicated LPDDR5X memory on a 128-bit bus, delivering 119.9 GB/s of bandwidth. The Z2 GPU's memory operates at 937 MHz with 7.5 Gbps effective transfer. The 680M's memory performance cannot be specified without knowing the host system's configuration.

Power characteristics also differ. The 680M has a TDP of 50 W, while the Z2 GPU has a lower TDP of 28 W despite its higher compute throughput. The 680M is an IGP with no slot width and no power connectors, while the Z2 GPU also has no power connectors and a single USB Type-C display output. The 680M connects via PCIe 4.0 x8, while the Z2 GPU's bus interface is not recorded.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries between these two GPUs. The wins column shows zero for both items. This means all comparative analysis must come from the 680M's recorded scores and the Z2 GPU's theoretical specifications.

The 680M's average benchmark score of 15,270 places it in a tight cluster of desktop GPUs. It sits 0.1% below the NVIDIA GeForce GTX 580's average of 15,283 and 0.1% below the NVIDIA GeForce RTX 2060's average of 15,290. It sits 0.5% above the NVIDIA GeForce RTX 3050 OEM's average of 15,199 and 0.7% above the AMD Radeon RX 7600's average of 15,171. These deltas are all within one percentage point, indicating that the 680M's measured performance is comparable to those discrete desktop cards from previous generations.

The Z2 GPU's lack of benchmark data leaves only its compute specifications for comparison. Its FP32 output of 8.294 TFLOPS is 2.45 times the 680M's 3.379 TFLOPS. Its texture rate of 129.6 GTexel/s is 22.7% higher than the 680M's 105.6 GTexel/s. Its pixel rate of 86.40 GPixel/s is 22.7% higher than the 680M's 70.40 GPixel/s. The Z2 GPU's boost clock of 2700 MHz exceeds the 680M's 2200 MHz by 500 MHz, while its base clock of 800 MHz is substantially lower than the 680M's 2000 MHz base.

The FP16 compute figures reveal an architectural difference. The 680M delivers 6.758 TFLOPS at 2:1 ratio, meaning it halves its FP32 rate for FP16 work. The Z2 GPU delivers 8.294 TFLOPS at 1:1 ratio, meaning its FP16 throughput equals its FP32 throughput. This indicates the Z2 GPU is better suited for workloads that leverage FP16 arithmetic, such as certain graphics effects and compute tasks.

The 680M's 3DMark Steel Nomad DX12 score of 378 is its only direct gaming-oriented measurement. Its Geekbench OpenCL score of 23,468 and Vulkan score of 21,965 provide compute and graphics API performance data. These are the only recorded measurements for either GPU in the database.

The Verdict

The data presents a clear split. The 680M is a proven, measured performer with an average benchmark score of 15,270 and a 57th percentile ranking. Its nearest rivals include the NVIDIA GeForce RTX 2060, with a delta of -0.1%, meaning the 680M essentially matches that discrete GPU's average score. This makes the 680M a known quantity for systems that rely on integrated graphics with system-shared memory.

The Ryzen Z2 GPU is an unmeasured specification sheet. Its 50th percentile ranking is based on no recorded benchmarks, and its average benchmark score is zero. The specifications, however, suggest substantial compute headroom: 8.294 TFLOPS FP32, 16 GB of dedicated LPDDR5X memory at 119.9 GB/s, and a 28 W TDP. The Z2 GPU's higher clock boost, higher pixel and texture rates, and larger memory allocation indicate it was designed for workloads that require sustained throughput beyond what the 680M can deliver.

For users who need verified performance, the 680M is the only option with recorded data. Its scores place it alongside the RTX 2060 and GTX 580 in average benchmark terms. For users who prioritize raw compute specifications and dedicated memory, the Z2 GPU's architecture suggests it can exceed the 680M's throughput, but the database provides no measured confirmation. The Z2 GPU's 4 nm process and RDNA 3.0 architecture represent a newer design, while the 680M's 6 nm process and RDNA 2.0 architecture represent the previous generation.

The power efficiency picture favors the Z2 GPU. It delivers higher theoretical compute at 28 W versus the 680M's 50 W TDP. This means the Z2 GPU's specification advantage comes with lower power draw, which is notable for portable or console-style applications. The 680M's higher TDP and system-shared memory make it dependent on the host system's configuration for memory bandwidth and capacity.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon 680M has an average benchmark score of 15,270. The AMD Ryzen Z2 GPU has an average benchmark score of zero because it has no recorded benchmarks in the database.

Q: How does the Radeon 680M compare to the NVIDIA GeForce RTX 2060?

A: The 680M's average benchmark score of 15,270 is 0.1% below the RTX 2060's average score of 15,290. This places the two GPUs at essentially the same performance level.

Q: What memory configuration does each GPU use?

A: The Radeon 680M uses system-shared memory with system-dependent bandwidth. The Ryzen Z2 GPU uses 16 GB of LPDDR5X memory on a 128-bit bus with 119.9 GB/s bandwidth.

Q: How do the FP32 compute figures compare?

A: The Ryzen Z2 GPU delivers 8.294 TFLOPS FP32, which is 2.45 times the Radeon 680M's 3.379 TFLOPS.

Q: What are the process node differences?

A: The Radeon 680M uses TSMC's 6 nm process with 13,100 million transistors on a 208 mm² die. The Ryzen Z2 GPU uses TSMC's 4 nm process with 25,390 million transistors on a 178 mm² die.

Q: What is the TDP for each GPU?

A: The Radeon 680M has a TDP of 50 W. The Ryzen Z2 GPU has a TDP of 28 W.