GPU Comparison

The NVIDIA GeForce RTX 2060 and the AMD Radeon 680M are separated by a single percentage point in average benchmark score, yet their performance profiles could hardly be more different. The RTX 2060 wins every single head-to-head test by a massive margin, while the 680M counters with vastly superior power efficiency and a modern 6 nm manufacturing process. The data shows a clear winner for raw performance, but the 680M remains a compelling option for specific, power-constrained use cases.

Head-to-Head Benchmarks

The RTX 2060 dominates the 680M across all three shared benchmark tests, with margins ranging from 177% to 228.6%. In the 3DMark Steel Nomad DX12 test, the RTX 2060 scores 1242 against the 680M’s 378, a delta of 228.6% in favor of the NVIDIA card. This is not a close contest; the RTX 2060 more than triples the AMD integrated graphics’ output in this modern DirectX 12 workload.

The gap narrows slightly in compute-oriented tests but remains overwhelming. In Geekbench OpenCL, the RTX 2060 posts 65014 points versus 23468 for the 680M, a 177% advantage. The Vulkan result is similarly lopsided: 65846 for the RTX 2060 versus 21965 for the 680M, a 199.8% delta. Across all three tests, the RTX 2060 wins 3–0, with an average lead of roughly 200%.

The average benchmark scores tell a different story, however. The RTX 2060 averages 15290 across all its benchmarks, while the 680M averages 15270. The deltaPct between them is a mere 0.1%, placing them as near-identical in overall standing. The 680M’s closest rival is the NVIDIA GeForce GTX 580 at 15283 (deltaPct -0.1%), and the RTX 2060’s closest rival is the same GTX 580 at 15283 (deltaPct 0%). Both GPUs sit at the 57th–58th percentile of all GPUs, indicating they are mid-pack performers overall, despite the RTX 2060’s head-to-head superiority in the tests where they directly compete.

Architecture Differences

The architectural divide between these two is stark. The RTX 2060 is built on NVIDIA’s Turing architecture, using the TU106 chip fabricated on a 12 nm process at TSMC. It packs 10,800 million transistors into a 445 mm² die, yielding a transistor density of 24.3M per mm². In contrast, the 680M uses AMD’s RDNA 2.0 architecture with the Rembrandt+ chip, manufactured on a 6 nm process, also at TSMC. It contains 13,100 million transistors on a much smaller 208 mm² die, achieving a density of 63.0M per mm², nearly three times the density of the NVIDIA chip.

The RTX 2060 is a discrete graphics card with 1920 shading units, 120 texture mapping units, and 48 ROPs. It also includes 30 ray tracing cores and 240 tensor cores, enabling hardware-accelerated ray tracing and AI features. The 680M is an integrated graphics processor (IGP) with just 768 shading units, 48 TMUs, and 32 ROPs. It has 12 ray tracing cores but no tensor cores, limiting its feature set compared to the NVIDIA card.

Memory configurations are fundamentally different. The RTX 2060 has 6 GB of dedicated GDDR6 memory on a 192-bit bus, delivering 336.0 GB/s of bandwidth. The 680M uses System Shared memory, with its bandwidth described as System Dependent. This means the 680M’s performance is heavily influenced by the host system’s RAM, unlike the RTX 2060’s dedicated VRAM.

Clock speeds favor the 680M: its base clock is 2000 MHz and boost reaches 2200 MHz, versus the RTX 2060’s 1365 MHz base and 1680 MHz boost. However, the RTX 2060’s sheer number of cores and dedicated memory more than compensate for its lower clocks. The RTX 2060 delivers 6.451 TFLOPS of FP32 performance and 12.90 TFLOPS of FP16, while the 680M manages 3.379 TFLOPS FP32 and 6.758 TFLOPS FP16.

Where Each One Wins

The RTX 2060 wins decisively in scenarios demanding raw GPU throughput. Its 6 GB of dedicated GDDR6 memory with 336.0 GB/s bandwidth makes it suitable for high-resolution textures and complex scenes that would starve a shared-memory IGP. The data confirms this: its 3DMark Steel Nomad score of 1242 indicates strong DirectX 12 gaming capability, while the 680M’s 378 suggests it struggles with modern, demanding titles.

The 680M wins on efficiency and integration. Its 50 W TDP is dramatically lower than the RTX 2060’s 160 W, making it ideal for thin-and-light laptops where power draw and heat dissipation are critical. The 680M requires no power connectors and is described as an IGP, meaning it is built into the processor rather than occupying a slot. Its 6 nm process node and 63.0M / mm² transistor density indicate a modern, power-efficient design.

For gaming, the RTX 2060 is the clear choice based on benchmark data. For portable devices prioritizing battery life and low heat, the 680M’s system-dependent memory and 50 W TDP make it a practical option, provided the user accepts significantly lower frame rates in demanding games. The 680M’s PCIe 4.0 x8 interface also offers faster data transfer than the RTX 2060’s PCIe 3.0 x16, though this benefits system-level tasks more than GPU compute.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 2060 has an average benchmark score of 15290, while the AMD Radeon 680M scores 15270, a difference of just 0.1% in favor of the RTX 2060.

Q: How much faster is the RTX 2060 in 3DMark Steel Nomad DX12?

A: The RTX 2060 scores 1242 versus the 680M’s 378, a 228.6% advantage for the NVIDIA card.

Q: What is the TDP difference between these two GPUs?

A: The RTX 2060 has a TDP of 160 W, while the 680M draws just 50 W, making the AMD IGP far more power-efficient.

Q: Does the RTX 2060 support ray tracing?

A: Yes, the RTX 2060 includes 30 ray tracing cores. The 680M also has 12 ray tracing cores, but its lower overall performance limits their usefulness.

Q: What type of memory does each GPU use?

A: The RTX 2060 uses 6 GB of dedicated GDDR6 memory on a 192-bit bus. The 680M uses System Shared memory, with bandwidth described as System Dependent.

Q: Which GPU has a higher transistor density?

A: The 680M has 63.0M transistors per mm² on a 6 nm process, versus the RTX 2060’s 24.3M per mm² on a 12 nm process.

The Verdict

The data is unambiguous: the NVIDIA GeForce RTX 2060 is the superior performer in every benchmark where both are tested. Its 228.6% lead in 3DMark Steel Nomad, 177% lead in Geekbench OpenCL, and 199.8% lead in Geekbench Vulkan leave no doubt about which card delivers higher frame rates and faster compute. For anyone building a desktop or gaming laptop where performance is the priority, the RTX 2060 is the only choice from this comparison.

The AMD Radeon 680M, however, serves a different purpose. Its 50 W TDP, lack of power connectors, and IGP form factor make it suitable for ultraportable devices where the RTX 2060’s 160 W draw and dual-slot footprint would be impractical. The 680M’s 6 nm process and 63.0M / mm² density show it is a modern design optimized for efficiency rather than brute force. Its average benchmark score of 15270 is nearly identical to the RTX 2060’s 15290, but this is misleading, the 680M’s score is likely inflated by less demanding workloads, while the RTX 2060 dominates in the shared tests.

The RTX 2060 also offers advantages beyond raw speed: 6 GB of dedicated GDDR6 memory, 240 tensor cores for AI acceleration, and a 192-bit memory bus. The 680M counters with a newer PCIe 4.0 x8 interface and a substantially smaller die size (208 mm² versus 445 mm²). Ultimately, the verdict depends on the use case: the RTX 2060 for uncompromising performance, the 680M for power-constrained portable systems where gaming is secondary.

Specification Differences

| Specification | NVIDIA GeForce RTX 2060 | AMD Radeon 680M |

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

| Architecture | Turing | RDNA 2.0 |

| Process Node | 12 nm | 6 nm |

| Transistors | 10,800 million | 13,100 million |

| Die Size | 445 mm² | 208 mm² |

| Transistor Density | 24.3M / mm² | 63.0M / mm² |

| Base Clock | 1365 MHz | 2000 MHz |

| Boost Clock | 1680 MHz | 2200 MHz |

| Memory Size | 6 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 192 bit | System Shared |

| Memory Bandwidth | 336.0 GB/s | System Dependent |

| Shading Units | 1920 | 768 |

| TMUs | 120 | 48 |

| ROPs | 48 | 32 |

| RT Cores | 30 | 12 |

| Tensor Cores | 240 | null |

| Pixel Rate | 80.64 GPixel/s | 70.40 GPixel/s |

| Texture Rate | 201.6 GTexel/s | 105.6 GTexel/s |

| FP32 Performance | 6.451 TFLOPS | 3.379 TFLOPS |

| FP16 Performance | 12.90 TFLOPS (2:1) | 6.758 TFLOPS (2:1) |

| TDP | 160 W | 50 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 8-pin | None |

| Suggested PSU | 450 W | null |

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

| Display Outputs | 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, 1x USB Type-C | Portable Device Dependent |

| Production Status | End-of-life | Active |

| Release Date | 2019-01-06 | 2023-01-02 |

| Predecessor | GeForce 10 | Vega II IGP |

| Successor | GeForce 30 | Navi III IGP |