AMD Radeon RX 6600S vs NVIDIA GeForce GTX 1080 Comparison

The NVIDIA GeForce GTX 1080 and AMD Radeon RX 6600S represent two distinct generations of GPU design, separated by nearly six years of architectural evolution. The GTX 1080, a desktop flagship from the Pascal era, relies on raw shader throughput and memory bandwidth, while the RX 6600S, a mobile-first RDNA 2 part, counters with a modern process node and compute-centric features. Benchmark data shows a clear split: the GTX 1080 wins 6 of 8 head-to-head tests, but the RX 6600S takes the two most forward-looking workloads, revealing a trade-off between legacy rasterization strength and modern API efficiency.

Where Each One Wins

The GTX 1080 dominates in legacy DirectX workloads, which reflects its higher shading unit count and wider memory interface. In PassMark DirectX 9, the GTX 1080 scores 211 against the RX 6600S’s 176, a 19.9% margin. DirectX 11 shows a similar pattern: 124 versus 105, an 18.1% lead for NVIDIA. These older APIs rely heavily on raw geometry throughput and texture fill, areas where the GTX 1080’s 2560 shading units and 160 TMUs outperform the RX 6600S’s 1792 shading units and 112 TMUs. Even in DirectX 10, the GTX 1080 holds a 14.8% advantage (93 vs 81), confirming that its Pascal architecture scales better with pre-12 API workloads.

The RX 6600S wins only in compute and the newest API tier. In Geekbench OpenCL, it posts 66435 against the GTX 1080’s 51204, a 22.9% victory that highlights its 14.34 TFLOPS FP16 throughput versus the GTX 1080’s 138.6 GFLOPS FP16. The AMD card also edges out the GTX 1080 in PassMark DirectX 12 by a narrow 1.8% (56 vs 55), suggesting that its RDNA 2 architecture, with 28 ray accelerators and DirectX 12 Ultimate support, handles modern draw calls more efficiently. The GTX 1080’s DirectX 12_1 support is functionally limited compared to the RX 6600S’s 12_2 feature set.

For 2D and general compute, the GTX 1080 is decisively ahead. Its PassMark G2D score of 888 dwarfs the RX 6600S’s 647, a 37.2% gap, which is notable given the RX 6600S’s mobile IGP form factor. In PassMark GPU Compute, the GTX 1080’s 7614 score is 56.1% higher than the RX 6600S’s 4879, indicating that Pascal’s FP32-heavy design (8.873 TFLOPS) still outperforms RDNA 2’s compute when FP32 is the bottleneck. The overall PassMark G3D score reinforces this: 15586 for NVIDIA versus 12649 for AMD, a 23.2% lead.

Architecture Differences

The two GPUs are built on fundamentally different process nodes. The GTX 1080 uses TSMC’s 16 nm process, packing 7,200 million transistors into a 314 mm² die with a density of 22.9M transistors per mm². The RX 6600S uses the same foundry but on a 7 nm node, cramming 11,060 million transistors into a smaller 237 mm² die, achieving a density of 46.7M per mm². This density advantage allows AMD to deliver more than 50% more transistors despite a 24.5% smaller die, which directly enables the RX 6600S’s lower power envelope.

Clock speeds differ significantly. The RX 6600S has a higher base clock of 1700 MHz and a boost of 2000 MHz, with a game clock of 1881 MHz. The GTX 1080 runs at 1607 MHz base and 1733 MHz boost. Despite lower clocks, the GTX 1080 achieves higher peak throughput in FP32 (8.873 TFLOPS vs 7.168 TFLOPS) because it has 43% more shading units. However, the RX 6600S crushes FP16 performance: 14.34 TFLOPS at a 2:1 ratio versus the GTX 1080’s 138.6 GFLOPS at a 1:64 ratio, a 100-fold difference that explains its OpenCL dominance.

Memory subsystems are polar opposites. The GTX 1080 features 8 GB of GDDR5X on a 256-bit bus, yielding 320.3 GB/s of bandwidth. The RX 6600S has only 4 GB of GDDR6 on a 128-bit bus, capping at 224.0 GB/s — 30% less bandwidth. The GTX 1080’s memory clock is 1251 MHz (10 Gbps effective) versus the RX 6600S’s 1750 MHz (14 Gbps effective), but the wider bus compensates. Pixel rates favor the AMD card (128.0 GPixel/s vs 110.9 GPixel/s) due to higher clocks, while texture rates favor NVIDIA (277.3 GTexel/s vs 224.0 GTexel/s) due to more TMUs.

The RX 6600S includes 28 ray tracing cores and supports DirectX 12 Ultimate (12_2), while the GTX 1080 has no RT cores and is limited to DirectX 12_1. Both support OpenGL 4.6 and Vulkan 1.4. Power consumption is starkly different: the RX 6600S is rated at 80 W TDP with no power connectors, while the GTX 1080 draws 180 W and requires a single 8-pin connector. The GTX 1080 also uses PCIe 3.0 x16, whereas the RX 6600S uses PCIe 4.0 x8, which halves the electrical lanes but doubles the per-lane bandwidth.

The Verdict

For users prioritizing legacy DirectX performance, the GTX 1080 is the clear choice. Its wins in DirectX 9, 10, and 11 are substantial, ranging from 14.8% to 19.9%, and its PassMark G3D score of 15586 places it in the 51st percentile of all GPUs, above the RX 6600S’s 49th percentile. The GTX 1080 also excels in 2D workloads and FP32 compute, making it more versatile for older applications and general-purpose tasks. Its 8 GB memory buffer is twice the RX 6600S’s 4 GB, which matters for texture-heavy games that exceed the latter’s capacity.

The RX 6600S is the better choice for modern compute and efficiency-sensitive scenarios. Its 22.9% lead in Geekbench OpenCL and DirectX 12 superiority, however narrow, indicate that RDNA 2’s feature set—ray tracing, DirectX 12 Ultimate, and FP16 acceleration—will age better as software shifts toward these APIs. The 80 W TDP versus 180 W makes it far more suitable for thin-and-light laptops, and its 7 nm process delivers better performance per watt despite lower absolute scores. The RX 6600S’s average benchmark score of 10629 is 11.1% lower than the GTX 1080’s 11960, but its nearest rivals include the GTX 560 Ti (10690) and Quadro K2200 (10761), showing it competes with older desktop parts despite its mobile nature.

The data does not support a universal winner. The GTX 1080 wins 6 of 8 tests with an average lead of 28.3% across those wins, but the RX 6600S wins the two tests that point toward future workloads. If the primary use case is DirectX 11-era gaming or compute-heavy FP32 tasks, the GTX 1080 is superior. If the use case involves OpenCL compute, ray tracing, or power-constrained environments, the RX 6600S is the rational pick. The GTX 1080’s launch MSRP was 599 USD, which contextualizes its desktop-class performance; the RX 6600S has no listed launch MSRP, reflecting its OEM mobile focus.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 1080 has an average benchmark score of 11960, while the AMD Radeon RX 6600S scores 10629, a difference of 12.5% in NVIDIA’s favor.

Q: Does the RX 6600S support ray tracing?

A: Yes, the RX 6600S includes 28 ray tracing cores as part of its RDNA 2.0 architecture and supports DirectX 12 Ultimate (12_2). The GTX 1080 has no ray tracing cores.

Q: How do they compare in memory bandwidth?

A: The GTX 1080 offers 320.3 GB/s from 8 GB of GDDR5X on a 256-bit bus, while the RX 6600S provides 224.0 GB/s from 4 GB of GDDR6 on a 128-bit bus, giving NVIDIA a 43% bandwidth advantage.

Q: Which card wins in DirectX 9 performance?

A: The GTX 1080 wins decisively with a PassMark DirectX 9 score of 211 versus the RX 6600S’s 176, a 19.9% margin.

Q: What is the power consumption difference?

A: The RX 6600S is rated at 80 W TDP with no power connectors, while the GTX 1080 is rated at 180 W and requires a single 8-pin power connector.

Q: Which GPU has better OpenCL performance?

A: The RX 6600S leads in Geekbench OpenCL with a score of 66435 against the GTX 1080’s 51204, a 22.9% advantage, driven by its superior FP16 throughput.

Head-to-Head Benchmarks

The largest victory for the GTX 1080 comes in PassMark GPU Compute, where it scores 7614 against the RX 6600S’s 4879 — a 56.1% lead. This gap reflects the GTX 1080’s 8.873 TFLOPS FP32 versus the RX 6600S’s 7.168 TFLOPS, combined with the latter’s lower memory bandwidth. The second-biggest win is in PassMark G2D: 888 versus 647, a 37.2% margin that suggests Pascal’s 2D engine is far more efficient despite the RX 6600S’s higher clocks.

In DirectX 9, the GTX 1080 scores 211 against 176, a 19.9% win that aligns with its 160 TMUs and 64 ROPs. DirectX 11 shows an 18.1% lead (124 vs 105), and DirectX 10 a 14.8% lead (93 vs 81). The PassMark G3D score of 15586 versus 12649 is a 23.2% difference, which matches the GTX 1080’s percentile rank of 51 versus the RX 6600S’s 49.

The RX 6600S’s biggest win is Geekbench OpenCL: 66435 versus 51204, a 22.9% margin. This is directly attributable to its 14.34 TFLOPS FP16 performance, which is 103 times higher than the GTX 1080’s 138.6 GFLOPS. The RX 6600S also wins DirectX 12 by a slim 1.8% (56 vs 55), a negligible margin but one that indicates architectural parity in the newest API. The GTX 1080’s DirectX 12_1 support cannot match the RX 6600S’s 12_2 feature set, which includes ray tracing and variable rate shading.

Specification Differences

| Specification | NVIDIA GeForce GTX 1080 | AMD Radeon RX 6600S |

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

| Architecture | Pascal | RDNA 2.0 |

| Process Node | 16 nm | 7 nm |

| Transistors | 7,200 million | 11,060 million |

| Die Size | 314 mm² | 237 mm² |

| Transistor Density | 22.9M / mm² | 46.7M / mm² |

| Base Clock | 1607 MHz | 1700 MHz |

| Boost Clock | 1733 MHz | 2000 MHz |

| Memory Size | 8 GB GDDR5X | 4 GB GDDR6 |

| Memory Bus | 256 bit | 128 bit |

| Memory Bandwidth | 320.3 GB/s | 224.0 GB/s |

| Shading Units | 2560 | 1792 |

| TMUs | 160 | 112 |

| ROPs | 64 | 64 |

| RT Cores | None | 28 |

| FP32 Performance | 8.873 TFLOPS | 7.168 TFLOPS |

| FP16 Performance | 138.6 GFLOPS | 14.34 TFLOPS |

| Pixel Rate | 110.9 GPixel/s | 128.0 GPixel/s |

| Texture Rate | 277.3 GTexel/s | 224.0 GTexel/s |

| TDP | 180 W | 80 W |

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

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

| DirectX Support | 12 (12_1) | 12 Ultimate (12_2) |

| Release Date | 2016-05-26 | 2022-01-03 |

The GTX 1080 has twice the memory capacity and 43% more bandwidth, but the RX 6600S counters with 54% more transistors, 134% higher FP16 throughput, and 125% lower TDP. The RX 6600S also has a 15.4% higher base clock and 15.4% higher boost clock, while the GTX 1080 offers a 23.8% higher texture rate and 11.1% higher FP32. The GTX 1080’s average benchmark score of 11960 sits 11.1% above the RX 6600S’s 10629, but the RX 6600S’s nearest rivals include the GTX 560 Ti (10690) and R9 M275X (10582), while the GTX 1080’s rivals include the RX 7800 XT (11627) and GTX 1660 (11680), indicating the GTX 1080 competes in a higher performance tier.