AMD Radeon RX 6600M vs NVIDIA GeForce RTX 2080 SUPER Comparison

FAQ

Q: Which GPU is faster in raw compute performance?

A: The NVIDIA GeForce RTX 2080 SUPER leads in every benchmark recorded. Its FP32 throughput is 11.15 TFLOPS compared to 8.659 TFLOPS for the AMD Radeon RX 6600M, and it wins all 10 head-to-head tests.

Q: How large is the performance gap in the newest DirectX 12 test?

A: In the 3DMark Steel Nomad DX12 test, the RTX 2080 SUPER scores 1882 against 1495 for the RX 6600M, a 25.9% advantage. This is among the narrower gaps between the two.

Q: Does the RX 6600M have any advantage in memory capacity?

A: No. Both cards feature 8 GB of GDDR6 memory. They differ significantly in bus width and bandwidth: the NVIDIA card uses a 256-bit bus with 495.9 GB/s bandwidth, while AMD uses a 128-bit bus with 224.0 GB/s.

Q: Which card has a higher manufacturing process density?

A: The AMD RX 6600M uses a 7 nm process with 46.7M transistors per mm², while the NVIDIA RTX 2080 SUPER uses a 12 nm process with 25.0M per mm². Despite this, NVIDIA still wins all benchmarks.

Q: How do these cards rank against all other GPUs?

A: The RTX 2080 SUPER sits at the 69th percentile of all GPUs, while the RX 6600M sits at the 68th. Their average benchmark scores are 24170 and 23273, respectively, a difference of about 3.9%.

Q: Are both cards equally suited for modern API support?

A: Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA card additionally includes 384 tensor cores, while the AMD card has none listed.

Architecture Differences

The two GPUs come from different architectural generations and design philosophies. The NVIDIA GeForce RTX 2080 SUPER is built on the Turing architecture using the TU104 chip, fabricated by TSMC on a 12 nm process. It packs 13,600 million transistors into a 545 mm² die, yielding a transistor density of 25.0M per mm². The AMD Radeon RX 6600M is based on RDNA 2.0 architecture with the Navi 23 chip, also from TSMC but on a 7 nm process. It contains 11,060 million transistors on a much smaller 237 mm² die, achieving a superior transistor density of 46.7M per mm².

The compute resource allocation differs substantially. NVIDIA's chip carries 3072 shading units, 192 texture mapping units, and 64 ROPs. It also includes dedicated hardware for ray tracing with 48 RT cores and machine learning with 384 tensor cores. AMD's chip has 1792 shading units, 112 TMUs, and 64 ROPs, along with 28 ray accelerators (RT cores). Notably, the AMD card lists no tensor core equivalent.

Memory architecture is another major divergence. The RTX 2080 SUPER uses a 256-bit memory bus with memory clocked at 1937 MHz (15.5 Gbps effective), producing 495.9 GB/s of bandwidth. The RX 6600M halves the bus width to 128-bit and runs memory at 1750 MHz (14 Gbps effective), resulting in 224.0 GB/s bandwidth. Both have 8 GB of GDDR6, but the NVIDIA card has over twice the memory bandwidth.

Clock speeds favor AMD significantly. The RX 6600M has a base clock of 2068 MHz and boost of 2416 MHz, with a game clock of 2177 MHz. The RTX 2080 SUPER runs at a 1650 MHz base and 1815 MHz boost. Despite lower clocks, the NVIDIA card achieves higher performance through its larger shader count and wider memory bus. Pixel rates reflect this: the RTX 2080 SUPER hits 116.2 GPixel/s, while the RX 6600M reaches 154.6 GPixel/s — the AMD card actually leads here due to its higher clocks.

Power and physical characteristics are also contrasting. The RTX 2080 SUPER is a dual-slot desktop card with a 250 W TDP, requiring 1x 6-pin + 1x 8-pin power connectors and a 600 W suggested PSU. The RX 6600M is an integrated graphics processor for mobile platforms with a 100 W TDP and no external power connectors, making it fundamentally different in deployment scenario.

Head-to-Head Benchmarks

The data shows a clean sweep: the NVIDIA GeForce RTX 2080 SUPER wins all 10 recorded head-to-head benchmarks against the AMD Radeon RX 6600M, with a 10-0 win tally. The margins, however, vary meaningfully across different workloads.

The largest single advantage appears in PassMark DirectX 10, where NVIDIA scores 140 versus AMD's 87, a 60.9% lead. This suggests the older API layer is particularly favorable to the Turing architecture. The Vulkan compute test shows a 50.9% gap (111284 vs 73740), indicating strong raw compute performance in NVIDIA's favor. OpenCL results show a 46.4% lead (99226 vs 67765), and PassMark GPU Compute follows closely at 46.8% (8290 vs 5646).

The narrower gaps appear in more modern, balanced workloads. The 3DMark Steel Nomad DX12 test shows a 25.9% difference (1882 vs 1495), the smallest delta in the set. PassMark DirectX 11 shows a 21.3% gap (165 vs 136), and DirectX 9 shows 23.4% (227 vs 184). The 2D performance test (PassMark G2D) shows a 26.4% difference (920 vs 728), while the overall 3D test (PassMark G3D) shows a 39.9% gap (19490 vs 13929).

The average benchmark score tells a similar story: the RTX 2080 SUPER averages 24170 across all tests, while the RX 6600M averages 23273. This places the NVIDIA card at the 69th percentile of all GPUs, just one percentile point above AMD's 68th percentile. Interestingly, the RTX 2080 SUPER's nearest rivals include the AMD Radeon RX 6600 XT (which scores 24442, 1.1% higher) and the RX 6800S (24063, 0.4% lower). The RX 6600M's nearest rivals include the AMD Radeon R9 M290X (23276, 0% delta) and the NVIDIA P106-100 (23249, 0.1% delta), indicating it sits in a lower performance tier.

The Verdict

The benchmark data is unambiguous: the NVIDIA GeForce RTX 2080 SUPER outperforms the AMD Radeon RX 6600M across every measured workload. The average score difference of 897 points (24170 vs 23273) represents a meaningful gap, but the distribution of wins shows this is not a close contest — NVIDIA wins by double-digit percentages in all but the closest test.

The RTX 2080 SUPER is the choice for anyone prioritizing peak performance. Its 25.9% lead in the modern DX12 Steel Nomad test and 50.9% lead in Vulkan demonstrate that its advantage persists even in current-generation APIs. The 60.9% lead in DX10 suggests backward compatibility is also stronger.

The RX 6600M's case rests not on performance but on its fundamental design. As a 100 W integrated part with no power connectors, it is built for mobile platforms where the RTX 2080 SUPER's 250 W TDP and dual-slot form factor cannot apply. The data shows AMD's card still achieves 68th percentile ranking despite using a 7 nm process with higher clocks, which speaks to efficiency rather than raw speed.

For desktop users with power and space available, the RTX 2080 SUPER is clearly superior. For laptop implementations where power draw is the binding constraint, the RX 6600M is the only viable option of the two, and it delivers competitive performance relative to its power envelope.

Specification Differences

| Specification | NVIDIA GeForce RTX 2080 SUPER | AMD Radeon RX 6600M |

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

| Process Node | 12 nm | 7 nm |

| Transistors | 13,600 million | 11,060 million |

| Die Size | 545 mm² | 237 mm² |

| Transistor Density | 25.0M / mm² | 46.7M / mm² |

| Base Clock | 1650 MHz | 2068 MHz |

| Boost Clock | 1815 MHz | 2416 MHz |

| Game Clock | — | 2177 MHz |

| Memory Clock | 1937 MHz (15.5 Gbps) | 1750 MHz (14 Gbps) |

| Memory Bus Width | 256 bit | 128 bit |

| Memory Bandwidth | 495.9 GB/s | 224.0 GB/s |

| Shading Units | 3072 | 1792 |

| TMUs | 192 | 112 |

| RT Cores | 48 | 28 |

| Tensor Cores | 384 | — |

| Pixel Rate | 116.2 GPixel/s | 154.6 GPixel/s |

| Texture Rate | 348.5 GTexel/s | 270.6 GTexel/s |

| FP32 | 11.15 TFLOPS | 8.659 TFLOPS |

| FP16 | 22.30 TFLOPS (2:1) | 17.32 TFLOPS (2:1) |

| TDP | 250 W | 100 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 6-pin + 1x 8-pin | None |

| Suggested PSU | 600 W | — |

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

| Launch MSRP | 699 USD | — |

Where Each One Wins

NVIDIA GeForce RTX 2080 SUPER — Performance-Critical Desktop Use

Every benchmark category belongs to the RTX 2080 SUPER. The largest margins come in legacy DirectX workloads — 60.9% in DX10 and 23.4% in DX9 — making it the stronger choice for older game libraries and compatibility scenarios. Its 50.9% Vulkan lead and 46.4% OpenCL lead indicate superior compute throughput for applications that leverage these APIs. The 39.9% advantage in PassMark G3D shows it is also the better general-purpose 3D performer. With 384 tensor cores, it offers hardware acceleration for AI workloads that the AMD card cannot match.

AMD Radeon RX 6600M — Power-Constrained Mobile Deployment

The RX 6600M wins no performance benchmarks, but its design makes it the only option for thin-and-light laptops. Its 100 W TDP is 150 W lower than the NVIDIA card, and it requires no external power connectors. The 7 nm process with 46.7M transistors per mm² achieves higher density, and its pixel rate of 154.6 GPixel/s is actually 33% higher than the RTX 2080 SUPER's 116.2 GPixel/s — the only meaningful specification where AMD leads. Its PCIe 4.0 x8 interface is newer than NVIDIA's PCIe 3.0 x16, though this does not translate to benchmark wins. For users who need a discrete-class GPU in a mobile form factor, the RX 6600M's 68th percentile ranking shows it remains competitive within its power class.