AMD Radeon 660M vs NVIDIA GeForce GTX 1660 SUPER Comparison

The AMD Radeon 660M and NVIDIA GeForce GTX 1660 SUPER occupy different corners of the GPU landscape, and the benchmark data reflects that divide clearly. The 660M is an integrated processor for mobile devices, while the GTX 1660 SUPER is a dedicated, dual-slot desktop card. Across the two shared benchmark tests, the NVIDIA part wins decisively, but the AMD chip’s integrated nature and modern architecture offer a distinct set of trade-offs that go beyond raw score comparisons.

Head-to-Head Benchmarks

The head-to-head results are unambiguous. In Geekbench OpenCL, the GTX 1660 SUPER scores 52,490 against the Radeon 660M’s 12,876. That is a delta of -75.5% for the AMD part, meaning the NVIDIA card delivers roughly four times the compute performance in this workload. The gap is nearly identical in Geekbench Vulkan: the GTX 1660 SUPER posts 57,102 versus 14,748 for the 660M, a -74.2% delta. These are not close contests; the dedicated card dominates both API-level tests by a wide margin.

Interpreting these scores requires context. The 660M’s average benchmark score across all recorded tests is 13,812, placing it in the 55th percentile of all GPUs. Its nearest rival is the NVIDIA RTX A2000 Mobile, which averages 13,821 — a negligible -0.1% difference — and it sits just ahead of the AMD Radeon RX 7900 XT, which averages 13,745 and trails by 0.5%. This suggests the 660M punches well above its integrated-class weight, trading blows with mobile workstation and even flagship desktop parts in the aggregate benchmark pool.

The GTX 1660 SUPER, meanwhile, averages 12,986 across its ten recorded benchmarks, sitting in the 53rd percentile. Its nearest rival is the NVIDIA GeForce RTX 3050 Ti Mobile at 12,940, a 0.4% gap in favor of the GTX card. The AMD Radeon RX 580 is essentially tied at 12,928, also 0.4% behind. Interestingly, the older AMD Radeon 740M trails by 0.9% at 12,870. The GTX 1660 SUPER’s average is actually lower than the 660M’s average, but that is a function of the test suites: the NVIDIA card’s PassMark scores (e.g., 12,699 in G3D, 5,076 in GPU Compute) and a low 3DMark Steel Nomad DX12 score of 1,278 pull its average down, while the AMD chip only has two strong Geekbench results in its pool.

The key takeaway: in the two tests where both parts are measured, the GTX 1660 SUPER is between 74% and 76% faster. In the broader average, the 660M looks better relative to its class, but that is a data artifact of different benchmark mixes, not a sign of real-world parity. The head-to-head numbers are the truest apples-to-apples comparison available, and they favor NVIDIA overwhelmingly.

Architecture Differences

The two GPUs are built on fundamentally different foundations. The AMD Radeon 660M uses the RDNA 2.0 architecture, fabricated on a 6 nm TSMC process. It is part of the Navi II IGP generation, specifically the Rembrandt mobile chip, and it integrates 13,100 million transistors on a 208 mm² die — a transistor density of 63.0 million per square millimeter. The NVIDIA GeForce GTX 1660 SUPER, by contrast, uses the Turing architecture on a 12 nm TSMC process. Its TU116 chip packs 6,600 million transistors into a larger 284 mm² die, yielding just 23.2 million transistors per square millimeter. The AMD chip is denser by a factor of nearly three, reflecting its newer process node.

Core configurations diverge sharply. The 660M has 384 shading units, 24 texture mapping units, and 16 raster output pipelines. It also includes 6 ray tracing cores, a feature the GTX 1660 SUPER lacks entirely — the NVIDIA part lists no RT cores and no tensor cores. The GTX 1660 SUPER compensates with raw scale: 1,408 shading units, 88 TMUs, and 48 ROPs. That translates to a pixel rate of 85.68 GPixel/s and a texture rate of 157.1 GTexel/s, versus the 660M’s 30.40 GPixel/s and 45.60 GTexel/s. Compute throughput follows the same pattern: the NVIDIA card delivers 5.027 TFLOPS FP32 and 10.05 TFLOPS FP16 (2:1), while the AMD part manages 1,459.2 GFLOPS FP32 and 2.918 TFLOPS FP16 (2:1).

Memory is another major split. The 660M uses system-shared memory, with its size, type, bus width, and bandwidth all listed as “System Shared” or “System Dependent.” The GTX 1660 SUPER has dedicated 6 GB of GDDR6 on a 192-bit bus, delivering 336.0 GB/s of bandwidth. Clock speeds are comparable on paper — the AMD chip runs at a 1500 MHz base and 1900 MHz boost, while the NVIDIA card sits at 1530 MHz base and 1785 MHz boost — but the NVIDIA part’s dedicated VRAM and far wider memory interface give it a decisive advantage in bandwidth-bound workloads.

Feature support differs on the API front. The 660M supports DirectX 12 Ultimate (12_2), while the GTX 1660 SUPER is limited to DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The AMD part also uses a PCIe 4.0 x8 interface, whereas the NVIDIA card uses the older PCIe 3.0 x16 standard. Power and physical design are opposites: the 660M is an IGP with a 40 W TDP and no power connectors, while the GTX 1660 SUPER is a dual-slot card with a 125 W TDP, a single 8-pin connector, and a suggested 300 W PSU. The NVIDIA card measures 229 mm in length, 111 mm in height, and 35 mm in width.

The Verdict

The data supports a clear split decision. For anyone building or buying a dedicated desktop GPU, the NVIDIA GeForce GTX 1660 SUPER is the obvious choice on raw performance: it is 75.5% ahead in OpenCL and 74.2% ahead in Vulkan, with a massive memory bandwidth advantage (336.0 GB/s versus system-dependent shared memory) and nearly 3.5 times the FP32 throughput (5.027 TFLOPS versus 1,459.2 GFLOPS). Its 53rd percentile standing, with rivals like the RTX 3050 Ti Mobile and RX 580 within 0.4%, shows it remains competitive in its class despite being end-of-life since a 2019 release.

For a mobile or low-power integrated solution, the AMD Radeon 660M is the more sensible part. Its 40 W TDP and IGP form factor require no power connectors and fit in portable devices, while its 55th percentile average score — higher than the GTX 1660 SUPER’s 53rd — indicates it competes surprisingly well in aggregate benchmarks for its class. The inclusion of 6 ray tracing cores and DirectX 12 Ultimate support means it has modern feature parity that the older NVIDIA card lacks. The 660M’s nearest rival is the RTX A2000 Mobile at a -0.1% delta, which puts it in mobile workstation territory.

The verdict hinges on use case. If the workload demands maximum frame rates, compute throughput, or dedicated VRAM, the GTX 1660 SUPER wins outright. If the priority is an integrated, low-power solution with modern API support and a higher relative percentile standing, the 660M is the data-backed pick. They are not direct competitors; the benchmarks simply quantify how far apart they are.

Specification Differences

| Field | AMD Radeon 660M | NVIDIA GeForce GTX 1660 SUPER |

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

| Architecture | RDNA 2.0 | Turing |

| Process Node | 6 nm | 12 nm |

| Transistors | 13,100 million | 6,600 million |

| Die Size | 208 mm² | 284 mm² |

| Transistor Density | 63.0M / mm² | 23.2M / mm² |

| Base Clock | 1500 MHz | 1530 MHz |

| Boost Clock | 1900 MHz | 1785 MHz |

| Memory Size | System Shared | 6 GB |

| Memory Type | System Shared | GDDR6 |

| Memory Bus Width | System Shared | 192 bit |

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

| Shading Units | 384 | 1408 |

| TMUs | 24 | 88 |

| ROPs | 16 | 48 |

| RT Cores | 6 | null |

| Pixel Rate | 30.40 GPixel/s | 85.68 GPixel/s |

| Texture Rate | 45.60 GTexel/s | 157.1 GTexel/s |

| FP32 | 1,459.2 GFLOPS | 5.027 TFLOPS |

| FP16 | 2.918 TFLOPS (2:1) | 10.05 TFLOPS (2:1) |

| TDP | 40 W | 125 W |

| Slot Width | IGP | Dual-slot |

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

| Suggested PSU | null | 300 W |

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

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

| Release Date | 2022-01-03 | 2019-10-28 |

FAQ

Q: Which GPU wins in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 1660 SUPER wins decisively with a score of 52,490 against the AMD Radeon 660M’s 12,876, a delta of -75.5% for the AMD part.

Q: Does the AMD Radeon 660M support ray tracing?

A: Yes, the 660M includes 6 ray tracing cores as part of its RDNA 2.0 architecture. The GTX 1660 SUPER has no RT cores listed.

Q: How much memory does each GPU have?

A: The GTX 1660 SUPER has 6 GB of GDDR6 on a 192-bit bus with 336.0 GB/s bandwidth. The 660M uses system-shared memory, with size, type, and bandwidth all listed as system-dependent.

Q: What is the average benchmark score for each GPU?

A: The 660M averages 13,812 across its two Geekbench tests, placing it in the 55th percentile. The GTX 1660 SUPER averages 12,986 across ten tests, placing it in the 53rd percentile.

Q: Are there any power connector requirements for the 660M?

A: No, the 660M is an IGP with a 40 W TDP and no power connectors. The GTX 1660 SUPER requires a single 8-pin connector and a suggested 300 W PSU.

Q: Which GPU has a higher boost clock?

A: The AMD Radeon 660M has a boost clock of 1900 MHz, which is higher than the GTX 1660 SUPER’s 1785 MHz boost clock. However, the NVIDIA card’s base clock is slightly higher at 1530 MHz versus 1500 MHz.