NVIDIA GeForce GTX 1660 vs NVIDIA GeForce GTX 780M Comparison

The NVIDIA GeForce GTX 1660 and the NVIDIA GeForce GTX 780M represent two distinct eras of laptop and desktop graphics, separated by nearly six years of architectural evolution. The benchmark data shows a decisive generational leap, with the GTX 1660 dominating the shared test suite, yet the GTX 780M’s specifications reveal a design with its own historical strengths. This analysis compares the two based on available benchmark scores, architecture, and specifications.

Head-to-Head Benchmarks

The head-to-head data is unambiguous, with the GTX 1660 winning both shared benchmark tests by overwhelming margins. In the Geekbench OpenCL test, the GTX 1660 scores 47,850 points against the GTX 780M’s 12,769, a delta of 274.7% in favor of the newer card. This is not a marginal improvement; it is a near-tripling of compute throughput, reflecting the massive gains in shader efficiency and clock speeds across generations.

The Vulkan results are even more lopsided. The GTX 1660 posts 50,137 points, while the GTX 780M manages only 12,696, representing a 294.9% advantage for the GTX 1660. This suggests that the Turing architecture’s asynchronous compute and modern driver optimizations translate into outsized gains in low-level API workloads. For context, the GTX 1660’s average benchmark score across all tests is 11,680, placing it in the 51st percentile of all GPUs. Its closest rival, the AMD Radeon RX 7800 XT, averages 11,627 (a 0.5% difference), while the AMD Radeon RX 6500 XT scores 11,842 (1.4% higher). The GTX 1660 sits in a tightly contested mid-range bracket.

The GTX 780M, by contrast, has an average benchmark score of 11,261, placing it in the 50th percentile. Its nearest rivals are the AMD Radeon Pro WX 3200 (11,228, 0.3% lower) and the AMD FirePro W4300 (11,225, 0.3% lower). Notably, the two NVIDIA RTX PRO 6000 Blackwell Max-Q variants also appear as rivals, scoring 11,088 (1.6% lower), which is an unusual grouping for a mobile GPU from 2013. The delta between the two cards’ average scores is only 419 points (11,680 vs 11,261), yet the head-to-head Geekbench tests show a chasm. This discrepancy arises because the GTX 780M’s average is pulled up by its Geekbench Metal score of 8,319, which is not a test the GTX 1660 participates in, while the GTX 1660’s average includes additional DirectX and Passmark tests where it excels.

In practical terms, the GTX 1660’s 274.7% lead in OpenCL means it processes compute workloads nearly four times faster. For gaming, the Vulkan advantage of 294.9% indicates that modern titles using Vulkan will see dramatically higher frame rates on the GTX 1660. The GTX 780M, with its 50th percentile ranking, was competitive in its day but is now firmly in legacy territory. The data shows no single test where the GTX 780M wins, confirming the GTX 1660 as the superior performer across all measured metrics.

Architecture Differences

The architectural gap between these two GPUs is vast, stemming from a 6-year generational split. The GTX 1660 is built on the TU116 chip using the Turing architecture, fabricated on a 12 nm process at TSMC. It packs 6,600 million transistors into a 284 mm² die, yielding a transistor density of 23.2M per mm². In contrast, the GTX 780M uses the GK104 chip with the Kepler architecture, built on a 28 nm process, also at TSMC. It contains 3,540 million transistors on a slightly larger 294 mm² die, resulting in a much lower density of 12.0M per mm². The smaller process node allows the GTX 1660 to nearly double the transistor count while using a comparable die area.

Clock speeds tell a similar story of progress. The GTX 1660 runs at a base clock of 1530 MHz and boosts to 1785 MHz. The GTX 780M is dramatically slower, with a base of 771 MHz and a boost of just 797 MHz. This nearly 2x clock advantage, combined with the architectural improvements, explains the massive benchmark deltas.

Memory configurations also diverge. The GTX 1660 uses 6 GB of GDDR5 on a 192-bit bus, achieving 192.1 GB/s of bandwidth with an effective memory clock of 8 Gbps. The GTX 780M has 4 GB of GDDR5 on a wider 256-bit bus, but its memory runs at just 5 Gbps effective, yielding 160.0 GB/s. The GTX 1660’s higher bandwidth, despite the narrower bus, is a direct result of faster memory technology.

Shader resources favor the older card in raw counts but not in efficiency. The GTX 780M has 1536 shading units and 128 TMUs, compared to the GTX 1660’s 1408 shading units and 88 TMUs. However, the GTX 1660 has 48 ROPs versus the GTX 780M’s 32. The Kepler architecture’s higher unit counts are nullified by its low clocks and older design. The GTX 1660 achieves a pixel rate of 85.68 GPixel/s and a texture rate of 157.1 GTexel/s, versus the GTX 780M’s 25.50 GPixel/s and 102.0 GTexel/s. FP32 throughput is 5.027 TFLOPS on the GTX 1660, more than double the GTX 780M’s 2.448 TFLOPS. The GTX 1660 also supports FP16 at 10.05 TFLOPS (2:1), a feature the GTX 780M lacks entirely.

API support differs subtly. Both support DirectX 12, but the GTX 1660 reaches 12_1 while the GTX 780M is limited to 11_0. OpenGL is identical at 4.6, but Vulkan support is newer on the GTX 1660 (1.4) versus the GTX 780M (1.2.175). The GTX 1660 also has no dedicated RT or Tensor cores, matching the GTX 780M, which also lacks them.

Power and form factor are points of contrast. The GTX 1660 has a 120 W TDP, uses a dual-slot design, requires a 1x 8-pin power connector, and suggests a 300 W PSU. The GTX 780M has a 122 W TDP, nearly identical, but comes as an MXM Module with no power connectors, relying on the host laptop’s power delivery. The GTX 1660 is a desktop card with PCIe 3.0 x16 and standard display outputs (DVI, HDMI 2.0, DisplayPort 1.4a), while the GTX 780M uses an MXM-B (3.0) interface and has portable-device-dependent outputs.

FAQ

Q: Which GPU is faster in compute workloads?

A: The GTX 1660 is decisively faster. In Geekbench OpenCL, it scores 47,850 versus the GTX 780M’s 12,769, a 274.7% advantage. This aligns with its 5.027 TFLOPS FP32 throughput versus 2.448 TFLOPS.

Q: Does the GTX 780M have any advantages in raw specifications?

A: Yes, in two areas. It has more shading units (1536 vs 1408), more TMUs (128 vs 88), and a wider 256-bit memory bus versus 192-bit. However, these advantages do not translate into performance wins due to much lower clock speeds and older architecture.

Q: What is the difference in Vulkan performance?

A: The GTX 1660 leads by 294.9% in Geekbench Vulkan, scoring 50,137 against 12,696. The GTX 1660 supports Vulkan 1.4, while the GTX 780M supports Vulkan 1.2.175.

Q: How do their transistor densities compare?

A: The GTX 1660 has a density of 23.2M transistors per mm² on a 12 nm process, while the GTX 780M has 12.0M per mm² on a 28 nm process. The GTX 1660 fits 6,600 million transistors on a 284 mm² die, versus 3,540 million on a 294 mm² die.

Q: Are both GPUs end-of-life?

A: Yes. The GTX 1660’s production status is end-of-life, with a release date of 2019-03-13. The GTX 780M is also end-of-life, released on 2013-05-10.

Q: Which GPU has higher memory bandwidth?

A: The GTX 1660 achieves 192.1 GB/s, while the GTX 780M achieves 160.0 GB/s. The GTX 1660 uses faster GDDR5 at 8 Gbps effective, while the GTX 780M runs at 5 Gbps effective.

The Verdict

The data points to a clear winner for any modern workload. The GTX 1660 wins both head-to-head benchmarks by margins of 274.7% and 294.9%, and its average benchmark score of 11,680 is higher than the GTX 780M’s 11,261. The GTX 1660 also has a transistor density nearly double that of the GTX 780M, a 12 nm process versus 28 nm, and more than twice the FP32 throughput (5.027 TFLOPS vs 2.448 TFLOPS). For gamers or compute users, the GTX 1660 is the only rational choice, as it delivers modern API support (DirectX 12_1, Vulkan 1.4) and 6 GB of memory.

The GTX 780M, however, retains historical relevance. Its 1536 shading units and 128 TMUs were current-generation for 2013, and its MXM form factor was designed for high-end laptops. Its 4 GB of GDDR5 on a 256-bit bus was respectable at launch, and its 122 W TDP is comparable to the GTX 1660’s 120 W. But the data shows it cannot compete in any benchmark where both are measured. The GTX 780M’s percentile rank of 50 versus the GTX 1660’s 51 is misleading; the average scores are close only because the GTX 780M’s Metal score (8,319) is included in its average, while the GTX 1660’s average is dragged down by low DirectX 9 and 10 scores (61 and 49).

For a user picking between these two today, the GTX 1660 is the unequivocal choice for performance. It is 274.7% faster in OpenCL and 294.9% faster in Vulkan, making it suitable for current games and compute tasks. The GTX 780M should only be considered for legacy systems where the MXM form factor is required, as its performance is insufficient for modern applications. The GTX 1660’s launch MSRP is 219 USD, a figure that reflects its mid-range positioning at release.

Specification Differences

| Specification | NVIDIA GeForce GTX 1660 | NVIDIA GeForce GTX 780M |

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

| Architecture | Turing | Kepler |

| Process Node | 12 nm | 28 nm |

| Transistors | 6,600 million | 3,540 million |

| Die Size | 284 mm² | 294 mm² |

| Transistor Density | 23.2M / mm² | 12.0M / mm² |

| Base Clock | 1530 MHz | 771 MHz |

| Boost Clock | 1785 MHz | 797 MHz |

| Memory Size | 6 GB | 4 GB |

| Memory Bus | 192 bit | 256 bit |

| Memory Bandwidth | 192.1 GB/s | 160.0 GB/s |

| Memory Clock | 2001 MHz (8 Gbps effective) | 1250 MHz (5 Gbps effective) |

| Shading Units | 1408 | 1536 |

| TMUs | 88 | 128 |

| ROPs | 48 | 32 |

| Pixel Rate | 85.68 GPixel/s | 25.50 GPixel/s |

| Texture Rate | 157.1 GTexel/s | 102.0 GTexel/s |

| FP32 | 5.027 TFLOPS | 2.448 TFLOPS |

| FP16 | 10.05 TFLOPS (2:1) | None |

| TDP | 120 W | 122 W |

| Slot Width | Dual-slot | MXM Module |

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

| Bus Interface | PCIe 3.0 x16 | MXM-B (3.0) |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a | Portable Device Dependent |

| DirectX Support | 12 (12_1) | 12 (11_0) |

| Vulkan Support | 1.4 | 1.2.175 |

| Release Date | 2019-03-13 | 2013-05-10 |

| Predecessor | GeForce 10 | GeForce 600M |

| Successor | GeForce 20 | GeForce 800M |