NVIDIA GeForce GTX 680M vs NVIDIA GeForce GTX 770M Comparison

Where Each One Wins

The benchmark data splits cleanly between the two mobile Kepler parts, but not in the way the generation numbers might suggest. The NVIDIA GeForce GTX 680M wins both recorded head-to-head tests, making it the stronger performer despite being the older product. Its wins come in Geekbench Metal and Geekbench OpenCL, with the Metal gap being substantial at 39.6% and the OpenCL gap more modest at 7.9%.

The GTX 770M, despite its newer generation label, does not claim a single benchmark victory in the recorded data. This is unusual because the 770M belongs to the GeForce 700M family, which typically represents a refinement over the prior generation. However, the database shows the 680M holding a clear edge in both compute workloads.

Looking at average benchmark scores reinforces this split. The 680M posts an average score of 7023, placing it in the 39th percentile among all GPUs. The 770M averages 6000, sitting at the 35th percentile. That 17% gap in average performance mirrors the OpenCL result more closely than the Metal result, suggesting the 680M's advantage grows in certain API workloads but remains consistent across the board.

The use-case split is therefore not about workloads where one card wins and the other takes the rest. Instead, the 680M simply outperforms the 770M everywhere measured. The real distinction lies in efficiency: the 770M draws 75 W against the 680M's 100 W, making it the choice for thermally constrained laptops where the performance deficit is acceptable.

Architecture Differences

Both GPUs share the Kepler architecture and are fabricated by TSMC on a 28 nm process, but they are fundamentally different chips. The 680M uses the GK104 chip, a larger design with 3,540 million transistors on a 294 mm² die. The 770M uses the GK106 chip, which is smaller at 2,540 million transistors and 221 mm². Transistor density is similar, 12.0M per mm² for the 680M versus 11.5M per mm² for the 770M, but the 680M packs considerably more hardware.

The compute configuration differs sharply. The 680M carries 1344 shading units, 112 texture mapping units, and 32 raster operation pipelines. The 770M has 960 shading units, 80 TMUs, and 24 ROPs. That means the 680M has 40% more shaders, 40% more texture units, and 33% more ROPs, which explains its consistent performance lead.

Memory subsystems also diverge. The 680M uses a 256 bit bus with 4 GB of GDDR5 and reaches 115.2 GB/s of bandwidth. The 770M narrows to a 192 bit bus with 3 GB of GDDR5 and 96.19 GB/s. Effective memory speed is higher on the 770M at 4 Gbps versus 3.6 Gbps on the 680M, but the wider bus on the older card still delivers more total bandwidth.

Clock behavior is interesting. The 680M has a higher base clock at 719 MHz versus 706 MHz, but the 770M boosts higher at 797 MHz against 758 MHz. The 770M's boost advantage does not overcome the 680M's raw hardware advantage. Both cards use the MXM-B (3.0) interface, draw power through no external connectors, and output to portable device dependent displays. Their API support is identical: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175.

Head-to-Head Benchmarks

The Geekbench Metal test shows the largest divergence. The 680M scores 4815 against 3448 for the 770M, a 39.6% advantage. This is a decisive margin that indicates the 680M's larger shader count and wider memory bus translate directly into Metal compute performance. For users running Metal-accelerated applications, the 680M is clearly the superior choice.

The Geekbench OpenCL test is closer but still favors the 680M. It scores 9230 versus 8552, a 7.9% lead. This narrower gap suggests OpenCL workloads are less sensitive to the 680M's extra hardware, or that the 770M's higher boost clock helps it partially close the distance. Still, the 680M wins outright.

The nearest rival data provides additional context for each card. The 680M sits just below the NVIDIA T600 (7035 average, 0.2% behind) and just above the AMD Radeon R5 M240 (6975, 0.7% ahead). It also trails the NVIDIA GeForce GTX 970 (7157) by 1.9% and leads the GeForce GTX 675M (6946) by 1.1%. These are tight margins, placing the 680M in a crowded mid-range cluster.

The 770M's nearest rivals tell a different story. It is essentially tied with the AMD Radeon RX 6400 (6001, 0% delta), the NVIDIA RTX PRO 6000 Blackwell Server (5996, 0.1% ahead), the AMD FirePro W4100 (5987, 0.2% ahead), and the NVIDIA Quadro K4000M (5986, 0.2% ahead). The 770M is the reference point in its group, neither clearly ahead nor behind its closest competitors.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 680M averages 7023 points, while the GTX 770M averages 6000 points. The 680M outperforms the 770M by roughly 17% on average.

Q: How large is the performance gap in the Metal benchmark?

A: The 680M scores 4815 in Geekbench Metal, which is 39.6% higher than the 770M's 3448. This is the largest single-test margin between the two cards.

Q: Does the newer GTX 770M have any architectural advantages?

A: The 770M has a higher boost clock (797 MHz versus 758 MHz) and faster effective memory (4 Gbps versus 3.6 Gbps). However, it has fewer shading units, TMUs, ROPs, and a narrower memory bus, so these advantages do not translate into higher benchmark scores.

Q: What is the power consumption difference between the two cards?

A: The GTX 680M has a TDP of 100 W, while the GTX 770M has a TDP of 75 W. The 770M is the more power-efficient option, which matters for laptop thermals and battery life.

Q: How do the memory configurations compare?

A: The 680M offers 4 GB of GDDR5 on a 256 bit bus with 115.2 GB/s bandwidth. The 770M offers 3 GB of GDDR5 on a 192 bit bus with 96.19 GB/s bandwidth. The 680M has both more capacity and more bandwidth.

Q: Which card sits higher in the overall GPU percentile ranking?

A: The 680M is in the 39th percentile of all GPUs, while the 770M is in the 35th percentile. The 680M ranks higher despite being the older product.

Specification Differences

| Specification | NVIDIA GeForce GTX 680M | NVIDIA GeForce GTX 770M |

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

| Chip | GK104 | GK106 |

| Generation | GeForce 600M | GeForce 700M |

| Transistors | 3,540 million | 2,540 million |

| Die Size | 294 mm² | 221 mm² |

| Transistor Density | 12.0M / mm² | 11.5M / mm² |

| Base Clock | 719 MHz | 706 MHz |

| Boost Clock | 758 MHz | 797 MHz |

| Memory Clock | 900 MHz (3.6 Gbps effective) | 1002 MHz (4 Gbps effective) |

| Memory Size | 4 GB | 3 GB |

| Memory Bus Width | 256 bit | 192 bit |

| Memory Bandwidth | 115.2 GB/s | 96.19 GB/s |

| Shading Units | 1344 | 960 |

| TMUs | 112 | 80 |

| ROPs | 32 | 24 |

| Pixel Rate | 21.22 GPixel/s | 15.94 GPixel/s |

| Texture Rate | 84.90 GTexel/s | 63.76 GTexel/s |

| FP32 Performance | 2.038 TFLOPS | 1.530 TFLOPS |

| TDP | 100 W | 75 W |

| Release Date | 2012-06-03 | 2013-05-29 |

| Predecessor | GeForce 500M | GeForce 600M |

| Successor | GeForce 700M | GeForce 800M |

The Verdict

The data points to a clear performance hierarchy: the GTX 680M is the faster card in every recorded benchmark. Its 39.6% lead in Metal and 7.9% lead in OpenCL establish it as the stronger compute performer, and its higher average score of 7023 versus 6000 confirms this across the board.

The GTX 770M's case rests entirely on efficiency. It draws 75 W compared to the 680M's 100 W, a 25% reduction in power consumption. For thin-and-light laptops where thermal limits are strict, that efficiency advantage may justify the performance loss. The 770M also boosts to a higher clock (797 MHz versus 758 MHz) and runs faster memory (4 Gbps versus 3.6 Gbps), which narrows the gap in some workloads.

Users who need maximum compute throughput should choose the GTX 680M. Its larger shader array, wider memory bus, and higher FP32 throughput (2.038 TFLOPS versus 1.530 TFLOPS) make it the better option for GPU-accelerated tasks. Users who prioritize battery life and cooler operation should choose the GTX 770M, accepting its lower benchmark scores in exchange for substantially lower power draw.

The generation numbering is misleading here. The 700M series did not deliver a performance uplift over the 600M series in this comparison. The 680M remains the stronger mobile GPU, and the 770M's only meaningful advantage is its efficiency profile. Both cards are end-of-life, so the decision rests on whether raw performance or power efficiency matters more in the target laptop chassis.