NVIDIA GeForce GTX 780M vs NVIDIA GeForce GTX 960A Comparison

# Head-to-Head Benchmarks

The GeForce GTX 960A and GeForce GTX 780M belong to different generations and architectural families, yet their actual benchmark performance lands surprisingly close. In the single head-to-head benchmark available, the OpenCL compute test, the older GTX 780M posts a score of 12,769 against the GTX 960A’s 11,998. That works out to a 6% advantage for the GTX 780M. The margin is modest but real, indicating that the Kepler-based mobile flagship retains an edge in raw compute throughput over its Maxwell successor.

Contextualizing these scores against their respective peer groups sharpens the picture. The GTX 960A’s 11,998 OpenCL result sits just 0.3% above the NVIDIA GeForce GTX 1080’s 11,960 and 1.3% above the AMD Radeon RX 6500 XT’s 11,842. It also edges out the GTX 1660 by 2.7% and the RX 7800 XT by 3.2%. This is a striking outcome: a mobile MXM part from the GeForce 900A series keeps pace with desktop GPUs that carry far more recent branding and, presumably, more modern feature sets. The data shows the GTX 960A’s compute efficiency is strong enough to put it in the 51st percentile of all GPUs.

The GTX 780M, meanwhile, records a 12,769 OpenCL score that places it 0.3% ahead of the AMD Radeon Pro WX 3200 and the AMD FirePro W4300, both at 11,228 and 11,225 respectively. It also leads the NVIDIA RTX PRO 6000 Blackwell Max-Q and the RTX PRO 6000D Blackwell Max-Q, each scoring 11,088, by 1.6%. The GTX 780M’s 50th percentile ranking is one point lower than the GTX 960A’s 51st, despite the higher raw score. That suggests the GTX 780M faces slightly stiffer competition in the broader GPU landscape, even though it wins the head-to-head contest against the GTX 960A.

The delta between the two is consistent across other metrics. The GTX 780M also has additional benchmark data points: it scores 8,319 in Geekbench Metal and 12,696 in Geekbench Vulkan. These are not directly comparable to the GTX 960A, which only has the OpenCL entry, but they do indicate that the GTX 780M’s compute advantage is not confined to a single API. The OpenCL result is the only apples-to-apples comparison, and it favors the GTX 780M by 6%.

One way to frame the head-to-head is that the GTX 960A loses the only direct comparison, but loses narrowly. A 6% gap is well within the range of driver optimization, thermal state, or even measurement variance for mobile parts. The GTX 960A’s superior per-clock efficiency, a hallmark of Maxwell, partially compensates for its smaller silicon. The GTX 780M’s larger, more heavily configured chip ultimately wins on brute force.

# Where Each One Wins

The GTX 780M is the clear winner in compute-heavy workloads based on the available data. Its OpenCL score of 12,769 beats the GTX 960A’s 11,998 by 6%. This advantage extends to its other benchmark results: the Vulkan score of 12,696 nearly matches its OpenCL performance, and even the Metal score of 8,319, while lower, shows the chip can deliver across multiple APIs. For tasks that leverage raw FP32 throughput, texture fill, or memory bandwidth — all areas where the GTX 780M’s specification sheet shows substantial leads — this GPU is the stronger choice.

The GTX 960A, conversely, finds its strengths in efficiency and architectural maturity. It achieves 92% of the GTX 780M’s OpenCL score (11,998 versus 12,769) while drawing 75 W against the GTX 780M’s 122 W. That is a 39% power consumption reduction for a 6% performance deficit in compute. For thermally constrained mobile chassis, the GTX 960A is the more sensible option. Its Maxwell architecture also supports Vulkan 1.4, whereas the GTX 780M’s Kepler design tops out at Vulkan 1.2.175. Applications that require newer Vulkan features will run on the GTX 960A but not on the GTX 780M. Both GPUs support DirectX 12 (11_0) and OpenGL 4.6, so API-level feature parity exists at those levels.

For gaming, the data does not provide a direct frame-rate comparison. However, the GTX 780M’s higher pixel rate (25.50 GPixel/s versus 18.82 GPixel/s) and texture rate (102.0 GTexel/s versus 47.04 GTexel/s) suggest it will handle resolution and texture-heavy scenes better. The GTX 960A’s higher clock speeds — 1097 MHz base and 1176 MHz boost versus 771 MHz base and 797 MHz boost — help it close the gap in latency-sensitive scenarios, but the GTX 780M’s 128 texture mapping units and 32 ROPs dwarf the GTX 960A’s 40 TMUs and 16 ROPs. The GTX 780M also doubles the memory bus width to 256 bit and doubles the memory capacity to 4 GB, offering 160.0 GB/s of bandwidth against the GTX 960A’s 80.19 GB/s. In memory-bound workloads, the GTX 780M wins decisively.

The GTX 960A wins where power envelope and API modernity matter. It is the more future-proof part for software that adopts newer Vulkan extensions, and it fits into lower-power designs. The GTX 780M wins where raw compute, memory throughput, and pixel processing dominate. There is no single winner across all use cases; the choice depends on the workload priority.

# Architecture Differences

The two GPUs come from different architectural lineages. The GTX 960A uses the GM107 chip, built on NVIDIA’s Maxwell architecture and fabricated by TSMC on a 28 nm process. The GTX 780M uses the GK104 chip, built on the older Kepler architecture, also on TSMC’s 28 nm node. Both are end-of-life products, but they represent different design philosophies from NVIDIA.

The transistor counts tell a stark story. The GTX 780M packs 3,540 million transistors onto a 294 mm² die, while the GTX 960A contains 1,870 million transistors on a 148 mm² die. The GTX 780M’s die is nearly twice the size and holds nearly twice the transistors. Transistor density is almost identical: 12.6 million per mm² for the GTX 960A and 12.0 million per mm² for the GTX 780M. This indicates that the architectural differences are not due to process improvements — both use the same 28 nm node — but rather to how the chips are organized internally.

The GTX 780M’s Kepler design emphasizes parallel throughput. It features 1,536 shading units, 128 TMUs, and 32 ROPs. The GTX 960A’s Maxwell design, by contrast, uses 640 shading units, 40 TMUs, and 16 ROPs. Maxwell achieves higher per-clock efficiency, which is why the GTX 960A can run at 1097 MHz base and 1176 MHz boost, compared to the GTX 780M’s 771 MHz base and 797 MHz boost. The GTX 960A’s higher clocks partially offset its fewer execution units. Its FP32 throughput is 1.505 TFLOPS, while the GTX 780M reaches 2.448 TFLOPS — a 63% advantage for the Kepler part.

Memory architecture diverges as well. The GTX 960A has 2 GB of GDDR5 on a 128-bit bus, yielding 80.19 GB/s of bandwidth. The GTX 780M has 4 GB of GDDR5 on a 256-bit bus, yielding 160.0 GB/s. Both use the same effective memory speed of 5 Gbps, so the bandwidth difference comes entirely from the wider bus. The GTX 780M also doubles the frame buffer capacity, which matters for high-resolution textures or multi-monitor setups.

API support shows another generational split. Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 960A supports Vulkan 1.4, while the GTX 780M supports Vulkan 1.2.175. This is a meaningful difference for modern applications that rely on the latest Vulkan features. Neither GPU has ray tracing cores or tensor cores, so those features are absent from both.

Power consumption reflects the design differences. The GTX 960A is rated at 75 W, while the GTX 780M is rated at 122 W. Both use MXM-B (3.0) as the bus interface and come in MXM Module slot width, with no power connectors and display outputs that are portable-device dependent. The GTX 960A’s lower power draw is a direct consequence of its smaller chip and fewer active units.

Release timing also differs. The GTX 780M launched on 2013-05-10, while the GTX 960A arrived on 2015-03-12. The GTX 780M’s predecessor is the GeForce 600M and its successor is the GeForce 800M. The GTX 960A’s predecessor is the GeForce 800A, with no successor listed. This places the GTX 960A at the tail end of the 900A series, while the GTX 780M sits in the middle of the 700M generation.

# FAQ

Q: Which GPU has a higher OpenCL benchmark score?

A: The NVIDIA GeForce GTX 780M scores 12,769 in Geekbench OpenCL, which is 6% higher than the NVIDIA GeForce GTX 960A’s 11,998.

Q: How do the two GPUs compare in terms of power consumption?

A: The GTX 960A has a TDP of 75 W, while the GTX 780M has a TDP of 122 W. The GTX 960A consumes 47 W less.

Q: Do both GPUs support the same versions of DirectX and OpenGL?

A: Yes, both support DirectX 12 (11_0) and OpenGL 4.6. They differ in Vulkan support: the GTX 960A supports Vulkan 1.4, while the GTX 780M supports Vulkan 1.2.175.

Q: Which GPU has more memory bandwidth?

A: The GTX 780M has 160.0 GB/s of bandwidth from a 256-bit bus with 4 GB of GDDR5. The GTX 960A has 80.19 GB/s from a 128-bit bus with 2 GB of GDDR5.

Q: What are the core clock speeds of each GPU?

A: The GTX 960A runs at 1097 MHz base and 1176 MHz boost. The GTX 780M runs at 771 MHz base and 797 MHz boost.

Q: Which GPU has more shading units and texture mapping units?

A: The GTX 780M has 1,536 shading units and 128 TMUs. The GTX 960A has 640 shading units and 40 TMUs.

# Specification Differences

| Specification | NVIDIA GeForce GTX 960A | NVIDIA GeForce GTX 780M |

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

| Chip | GM107 | GK104 |

| Architecture | Maxwell | Kepler |

| Generation | GeForce 900A | GeForce 700M |

| Transistors | 1,870 million | 3,540 million |

| Die Size | 148 mm² | 294 mm² |

| Transistor Density | 12.6M / mm² | 12.0M / mm² |

| Base Clock | 1097 MHz | 771 MHz |

| Boost Clock | 1176 MHz | 797 MHz |

| Memory Size | 2 GB | 4 GB |

| Memory Bus Width | 128 bit | 256 bit |

| Memory Bandwidth | 80.19 GB/s | 160.0 GB/s |

| Shading Units | 640 | 1536 |

| TMUs | 40 | 128 |

| ROPs | 16 | 32 |

| Pixel Rate | 18.82 GPixel/s | 25.50 GPixel/s |

| Texture Rate | 47.04 GTexel/s | 102.0 GTexel/s |

| FP32 | 1.505 TFLOPS | 2.448 TFLOPS |

| TDP | 75 W | 122 W |

| Vulkan Version | 1.4 | 1.2.175 |

| Release Date | 2015-03-12 | 2013-05-10 |

| Predecessor | GeForce 800A | GeForce 600M |

| Successor | None | GeForce 800M |

| Geekbench OpenCL Score | 11998 | 12769 |

| Avg Benchmark Score | 11998 | 11261 |

| Percentile vs All GPUs | 51 | 50 |