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

Head-to-Head Benchmarks

The recorded database contains two direct comparison tests between the NVIDIA GeForce GTX 780M and the NVIDIA GeForce GTX 960M: Geekbench OpenCL and Geekbench Vulkan. The GTX 780M wins both tests, giving it a clean 2-0 record in head-to-head matchups.

In Geekbench OpenCL, the GTX 780M scores 12,769 against the GTX 960M's 11,045. That is a 15.6% advantage for the older Kepler-based part. The gap is substantial, but the OpenCL result still shows the GTX 960M operating in the same performance tier, only about one-seventh behind.

The Vulkan test tells a different story. The GTX 780M scores 12,696, while the GTX 960M manages only 8,245. That is a 54% delta, a commanding lead for the GTX 780M. In Vulkan workloads, the GTX 960M falls far behind, suggesting that the Maxwell architecture's lower shader count and narrower memory bus are severe limitations in this API.

Looking at the average benchmark score across all recorded tests, the GTX 780M averages 11,261, placing it in the 50th percentile of all GPUs in the database. The GTX 960M averages 9,645, which lands at the 46th percentile. The difference in average scores is 1,616 points, roughly 16.8% in favor of the GTX 780M.

The nearest rivals for the GTX 780M illustrate its positioning. The AMD Radeon Pro WX 3200 averages 11,228, just 0.3% behind. The AMD FirePro W4300 averages 11,225, also 0.3% behind. The NVIDIA RTX PRO 6000 Blackwell Max-Q and NVIDIA RTX PRO 6000D Blackwell Max-Q both average 11,088, sitting 1.6% lower. These deltas show the GTX 780M clustered with a group of workstation-class GPUs, all within a narrow 1.6% band.

For the GTX 960M, the rival set is similarly tight. The NVIDIA Quadro K5000 averages 9,637, just 0.1% behind. The AMD Radeon Pro WX 2100 averages 9,653, a 0.1% edge over the GTX 960M. The NVIDIA Quadro P4000 averages 9,665, 0.2% ahead. The NVIDIA Tesla C2070 averages 9,716, 0.7% ahead. The GTX 960M sits in the middle of this group, with the closest rival on either side separated by less than 1%.

The head-to-head data makes the hierarchy clear: the GTX 780M is the faster part in both recorded API tests, with a particularly large margin in Vulkan. The GTX 960M, despite being a later generation, cannot close the gap in these specific workloads.

FAQ

Q: Which GPU wins in Geekbench Vulkan, and by how much?

A: The NVIDIA GeForce GTX 780M wins with a score of 12,696 against 8,245 for the GTX 960M, a 54% advantage.

Q: What is the average benchmark score difference between the two cards?

A: The GTX 780M averages 11,261 across all recorded tests, while the GTX 960M averages 9,645, a difference of 1,616 points.

Q: How does the GTX 960M compare to its nearest rival, the NVIDIA Quadro P4000?

A: The Quadro P4000 averages 9,665, which is 0.2% higher than the GTX 960M's 9,645 average.

Q: Is the GTX 780M closer to workstation GPUs or gaming GPUs in performance?

A: The GTX 780M's average score of 11,261 places it within 0.3% of the AMD Radeon Pro WX 3200 and AMD FirePro W4300, and within 1.6% of the NVIDIA RTX PRO 6000 Blackwell Max-Q, indicating strong alignment with workstation-class parts.

Q: How many benchmark wins does each GPU have in the head-to-head comparison?

A: The GTX 780M wins both recorded tests (Geekbench OpenCL and Geekbench Vulkan), giving it 2 wins. The GTX 960M has 0 wins.

Q: Where does the GTX 960M rank in the overall GPU percentile?

A: The GTX 960M sits in the 46th percentile of all GPUs, while the GTX 780M sits in the 50th percentile.

Architecture Differences

The GTX 780M is built on the Kepler architecture using the GK104 chip, fabricated on a 28 nm process at TSMC. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0 million per square millimeter. The GTX 960M uses the Maxwell architecture with the GM107 chip, also on a 28 nm TSMC process, but with 1,870 million transistors on a much smaller 148 mm² die, giving a slightly higher density of 12.6 million per square millimeter.

The shader configuration differs dramatically. The GTX 780M has 1,536 shading units, 128 texture mapping units, and 32 raster output units. The GTX 960M has only 640 shading units, 40 TMUs, and 16 ROPs. That is a 58% reduction in shader count, a 69% reduction in TMUs, and a 50% reduction in ROPs for the Maxwell part.

The memory subsystems are also fundamentally different. Both cards use 4 GB of GDDR5, but the GTX 780M runs a 256-bit bus with 160.0 GB/s of bandwidth, while the GTX 960M uses a 128-bit bus with 80.19 GB/s. The GTX 780M has exactly double the memory bandwidth. The memory clock is nearly identical: 1250 MHz (5 Gbps effective) on the GTX 780M versus 1253 MHz (5 Gbps effective) on the GTX 960M, so the bandwidth gap comes entirely from the bus width.

Clock speeds favor the GTX 960M. Its base clock is 1097 MHz with a boost of 1176 MHz, while the GTX 780M runs at 771 MHz base and 797 MHz boost. The GTX 960M's clocks are roughly 42% higher at base and 48% higher at boost. Despite this, the GTX 780M's larger shader array and wider memory bus deliver higher throughput in the recorded benchmarks.

Pixel and texture rates reflect the hardware differences. The GTX 780M achieves 25.50 GPixel/s and 102.0 GTexel/s. The GTX 960M achieves 18.82 GPixel/s and 47.04 GTexel/s. The GTX 780M is 35% faster in pixel rate and 117% faster in texture rate.

FP32 performance follows the same pattern: the GTX 780M delivers 2.448 TFLOPS, while the GTX 960M delivers 1.505 TFLOPS, a 63% advantage for the Kepler part.

The GTX 780M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The GTX 960M supports the same DirectX and OpenGL versions but a newer Vulkan 1.4. Neither card has ray tracing cores or tensor cores.

Specification Differences

The two GPUs differ in nearly every compute-relevant specification. The GTX 780M uses the GK104 chip on the Kepler architecture, while the GTX 960M uses the GM107 chip on Maxwell. Both use 28 nm TSMC fabrication, but the GTX 780M has 3,540 million transistors versus 1,870 million for the GTX 960M. The die size is 294 mm² versus 148 mm².

Clock speeds are higher on the GTX 960M: base 1097 MHz versus 771 MHz, boost 1176 MHz versus 797 MHz. Memory clocks are nearly equal, with the GTX 960M at 1253 MHz and the GTX 780M at 1250 MHz.

Memory bus width is 256-bit on the GTX 780M and 128-bit on the GTX 960M. Bandwidth is 160.0 GB/s versus 80.19 GB/s. Both have 4 GB of GDDR5.

Shader units: 1,536 versus 640. TMUs: 128 versus 40. ROPs: 32 versus 16. Pixel rate: 25.50 GPixel/s versus 18.82 GPixel/s. Texture rate: 102.0 GTexel/s versus 47.04 GTexel/s. FP32: 2.448 TFLOPS versus 1.505 TFLOPS.

TDP is 122 W for the GTX 780M and 75 W for the GTX 960M. Both use MXM Module slot width, have no power connectors, and use an MXM-B (3.0) bus interface. Display outputs are portable device dependent for both.

The GTX 780M belongs to the GeForce 700M generation with a release date of May 10, 2013, a predecessor of GeForce 600M and successor of GeForce 800M. The GTX 960M belongs to the GeForce 900M generation, released March 12, 2015, with predecessor GeForce 800M and successor GeForce 10 Mobile. Production status is end-of-life for both.

Vulkan support differs: the GTX 780M supports Vulkan 1.2.175, while the GTX 960M supports Vulkan 1.4.

The Verdict

The benchmark data points to a clear winner: the NVIDIA GeForce GTX 780M outperforms the GTX 960M in both recorded head-to-head tests. The OpenCL margin of 15.6% and the Vulkan margin of 54% are decisive. The GTX 780M also holds a higher average score (11,261 versus 9,645) and a higher percentile rank (50th versus 46th).

Users who prioritize compute performance in OpenCL or Vulkan workloads should select the GTX 780M. Its 256-bit memory bus and 2.448 TFLOPS of FP32 throughput provide a substantial edge over the GTX 960M's 128-bit bus and 1.505 TFLOPS. The GTX 780M's nearest rivals are all workstation-class GPUs within 1.6%, which speaks to its sustained compute relevance.

The GTX 960M, however, offers a lower TDP of 75 W versus 122 W, a 47 W reduction. For systems where power draw is a constraint, the GTX 960M is the more efficient choice. Its higher clock speeds (1097 MHz base versus 771 MHz) do not compensate for the reduced shader count in these tests, but they do indicate a more efficient per-core design.

The GTX 960M also supports Vulkan 1.4, a newer API version than the GTX 780M's Vulkan 1.2.175. This could matter for software that relies on newer Vulkan features, though the recorded Vulkan benchmark shows the GTX 780M far ahead in raw performance.

There is no winner on price, as neither card has a recorded launch MSRP in the database. The choice comes down to performance versus power efficiency and API version support.

Where Each One Wins

The GTX 780M wins in raw compute performance. Its OpenCL score of 12,769 against 11,045 and Vulkan score of 12,696 against 8,245 make it the stronger choice for any workload that stresses the GPU's shader array or memory bandwidth. The 160.0 GB/s bandwidth is double the GTX 960M's 80.19 GB/s, which directly benefits memory-intensive tasks. The GTX 780M also wins in texture throughput (102.0 GTexel/s versus 47.04 GTexel/s) and pixel throughput (25.50 GPixel/s versus 18.82 GPixel/s).

The GTX 960M wins in power efficiency. Its 75 W TDP is 38.5% lower than the GTX 780M's 122 W. For laptops or compact MXM systems with limited cooling or battery budgets, the GTX 960M draws significantly less power while still delivering serviceable OpenCL performance (within 15.6%). The GTX 960M also wins on clock speed, with a boost of 1176 MHz versus 797 MHz, and on Vulkan API version support (1.4 versus 1.2.175).

In percentile terms, the GTX 780M sits at the 50th percentile of all GPUs, while the GTX 960M sits at the 46th. That four-percentile gap reflects the average score difference of 1,616 points. The GTX 780M's rival cluster includes the AMD Radeon Pro WX 3200 and AMD FirePro W4300 at 0.3% behind, while the GTX 960M's cluster includes the AMD Radeon Pro WX 2100 at 0.1% ahead, making the GTX 960M the lower performer in its own peer group.

The practical split is straightforward: the GTX 780M for compute-heavy tasks and the GTX 960M for power-constrained environments. Neither card supports ray tracing or tensor cores, so those workloads are not applicable. Both use MXM modules and lack power connectors, so system integration is similar. The deciding factors are the 47 W power difference and the 54% Vulkan performance gap, which together define the use cases for each card.