GPU Comparison

Head-to-Head Benchmarks

The recorded data shows a decisive performance gap between the NVIDIA RTX A400 and the NVIDIA GeForce GTX 765M in the two shared benchmark tests. The RTX A400 leads by an enormous margin in both OpenCL and Vulkan workloads, making this comparison less about close competition and more about generational distance.

In the Geekbench OpenCL test, the RTX A400 scores 22,844 points against the GTX 765M’s 7,176 points. That is a 218.3% advantage for the RTX A400, meaning the older mobile GPU delivers less than a third of the raw compute throughput. The Geekbench Vulkan test tells a similar story: the RTX A400 posts 22,237 points, while the GTX 765M manages only 6,714 points. The delta here is even steeper at 231.2%. Across the two head-to-head tests, the RTX A400 wins both, giving it a clean 2-0 record with zero wins for the GTX 765M.

These are not marginal improvements. The RTX A400 more than triples the GTX 765M’s OpenCL result and more than triples its Vulkan result. Even accounting for the fact that the GTX 765M is a mobile part from 2013, the scale of the gap is striking. The average benchmark score for the RTX A400 sits at 6,078, while the GTX 765M averages 5,501, a difference of roughly 10.5%. That average, however, is pulled by different test sets: the RTX A400 has nine recorded benchmarks, while the GTX 765M only has three. For direct comparison, the shared tests are the more reliable signal.

The RTX A400 also places in the 35th percentile of all GPUs in the database, while the GTX 765M sits in the 32nd percentile. That three-point gap in percentile ranking confirms that the RTX A400 is not merely ahead of this specific rival, but also better positioned relative to the broader GPU landscape.

Where Each One Wins

The RTX A400 wins every workload category where both have data. In OpenCL, its 22,844 score reflects strong general-purpose compute performance, which matters for tasks like physics simulation, image processing, and other GPU-accelerated applications that do not rely on proprietary APIs. The Vulkan score of 22,237 reinforces this, showing that the RTX A400 handles modern cross-platform graphics APIs with ease.

The GTX 765M has no wins in the head-to-head data. Its strengths, if any, would have to be inferred from its architecture and feature set rather than from benchmark victories. It does support Metal via Geekbench, scoring 2,612, but the RTX A400 has no recorded Metal score to compare against, so this cannot be counted as a competitive advantage. The GTX 765M’s only other recorded benchmarks are the same OpenCL and Vulkan tests where it loses decisively.

For use-case planning, the data points to the RTX A400 as the choice for any workload leveraging OpenCL or Vulkan. That includes modern game engines, compute shaders, and GPU-accelerated productivity tools. The GTX 765M, by contrast, belongs to an era where those APIs were either less mature or less central to daily use. Its 2 GB memory capacity and 64.13 GB/s bandwidth will also bottleneck larger datasets, whereas the RTX A400’s 4 GB GDDR6 at 96.00 GB/s provides a more comfortable buffer.

Architecture Differences

The architectural gap between these two GPUs is generational. The RTX A400 uses the GA107 chip on an 8 nm Samsung process, while the GTX 765M uses the GK106 chip on a 28 nm TSMC process. The process shrink is significant: 8 nm versus 28 nm means the RTX A400 packs far more transistors into a smaller area. The RTX A400 has 8,700 million transistors on a 200 mm² die, giving a transistor density of 43.5 million per mm². The GTX 765M has 2,540 million transistors on a 221 mm² die, for a density of only 11.5 million per mm². That is nearly a fourfold density advantage for the newer part.

The memory subsystems differ sharply as well. The RTX A400 uses 4 GB of GDDR6 on a 64-bit bus, delivering 96.00 GB/s of bandwidth. The GTX 765M uses 2 GB of GDDR5 on a 128-bit bus, delivering only 64.13 GB/s. Despite having twice the bus width, the older GPU loses on bandwidth due to its slower memory clock. The RTX A400’s memory runs at 1500 MHz (12 Gbps effective), while the GTX 765M’s runs at 1002 MHz (4 Gbps effective).

Compute resources show a mixed picture. Both GPUs have 768 shading units and 16 ROPs, but the GTX 765M has 64 TMUs against the RTX A400’s 24. The RTX A400 compensates with far higher clocks: 1417 MHz base and 1762 MHz boost versus 797 MHz base and 863 MHz boost for the GTX 765M. The result is that the RTX A400 achieves 28.19 GPixel/s pixel rate and 42.29 GTexel/s texture rate, while the GTX 765M manages 13.81 GPixel/s and 55.23 GTexel/s. The RTX A400 wins on pixel throughput, but the GTX 765M actually has higher texture throughput due to its extra TMUs.

The RTX A400 also brings features the GTX 765M simply lacks. It has 6 RT cores and 24 tensor cores, enabling hardware-accelerated ray tracing and AI workloads. The GTX 765M has neither. The RTX A400 supports DirectX 12 Ultimate (12_2), while the GTX 765M only reaches DirectX 12 (11_0). Vulkan support is also newer on the RTX A400: version 1.4 versus 1.2.175. OpenGL support is identical at 4.6.

Power and form factor differences are notable. The RTX A400 has a 50 W TDP and is a single-slot card with no power connectors, drawing all its power from the PCIe slot. The GTX 765M has a 75 W TDP and uses an MXM module form factor, which means it was designed for laptops and small-form-factor systems. The RTX A400 is an active production card released in April 2024, while the GTX 765M is end-of-life and was released in May 2013.

The Verdict

The data points to a simple conclusion: the NVIDIA RTX A400 is the superior GPU in every measurable way. It wins both shared benchmarks by more than 200%, has a higher average benchmark score, sits in a higher percentile, and brings modern architectural features like RT cores, tensor cores, and DirectX 12 Ultimate support. Anyone choosing between these two for a new build or upgrade should pick the RTX A400 without hesitation.

That said, the GTX 765M is not without a niche. Its 64 TMUs give it higher texture fill rate, which could theoretically help in older texture-bound workloads. Its MXM form factor means it might be the only option in certain legacy laptops where the RTX A400 cannot physically fit. But those are edge cases. For any general-purpose compute, modern gaming, or professional workstation task, the RTX A400 is the clear winner.

The RTX A400’s 50 W TDP is also lower than the GTX 765M’s 75 W, meaning it delivers more performance while consuming less power. That efficiency gain is a direct result of the 8 nm process versus 28 nm. The RTX A400 also offers four mini-DisplayPort 1.4a outputs, while the GTX 765M’s outputs are portable-device dependent, which is a practical advantage for multi-monitor setups.

FAQ

Q: Which GPU is faster in OpenCL?

A: The NVIDIA RTX A400 scores 22,844 in Geekbench OpenCL, while the NVIDIA GeForce GTX 765M scores 7,176. That is a 218.3% advantage for the RTX A400.

Q: Does the GTX 765M win any benchmark against the RTX A400?

A: No. In the shared Geekbench OpenCL and Vulkan tests, the RTX A400 wins both. The head-to-head record is 2 wins for the RTX A400 and 0 for the GTX 765M.

Q: What is the memory capacity difference?

A: The RTX A400 has 4 GB of GDDR6, while the GTX 765M has 2 GB of GDDR5. The RTX A400 also has higher bandwidth at 96.00 GB/s versus 64.13 GB/s.

Q: Does the RTX A400 support ray tracing?

A: Yes, it has 6 RT cores. The GTX 765M has no RT cores and cannot do hardware-accelerated ray tracing.

Q: What is the TDP difference?

A: The RTX A400 has a 50 W TDP, while the GTX 765M has a 75 W TDP. The RTX A400 delivers more performance at lower power draw.

Q: Which GPU has better Vulkan support?

A: The RTX A400 supports Vulkan 1.4, while the GTX 765M supports Vulkan 1.2.175. The RTX A400 also scores 22,237 in Geekbench Vulkan versus 6,714 for the GTX 765M.

Specification Differences

| Specification | NVIDIA RTX A400 | NVIDIA GeForce GTX 765M |

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

| Architecture | Ampere | Kepler |

| Process Node | 8 nm (Samsung) | 28 nm (TSMC) |

| Transistors | 8,700 million | 2,540 million |

| Die Size | 200 mm² | 221 mm² |

| Transistor Density | 43.5M / mm² | 11.5M / mm² |

| Base Clock | 1417 MHz | 797 MHz |

| Boost Clock | 1762 MHz | 863 MHz |

| Memory Size | 4 GB | 2 GB |

| Memory Type | GDDR6 | GDDR5 |

| Memory Bus Width | 64 bit | 128 bit |

| Memory Bandwidth | 96.00 GB/s | 64.13 GB/s |

| Memory Clock | 1500 MHz (12 Gbps effective) | 1002 MHz (4 Gbps effective) |

| Shading Units | 768 | 768 |

| TMUs | 24 | 64 |

| ROPs | 16 | 16 |

| RT Cores | 6 | None |

| Tensor Cores | 24 | None |

| Pixel Rate | 28.19 GPixel/s | 13.81 GPixel/s |

| Texture Rate | 42.29 GTexel/s | 55.23 GTexel/s |

| FP32 Performance | 2.706 TFLOPS | 1,325.6 GFLOPS |

| FP16 Performance | 2.706 TFLOPS (1:1) | None |

| TDP | 50 W | 75 W |

| Slot Width | Single-slot | MXM Module |

| Power Connectors | None | None |

| Suggested PSU | 250 W | None |

| Bus Interface | PCIe 4.0 x8 | MXM-B (3.0) |

| Display Outputs | 4x mini-DisplayPort 1.4a | Portable Device Dependent |

| DirectX Support | 12 Ultimate (12_2) | 12 (11_0) |

| Vulkan Support | 1.4 | 1.2.175 |

| Production Status | Active | End-of-life |

| Release Date | 2024-04-15 | 2013-05-29 |

| Predecessor | Quadro Turing | GeForce 600M |

| Successor | Workstation Ada | GeForce 800M |