NVIDIA GeForce GTX 765M vs NVIDIA GeForce MX230 Comparison

Where Each One Wins

The benchmark data splits cleanly between the two NVIDIA mobile GPUs, with the older GeForce GTX 765M taking both recorded victories. In OpenCL compute workloads, the GTX 765M posts a score of 7,176 against the MX230’s 5,739, a 20% margin that places it firmly ahead in raw parallel processing. The Vulkan results are closer, with the GTX 765M scoring 6,714 versus 6,414 for the MX230, a 4.5% edge that still favors the Kepler-based part.

The MX230’s strengths lie elsewhere: it draws only 10 W compared to the GTX 765M’s 75 W, making it the clear choice for thin-and-light notebooks where thermal and power budgets are tight. Its Pascal architecture supports Vulkan 1.4, while the GTX 765M tops out at Vulkan 1.2.175, so newer graphics API titles may run more efficiently on the MX230 despite its lower raw throughput. The MX230 also achieves a higher pixel rate of 24.50 GPixel/s versus 13.81 GPixel/s, indicating better per-clock fill-rate efficiency for certain 2D or light 3D tasks.

For gaming, the GTX 765M’s superior texture rate (55.23 GTexel/s versus 24.50 GTexel/s) and larger 128-bit memory bus provide a clear advantage in bandwidth-sensitive scenes. The MX230’s 64-bit bus and 48.06 GB/s bandwidth will bottleneck in modern titles, while the GTX 765M’s 64.13 GB/s, though modest by current standards, gives it more headroom. The GTX 765M also has 768 shading units versus 256, a 3x difference that directly translates to higher FP32 throughput (1,325.6 GFLOPS versus 783.9 GFLOPS).

Architecture Differences

The two GPUs represent distinct generations of NVIDIA mobile silicon. The MX230 uses the GP108 chip built on a 14 nm Samsung process, packing 1,800 million transistors into a 74 mm² die. This yields a transistor density of 24.3M per mm², reflecting the efficiency gains of a newer node. The GTX 765M uses the GK106 chip on a 28 nm TSMC process, with 2,540 million transistors spread across a much larger 221 mm² die, resulting in a density of only 11.5M per mm².

Architecturally, the MX230 is Pascal, while the GTX 765M is Kepler. This generational gap explains several key differences: the MX230 supports DirectX 12 (12_1) and Vulkan 1.4, whereas the GTX 765M offers DirectX 12 (11_0) and Vulkan 1.2.175. The Pascal part also has a higher base clock of 1,519 MHz and boost of 1,531 MHz, compared to the GTX 765M’s 797 MHz base and 863 MHz boost. Despite lower clocks, the GTX 765M compensates with far more execution resources: 768 shading units and 64 texture mapping units versus 256 and 16, respectively. Both have 16 ROPs.

Memory configurations differ in bus width: the MX230 uses a 64-bit interface, the GTX 765M a 128-bit interface. Both have 2 GB of GDDR5, but the GTX 765M’s memory runs at 1,002 MHz (4 Gbps effective) yielding 64.13 GB/s, while the MX230’s memory runs at 1,502 MHz (6 Gbps effective) but only achieves 48.06 GB/s due to the narrower bus. The MX230 supports FP16 at 12.25 GFLOPS (1:64 ratio), while the GTX 765M has no recorded FP16 capability. Power delivery also diverges: the MX230 is an IGP with no power connectors, while the GTX 765M is an MXM module, also without connectors, but with a 65 W higher TDP.

Head-to-Head Benchmarks

The database records two direct comparisons between these GPUs. In Geekbench OpenCL, the GTX 765M wins decisively with 7,176 points against the MX230’s 5,739, a delta of -20% from the MX230’s perspective. This aligns with the GTX 765M’s 1.7x higher FP32 throughput and 2.25x more texture units. The OpenCL test stresses compute shaders and memory bandwidth, both areas where the Kepler part’s wider memory bus and larger shader count matter.

The Vulkan benchmark is a tighter contest. The GTX 765M scores 6,714, just 4.5% above the MX230’s 6,414. The smaller gap suggests that the MX230’s newer architecture and higher clocks partially offset its hardware deficit in API-bound workloads. Pascal’s improved command processor and driver overhead handling likely narrow the gap, even though the GTX 765M retains the overall win.

In aggregate, the GTX 765M holds a higher average benchmark score of 5,501 across all recorded tests, while the MX230 averages 6,077. However, this average is skewed by the MX230’s lack of a Metal benchmark; the GTX 765M’s Metal score of 2,612 drags its average down. Looking only at shared tests (OpenCL and Vulkan), the GTX 765M wins both, but the margin in Vulkan is small enough that real-world differences in optimized games could be negligible. The MX230’s percentile rank among all GPUs is 35, versus 32 for the GTX 765M, indicating that despite losing the head-to-head, the MX230 sits slightly higher in the overall distribution.

FAQ

Q: Which GPU is faster in raw compute performance?

A: The GeForce GTX 765M wins both recorded benchmarks. In Geekbench OpenCL, it scores 7,176 versus 5,739 for the MX230 (a 20% lead). In Vulkan, it scores 6,714 versus 6,414 (a 4.5% lead).

Q: Does the MX230 have any advantages over the GTX 765M?

A: Yes. The MX230 has a higher pixel rate (24.50 GPixel/s versus 13.81 GPixel/s), a much lower TDP (10 W versus 75 W), and supports newer APIs including DirectX 12_1 and Vulkan 1.4. It also has a higher base clock (1,519 MHz versus 797 MHz).

Q: Why does the GTX 765M perform better despite being older?

A: The GTX 765M has 768 shading units versus 256, and 64 texture units versus 16. It also uses a 128-bit memory bus versus 64-bit, giving it 64.13 GB/s bandwidth versus 48.06 GB/s. These hardware resources outweigh the MX230’s clock speed and process node advantages.

Q: Are these GPUs suitable for modern gaming?

A: Both are end-of-life products. The GTX 765M’s higher texture rate and bandwidth make it more capable for gaming, but its DirectX 12 support is limited to 11_0. The MX230 supports DirectX 12_1 but has only 2 GB of VRAM on a 64-bit bus, which will limit texture quality and resolution.

Q: How do these GPUs compare to their nearest rivals?

A: The MX230’s average score of 6,077 is within 0.5% of the NVIDIA RTX A400 (6,078) and 0.5% ahead of the Quadro P2000 (6,049). The GTX 765M’s average of 5,501 is 0.1% behind the GeForce MX130 (5,508) and 0.6% ahead of the Quadro M4000 (5,467).

Q: Which GPU should I choose for a laptop?

A: For pure performance, the GTX 765M wins, but it requires a 75 W power budget and MXM form factor. The MX230’s 10 W TDP and IGP design make it suitable for ultraportables, where battery life and thermals matter more than raw FPS.

Specification Differences

| Specification | NVIDIA GeForce MX230 | NVIDIA GeForce GTX 765M |

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

| Chip | GP108 | GK106 |

| Architecture | Pascal | Kepler |

| Process node | 14 nm (Samsung) | 28 nm (TSMC) |

| Transistors | 1,800 million | 2,540 million |

| Die size | 74 mm² | 221 mm² |

| Transistor density | 24.3M / mm² | 11.5M / mm² |

| Base clock | 1519 MHz | 797 MHz |

| Boost clock | 1531 MHz | 863 MHz |

| Memory clock | 1502 MHz (6 Gbps effective) | 1002 MHz (4 Gbps effective) |

| Memory bus width | 64 bit | 128 bit |

| Memory bandwidth | 48.06 GB/s | 64.13 GB/s |

| Shading units | 256 | 768 |

| Texture mapping units | 16 | 64 |

| Pixel rate | 24.50 GPixel/s | 13.81 GPixel/s |

| Texture rate | 24.50 GTexel/s | 55.23 GTexel/s |

| FP32 performance | 783.9 GFLOPS | 1,325.6 GFLOPS |

| FP16 performance | 12.25 GFLOPS (1:64) | None recorded |

| TDP | 10 W | 75 W |

| Slot width | IGP | MXM Module |

| Bus interface | PCIe 3.0 x4 | MXM-B (3.0) |

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

| Vulkan support | 1.4 | 1.2.175 |

| Release date | 2019-02-20 | 2013-05-29 |

| Predecessor | None recorded | GeForce 600M |

| Successor | None recorded | GeForce 800M |

| Geekbench OpenCL | 5,739 | 7,176 |

| Geekbench Vulkan | 6,414 | 6,714 |

| Geekbench Metal | Not recorded | 2,612 |

| Average benchmark score | 6,077 | 5,501 |

| Percentile vs all GPUs | 35 | 32 |

The GTX 765M also supports OpenGL 4.6, matching the MX230. Both have 2 GB GDDR5 memory and 16 ROPs. The MX230’s memory runs at a higher effective speed (6 Gbps versus 4 Gbps), but the GTX 765M’s wider bus delivers more bandwidth. Neither GPU has ray tracing or tensor cores, and both are marked as end-of-life in the database.