NVIDIA GeForce 840M vs NVIDIA GeForce 940MX Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 840M leads with an average score of 5322, while the NVIDIA GeForce 940MX trails at 4844. That is a 9.9% gap in favor of the older chip.

Q: How do the two compare in OpenCL performance?

A: The 840M wins decisively in the Geekbench OpenCL test, scoring 5764 against the 940MX's 4939, a 16.7% advantage.

Q: Is the 940MX more efficient despite losing benchmarks?

A: Yes. The 940MX has a TDP of 23 W compared to the 840M's 33 W, meaning it delivers its lower performance with roughly 30% less power draw.

Q: What memory configuration does each GPU use?

A: Both have 2 GB of memory and a 64-bit bus, but the 840M uses DDR3 at 16.02 GB/s bandwidth, while the 940MX uses GDDR5 at 40.10 GB/s bandwidth.

Q: Which GPU has more shading units?

A: The 940MX has 512 shading units versus the 840M's 384, a 33% higher count, yet it still loses in raw compute benchmarks.

Q: What is the release timeline for these two chips?

A: The 840M launched on 2014-03-11 under the GeForce 800M generation, while the 940MX arrived on 2016-06-27 under the GeForce 900M generation.

Architecture Differences

The two GPUs share the Maxwell architecture and are both fabricated by TSMC on a 28 nm process, but the physical implementations diverge significantly. The 840M uses the GM108S chip with 1,020 million transistors on a 77 mm² die, giving a transistor density of 13.2M per mm². The 940MX steps up to the GM107 chip, which packs 1,870 million transistors onto a 148 mm² die, yielding a density of 12.6M per mm². The 940MX therefore has roughly 83% more transistors and a die nearly twice the size, but the density is slightly lower.

The compute resources tell a similar story. The 940MX carries 512 shading units and 32 texture mapping units, while the 840M has 384 shading units and 16 TMUs. Both have 8 ROPs. The 940MX's larger shader count does not translate into higher clock speeds: its base clock is 795 MHz with a boost of 861 MHz, whereas the 840M runs at 1029 MHz base and 1124 MHz boost. This clock deficit explains much of the benchmark outcome.

Memory architecture is another clear split. The 840M pairs its 2 GB DDR3 memory with a 64-bit bus, producing 16.02 GB/s of bandwidth. The 940MX uses 2 GB of GDDR5 on the same 64-bit bus, more than doubling bandwidth to 40.10 GB/s. The memory clock also differs: the 840M runs at 1001 MHz (2 Gbps effective), while the 940MX runs at 1253 MHz (5 Gbps effective).

The 940MX is built for a different physical form factor. Its slot width is listed as MXM Module, whereas the 840M is an IGP (integrated graphics processor) solution. Both use PCIe 3.0 x8 and have no power connectors, but the 940MX's lower TDP of 23 W versus 33 W suggests it is designed for thinner, more power-conscious laptops. The API support is identical: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. Both are end-of-life products; the 840M's predecessor is GeForce 700M with successor GeForce 900M, while the 940MX's predecessor is GeForce 800M with successor GeForce 10 Mobile.

Head-to-Head Benchmarks

The recorded data shows a clean sweep for the 840M in the two available tests. In Geekbench OpenCL, the 840M scores 5764 against the 940MX's 4939, a delta of 16.7%. This is a substantial margin for a chip that is two years older. The 840M's higher boost clock of 1124 MHz, compared to 861 MHz on the 940MX, appears to outweigh the 940MX's extra 128 shading units and doubled TMU count.

In Geekbench Vulkan, the gap narrows considerably. The 840M scores 4880, and the 940MX scores 4749, a delta of just 2.8%. The 940MX's GDDR5 memory and higher bandwidth may help it close the distance in this API, but it still does not overtake the 840M. The Vulkan result is close enough that driver and workload variability could matter, but the recorded data gives the win to the 840M.

The 840M's average benchmark score of 5322 places it in the 31st percentile of all GPUs in the database. The 940MX's average of 4844 places it in the 28th percentile. The 840M's nearest rivals include the GeForce 930A (5317, 0.1% slower), the GTX 980M (5308, 0.3% slower), the Radeon R7 M445 (5358, 0.7% faster), and the GeForce 940M (5284, 0.7% slower). The 940MX's nearest rivals include the GTX 560M (4855, 0.2% faster), the Radeon R6 M255DX (4867, 0.5% faster), the GTS 450 (4893, 1% faster), and the RTX 3080 12 GB (4791, 1.1% slower). These lists underscore that the 840M sits in a slightly higher performance tier, while the 940MX rubs shoulders with older desktop parts.

The wins tally is decisive: the 840M takes 2 wins and the 940MX takes 0. However, the narrow Vulkan margin suggests that the 940MX is competitive in certain workloads, and its architectural advantages in memory and shader count are real even if they do not show up in these aggregate scores.

Specification Differences

| Specification | NVIDIA GeForce 840M | NVIDIA GeForce 940MX |

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

| Chip | GM108S | GM107 |

| Generation | GeForce 800M | GeForce 900M |

| Transistors | 1,020 million | 1,870 million |

| Die Size | 77 mm² | 148 mm² |

| Transistor Density | 13.2M / mm² | 12.6M / mm² |

| Base Clock | 1029 MHz | 795 MHz |

| Boost Clock | 1124 MHz | 861 MHz |

| Memory Clock | 1001 MHz (2 Gbps effective) | 1253 MHz (5 Gbps effective) |

| Memory Type | DDR3 | GDDR5 |

| Memory Bandwidth | 16.02 GB/s | 40.10 GB/s |

| Shading Units | 384 | 512 |

| TMUs | 16 | 32 |

| Pixel Rate | 8.992 GPixel/s | 6.888 GPixel/s |

| Texture Rate | 17.98 GTexel/s | 27.55 GTexel/s |

| FP32 | 863.2 GFLOPS | 881.7 GFLOPS |

| TDP | 33 W | 23 W |

| Slot Width | IGP | MXM Module |

| Release Date | 2014-03-11 | 2016-06-27 |

| Predecessor | GeForce 700M | GeForce 800M |

| Successor | GeForce 900M | GeForce 10 Mobile |

Shared specifications include the 28 nm process node, TSMC foundry, 2 GB memory size, 64-bit bus width, 8 ROPs, PCIe 3.0 x8 interface, no power connectors, portable-device-dependent display outputs, and identical API lists (DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.4). Neither GPU has a listed launch MSRP.

Where Each One Wins

The 840M wins in raw compute benchmarks. Its OpenCL score of 5764 is 16.7% higher than the 940MX, and its Vulkan score of 4880 is 2.8% higher. The 840M also wins on pixel rate (8.992 GPixel/s versus 6.888 GPixel/s) and has a higher average benchmark score (5322 versus 4844) and a better percentile rank (31st versus 28th). For users who prioritize raw throughput in compute-heavy tasks, the 840M is the stronger choice.

The 940MX wins on efficiency and memory bandwidth. Its TDP of 23 W is 10 W lower than the 840M's 33 W, which is a meaningful advantage for thin-and-light laptops where thermals and battery life matter. Its GDDR5 memory provides 40.10 GB/s of bandwidth versus 16.02 GB/s, a 2.5x advantage that could help in memory-bound scenarios. The 940MX also has more shading units (512 versus 384) and more TMUs (32 versus 16), which may benefit certain workloads that scale with these resources. Its texture rate of 27.55 GTexel/s is 53% higher than the 840M's 17.98 GTexel/s, and its FP32 output of 881.7 GFLOPS slightly exceeds the 840M's 863.2 GFLOPS.

The 940MX's nearest rivals include the GTX 560M and GTS 450, both desktop-era parts, indicating it competes in a lower performance tier. The 840M's rivals include the GTX 980M and the GeForce 940M, showing it punches above its generation. For a laptop buyer who values battery life and quiet operation over peak performance, the 940MX is the more sensible engineering choice. For someone who wants maximum compute output from an older, higher-power part, the 840M delivers.

The Verdict

The data is clear: the NVIDIA GeForce 840M outperforms the NVIDIA GeForce 940MX in both recorded benchmarks. The 16.7% OpenCL lead and the 2.8% Vulkan lead are consistent, and the 840M's higher average score (5322 versus 4844) and higher percentile rank (31st versus 28th) reinforce the conclusion. The 840M wins 2 out of 2 head-to-head tests, and its higher base and boost clocks appear to be the decisive factor.

However, the 940MX is not without merit. Its 23 W TDP makes it a far more power-efficient part, and its GDDR5 memory with 40.10 GB/s bandwidth is a major upgrade over the 840M's DDR3 at 16.02 GB/s. In workloads that are memory-bound rather than compute-bound, the 940MX could plausibly close the gap or even overtake the 840M, though the recorded data does not show that result. Its larger shader count and TMU count also give it headroom in texture-heavy scenarios.

For a user choosing between these two in a laptop, the decision comes down to priorities. If the goal is maximum benchmark performance, the 840M is the pick, despite being from an older generation. If the goal is a cooler, quieter, and more efficient system with faster memory, the 940MX is the better fit. The 840M's release date of 2014-03-11 versus the 940MX's 2016-06-27 also means the 940MX is a later product, but later does not mean faster in this case. The 840M's average score of 5322 places it 9.9% ahead of the 940MX's 4844, and that is the headline number from the database.