NVIDIA GeForce 940MX vs NVIDIA GeForce GTX 765M Comparison

# NVIDIA GeForce GTX 765M vs NVIDIA GeForce 940MX

The GTX 765M and the 940MX are two mobile GPUs from different NVIDIA generations, and the data shows a clear performance hierarchy. The GTX 765M, built on the Kepler architecture, is the older but significantly faster part. The 940MX, from the Maxwell family, is the more recent but slower option. Across the database's average benchmark scores, the GTX 765M scores 5501, while the 940MX scores 4844. That is a 13.6% advantage for the Kepler part. In the two specific head-to-head tests, the result is even more one-sided. The GTX 765M wins both, with a 45.3% lead in the OpenCL test and a 41.4% lead in the Vulkan test.

Head-to-Head Benchmarks

The recorded head-to-head data contains two direct comparisons. Both go to the GTX 765M.

In the geekbench_opencl test, the GTX 765M scores 7176, against 4939 for the 940MX, a delta of 45.3%. This is the largest win for the older part. The gap narrows slightly in the geekbench_vulkan test, where the GTX 765M scores 6714 versus the 940MX's 4749, a 41.4% difference. The 940MX never gets closer than 41% behind in these direct tests.

Looking at the wider database, the GTX 765M sits in the 32nd percentile of all GPUs, meaning roughly two-thirds of recorded parts are faster. The 940MX sits in the 28th percentile, a similar positioning. The two parts are therefore in the same performance tier relative to the entire database, but the GTX 765M is consistently ahead.

Where Each One Wins

The benchmark wins are all on the GTX 765M side. It wins both the OpenCL and Vulkan tests. There is no test in the data where the 940MX comes out ahead. If the workload is heavily compute-bound and can use OpenCL or Vulkan, the GTX 765M is the stronger choice by a significant margin, with a 40% plus advantage in both recorded APIs.

The 940MX's advantage is not in raw compute performance. Its edge lies elsewhere, namely in platform features and driver support. The data shows the 940MX supports DirectX 12 (11_0) and OpenGL 4.6, but the GTX 765M is not far behind with DirectX 12 (11_0) and OpenGL 4.6. Both support Vulkan 1.2 with the 940MX listing version 1.4. For a system that only needs the newer APIs and where the workload does not stress raw shading throughput, the 940MX is the more modern option. But for raw number crunching, the GTX 765M is the clear winner.

Architecture Differences

The two parts come from different design schools. The GTX 765M uses the GK106 chip on the 28 nm process at TSMC. It packs 2,540 million transistors on a 221 mm² die, for a density of 11.5 million transistors per square millimeter. The 940MX uses the GM107 chip, also on a 28 nm process at TSMC, with 1,870 million transistors on a die of 148 mm². That gives it a higher transistor density of 12.6M per mm².

The GTX 765M is the larger chip with more transistors, but the 940MX is the denser design. The memory subsystem shows two different approaches. The GTX 765M uses 2 GB of GDDR5 memory on a 128-bit bus, delivering a bandwidth of 64.13 GB/s. The 940MX also has 2 GB of GDDR5, but on a narrower 64-bit bus, which cuts bandwidth to 40.10 GB/s. For memory-heavy workloads, the GTX 765M's wider memory bus is a substantial advantage.

The shading config also favors the older part. The GTX 765M has 768 shading units and 64 texture mapping units. The 940MX comes with 512 shading units and 32 TMUs. Both parts have 16 ROPs, so pixel fill and raster operations are less distinct. The GTX 765M also has a higher pixel rate of 13.81 GPixel/s versus 6.888 GPixel/s for the 940MX, and a higher texture rate of 55.23 GTexel/s versus 27.55 GTexel/s. The GTX 765M is roughly double the speed in both fill and texturing throughput.

The other side of the ledger is power. The GTX 765M has a TDP of 75 W. The 940MX is a much more efficient part with a TDP of 23 W. This is a large gap. For a laptop, this matters. The GTX 765M requires more cooling and more power delivery, which in a portable chassis is a challenge. The 940MX will be easier to cool and allow thinner designs.

The GTX 765M features a GK106 chip with a boost clock of 863 MHz and a base of 797 MHz. Its memory runs at 1002 MHz. The 940MX clocks are lower: 861 MHz boost, 795 MHz base, and 1253 MHz memory. The 940MX clocks are lower across the board.

The GTX 765M is an older part, launched in May 2013, and is now End-of-life. The 940MX came later, in June 2016, and is also End-of-life. The GTX 765M is the successor of the GeForce 600M, and its successor line is the GeForce 10 Mobile. The 940MX's predecessor is the GeForce 800M and its successor is the GeForce 10 Mobile. Both parts use an MXM module: the GTX 765M uses MXM-B (3.0), and the 940MX uses the newer MXM Module. Both are modular, not soldered.

In terms of API support, the 940MX has a slight edge in DirectX (12 vs 12 (11_0) for the GTX). Both support OpenGL 4.6 and Vulkan 1.2. The 940MX's support for newer DirectX could be the deciding factor for applications that require it.

FAQ

Q: Which GPU is faster in synthetic benchmarks?

A: The NVIDIA GeForce GTX 765M. Its average benchmark score is 5501 versus 4844 for the 940MX. In the head-to-head tests, the GTX 765M leads by 45.3% (OpenCL) and 41.4% (Vulkan).

Q: Are both parts compatible with modern APIs?

A: Both support Vulkan 1.2 and OpenGL 4.6. The 940MX supports DirectX 12 (11_0), while the GTX 765M supports DirectX 12 (11_0). So both are fine for modern DirectX titles, but the 940MX has a slightly newer feature level listing.

Q: Which one has more memory?

A: Both have 2 GB of GDDR5 memory. The GTX 765M uses a 128-bit interface, the 940MX a 64-bit interface. The GTX 765M has a much higher bandwidth of 64.13 GB/s versus 40.10 GB/s.

Q: Which part has higher clock speeds?

A: The GTX 765M. It has a base clock of 797 MHz, a boost of 863 MHz, and a memory clock of 1002 MHz. The 940MX has a base of 795, boost of 861, and memory of 1253 MHz. The GTX 765M runs higher core clocks, but the 940MX has a higher memory clock.

Q: Are there differences in the die size or transistor count?

A: Yes. The GTX 765M has a die size of 254 mm² and 2,540 million transistors. The 940MX has a smaller 148 mm² die and 1,870 million transistors. The 940MX is the denser die.

Specification Differences

The two parts differ in several key recorded fields. The GTX 765M has a 128-bit memory bus, the 940MX has a 64-bit bus. Both have 2 GB of GDDR5. The GTX 765M has more shading units (768 vs 512) and more TMUs (64 vs 32). Both have 16 ROPs. The GTX 765M has higher pixel and texture rates, 13.81 GPixel/s and 55.23 GTexel/s versus 6.888 GPixel/s and 27.55 GTexel/s for the 940MX.

The process node is 28 nm for both, but the chips are different: GK106 (Kepler) for the GTX, GM107 (Maxwell) for the 940MX. The GTX 765M has a higher TDP of 75 W, the 940MX is rated at 23 W. The older part uses an MXM-B (3.0) socket, the newer one uses MXM. Both have no power connectors.

In API support, the 940MX lists DirectX 12 (11_0) while the 765M lists 12 (11_0). Both list OpenGL 4.6 and Vulkan 1.2. The 940MX has a slight lead on the DirectX front.

The Verdict

The data is unambiguous. The NVIDIA GeForce GTX 765M is the faster part in every benchmark recorded. It wins the OpenCL test by 45.3% and the Vulkan test by 41.4%. Its average benchmark score is 5501, which is 940% higher than the 940MX's 4844. If you need raw throughput, for compute workloads that use OpenCL or Vulkan, the GTX 765M is the choice.

The GeForce 940MX is not about speed. It is a newer, denser chip on a smaller process, with a lower TDP of 23 W, support for DirectX 12, and a narrower 64-bit memory bus. It will run cooler, fit in thinner chassis, and support the newest Microsoft graphics API. But its throughput is roughly half that of the GTX 765M in shading and texturing. If your workload is light and you care about driver modernity and power efficiency, the 940MX. If your workload is heavy and you want the highest frames, the GTX 765M, despite being the older part, is the one that shows its teeth in the actual benchmarks.