NVIDIA GeForce 940MX vs NVIDIA GeForce GT 755M Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is Geekbench OpenCL, and the result is remarkably close. The NVIDIA GeForce 940MX scores 4939, while the NVIDIA GeForce GT 755M scores 4934. That is a delta of just 0.1% in favor of the 940MX, a margin that falls well within normal run-to-run variation for mobile GPUs. In practical terms, the two are effectively identical in raw OpenCL compute throughput.

Looking at the broader benchmark averages, the picture changes somewhat. The 940MX posts an average benchmark score of 4844, while the GT 755M averages 4033. That is an 811-point gap, roughly 20% separating the two cards in the database's aggregate metric. However, the GT 755M's average is dragged down by its Geekbench Metal result of 3132, a test the 940MX does not appear in. When comparing only the shared OpenCL workload, the two are near-tied.

The percentile rankings reinforce this split. The 940MX sits at the 28th percentile among all GPUs, while the GT 755M sits at the 24th percentile. Both are entry-level mobile parts by the database's classification, but the 940MX edges ahead in overall standing. The 940MX's nearest rivals include the NVIDIA GeForce GTX 560M at 4855 (0.2% higher), the AMD Radeon R6 M255DX at 4867 (0.5% higher), and the NVIDIA GeForce GTS 450 at 4893 (1% higher). Interestingly, the NVIDIA GeForce RTX 3080 12 GB appears in its rival list with an average score of 4791, which is 1.1% lower than the 940MX, though this likely reflects the RTX 3080's sparse benchmark representation rather than a real performance comparison.

The GT 755M's nearest rivals tell a similar story at a lower tier. The AMD FirePro M4150 scores 4013 (0.5% below the GT 755M), while the Intel HD Graphics 630 scores 4075 (1% above). The AMD Radeon HD 6850 X2 scores 3977 (1.4% below), and the AMD Radeon RX 9060 XT 8 GB scores 4093 (1.5% above). These deltas indicate the GT 755M sits in a narrow performance band, surrounded by integrated graphics and older discrete parts.

The head-to-head table in the database records exactly one shared benchmark, Geekbench OpenCL. The 940MX wins that single test by 0.1%. The GT 755M has zero head-to-head wins. This is a razor-thin victory, and it would be misleading to declare either card a clear winner based on this data alone.

Where Each One Wins

The 940MX wins the only direct comparison available, Geekbench OpenCL, with a score of 4939 against 4934. That is the entirety of its measured head-to-head advantage. In the aggregate, the 940MX also holds a higher average benchmark score of 4844 versus 4033 for the GT 755M, and a higher percentile rank of 28 versus 24.

The GT 755M's case rests on its Geekbench Metal performance. It scores 3132 in that test, which is a workload the 940MX has no recorded result for. If Metal compute is the target application, the GT 755M is the only one of the two with demonstrated capability in that API. The 940MX has no Metal result in the database, so any comparison there is impossible.

Beyond raw scores, the GT 755M offers a wider memory bus. The database lists a 128-bit bus for the GT 755M versus a 64-bit bus for the 940MX. This does not show up in the OpenCL scores, where the two are nearly identical, but it suggests the GT 755M may have an advantage in memory-bandwidth-sensitive workloads that the current benchmark set does not capture. The bandwidth figures support this: the GT 755M delivers 86.40 GB/s versus 40.10 GB/s for the 940MX. That is more than double the memory bandwidth, a significant architectural difference that the single shared benchmark does not reflect.

The 940MX counters with higher raw compute throughput. Its FP32 rating is 881.7 GFLOPS versus 783.4 GFLOPS for the GT 755M, a 12.5% advantage. The 940MX also has more shading units at 512 versus 384, which explains the FP32 lead despite lower clock speeds. Texture rate favors the GT 755M at 32.64 GTexel/s versus 27.55 GTexel/s, while pixel rate favors the GT 755M as well at 8.160 GPixel/s versus 6.888 GPixel/s.

Architecture Differences

The two GPUs come from different NVIDIA architectures and different generations. The 940MX is built on Maxwell, uses the GM107 chip, and belongs to the GeForce 900M generation. The GT 755M is built on Kepler, uses the GK107 chip, and belongs to the GeForce 700M generation. Both are manufactured by TSMC on a 28 nm process, so the underlying fabrication node is identical.

Transistor counts differ notably. The 940MX packs 1,870 million transistors into a 148 mm² die, yielding a transistor density of 12.6 million per square millimeter. The GT 755M carries 1,270 million transistors on a smaller 118 mm² die, for a density of 10.8 million per square millimeter. The 940MX is the denser chip, which aligns with its newer architecture and higher transistor budget.

Compute resources are arranged differently. The 940MX has 512 shading units, 32 texture mapping units, and 8 raster output units. The GT 755M has 384 shading units, 32 texture mapping units, and 16 raster output units. The 940MX has a third more shaders, but the GT 755M has double the ROPs. This explains the split in fill rates: the GT 755M leads in pixel rate and texture rate, while the 940MX leads in FP32 compute.

Clock speeds favor the GT 755M. Its base clock is 980 MHz with a boost of 1020 MHz, while the 940MX runs at 795 MHz base and 861 MHz boost. The GT 755M also runs its memory at 1350 MHz (5.4 Gbps effective) versus 1253 MHz (5 Gbps effective) for the 940MX. Despite lower clocks, the 940MX achieves higher FP32 throughput thanks to its wider shader array.

Memory configuration is a major architectural divergence. The 940MX uses a 64-bit memory bus with 2 GB of GDDR5 and 40.10 GB/s of bandwidth. The GT 755M uses a 128-bit bus with 2 GB of GDDR5 and 86.40 GB/s of bandwidth. Both have the same 2 GB capacity and the same memory type, but the GT 755M has exactly double the bus width and more than double the bandwidth.

API support differs on Vulkan. The 940MX supports Vulkan 1.4, while the GT 755M supports Vulkan 1.2.175. Both support DirectX 12 (11_0) and OpenGL 4.6. Power consumption is dramatically different: the 940MX has a TDP of 23 W, while the GT 755M has a TDP of 50 W, more than double. Both use MXM modules with no power connectors, and both have portable-device-dependent display outputs.

The GT 755M connects via PCIe 3.0 x16, while the 940MX uses PCIe 3.0 x8. This half-width bus interface may limit the 940MX in scenarios that rely on host-to-device transfers, though it does not appear in the current benchmark results.

Specification Differences

The following fields differ between the two GPUs in the database:

• Architecture: Maxwell (940MX) versus Kepler (GT 755M)
• Chip: GM107 versus GK107
• Generation: GeForce 900M versus GeForce 700M
• Transistors: 1,870 million versus 1,270 million
• Die size: 148 mm² versus 118 mm²
• Transistor density: 12.6M / mm² versus 10.8M / mm²
• Base clock: 795 MHz versus 980 MHz
• Boost clock: 861 MHz versus 1020 MHz
• Memory clock: 1253 MHz (5 Gbps effective) versus 1350 MHz (5.4 Gbps effective)
• Memory bus width: 64 bit versus 128 bit
• Memory bandwidth: 40.10 GB/s versus 86.40 GB/s
• Shading units: 512 versus 384
• ROPs: 8 versus 16
• Pixel rate: 6.888 GPixel/s versus 8.160 GPixel/s
• Texture rate: 27.55 GTexel/s versus 32.64 GTexel/s
• FP32: 881.7 GFLOPS versus 783.4 GFLOPS
• TDP: 23 W versus 50 W
• Bus interface: PCIe 3.0 x8 versus PCIe 3.0 x16
• Vulkan version: 1.4 versus 1.2.175
• Release date: 2016-06-27 versus 2013-06-24
• Predecessor: GeForce 800M versus GeForce 600M
• Successor: GeForce 10 Mobile versus GeForce 800M
• Average benchmark score: 4844 versus 4033
• Percentile: 28 versus 24

Fields that are identical include the manufacturer (NVIDIA), process node (28 nm), foundry (TSMC), memory size (2 GB), memory type (GDDR5), TMUs (32), DirectX support (12, 11_0), OpenGL support (4.6), slot width (MXM Module), power connectors (none), and display outputs (portable device dependent). Neither card has RT cores or tensor cores, and neither has a launch MSRP in the database.

FAQ

Q: Which GPU is faster in OpenCL?

A: The NVIDIA GeForce 940MX scores 4939 in Geekbench OpenCL versus 4934 for the NVIDIA GeForce GT 755M, a 0.1% difference. The 940MX wins the only head-to-head benchmark in the database.

Q: Does the GT 755M have more memory bandwidth?

A: Yes. The GT 755M has a 128-bit bus and 86.40 GB/s of bandwidth, while the 940MX has a 64-bit bus and 40.10 GB/s. The GT 755M delivers more than double the bandwidth.

Q: Which GPU has higher compute throughput?

A: The 940MX has higher FP32 performance at 881.7 GFLOPS versus 783.4 GFLOPS for the GT 755M, a 12.5% advantage. It achieves this with 512 shading units versus 384, despite lower clock speeds.

Q: Which GPU is more power efficient?

A: The 940MX has a TDP of 23 W, while the GT 755M has a TDP of 50 W. The 940MX delivers comparable OpenCL performance at less than half the power draw.

Q: Do both GPUs support DirectX 12?

A: Yes, both support DirectX 12 (11_0) and OpenGL 4.6. The 940MX supports Vulkan 1.4, while the GT 755M supports Vulkan 1.2.175.

Q: Which GPU has a higher percentile ranking?

A: The 940MX ranks at the 28th percentile among all GPUs, while the GT 755M ranks at the 24th percentile. The 940MX also has a higher average benchmark score of 4844 versus 4033.

The Verdict

The data supports a narrow overall win for the NVIDIA GeForce 940MX. It wins the only shared benchmark, Geekbench OpenCL, by 0.1%. It has a higher average benchmark score at 4844 versus 4033, a higher percentile rank at 28 versus 24, and a 12.5% advantage in FP32 compute at 881.7 GFLOPS. It also does all of this at a much lower TDP of 23 W versus 50 W, making it the more efficient part.

The GT 755M is not without strengths. It has double the memory bandwidth at 86.40 GB/s versus 40.10 GB/s, double the ROP count at 16 versus 8, higher pixel and texture rates, and a wider PCIe 3.0 x16 interface versus x8. It also has a recorded Geekbench Metal score of 3132, a workload where the 940MX has no data. For applications that depend on memory bandwidth or Metal compute, the GT 755M may be the better choice.

However, the aggregate benchmark data places the 940MX ahead. Its average score is roughly 20% higher, and its nearest rivals sit at a higher performance tier. The GT 755M's rival list includes integrated graphics like the Intel HD Graphics 630, which underscores its lower standing.

Buyers should pick the 940MX if they want the better overall benchmark performer, the higher compute throughput, and the more power-efficient design. The GT 755M makes sense only for specific use cases that exploit its memory bandwidth advantage or require Metal support, since it is the only one of the two with a Metal benchmark result in the database. Both cards are end-of-life products, and neither has a launch MSRP recorded, but the recorded data consistently favors the Maxwell-based 940MX in general-purpose compute.