NVIDIA GeForce 940M vs NVIDIA GeForce MX130 Comparison

NVIDIA’s MX130 and 940M are both end-of-life Maxwell parts aimed at thin-and-light laptops, but the benchmark data shows a clear, if modest, edge for the newer MX130. Across the two available head-to-head tests, the MX130 wins both, with its largest advantage coming in the Vulkan workload. The 940M, however, retains a structural advantage in raw shading power and memory bandwidth architecture, making the choice between them dependent on whether you prioritize synthetic compute scores or theoretical rendering throughput.

Where Each One Wins

The MX130 wins every head-to-head benchmark recorded in the data. In Geekbench OpenCL, it scores 6102 against the 940M’s 6018, a 1.4% lead. The gap widens substantially in Geekbench Vulkan, where the MX130 posts 4914 versus 4549, an 8% advantage. That Vulkan result is the single biggest differentiator between the two; it suggests the MX130 has better driver optimization or scheduling efficiency for modern low-level graphics APIs, even though both chips share the same Maxwell architecture and identical DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4 API support.

The 940M, by contrast, does not win any recorded benchmark. Its average benchmark score is 5284, which falls 4.2% behind the MX130’s 5508 average. That said, the 940M’s closest rival data reveals it sits in a slightly different performance class: its nearest neighbor is the NVIDIA GeForce GTX 980M at 5308 (delta -0.4%), while the MX130’s nearest rival is the NVIDIA GeForce GTX 765M at 5501 (delta 0.1%). In practical terms, the MX130 aligns with mid-range mobile GTX parts of an earlier generation, whereas the 940M aligns with lower-tier 900M and 800M products.

For users who run OpenCL compute workloads, the MX130’s 6102 score is the better result, but only marginally. For Vulkan-based gaming or compute, the 8% delta is more meaningful and gives the MX130 a clear win. The 940M’s only theoretical advantage lies in its larger shading core count and higher peak pixel/texture rates, which do not translate into a single benchmark victory in this data set.

Architecture Differences

Both GPUs are built on the same 28 nm TSMC process and use NVIDIA’s Maxwell architecture, but they are fundamentally different chips. The MX130 uses the GM108S die, which packs 1,020 million transistors into a 77 mm² area, yielding a transistor density of 13.2 million per mm². The 940M uses the larger GM107 die, with 1,870 million transistors across 148 mm², for a lower density of 12.6 million per mm². That means the 940M has roughly 83% more transistors and a 92% larger die, yet it does not convert that extra silicon into benchmark wins.

The 940M compensates with more execution resources: 512 shading units, 32 texture mapping units, and 16 ROPs, versus the MX130’s 384 shading units, 24 TMUs, and 8 ROPs. Consequently, the 940M’s theoretical peak pixel rate is 17.57 GPixel/s and its texture rate is 35.14 GTexel/s, both far above the MX130’s 9.512 GPixel/s and 28.54 GTexel/s. The 940M also leads in FP32 compute at 1,124.4 GFLOPS versus 913.2 GFLOPS. Despite this, the MX130’s faster clock speeds — 1109 MHz base and 1189 MHz boost versus 1020 MHz base and 1098 MHz boost — help close some of that gap, but not fully.

Memory is where the two diverge sharply. The MX130 uses 2 GB of GDDR5 on a 64-bit bus, running at 1253 MHz with 5 Gbps effective speed, delivering 40.10 GB/s of bandwidth. The 940M also has 2 GB and a 64-bit bus, but it uses DDR3 at 900 MHz (1800 Mbps effective), yielding only 14.40 GB/s. That is a 2.8x bandwidth advantage for the MX130, which likely explains its Vulkan win despite fewer cores. The 940M’s much higher TDP — 75 W versus 30 W — is a major practical difference: the MX130 is an IGP with no power connectors, while the 940M is an MXM module, also without power connectors but consuming far more power.

The Verdict

The data is unambiguous: the NVIDIA GeForce MX130 is the faster GPU in every measured benchmark. It wins OpenCL by 1.4% and Vulkan by 8%, and its average benchmark score of 5508 sits 4.2% above the 940M’s 5284. The MX130’s GDDR5 memory and higher clocks overcome the 940M’s larger core count and compute ceiling. For any user choosing between these two in a laptop, the MX130 is the correct pick based on performance data alone.

The 940M is not without a rationale, but it is not a performance one. Its 512 shading units and 16 ROPs give it a theoretical pixel rate of 17.57 GPixel/s, which is 85% higher than the MX130’s 9.512 GPixel/s. In fill-rate-bound scenarios, such as high-resolution texture blitting or certain legacy OpenGL workloads, the 940M could in principle pull ahead, but no benchmark in the fact pack confirms that. The 940M also has a higher FP32 throughput (1,124.4 GFLOPS vs. 913.2 GFLOPS), but again, this does not translate into a single recorded win.

The verdict is simple: pick the MX130 for better real-world compute and graphics scores, lower power draw (30 W vs. 75 W), and superior memory bandwidth. Pick the 940M only if you specifically need the larger die’s raw shading resources for a niche fill-rate workload, and you are prepared to accept its much higher power consumption and slower memory. Given that both are end-of-life products, the MX130’s newer release (November 2017 vs. March 2015) also suggests better driver longevity.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The MX130 averages 5508, while the 940M averages 5284, giving the MX130 a 4.2% advantage.

Q: How much faster is the MX130 in Vulkan?

A: The MX130 scores 4914 in Geekbench Vulkan, compared to the 940M’s 4549, an 8% lead.

Q: Does the 940M have more shading units?

A: Yes, the 940M has 512 shading units versus the MX130’s 384, but it still loses both head-to-head benchmarks.

Q: What is the memory bandwidth difference?

A: The MX130 delivers 40.10 GB/s via GDDR5, while the 940M provides 14.40 GB/s via DDR3, a 2.8x gap.

Q: Which GPU consumes less power?

A: The MX130 has a 30 W TDP, while the 940M is rated at 75 W, making the MX130 significantly more efficient.

Q: Are they the same architecture?

A: Both use NVIDIA’s Maxwell architecture on a 28 nm TSMC process, but the MX130 uses the GM108S chip and the 940M uses the GM107 chip.

Head-to-Head Benchmarks

The two recorded head-to-head tests both favor the MX130, but the margins tell different stories. In Geekbench OpenCL, the MX130 scores 6102 against 6018 for the 940M, a 1.4% delta. That is a close race — within the margin of typical driver variation — and it suggests that in compute workloads that are not memory-bound, the 940M’s extra shading units nearly neutralize the MX130’s bandwidth advantage. The MX130’s higher boost clock (1189 MHz vs. 1098 MHz) and faster memory are just enough to edge out the 940M.

The Vulkan test is more decisive. The MX130’s 4914 score beats the 940M’s 4549 by 8%, which is a substantial gap in GPU terms. Vulkan is a low-overhead API that tends to reward memory bandwidth and efficient scheduling over raw core counts, and this is exactly where the MX130’s GDDR5 memory shines. With 40.10 GB/s of bandwidth versus 14.40 GB/s, the MX130 can feed its 384 shading units far more effectively than the 940M can feed its 512 units. The 940M’s theoretical FP32 advantage (1,124.4 GFLOPS vs. 913.2 GFLOPS) is negated by the memory bottleneck, and the MX130’s 8% win reflects that.

Notably, the MX130’s average benchmark score of 5508 places it 0.1% above the GTX 765M (5501) and 0.5% above the AMD Radeon R7 M440 (5483). The 940M’s 5284 average places it 0.4% below the GTX 980M (5308) and 0.6% below the GeForce 930A (5317). This shows that the MX130 competes with a higher tier of mobile GPUs despite its smaller die, while the 940M is anchored to lower-tier parts. The MX130’s 32nd percentile among all GPUs versus the 940M’s 31st percentile is a small but consistent margin across the board.

Specification Differences

The two GPUs differ in nearly every silicon and memory specification. The MX130 uses the GM108S chip with 1,020 million transistors on a 77 mm² die, while the 940M uses the GM107 with 1,870 million transistors on a 148 mm² die. Transistor density is 13.2M per mm² for the MX130 and 12.6M per mm² for the 940M. Clock speeds favor the MX130: base 1109 MHz and boost 1189 MHz, versus 1020 MHz base and 1098 MHz boost on the 940M. Memory type is the biggest split — GDDR5 at 1253 MHz (5 Gbps effective) for the MX130, DDR3 at 900 MHz (1800 Mbps effective) for the 940M. Both have 2 GB capacity and a 64-bit bus, but bandwidth is 40.10 GB/s versus 14.40 GB/s.

Compute resources heavily favor the 940M: 512 shading units, 32 TMUs, and 16 ROPs, versus 384 shading units, 24 TMUs, and 8 ROPs on the MX130. Peak rates follow suit: the 940M’s pixel rate is 17.57 GPixel/s and texture rate is 35.14 GTexel/s, while the MX130 manages 9.512 GPixel/s and 28.54 GTexel/s. FP32 compute is 1,124.4 GFLOPS for the 940M and 913.2 GFLOPS for the MX130. Power draw is dramatically different: 75 W for the 940M as an MXM module, versus 30 W for the MX130 as an IGP. The MX130 uses a PCIe 3.0 x4 bus interface, while the 940M uses MXM-B (3.0). Both have no power connectors, portable-device-dependent display outputs, and identical API support (DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.4). Release dates differ by over two and a half years, with the MX130 arriving on 2017-11-16 and the 940M on 2015-03-12, and the 940M lists its predecessor as GeForce 800M and successor as GeForce 10 Mobile.