NVIDIA GeForce 930A vs NVIDIA GeForce 940M Comparison

The Verdict

The data presents a clear split decision. The NVIDIA GeForce 940M is the outright performance winner, delivering a 13.2% higher Geekbench OpenCL score (6018 vs. 5317) than the GeForce 930A. For any workload that leans on raw compute throughput, the 940M is the definitive choice. However, the 930A is the efficiency pick. Its 33 W TDP is less than half of the 940M's 75 W, and its IGP slot width indicates a design intended for compact, low-power notebooks. The 930A is for users who prioritize battery life and slim chassis over frame rates; the 940M is for those who need the maximum performance available in a mobile discrete GPU of this generation. The 930A also holds a marginal average benchmark edge over the 940M in aggregate scoring (5317 vs. 5284), but that is a statistical artifact of the 940M's Vulkan test dragging its average down, not a sign of real-world superiority.

FAQ

Q: Which GPU has the higher raw compute performance?

A: The GeForce 940M. Its FP32 rating of 1,124.4 GFLOPS is significantly higher than the 930A's 722.7 GFLOPS, and it wins the head-to-head Geekbench OpenCL test with a score of 6018 versus 5317, a delta of -11.6% from the 930A's perspective.

Q: How do their power requirements differ?

A: The GeForce 930A has a 33 W TDP and is listed as an IGP (integrated graphics processor) with a slot width of "IGP," while the GeForce 940M has a 75 W TDP and uses an MXM Module slot width. This makes the 930A considerably more suitable for thinner, cooler-running systems.

Q: Are they based on the same architecture?

A: Yes, both use the Maxwell architecture and are fabricated by TSMC on a 28 nm process node. They also share the same DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4 API support.

Q: What are the key memory differences?

A: Both have 2 GB of DDR3 memory on a 64-bit bus. However, the 930A runs its memory at 1001 MHz (2 Gbps effective) providing 16.02 GB/s of bandwidth, while the 940M runs at 900 MHz (1800 Mbps effective) providing 14.40 GB/s. The 930A actually has a slight memory bandwidth advantage.

Q: Which GPU has more shading units and texture units?

A: The GeForce 940M has more of both. It features 512 shading units and 32 TMUs, compared to the 930A's 384 shading units and 24 TMUs. The 940M also doubles the ROP count to 16 versus 8.

Q: What is the production status of these GPUs?

A: Both are listed as "End-of-life" and share the same release date of 2015-03-12. The 940M's successor is listed as "GeForce 10 Mobile," while the 930A's successor is not specified.

Architecture Differences

Both the GeForce 930A and 940M are built on NVIDIA's Maxwell architecture, but they leverage fundamentally different chips. The 930A uses the GM108 chip, a small, power-efficient die measuring 77 mm² and containing 1,020 million transistors. The 940M uses the GM107 chip, which is a substantially larger and more complex design at 148 mm² and containing 1,870 million transistors. This difference in physical size and transistor count is the root of their performance gap. The GM107's larger die allows for 512 shading units, 32 texture units, and 16 ROPs, whereas the GM108 in the 930A is limited to 384 shading units, 24 texture units, and 8 ROPs. The transistor density is slightly higher on the 930A (13.2M / mm² versus 12.6M / mm²), but the sheer scale of the 940M's chip overcomes that density advantage. Both chips are fabricated by TSMC on a 28 nm process, and neither includes dedicated ray tracing or tensor cores, as those features are absent from the data. The architectural lineage is clear: the 940M's chip is designed for higher performance ceilings, while the 930A's chip is optimized for minimal power draw and physical footprint.

Specification Differences

The specification sheets reveal a consistent pattern of the 940M being a larger, more capable part, with one notable exception in memory clocks. The most significant difference is in the chip itself: the 930A uses the GM108 (77 mm², 1,020 million transistors), while the 940M uses the GM107 (148 mm², 1,870 million transistors). This leads to major disparities in core resources: the 940M offers 512 shading units versus 384, 32 TMUs versus 24, and 16 ROPs versus 8. Clock speeds also favor the 940M, with a base clock of 1020 MHz and boost of 1098 MHz, compared to the 930A's 928 MHz base and 941 MHz boost. The compute output reflects this: the 940M delivers 1,124.4 GFLOPS of FP32 performance and a texture rate of 35.14 GTexel/s, while the 930A manages 722.7 GFLOPS and 22.58 GTexel/s. Pixel rate also heavily favors the 940M at 17.57 GPixel/s versus 7.528 GPixel/s.

However, the 930A has two advantages. First, its memory runs at a higher effective speed: 2 Gbps versus 1800 Mbps, giving it 16.02 GB/s of bandwidth versus the 940M's 14.40 GB/s. Second, its power envelope is drastically lower at 33 W versus 75 W. The physical form factors differ as well, with the 930A being an IGP and the 940M being an MXM Module. Both connect via PCIe 3.0, but the 930A uses an x8 interface while the 940M uses an MXM-B (3.0) connector. Neither GPU has a power connector requirement, and display outputs are "Portable Device Dependent" for both.

Head-to-Head Benchmarks

The single head-to-head benchmark in the data is the Geekbench OpenCL test, and it decisively favors the GeForce 940M. The 940M scores 6018 points, while the 930A scores 5317 points. This represents a delta of -11.6% for the 930A, meaning the 940M is roughly 13.2% faster in this compute-bound workload. That margin is consistent with the theoretical specifications: the 940M has 33% more shading units, a higher boost clock (1098 MHz vs. 941 MHz), and more than double the pixel rate. The 940M's win in this test is its only head-to-head victory, as it holds a 1-0 win record against the 930A.

Yet the broader benchmark picture is more nuanced. Looking at the average benchmark scores and nearest rivals, the 930A actually posts a higher average score (5317) than the 940M (5284) when all their respective benchmarks are considered. This is because the 940M has a second benchmark result, Geekbench Vulkan, where it scores 4549 — a significantly lower number that pulls its average down. The 930A has only the one OpenCL result. In the nearest rivals data, the 930A is listed as being 0.6% ahead of the 940M in average score, while the 940M is listed as being 0.6% behind the 930A. This reciprocal deltaPct values (-0.6% from the 940M's perspective) confirm that the average scores are nearly identical, despite the 940M's clear win in the OpenCL test.

The data also places both GPUs in similar competitive territory. The 930A's nearest rivals include the GeForce 840M (5322, -0.1% delta), the GeForce GTX 980M (5308, +0.2% delta), and the AMD Radeon R7 M445 (5358, -0.8% delta). The 940M's rivals include the GTX 980M (5308, -0.4% delta), the 930A (5317, -0.6% delta), the 840M (5322, -0.7% delta), and the GeForce GTX 760M (5236, +0.9% delta). Both GPUs sit at the 31st percentile among all GPUs, indicating they are mid-pack performers in the broader market. The practical takeaway is that while the 940M is the stronger chip on paper and in the direct OpenCL comparison, the real-world aggregate performance of these two cards is nearly indistinguishable in the benchmark database — the 940M's advantage appears specifically in compute-heavy OpenCL workloads, not in the overall average.