GPU Comparison

The NVIDIA GeForce 930M and GeForce 830M are two closely related mobile graphics processors from NVIDIA’s Maxwell generation, both aimed at thin-and-light laptops. While they share the same fundamental architecture and manufacturing process, they are differentiated by core configuration and clock speeds, leading to measurable performance gaps in synthetic benchmarks. The data shows the 930M as the faster part in both tested workloads, but the 830M holds its own in specific metrics due to its higher operating frequencies. This analysis breaks down their architectural differences, benchmark results, and specification variances to determine which GPU is better suited for specific use cases.

FAQ

Q: Which GPU is faster in Geekbench OpenCL?

A: The NVIDIA GeForce 930M wins decisively, scoring 5046 compared to the 830M’s 4324. This represents a 16.7% advantage for the 930M in this compute workload.

Q: How do the two cards compare in Vulkan performance?

A: The 930M again leads, with a score of 3729 versus the 830M’s 3590. The margin is much narrower here, at only 3.9%, suggesting the 830M is relatively more competitive in this API.

Q: What is the average benchmark score for each GPU?

A: The 930M has an average benchmark score of 4388, while the 830M averages 3957. This puts the 930M in the 26th percentile of all GPUs, slightly ahead of the 830M’s 24th percentile.

Q: Are both GPUs based on the same architecture?

A: Yes, both are built on NVIDIA’s Maxwell architecture. They also share the same 28 nm manufacturing process at TSMC, with identical transistor counts of 1,020 million and die sizes of 77 mm².

Q: How do their memory subsystems compare?

A: Both feature 2 GB of DDR3 memory on a 64-bit bus. However, the 830M has a higher memory clock of 900 MHz (1800 Mbps effective) versus the 930M’s 800 MHz (1600 Mbps effective), giving the 830M a bandwidth advantage of 14.40 GB/s over 12.80 GB/s.

Q: Which GPU has more shading units?

A: The 930M has significantly more, with 384 shading units compared to the 830M’s 256. This is a key factor in its superior compute performance despite lower clocks.

Architecture Differences

Both GPUs are built on the same Maxwell architecture and produced on a 28 nm process at TSMC, using an identical die of 77 mm² with 1,020 million transistors. The fundamental difference lies in how NVIDIA configured the silicon. The 930M utilizes a fuller implementation of the GM108S chip, featuring 384 shading units, 24 texture mapping units (TMUs), and 8 raster output units (ROPs). In contrast, the 830M, based on the GM108 chip, is cut down to 256 shading units and 16 TMUs, while retaining the same 8 ROPs. This means the 930M has 50% more shading units and TMUs than the 830M, giving it a theoretical advantage in parallel compute and texturing workloads.

However, the 830M compensates with much higher clock speeds. Its base clock runs at 1082 MHz and boosts to 1150 MHz, while the 930M is locked at a flat 549 MHz for both base and boost. This clock disparity is substantial, with the 830M operating at nearly double the frequency. Consequently, the 830M achieves higher pixel and texture fill rates: 9.200 GPixel/s versus 4.392 GPixel/s, and 18.40 GTexel/s versus 13.18 GTexel/s. The 930M’s raw FP32 throughput of 421.6 GFLOPS is lower than the 830M’s 588.8 GFLOPS, a direct result of the clock speed difference overpowering the core count advantage in this metric.

Both GPUs share identical API support, including DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4, and use a PCIe 3.0 x8 bus interface. They also both have a 33 W TDP, are IGP-class (integrated) parts with no power connectors, and have display outputs that are portable device dependent. The 930M is from the GeForce 900M generation, released in March 2015, while the 830M is from the older GeForce 800M generation, released a year earlier in March 2014. Both are now end-of-life products, with the 930M succeeding the 800M series and the 830M succeeding the GeForce 700M line.

Head-to-Head Benchmarks

The head-to-head data is derived from two synthetic tests: Geekbench OpenCL and Geekbench Vulkan. In the OpenCL test, the 930M is the clear winner, posting a score of 5046 against the 830M’s 4324. This 16.7% delta is the largest performance gap between the two in any benchmark. This result is counterintuitive given the 830M’s higher clock speed and FP32 throughput, but it strongly suggests that the 930M’s additional 128 shading units and 8 TMUs are far more impactful for this type of compute workload. The 930M’s wider execution resources allow it to process more parallel instructions per cycle, which is the dominant factor in OpenCL’s highly parallel nature.

In the Vulkan test, the gap narrows considerably. The 930M scores 3729, while the 830M trails at 3590, a delta of only 3.9%. This smaller margin indicates that Vulkan’s lower-level API can better exploit the 830M’s higher clock speeds and fill rates, partially offsetting its core count disadvantage. The 830M’s superior texture rate of 18.40 GTexel/s and pixel rate of 9.200 GPixel/s become more relevant in graphics-centric workloads, which Vulkan typically represents. Still, the 930M maintains its lead, winning both benchmark tests. The data records 2 wins for the 930M and 0 for the 830M.

Looking at the overall picture, the 930M’s average benchmark score of 4388 places it in the 26th percentile of all GPUs. Its nearest rival is the NVIDIA GeForce GT 645M, which scores 4411, a marginal 0.5% difference. The 830M, with an average of 3957, sits in the 24th percentile, with its closest competitor being the AMD Radeon R5 M420 at 3956, a negligible 0% delta. This shows that while the 930M is faster, both are clustered in the lower-midrange of mobile GPU performance, and their real-world differences may be less pronounced in games that are not heavily compute-bound.

Specification Differences

The following table highlights the key specification differences between the two GPUs, based solely on the provided data.

| Specification | NVIDIA GeForce 930M | NVIDIA GeForce 830M |

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

| Chip | GM108S | GM108 |

| Generation | GeForce 900M | GeForce 800M |

| Release Date | March 2015 | March 2014 |

| Base Clock | 549 MHz | 1082 MHz |

| Boost Clock | 549 MHz | 1150 MHz |

| Memory Clock | 800 MHz / 1600 Mbps effective | 900 MHz / 1800 Mbps effective |

| Memory Bandwidth | 12.80 GB/s | 14.40 GB/s |

| Shading Units | 384 | 256 |

| TMUs | 24 | 16 |

| Pixel Rate | 4.392 GPixel/s | 9.200 GPixel/s |

| Texture Rate | 13.18 GTexel/s | 18.40 GTexel/s |

| FP32 Performance | 421.6 GFLOPS | 588.8 GFLOPS |

| Predecessor | GeForce 800M | GeForce 700M |

| Successor | GeForce 10 Mobile | GeForce 900M |

| Average Benchmark Score | 4388 | 3957 |

| Percentile vs All GPUs | 26 | 24 |

Identical specifications include the 28 nm process node, 1,020 million transistors, 77 mm² die size, 2 GB DDR3 memory, 64-bit bus width, 8 ROPs, 33 W TDP, PCIe 3.0 x8 interface, and API support for DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4.

The Verdict

From the data, the NVIDIA GeForce 930M is the superior performer. It wins both head-to-head benchmarks, with a significant 16.7% lead in OpenCL and a smaller but still decisive 3.9% lead in Vulkan. Its higher shading unit count and TMU count directly translate to better compute and texturing performance in these synthetic tests, despite its low clock speeds. The 930M’s average benchmark score of 4388 is about 10.9% higher than the 830M’s 3957, and its 26th percentile ranking is two points higher. For users prioritizing raw compute performance in applications that leverage OpenCL or Vulkan, the 930M is the clear choice.

However, the 830M is not without its merits. Its significantly higher clock speeds (1082 MHz base vs 549 MHz) and memory bandwidth (14.40 GB/s vs 12.80 GB/s) give it superior fill rates and FP32 throughput. This means that in workloads that are sensitive to clock speed and memory bandwidth, such as certain legacy games or lightly threaded graphics tasks, the 830M could potentially perform on par or better than the 930M. The data does not include such tests, so this remains speculative, but the specification differences point to it.

In terms of positioning, the 930M is a newer part from the GeForce 900M generation, while the 830M is from the preceding 800M series. Both are end-of-life mobile GPUs with a 33 W TDP, making them suitable for basic laptop use and light gaming. The 930M offers better overall compute performance and a higher average score, making it the better buy for general-purpose GPU acceleration. The 830M, with its higher clocks, might be preferred in scenarios where raw clock speed matters more than core count, but the benchmark evidence favors the 930M for most users. Ultimately, the 930M is the more capable GPU in the tests provided, and its victory in both OpenCL and Vulkan makes it the recommended part.