NVIDIA GeForce 930M vs NVIDIA GeForce GT 745M Comparison

Head-to-Head Benchmarks

The recorded head-to-head data presents a fascinating split, with each GPU taking one decisive victory. In the Geekbench OpenCL test, the NVIDIA GeForce 930M delivers a score of 5046, while the GeForce GT 745M manages 3580. That represents a 40.9% advantage for the newer Maxwell-based part, a substantial margin that highlights significant compute efficiency improvements. The OpenCL result is not a close call; the 930M simply outruns the older Kepler chip by a wide margin in this particular workload.

However, the Vulkan benchmark tells an entirely different story. Here, the GeForce GT 745M posts a strong 5502, leaving the 930M's 3729 far behind. The delta is 32.2% in favor of the GT 745M, making this a decisive reversal of fortunes. It is remarkable that the older GPU, with its lower overall average benchmark score, can so thoroughly outperform the newer part in a graphics API test. The data suggests that Vulkan workloads, which often stress memory bandwidth and raw rasterization throughput, play directly to the GT 745M's architectural strengths.

Looking at the aggregate data, the 930M holds a higher average benchmark score of 4388 compared to the GT 745M's 3953. This places the 930M in the 26th percentile of all GPUs, while the GT 745M sits at the 23rd percentile. The average score difference is about 11%, which is far smaller than the 40.9% gap seen in OpenCL, indicating that the GT 745M's Vulkan performance pulls its overall standing closer to its rival. The head-to-head wins are split at one each, but the magnitudes of those wins are asymmetric. A 40.9% win in one test versus a 32.2% win in the other means the 930M gains more total ground, but the GT 745M's Vulkan dominance is impossible to ignore.

Where Each One Wins

The benchmark results indicate a clear specialization. The NVIDIA GeForce 930M wins decisively in compute-oriented, cross-platform OpenCL workloads. Its 40.9% lead in that test suggests that applications leveraging general-purpose GPU compute, such as video encoding, image processing filters, or scientific calculations, will favor the 930M. The architecture appears better suited to extracting parallel throughput from shader units without being bottlenecked by memory bandwidth.

Conversely, the NVIDIA GeForce GT 745M is the victor in Vulkan-based scenarios. Vulkan is a low-overhead graphics API that often rewards higher memory bandwidth and greater texture and pixel processing capability. The GT 745M's 32.2% lead here indicates it is the stronger choice for modern game titles that utilize Vulkan, particularly those that push high fill rates or require rapid texture fetches. Users running Vulkan-native games or VR applications should expect noticeably smoother performance on the GT 745M.

The average scores further contextualize these wins. The 930M's nearest rival, the GeForce GT 645M, scores 4411, which is only 0.5% lower than the 930M's average. This shows the 930M is positioned in a tight cluster of similarly performing mobile GPUs. The GT 745M's nearest rival, the Radeon R5 M420, scores 3956, a mere 0.1% difference. Both cards are firmly in the entry-level mobile segment, but their distinct benchmark profiles mean user experience will vary dramatically depending on the software in use. For OpenCL-accelerated applications, the 930M is the clear pick. For Vulkan gaming, the GT 745M is the superior option.

Architecture Differences

The two GPUs represent distinct architectural generations from NVIDIA, and the recorded specifications illustrate how their designs diverge. The 930M is built on the Maxwell architecture using the GM108S chip, fabricated on a 28 nm process at TSMC. It contains 1,020 million transistors on a die size of 77 mm², resulting in a transistor density of 13.2 million per square millimeter. In contrast, the GT 745M is a Kepler part with the GK107 chip, also on a 28 nm TSMC process, but it packs 1,270 million transistors into a larger 118 mm² die, yielding a lower density of 10.8 million per square millimeter. The Maxwell design achieves higher density with fewer transistors, reflecting a more efficient layout.

Memory subsystems differ profoundly. The 930M uses 2 GB of DDR3 memory on a 64-bit bus, running at 800 MHz with 1600 Mbps effective speed. This configuration yields a bandwidth of 12.80 GB/s. The GT 745M, however, uses 2 GB of GDDR5 memory on a 128-bit bus, running at 1000 MHz with 4 Gbps effective speed. Its bandwidth is a much higher 64.00 GB/s, exactly five times that of the 930M. This disparity explains the GT 745M's Vulkan advantage, as high-bandwidth memory is critical for graphics-heavy workloads.

Core counts are identical for shading units: both have 384. However, the GT 745M has 32 texture mapping units (TMUs) and 16 raster output units (ROPs), while the 930M has only 24 TMUs and 8 ROPs. Consequently, the GT 745M achieves a texture rate of 17.57 GTexel/s versus the 930M's 13.18 GTexel/s. Both GPUs share the same pixel rate of 4.392 GPixel/s and identical FP32 compute of 421.6 GFLOPS. The GT 745M's extra TMUs and ROPs, combined with its massive bandwidth advantage, make it a more capable rasterization engine, while the 930M relies on architectural efficiency to match compute performance with far fewer resources.

The 930M also benefits from a newer API feature set, supporting Vulkan 1.4 compared to the GT 745M's Vulkan 1.2.175. Both support DirectX 12 (11_0) and OpenGL 4.6. The 930M has a lower TDP of 33 W versus 45 W for the GT 745M, making it more power-efficient. The 930M uses a PCIe 3.0 x8 interface, while the GT 745M uses the full x16 width. The GT 745M launched earlier, in 2013, while the 930M arrived in 2015, and each succeeded different prior generations: the 930M follows the GeForce 800M, while the GT 745M follows the GeForce 600M.

The Verdict

Based strictly on the recorded data, the choice between these two GPUs hinges entirely on the intended workload. For users prioritizing compute performance in OpenCL-based applications, the NVIDIA GeForce 930M is the definitive choice. Its 40.9% lead in that benchmark is overwhelming, and its higher average score of 4388 versus 3953 confirms it as the overall faster part in synthetic aggregate terms. The 930M also offers lower power consumption at 33 W, newer Vulkan 1.4 support, and a more advanced Maxwell architecture. For general productivity, media processing, or any task that leverages OpenCL acceleration, the 930M is the superior hardware.

However, for gaming scenarios, especially those using the Vulkan API, the NVIDIA GeForce GT 745M is the only rational pick. Its 32.2% Vulkan lead is substantial and directly translates to better frame rates in Vulkan-native titles. The GT 745M's 128-bit memory bus and GDDR5 memory provide 64.00 GB/s of bandwidth, which is critical for modern game engines. Its higher TMU and ROP counts further enhance its rendering capabilities. Users who predominantly play games released in recent years, many of which support Vulkan, will experience better performance on the GT 745M despite its older Kepler architecture.

The data does not support a universal winner. The 930M wins the compute crown, the GT 745M wins the graphics crown. The average benchmark scores suggest the 930M is marginally better overall, but the magnitude of the Vulkan deficit on the 930M is a serious drawback for gaming. The GT 745M's Vulkan score of 5502 is far beyond the 930M's 3729, and no amount of OpenCL superiority can compensate for that in a gaming-centric context. The verdict is clear: choose the 930M for compute-heavy tasks, choose the GT 745M for Vulkan gaming. Neither card excels at both.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce 930M has an average benchmark score of 4388, while the GeForce GT 745M scores 3953. This puts the 930M in the 26th percentile of all GPUs, compared to the GT 745M's 23rd percentile.

Q: How large is the OpenCL performance gap between the two cards?

A: In the Geekbench OpenCL test, the 930M scores 5046 against the GT 745M's 3580, giving the 930M a 40.9% advantage.

Q: Why does the GT 745M win in Vulkan benchmarks?

A: The GT 745M's memory configuration is the primary factor. It has a 128-bit bus with GDDR5 memory providing 64.00 GB/s bandwidth, versus the 930M's 64-bit DDR3 bus at 12.80 GB/s. The GT 745M also has 32 TMUs and 16 ROPs, compared to 24 TMUs and 8 ROPs on the 930M, which helps in graphics-heavy Vulkan workloads.

Q: Do both GPUs have the same compute power?

A: Yes, both the 930M and GT 745M have 384 shading units and deliver identical FP32 performance of 421.6 GFLOPS. They also share the same pixel rate of 4.392 GPixel/s.

Q: Which GPU is more power-efficient?

A: The 930M has a TDP of 33 W, while the GT 745M has a TDP of 45 W. The 930M's Maxwell architecture, built on a 28 nm process with 1,020 million transistors, achieves its performance at lower power consumption.

Q: What are the API differences between the two?

A: Both support DirectX 12 (11_0) and OpenGL 4.6. The 930M supports Vulkan 1.4, while the GT 745M supports the older Vulkan 1.2.175.

Specification Differences

| Specification | NVIDIA GeForce 930M | NVIDIA GeForce GT 745M |

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

| Architecture | Maxwell | Kepler |

| Chip | GM108S | GK107 |

| Generation | GeForce 900M | GeForce 700M |

| Transistors | 1,020 million | 1,270 million |

| Die Size | 77 mm² | 118 mm² |

| Transistor Density | 13.2M / mm² | 10.8M / mm² |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus Width | 64 bit | 128 bit |

| Memory Clock | 800 MHz, 1600 Mbps effective | 1000 MHz, 4 Gbps effective |

| Memory Bandwidth | 12.80 GB/s | 64.00 GB/s |

| TMUs | 24 | 32 |

| ROPs | 8 | 16 |

| Texture Rate | 13.18 GTexel/s | 17.57 GTexel/s |

| TDP | 33 W | 45 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 3.0 x16 |

| Vulkan Version | 1.4 | 1.2.175 |

| Release Date | 2015-03-12 | 2013-03-31 |

| Predecessor | GeForce 800M | GeForce 600M |

| Successor | GeForce 10 Mobile | GeForce 800M |