NVIDIA GeForce 920MX vs NVIDIA Quadro K3000M Comparison

Where Each One Wins

The recorded benchmark data splits cleanly between these two mobile GPUs based on workload type. The NVIDIA Quadro K3000M takes the single OpenCL compute test, scoring 4241 against the GeForce 920MX's 4068, a 4.3% advantage. That makes the Quadro the stronger choice for general-purpose GPU compute tasks that rely on OpenCL acceleration, such as rendering, simulation, or scientific workloads.

The GeForce 920MX, however, has a separate Vulkan benchmark score of 2987, which the Quadro K3000M has no recorded result for. This means the 920MX is the only one of the two with verified performance data under the Vulkan API. For applications that leverage Vulkan, the 920MX is the only option with confirmed benchmark coverage in the database.

Looking at the bigger picture, the Quadro K3000M sits at the 25th percentile among all GPUs, while the GeForce 920MX sits at the 21st percentile. The Quadro's single benchmark average is 4241, while the 920MX's average across its two recorded tests is 3528. That average is dragged down by the Vulkan result, which is substantially lower than its OpenCL score.

In terms of use-case split: the Quadro K3000M wins the OpenCL compute workload outright, while the GeForce 920MX offers the only Vulkan data. For users who prioritize OpenCL performance, the Quadro is the clear winner. For those targeting Vulkan-based workloads, the 920MX is the only candidate with recorded measurements, making it the practical choice despite its lower overall percentile ranking.

FAQ

Q: Which GPU is faster in OpenCL benchmarks?

A: The NVIDIA Quadro K3000M scores 4241 in Geekbench OpenCL, while the NVIDIA GeForce 920MX scores 4068. The Quadro leads by 4.3% in this test.

Q: Does the GeForce 920MX have any benchmark advantage?

A: The GeForce 920MX has a Geekbench Vulkan score of 2987. The Quadro K3000M has no recorded Vulkan benchmark result, so the 920MX is the only one with verified Vulkan performance data.

Q: How do these GPUs compare to their nearest rivals?

A: The Quadro K3000M is 0.6% behind the AMD Radeon Vega 3 (4268), 1% behind the NVIDIA GeForce GTX 460M (4282), 1.2% ahead of the NVIDIA GeForce GTX 1050 Ti (4193), and 1.3% behind the AMD FirePro W2100 (4295). The GeForce 920MX is 1.8% behind the NVIDIA GeForce GT 545 (3594), 1.9% behind the NVIDIA RTX 5000 Mobile Ada Generation (3596), 2.4% behind the NVIDIA GeForce GT 735M (3616), and 2.7% ahead of the NVIDIA Quadro 2000M (3434).

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

A: The Quadro K3000M has an average benchmark score of 4241 based on its single OpenCL result. The GeForce 920MX has an average of 3528, combining its OpenCL score of 4068 and Vulkan score of 2987.

Q: Which GPU has the higher memory bandwidth?

A: The Quadro K3000M has 89.60 GB/s of bandwidth from its 256-bit GDDR5 interface. The GeForce 920MX has 14.40 GB/s from its 64-bit DDR3 interface. The Quadro's bandwidth advantage is substantial.

Q: What is the power consumption difference?

A: The Quadro K3000M has a TDP of 75 W, while the GeForce 920MX has a TDP of 16 W. The 920MX consumes significantly less power, making it more suitable for thin and light portable devices.

Head-to-Head Benchmarks

The only direct head-to-head benchmark in the database is Geekbench OpenCL. The Quadro K3000M scores 4241, and the GeForce 920MX scores 4068. That is a 173-point gap, translating to a 4.3% delta in favor of the Quadro.

This result is consistent with the underlying compute resources. The Quadro K3000M has 576 shading units, 48 texture mapping units, and 32 ROPs. The GeForce 920MX has 256 shading units, 24 TMUs, and 8 ROPs. The Quadro's shading unit count is more than double the 920MX's, and its pixel rate is nearly identical at 7.848 GPixel/s versus 7.944 GPixel/s, despite the 920MX having far fewer ROPs. The texture rate favors the Quadro as well: 31.39 GTexel/s versus 23.83 GTexel/s.

The Quadro's FP32 throughput is 753.4 GFLOPS, against the 920MX's 508.4 GFLOPS. That is a 48% theoretical compute advantage, yet the real-world OpenCL gap is only 4.3%. The 920MX's higher clock speeds (965 MHz base, 993 MHz boost) compared to the Quadro's 654 MHz base and boost partially offset the Quadro's architectural advantages, but not enough to close the gap.

The 920MX's Vulkan score of 2987 cannot be directly compared to the Quadro, as no Vulkan result exists for the K3000M. However, placing that number in context, it sits well below the 920MX's own OpenCL score of 4068. This suggests that the 920MX is notably weaker under Vulkan than under OpenCL, a pattern that may reflect driver optimization or architectural limitations in the Maxwell design.

Overall, the head-to-head data shows a narrow but consistent win for the Quadro K3000M in the one shared benchmark. The deltas are small in percentage terms, but the Quadro's lead is real and measurable. The 920MX counters with exclusive Vulkan coverage, which gives it a distinct niche but not a comparative win.

Specification Differences

The two GPUs differ across nearly every major specification category. The table below highlights the key fields where they diverge:

| Specification | NVIDIA Quadro K3000M | NVIDIA GeForce 920MX |

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

| Architecture | Kepler | Maxwell |

| Generation | Quadro Kepler-M (Kx000M) | GeForce 900M |

| Chip | GK104 | GM108S |

| Process Node | 28 nm | 28 nm |

| Transistors | 3,540 million | 1,020 million |

| Die Size | 294 mm² | 77 mm² |

| Base Clock | 654 MHz | 965 MHz |

| Boost Clock | 654 MHz | 993 MHz |

| Memory Clock | 2.8 Gbps effective | 1800 Mbps effective |

| Memory Type | GDDR5 | DDR3 |

| Memory Bus | 256 bit | 64 bit |

| Memory Bandwidth | 89.60 GB/s | 14.40 GB/s |

| Shading Units | 576 | 256 |

| TMUs | 48 | 24 |

| ROPs | 32 | 8 |

| Pixel Rate | 7.848 GPixel/s | 7.944 GPixel/s |

| Texture Rate | 31.39 GTexel/s | 23.83 GTexel/s |

| FP32 | 753.4 GFLOPS | 508.4 GFLOPS |

| TDP | 75 W | 16 W |

| Bus Interface | MXM-B (3.0) | PCIe 3.0 x8 |

| Vulkan API | 1.2.175 | 1.4 |

| Release Date | 2012-05-31 | 2016-03-24 |

| Predecessor | Quadro Fermi-M | GeForce 800M |

| Successor | Quadro Maxwell-M | GeForce 10 Mobile |

| Transistor Density | 12.0M / mm² | 13.2M / mm² |

| DirectX | 12 (11_0) | 12 (11_0) |

| OpenGL | 4.6 | 4.6 |

| Slot Width | MXM Module | MXM Module |

| Power Connectors | None | None |

| Display Outputs | Portable Device Dependent | Portable Device Dependent |

Both share the same 28 nm TSMC process, 2 GB memory capacity, and identical DirectX and OpenGL API support. The Quadro uses GDDR5 memory with a 256-bit bus, while the 920MX uses DDR3 with a 64-bit bus. The Quadro's memory bandwidth is more than six times higher. The 920MX compensates with higher clocks: its base clock is 965 MHz versus 654 MHz, and its boost clock is 993 MHz versus 654 MHz.

The Quadro has a much larger die (294 mm² versus 77 mm²) and more than three times the transistor count (3,540 million versus 1,020 million). Despite this, the 920MX has a higher transistor density at 13.2M per mm² versus 12.0M per mm². The 920MX also has a significantly lower TDP at 16 W versus 75 W.

Architecture Differences

The architectural gap between these two GPUs spans multiple generations of NVIDIA design. The Quadro K3000M is built on the Kepler architecture with the GK104 chip, while the GeForce 920MX uses the Maxwell architecture with the GM108S chip. Kepler was NVIDIA's high-performance design for the Quadro professional line, while Maxwell was optimized for power efficiency in the consumer GeForce mobile segment.

The compute core layout differs substantially. The Quadro K3000M packs 576 shading units into its design, organized around 48 TMUs and 32 ROPs. The GeForce 920MX has 256 shading units, 24 TMUs, and just 8 ROPs. This gives the Quadro a 2.25x advantage in shading units, a 2x advantage in TMUs, and a 4x advantage in ROPs. Despite this, the pixel rates are nearly identical (7.848 GPixel/s versus 7.944 GPixel/s), because the 920MX's higher clock speed compensates for its lower ROP count.

Memory architecture is a major differentiator. The Quadro uses a 256-bit GDDR5 interface running at 2.8 Gbps effective, yielding 89.60 GB/s of bandwidth. The 920MX uses a 64-bit DDR3 interface at 1800 Mbps effective, yielding only 14.40 GB/s. The Quadro's memory subsystem is clearly designed for bandwidth-hungry professional workloads, while the 920MX's narrow bus is suited to low-power consumer tasks.

The transistor budget tells the story of design philosophy. The Quadro's GK104 die contains 3,540 million transistors across 294 mm², a density of 12.0M per mm². The 920MX's GM108S die has 1,020 million transistors across 77 mm², a density of 13.2M per mm². The Maxwell chip is more transistor-dense, reflecting its newer design, but it has far fewer total transistors because it targets a much lower power envelope.

The power difference is stark: 75 W TDP for the Quadro versus 16 W for the 920MX. This nearly 5x difference explains the clock strategy. The Quadro runs at a fixed 654 MHz for both base and boost, with no dynamic range. The 920MX runs at 965 MHz base and 993 MHz boost, leveraging its lower power draw to sustain higher clocks.

API support also differs. The Quadro supports Vulkan 1.2.175, while the 920MX supports Vulkan 1.4. Both support DirectX 12 (11_0) and OpenGL 4.6. The 920MX's newer Vulkan version reflects its later release date and more modern driver stack.

The bus interface differs as well: the Quadro uses MXM-B (3.0), while the 920MX uses PCIe 3.0 x8. Both are MXM modules with no power connectors and portable-device-dependent display outputs.

Release timing places these GPUs in different eras. The Quadro launched on 2012-05-31, succeeding the Quadro Fermi-M and preceding the Quadro Maxwell-M. The 920MX launched on 2016-03-24, succeeding the GeForce 800M and preceding the GeForce 10 Mobile. Both are now end-of-life products.

The architectural summary is clear: Kepler prioritizes raw compute and memory bandwidth at high power cost, while Maxwell prioritizes efficiency and clock speed at reduced resource counts. The benchmark results reflect this tradeoff, with the Quadro winning the shared OpenCL test but consuming far more power to do so.