NVIDIA GeForce GTX 670MX vs NVIDIA Quadro M4000 Comparison

The NVIDIA GeForce GTX 670MX and the NVIDIA Quadro M4000 represent two distinct generations of NVIDIA mobile and professional graphics. The data places them in adjacent performance percentiles, yet their benchmark scores reveal a significant gulf in raw compute capability that defines their respective use cases. This analysis draws strictly from the provided benchmark results and specification data.

Head-to-Head Benchmarks

The head-to-head comparison between the GTX 670MX and Quadro M4000 is limited to two common benchmark tests, and the results are decisively one-sided. In Geekbench OpenCL, the Quadro M4000 scores 19,118 against the GTX 670MX’s 6,125. This represents a -68% delta percentage for the GTX 670MX, meaning the Quadro M4000 delivers roughly three times the compute performance in this workload. The margin is even wider in Geekbench Vulkan, where the Quadro M4000 posts 24,640 points versus 5,316 for the GTX 670MX, a -78.4% delta. The Quadro M4000 wins both head-to-head tests, securing a clean 2-0 victory.

Looking beyond the direct head-to-head, the average benchmark scores corroborate this picture. The GTX 670MX averages 5,721 across its two benchmark entries, while the Quadro M4000 averages 5,467 across ten. This apparent contradiction — the GTX 670MX having a higher average despite losing both shared tests — stems from the fact that the Quadro M4000’s average is dragged down by low-scoring DirectX 9, 10, 11, and 12 Passmark tests (scores of 113, 33, 49, and 26, respectively). These legacy API tests do not reflect the GPU’s modern compute strengths. In the two tests where both cards are measured, the Quadro M4000 is faster by a factor of 3.1x in OpenCL and 4.6x in Vulkan.

The percentile rankings show both cards clustered near the bottom of the database: the GTX 670MX sits at the 33rd percentile, while the Quadro M4000 is at the 32nd. Their nearest rivals confirm this tier. The GTX 670MX is within 0.2% of the Intel Iris Pro Graphics P6300 (5,712) and -0.2% of the NVIDIA GeForce GTX 550 Ti (5,731). The Quadro M4000 sits between the AMD Radeon R7 M440 (5,483, -0.3%) and the AMD Radeon 610M (5,444, 0.4%). These deltas indicate that neither card is a standout in its class, but the Quadro M4000’s raw compute scores place it in a different performance realm than its average suggests.

Architecture Differences

The architectural gap between these two NVIDIA GPUs is substantial. The GTX 670MX is built on the GK104 chip using the Kepler architecture, fabricated on a 28 nm process at TSMC. It packs 3,540 million transistors onto a 294 mm² die, yielding a transistor density of 12.0M / mm². The Quadro M4000 uses the GM204 chip with the newer Maxwell 2.0 architecture, also on 28 nm TSMC, but with 5,200 million transistors on a larger 398 mm² die, achieving 13.1M / mm² density. The Maxwell 2.0 design provides a 47% increase in transistor count and a 35% larger die area over the Kepler part.

The compute resources diverge sharply. The GTX 670MX has 960 shading units, 80 texture mapping units (TMUs), and 24 raster operation units (ROPs). The Quadro M4000 more than doubles the shading units to 1,664, raises TMUs to 104, and substantially increases ROPs to 64. These differences translate directly to fill rates: the GTX 670MX achieves 12.02 GPixel/s pixel rate and 48.08 GTexel/s texture rate, while the Quadro M4000 reaches 49.47 GPixel/s and 80.39 GTexel/s. In floating-point performance, the GTX 670MX delivers 1,153.9 GFLOPS of FP32 compute, whereas the Quadro M4000 reaches 2.573 TFLOPS — a 2.2x advantage.

Clock speeds and memory configurations also differ. The GTX 670MX runs at a fixed 601 MHz base and boost clock, with memory at 700 MHz (or 2.8 Gbps effective). The Quadro M4000’s core clocks are not listed, but its memory runs at 1,502 MHz (6 Gbps effective). Memory capacity jumps from 3 GB GDDR5 on a 192-bit bus with 67.20 GB/s bandwidth to 8 GB GDDR5 on a 256-bit bus with 192.3 GB/s bandwidth — a 2.9x increase in memory throughput. The Quadro M4000 also supports a newer DirectX feature level: 12 (12_1) versus 12 (11_0) for the GTX 670MX, and a newer Vulkan version (1.4 vs 1.2.175). Both support OpenGL 4.6.

Where Each One Wins

The benchmark data offers a clear picture of where each card excels. The Quadro M4000 dominates in every shared benchmark. Its 19,118 OpenCL score indicates strong general-purpose compute capability, suitable for tasks like GPU-accelerated rendering, scientific simulations, or video encoding. The 24,640 Vulkan score further suggests it handles modern graphics APIs with ease, making it viable for current-generation game engines or compute workloads that leverage Vulkan’s low-level access. The Quadro M4000 also leads in memory capacity and bandwidth, with 8 GB versus 3 GB and 192.3 GB/s versus 67.20 GB/s, which directly benefits large dataset handling and high-resolution textures.

The GTX 670MX, by contrast, does not win any of the two head-to-head tests. Its strengths lie in its efficiency profile rather than raw performance. It draws a lower 75 W TDP compared to the Quadro M4000’s 120 W, and it requires no power connectors versus a 1x 6-pin connector. Its Portable Device Dependent display outputs indicate it was designed for laptops where space and power are at a premium. Its average benchmark score of 5,721 is actually higher than the Quadro M4000’s 5,467, driven by the Quadro’s weak legacy DirectX scores, which suggests the GTX 670MX may feel more consistent in older DirectX 9/10/11 applications. The GTX 670MX’s nearest rival, the Quadro M500M, sits just 2.1% behind, placing it in a similar professional-mobile tier.

For specific workloads, the data points to clear winners. The Quadro M4000 wins decisively in any modern compute or Vulkan-based task, with margins of 68% and 78.4% respectively. The GTX 670MX wins on power efficiency and portability, being a lower-power, connector-free mobile solution. Neither card wins on legacy API performance — the Passmark DirectX scores for the Quadro M4000 are uniformly low (33 for DX10, 49 for DX11, 26 for DX12, 113 for DX9), and the GTX 670MX lacks such data entirely, but its older Kepler architecture would not be expected to outperform the Maxwell part in newer APIs.

The Verdict

The data is unambiguous: the NVIDIA Quadro M4000 is the superior performer in every measured benchmark. Its 2.573 TFLOPS FP32 output, 49.47 GPixel/s pixel rate, and 192.3 GB/s memory bandwidth place it far ahead of the GTX 670MX’s 1,153.9 GFLOPS, 12.02 GPixel/s, and 67.20 GB/s. The Quadro M4000 wins both head-to-head tests by margins of 68% and 78.4%, making it the clear choice for any application that leverages OpenCL or Vulkan compute. Its 8 GB memory capacity also makes it suitable for larger workloads than the GTX 670MX’s 3 GB can handle.

Users should pick the Quadro M4000 if their priority is raw compute performance, modern API support (DirectX 12_1, Vulkan 1.4), or high-bandwidth memory. Its Single-slot design and 241 mm length make it a desktop workstation card, and its 300 W suggested PSU requirement indicates a full-sized system. The GTX 670MX, conversely, is the choice for portable, low-power systems. Its 75 W TDP and lack of power connectors make it a drop-in mobile solution, and its Portable Device Dependent outputs confirm its laptop orientation. However, its 33rd percentile ranking versus the Quadro’s 32nd shows both cards are near the bottom of the overall GPU hierarchy. The GTX 670MX is best suited for legacy applications or basic 3D tasks on the go, while the Quadro M4000 is the only one of the two that can handle demanding modern compute workloads with reasonable performance.

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The NVIDIA Quadro M4000 scores 19,118 in Geekbench OpenCL, versus 6,125 for the GTX 670MX, a 68% lead.

Q: How much memory bandwidth does each card have?

A: The GTX 670MX has 67.20 GB/s bandwidth on a 192-bit bus, while the Quadro M4000 has 192.3 GB/s on a 256-bit bus.

Q: What are the power requirements for each card?

A: The GTX 670MX has a 75 W TDP and no power connectors; the Quadro M4000 has a 120 W TDP and requires a 1x 6-pin connector with a 300 W suggested PSU.

Q: Do both cards support the same DirectX version?

A: No. The GTX 670MX supports DirectX 12 (11_0), while the Quadro M4000 supports DirectX 12 (12_1), a higher feature level.

Q: Which card has a higher average benchmark score?

A: The GTX 670MX averages 5,721, slightly above the Quadro M4000’s 5,467, despite losing both head-to-head tests.

Q: What is the transistor count difference?

A: The GTX 670MX has 3,540 million transistors on a 294 mm² die, while the Quadro M4000 has 5,200 million transistors on a 398 mm² die.

Specification Differences

| Specification | NVIDIA GeForce GTX 670MX | NVIDIA Quadro M4000 |

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

| Chip | GK104 | GM204 |

| Architecture | Kepler | Maxwell 2.0 |

| Generation | GeForce 600M | Quadro Maxwell (Mx000) |

| Transistors | 3,540 million | 5,200 million |

| Die Size | 294 mm² | 398 mm² |

| Transistor Density | 12.0M / mm² | 13.1M / mm² |

| Base Clock | 601 MHz | Not listed |

| Boost Clock | 601 MHz | Not listed |

| Memory Clock | 700 MHz (2.8 Gbps effective) | 1,502 MHz (6 Gbps effective) |

| Memory Size | 3 GB | 8 GB |

| Memory Bus Width | 192 bit | 256 bit |

| Memory Bandwidth | 67.20 GB/s | 192.3 GB/s |

| Shading Units | 960 | 1,664 |

| TMUs | 80 | 104 |

| ROPs | 24 | 64 |

| Pixel Rate | 12.02 GPixel/s | 49.47 GPixel/s |

| Texture Rate | 48.08 GTexel/s | 80.39 GTexel/s |

| FP32 Performance | 1,153.9 GFLOPS | 2.573 TFLOPS |

| TDP | 75 W | 120 W |

| Power Connectors | None | 1x 6-pin |

| Suggested PSU | Not listed | 300 W |

| Display Outputs | Portable Device Dependent | 4x DisplayPort 1.2 |

| DirectX Support | 12 (11_0) | 12 (12_1) |

| Vulkan Support | 1.2.175 | 1.4 |

| Slot Width | Not listed | Single-slot |

| Length | Not listed | 241 mm (9.5 inches) |

| Height | Not listed | 111 mm (4.4 inches) |