AMD FirePro W5130M vs NVIDIA GeForce GTX 670MX Comparison

Where Each One Wins

The benchmark data splits cleanly between these two mobile GPUs. The AMD FirePro W5130M records a single OpenCL score of 4,904, while the NVIDIA GeForce GTX 670MX posts an OpenCL score of 6,125 and a Vulkan score of 5,316. That gives NVIDIA one outright win in the head-to-head comparison, with AMD taking zero.

Where the AMD part shows its character is in the context of its nearest rivals. The FirePro W5130M sits at the 29th percentile among all GPUs in the database. Its closest competitors are the NVIDIA GeForce RTX 5060 Ti 8 GB (4,901, 0.1% behind), the NVIDIA GeForce GTS 450 (4,893, 0.2% behind), the AMD Radeon R7 M265 (4,929, 0.5% ahead), and the AMD Radeon R7 M360 (4,931, 0.5% ahead). The data shows the FirePro is essentially locked in a statistical tie with that cluster. It is not a standout, but it is not an outlier either.

The NVIDIA GeForce GTX 670MX, by contrast, reaches the 33rd percentile. Its average benchmark score across two tests is 5,721. Its nearest rivals include the Intel Iris Pro Graphics P6300 (5,712, 0.2% behind), the NVIDIA GeForce GTX 550 Ti (5,731, 0.2% ahead), the AMD Radeon HD 8790M (5,691, 0.5% behind), and the NVIDIA Quadro M500M (5,604, 2.1% behind). The GTX 670MX leads three of those four rivals, with only the GTX 550 Ti edging it out by a hair.

In terms of workload split, the NVIDIA part wins the only directly comparable test (OpenCL) by a substantial margin. The Vulkan result for the GTX 670MX is lower than its OpenCL score, but it still exceeds the FirePro's OpenCL figure. The FirePro has no Vulkan score in the database, so there is no direct cross-API comparison. The practical takeaway: NVIDIA wins compute workloads where OpenCL is the API, and it also demonstrates Vulkan capability that the AMD part does not show in the recorded data.

Architecture Differences

The two GPUs come from different architectural generations and different design philosophies. The AMD FirePro W5130M uses the Tropo chip built on GCN 1.0 architecture, fabricated on a 28 nm process at TSMC. It packs 1,500 million transistors on a 123 mm² die, yielding a transistor density of 12.2 million per square millimeter. The NVIDIA GeForce GTX 670MX uses the GK104 chip with Kepler architecture, also on TSMC's 28 nm node, but with 3,540 million transistors on a 294 mm² die, giving a density of 12.0 million per square millimeter. The die size difference is substantial: the NVIDIA chip is more than twice the physical area, and it holds more than double the transistor count.

Core counts differ sharply. The FirePro W5130M has 512 shading units, 32 texture mapping units, and 16 ROPs. The GTX 670MX has 960 shading units, 80 TMUs, and 24 ROPs. That is 87.5% more shading units, 150% more TMUs, and 50% more ROPs on the NVIDIA side. Neither chip has ray tracing cores or tensor cores, as both predate those features.

Clock behavior also diverges. The AMD part runs at a 900 MHz base with a 925 MHz boost. The NVIDIA part runs at a flat 601 MHz for both base and boost, with no boost headroom. Despite the much lower clock, the NVIDIA part's raw throughput is higher because of its large core count. The GTX 670MX also carries a TDP of 75 W and requires no power connectors, while the FirePro has no TDP listed in the database.

Memory configurations differ in capacity and bus width. The FirePro has 2 GB of GDDR5 on a 128 bit bus, with memory clocked at 1000 MHz (4 Gbps effective), yielding 64.00 GB/s bandwidth. The GTX 670MX has 3 GB of GDDR5 on a 192 bit bus, with memory at 700 MHz (2.8 Gbps effective), yielding 67.20 GB/s. The NVIDIA part has 50% more capacity and slightly higher bandwidth, despite lower memory clocks, because of the wider bus.

API support shows a split. The FirePro supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The GTX 670MX supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The NVIDIA part has a marginally newer Vulkan version, while the AMD part has a slightly higher DirectX feature level.

Release timing is also relevant. The NVIDIA part launched in late September 2012, while the AMD part launched in early October 2015. Both are end-of-life products now. The NVIDIA part belongs to the GeForce 600M generation, with the GeForce 500M as predecessor and GeForce 700M as successor. The AMD part belongs to the FirePro Mobile (Wx100M) generation, with FirePro Mobility as predecessor and Radeon Pro Mobile as successor.

Head-to-Head Benchmarks

The database records one direct comparison between these two GPUs: the Geekbench OpenCL test. The AMD FirePro W5130M scores 4,904, while the NVIDIA GeForce GTX 670MX scores 6,125. The delta is -19.9% for the AMD part, meaning the NVIDIA part outperforms it by nearly 20%. That is a decisive margin, not a marginal one.

To put that in perspective, the NVIDIA part's OpenCL score of 6,125 is 24.9% higher than the AMD part's 4,904. In practical terms, the GTX 670MX delivers roughly a quarter more compute performance in OpenCL workloads. The NVIDIA part also posts a Vulkan score of 5,316, which is 8.4% higher than the AMD part's OpenCL score. Even the NVIDIA part's weaker API result beats the AMD part's only recorded result.

The AMD part's average benchmark score is 4,904, which equals its single OpenCL score. The NVIDIA part's average is 5,721, which is the mean of its two scores (6,125 and 5,316). That average is 16.7% higher than the AMD part's average. The percentile spread is 29th versus 33rd, a four point gap that reflects the NVIDIA part's higher overall standing in the database.

Looking at the nearest rival clusters reinforces the gap. The FirePro's closest rivals all sit within 0.5% of its score, forming a tight group of similarly performing parts. The GTX 670MX's nearest rivals are more spread out, with the Quadro M500M trailing by 2.1%. The NVIDIA part is not just faster than the AMD part; it also sits in a slightly higher performance tier relative to the broader GPU landscape.

FAQ

Q: Which GPU has the higher raw compute throughput?

A: The NVIDIA GeForce GTX 670MX, with 1,153.9 GFLOPS FP32, compared to 947.2 GFLOPS for the AMD FirePro W5130M. The NVIDIA part leads by 206.7 GFLOPS.

Q: How do the memory bandwidth figures compare?

A: The NVIDIA part has 67.20 GB/s, while the AMD part has 64.00 GB/s. The NVIDIA part leads by 3.20 GB/s, or 5%.

Q: What is the pixel rate difference?

A: The AMD FirePro W5130M has a higher pixel rate at 14.80 GPixel/s, while the NVIDIA GTX 670MX has 12.02 GPixel/s. The AMD part leads by 2.78 GPixel/s.

Q: Which GPU has more texture mapping units?

A: The NVIDIA GeForce GTX 670MX has 80 TMUs, compared to 32 for the AMD FirePro W5130M. That is 48 more TMUs on the NVIDIA side.

Q: Do both GPUs support Vulkan?

A: Yes. The AMD part supports Vulkan 1.2.170, and the NVIDIA part supports Vulkan 1.2.175. The NVIDIA part has a slightly newer version.

Q: What is the transistor density difference?

A: The AMD part has 12.2 million transistors per square millimeter, and the NVIDIA part has 12.0 million. The AMD part is marginally denser, by 0.2 million per square millimeter.

The Verdict

The data points to a clear winner for general compute performance: the NVIDIA GeForce GTX 670MX. It wins the only direct head-to-head benchmark by 19.9%, has a higher average benchmark score (5,721 versus 4,904), and sits at a higher percentile (33rd versus 29th). It also has more shading units, more TMUs, more ROPs, more memory capacity, and higher FP32 throughput. For anyone prioritizing raw OpenCL or Vulkan performance, the NVIDIA part is the better choice.

The AMD FirePro W5130M does have some advantages. It has a higher base clock (900 MHz versus 601 MHz), a higher boost clock (925 MHz versus 601 MHz), a higher pixel rate (14.80 GPixel/s versus 12.02 GPixel/s), and a smaller die (123 mm² versus 294 mm²). It also supports a slightly higher DirectX feature level (11_1 versus 11_0). These advantages matter in specific scenarios, such as fill-rate bound workloads or applications that leverage the newer DirectX feature set.

However, the AMD part's advantages do not translate into benchmark wins. Its higher pixel rate does not compensate for the NVIDIA part's massive lead in shading units and TMUs. Its smaller die and lower transistor count suggest better manufacturing efficiency, but the database records no power consumption figure for the AMD part, so efficiency cannot be quantified. The NVIDIA part's 75 W TDP is the only power figure available.

For a user choosing between these two end-of-life mobile GPUs, the decision rests on workload. If the task is OpenCL compute, the GTX 670MX is clearly superior. If the task is fill-rate sensitive and the application can use the AMD part's DirectX 11_1 features, the FirePro has a niche. But the recorded data shows only one winner in direct comparison, and that is NVIDIA.

Specification Differences

The following table lists only the fields where the two GPUs differ.

| Field | AMD FirePro W5130M | NVIDIA GeForce GTX 670MX |

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

| Chip | Tropo | GK104 |

| Architecture | GCN 1.0 | Kepler |

| Generation | FirePro Mobile (Wx100M) | GeForce 600M |

| Transistors | 1,500 million | 3,540 million |

| Die Size | 123 mm² | 294 mm² |

| Transistor Density | 12.2M / mm² | 12.0M / mm² |

| Base Clock | 900 MHz | 601 MHz |

| Boost Clock | 925 MHz | 601 MHz |

| Memory Clock | 1000 MHz, 4 Gbps effective | 700 MHz, 2.8 Gbps effective |

| Memory Size | 2 GB | 3 GB |

| Bus Width | 128 bit | 192 bit |

| Bandwidth | 64.00 GB/s | 67.20 GB/s |

| Shading Units | 512 | 960 |

| TMUs | 32 | 80 |

| ROPs | 16 | 24 |

| Pixel Rate | 14.80 GPixel/s | 12.02 GPixel/s |

| Texture Rate | 29.60 GTexel/s | 48.08 GTexel/s |

| FP32 | 947.2 GFLOPS | 1,153.9 GFLOPS |

| TDP | Not listed | 75 W |

| Power Connectors | Not listed | None |

| Display Outputs | Not listed | Portable Device Dependent |

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

| Vulkan | 1.2.170 | 1.2.175 |

| Release Date | 2015-10-01 | 2012-09-30 |

| Predecessor | FirePro Mobility | GeForce 500M |

| Successor | Radeon Pro Mobile | GeForce 700M |

| OpenCL Score | 4,904 | 6,125 |

| Vulkan Score | Not listed | 5,316 |

| Average Benchmark Score | 4,904 | 5,721 |

| Percentile vs All GPUs | 29 | 33 |