NVIDIA GeForce GT 555M vs NVIDIA GeForce GTX 670MX Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct benchmark comparison between the NVIDIA GeForce GT 555M and the NVIDIA GeForce GTX 670MX, using the Geekbench OpenCL test. In this head-to-head measurement, the GT 555M scores 6493 points, while the GTX 670MX scores 6125 points. This gives the GT 555M a 6% advantage in the OpenCL workload. This is a notable outcome, as it reverses the expected hierarchy based on the two chips' specifications. The GTX 670MX is built on a newer architecture, has a larger memory pool, and carries a substantially higher transistor count, yet the older Fermi-based part emerges ahead in this specific compute test.

The delta between the two is not trivial. A 6% lead in a synthetic OpenCL benchmark can translate to meaningful differences in compute-heavy applications, though it is a single data point. The database shows only one recorded head-to-head test, with the GT 555M claiming the sole win. There are no Vulkan results for the GT 555M, while the GTX 670MX has a separate Vulkan score of 5316, but no direct comparison is available for that API.

When examining the broader context, the GT 555M's OpenCL score of 6493 places it at the 37th percentile among all GPUs, with an average benchmark score of 6493. Its nearest rivals include the NVIDIA Quadro M5000M at 6481 (0.2% slower), the AMD Radeon Vega 10 Mobile at 6476 (0.3% slower), the NVIDIA GeForce GTX 670M at 6513 (0.3% faster), and the Intel UHD Graphics P750 at 6554 (0.9% faster). This clustering indicates that the GT 555M sits in a tight performance band, where differences of less than 1% separate it from its closest competitors.

The GTX 670MX, despite its newer Kepler architecture, records an OpenCL score of 6125, which is 6% lower than the GT 555M. Its average benchmark score across both OpenCL and Vulkan is 5721, a figure dragged down by its Vulkan result of 5316. The GTX 670MX sits at the 33rd percentile among all GPUs, four points below the GT 555M. Its nearest rivals include the Intel Iris Pro Graphics P6300 at 5712 (0.2% faster), the NVIDIA GeForce GTX 550 Ti at 5731 (0.2% slower), the AMD Radeon HD 8790M at 5691 (0.5% faster), and the NVIDIA Quadro M500M at 5604 (2.1% faster). The average score gap between the two parts is 772 points, or roughly 12% in favor of the GT 555M when considering the GTX 670MX's combined average.

Where Each One Wins

The GT 555M wins the only direct benchmark recorded, the Geekbench OpenCL test, with a 6% margin. This suggests that in compute-oriented workloads that rely on OpenCL acceleration, the older Fermi architecture can still hold its own against the Kepler-based GTX 670MX. The GT 555M's shading unit count of 144, combined with its 24 texture mapping units and 16 render output units, appears to be well-optimized for the specific instructions used in the OpenCL test. Its memory bandwidth of 28.80 GB/s, while lower than the GTX 670MX's 67.20 GB/s, does not hamper it in this particular workload.

The GTX 670MX, on the other hand, wins in areas not covered by the direct comparison. It has a Vulkan benchmark score of 5316, which is a separate measurement. The GT 555M has no recorded Vulkan score, meaning the GTX 670MX is the only one of the two with measurable support for that API in the database. The GTX 670MX also lists Vulkan 1.2.175 support, while the GT 555M has no Vulkan version listed at all. For applications that leverage Vulkan, the GTX 670MX is the only option with recorded performance data.

Beyond benchmarks, the GTX 670MX offers a larger memory configuration: 3 GB of GDDR5 on a 192-bit bus, versus 1 GB of DDR3 on a 128-bit bus for the GT 555M. This gives the GTX 670MX a bandwidth advantage of 67.20 GB/s versus 28.80 GB/s, which is 133% higher. For workloads that are memory-bound, such as high-resolution texture streaming or large data buffers, the GTX 670MX is structurally better positioned even if the synthetic OpenCL score does not reflect it. The GTX 670MX also has a higher pixel rate of 12.02 GPixel/s versus 3.540 GPixel/s for the GT 555M, and a texture rate of 48.08 GTexel/s versus 14.16 GTexel/s. These raw throughput numbers suggest that in rasterization-heavy tasks, the GTX 670MX should outperform, even though the database does not include a direct gaming benchmark to confirm this.

The Verdict

Based strictly on the recorded data, the NVIDIA GeForce GT 555M is the winner of the direct head-to-head comparison. It scores 6% higher in Geekbench OpenCL, has a higher average benchmark score by 772 points, and sits at a higher percentile rank (37th versus 33rd). If the selection criterion is purely the single available benchmark, the GT 555M is the correct choice.

However, the data also reveals a more nuanced picture. The GTX 670MX is the only part with a Vulkan score, and it supports a newer API version. It also offers a much larger memory pool and significantly higher memory bandwidth, pixel rate, and texture rate. For users who prioritize compute performance in OpenCL, the GT 555M is the better pick. For users who need Vulkan support or expect to run applications that benefit from higher memory bandwidth and larger frame buffers, the GTX 670MX is the only option with recorded evidence in those areas.

The GT 555M's nearest rivals include the GTX 670M, which is 0.3% faster. The GTX 670MX's nearest rivals include the GTX 550 Ti, which is 0.2% faster. Neither part dominates its immediate competitive set. The GT 555M is effectively tied with the Quadro M5000M and the Radeon Vega 10 Mobile, while the GTX 670MX is effectively tied with the Iris Pro P6300 and the Radeon HD 8790M. This indicates that both parts are mid-pack performers in their respective tiers, with the GT 555M holding a slight edge in the OpenCL metric.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The NVIDIA GeForce GT 555M scores 6493 in Geekbench OpenCL, while the GTX 670MX scores 6125, giving the GT 555M a 6% lead.

Q: Does the GTX 670MX have any benchmark advantage over the GT 555M?

A: Yes, the GTX 670MX has a recorded Vulkan score of 5316, while the GT 555M has no Vulkan benchmark in the database.

Q: What is the memory bandwidth difference between the two GPUs?

A: The GTX 670MX has a memory bandwidth of 67.20 GB/s, while the GT 555M has 28.80 GB/s, a difference of 38.40 GB/s in favor of the GTX 670MX.

Q: Which GPU has a higher pixel fill rate?

A: The GTX 670MX has a pixel rate of 12.02 GPixel/s, compared to 3.540 GPixel/s for the GT 555M.

Q: How do the two GPUs compare in terms of average benchmark score?

A: The GT 555M has an average benchmark score of 6493, while the GTX 670MX has an average of 5721, a 772-point difference in favor of the GT 555M.

Q: Which GPU supports the newer Vulkan API version?

A: The GTX 670MX supports Vulkan 1.2.175, while the GT 555M has no Vulkan version listed in the database.

Architecture Differences

The two GPUs come from different architectural generations. The GT 555M is based on the GF106 chip, using the Fermi architecture, and belongs to the GeForce 500M series. It is manufactured on a 40 nm process at TSMC, with 1,170 million transistors packed into a die size of 238 mm². This results in a transistor density of 4.9 million transistors per square millimeter. The GTX 670MX, in contrast, uses the GK104 chip with the Kepler architecture, belonging to the GeForce 600M series. It is built on a 28 nm process, also at TSMC, with 3,540 million transistors on a 294 mm² die, giving a density of 12.0 million transistors per square millimeter. The newer process node allows the GTX 670MX to pack three times more transistors into a slightly larger die.

The memory subsystems differ significantly. The GT 555M has 1024 MB of DDR3 memory on a 128-bit bus, with a memory clock of 900 MHz and an effective data rate of 1800 Mbps, yielding a bandwidth of 28.80 GB/s. The GTX 670MX has 3 GB of GDDR5 memory on a 192-bit bus, with a base clock of 601 MHz and a boost clock of 601 MHz, and a memory clock of 700 MHz with an effective data rate of 2.8 Gbps, yielding a bandwidth of 67.20 GB/s. The GTX 670MX's memory bandwidth is more than double that of the GT 555M.

The compute resources are also unevenly distributed. The GT 555M has 144 shading units, 24 texture mapping units, and 16 render output units. The GTX 670MX has 960 shading units, 80 texture mapping units, and 24 render output units. This translates to a texture rate of 14.16 GTexel/s for the GT 555M versus 48.08 GTexel/s for the GTX 670MX, and a pixel rate of 3.540 GPixel/s versus 12.02 GPixel/s. The floating-point performance, measured in FP32, is 339.8 GFLOPS for the GT 555M and 1,153.9 GFLOPS for the GTX 670MX, a 3.4x difference in theoretical compute capacity.

The power envelopes differ as well. The GT 555M has a TDP of 35 W, while the GTX 670MX has a TDP of 75 W. The GTX 670MX also has no power connectors listed, while the GT 555M's power connector configuration is not specified. Both use a PCIe x16 interface, but the GT 555M uses PCIe 2.0, while the GTX 670MX uses PCIe 3.0, which offers double the per-lane bandwidth.

Both GPUs support DirectX 12 (11_0) and OpenGL 4.6. The GTX 670MX additionally supports Vulkan 1.2.175, while the GT 555M has no Vulkan support listed. The GT 555M was released on 2011-07-01, while the GTX 670MX was released on 2012-09-30, a gap of roughly 15 months. The GT 555M's predecessor is the GeForce 400M and its successor is the GeForce 600M. The GTX 670MX's predecessor is the GeForce 500M and its successor is the GeForce 700M. Both are marked as end-of-life production status. The GT 555M has no launch MSRP recorded, and the GTX 670MX also has no launch MSRP recorded.