AMD Radeon Pro WX 3100 vs NVIDIA GeForce GTX 670M Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is Geekbench OpenCL, and the result is clear: the AMD Radeon Pro WX 3100 leads with a score of 7,333 against the NVIDIA GeForce GTX 670M's 6,513. That is a 12.6% advantage for the AMD card. This is not a marginal gap; it is a decisive margin that shows the newer architecture pulling ahead in compute workloads. The GTX 670M has no wins in the head-to-head data.

Looking at the broader database, the AMD Radeon Pro WX 3100 sits in the 41st percentile of all GPUs, with an average benchmark score of 7,580. Its closest rival, the AMD Radeon R7 250, scores 7,557, putting the WX 3100 just 0.3% ahead. The Intel Arc A310 is effectively tied at 7,550, a 0.4% gap. The NVIDIA GeForce GTX 1650 trails by 1.4% with a 7,472 average. However, the AMD Radeon 540 is 1.2% faster at 7,673. This places the WX 3100 in a tight cluster of mid-range cards, but clearly above the GTX 670M.

The NVIDIA GeForce GTX 670M, by contrast, ranks in the 38th percentile with an average score of 6,513. Its closest rivals include the NVIDIA GeForce GT 555M at 6,493 (0.3% slower), the NVIDIA Quadro M5000M at 6,481 (0.5% slower), and the AMD Radeon Vega 10 Mobile at 6,476 (0.6% slower). The Intel UHD Graphics P750 is 0.6% ahead at 6,554. This shows the GTX 670M is not only behind the WX 3100, but also near the bottom of its own performance tier, barely outrunning older mobile parts.

In practice, the 12.6% OpenCL delta translates to roughly 820 points, which in real-world compute tasks means the AMD card will finish rendering, filtering, or simulation workloads noticeably faster. The GTX 670M's only benchmark result is 13% lower, and it lacks any Vulkan score, while the WX 3100 posts a 7,827 in Vulkan. That Vulkan result is higher than its own OpenCL score, suggesting the AMD card scales better in Vulkan-based applications.

Architecture Differences

The two GPUs come from opposite eras. The AMD Radeon Pro WX 3100 uses the Lexa chip on the GCN 4.0 architecture, built on a 14 nm process at GlobalFoundries. It packs 2,200 million transistors into a 103 mm² die, giving a transistor density of 21.4 million per mm². The NVIDIA GeForce GTX 670M, by contrast, uses the GF114 chip on the Fermi 2.0 architecture, made on TSMC's 40 nm node. It has 1,950 million transistors on a 332 mm² die, for a density of just 5.9 million per mm². The AMD chip is over three times denser, which explains much of its efficiency and performance advantage.

The memory configurations also differ. The WX 3100 has 4 GB of GDDR5 on a 128-bit bus, yielding 96.00 GB/s of bandwidth. The GTX 670M has only 1536 MB of GDDR5, but on a wider 192-bit bus, resulting in 72.00 GB/s. Despite the wider bus, the GTX 670M's lower memory clock (750 MHz, 3 Gbps effective) holds it back; the WX 3100 runs at 1500 MHz (6 Gbps effective) and delivers 33% more bandwidth. For compute tasks, this bandwidth advantage helps keep the AMD's 512 shading units fed.

Compute resources are also lopsided. The WX 3100 has 512 shading units, 32 TMUs, and 16 ROPs. The GTX 670M has 336 shading units, 56 TMUs, and 24 ROPs. While the GTX 670M has more TMUs and ROPs, its pixel rate is only 8.372 GPixel/s versus 19.50 GPixel/s for the AMD, and its texture rate is 33.49 GTexel/s versus 39.01 GTexel/s. The AMD also dominates floating point: 1,248.3 GFLOPS FP32 versus 803.7 GFLOPS. The AMD supports 1:1 FP16, the GTX 670M does not list FP16 support at all.

The feature sets differ as well. The WX 3100 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. The GTX 670M supports DirectX 12 (11_0) and OpenGL 4.6, but no Vulkan. This is a major gap for modern gaming and compute. The WX 3100 also has a 65 W TDP versus the GTX 670M's 75 W, despite being far more powerful. The AMD is a single-slot card with no power connector, while the GTX 670M is an MXM module with no power connector.

Where Each One Wins

The AMD Radeon Pro WX 3100 wins every benchmark where the two overlap, and it holds advantages across architecture, memory, and API support. Its 12.6% lead in OpenCL over the GTX 670M means it is the clear choice for compute tasks like video encoding, 3D rendering, and scientific calculations. Its Vulkan 1.3 support opens the door to modern games and applications that rely on Vulkan, a capability the GTX 670M lacks entirely. The 4 GB frame buffer and higher bandwidth also allow it to handle larger textures and data sets than the GTX 670M's 1536 MB.

The NVIDIA GeForce GTX 670M wins in a few narrow areas. It has a wider memory bus (192 bit versus 128 bit), which theoretically benefits certain bandwidth-bound tasks, but the actual bandwidth is lower. It has more TMUs (56 versus 32) and more ROPs (24 versus 16), which helps in some fill-rate-limited scenarios, but its lower clock speeds and older architecture negate that advantage. The GTX 670M also has a slightly lower TDP (75 W versus 65 W), but the AMD has the better performance per watt.

The GTX 670M is a mobile part, designed for laptops as an MXM module, so it might fit in systems that require that specific form factor. But for raw performance, the AMD wins outright. The GTX 670M's nearest rivals are all near its own score, and it is only 0.3% faster than the GT 555M, meaning it barely outruns older mid-range parts. The AMD, meanwhile, is competitive with much newer GPUs like the GTX 1650 (1.4% faster) and the Radeon 540 (1.2% slower).

FAQ

Q: Which GPU is faster in OpenCL?

A: The AMD Radeon Pro WX 3100 scores 7,333, a 12.6% lead over the NVIDIA GeForce GTX 670M's 6,513.

Q: Does the GTX 670M support Vulkan?

A: No. The GTX 670M supports DirectX 12 (11_0) and OpenGL 4.6, but Vulkan is not listed. The WX 3100 supports Vulkan 1.3 and has a Vulkan score of 7,827.

Q: How do their memory bandwidths compare?

A: The WX 3100 has 96.00 GB/s, while the GTX 670M has 72.00 GB/s. The AMD card has a 33% higher bandwidth despite using a narrower 128-bit bus.

Q: Which card has a higher TDP?

A: The GTX 670M has a 75 W TDP, while the WX 3100 has 65 W. The AMD is more power-efficient despite higher performance.

Q: What is the average benchmark score for each?

A: The WX 3100 averages 7,580 across all benchmarks, while the GTX 670M averages 6,513, a difference of about 16%.

Q: How does the WX 3100 compare to its own rivals?

A: The WX 3100 is 0.3% ahead of the Radeon R7 250, 0.4% ahead of the Intel Arc A310, and 1.4% ahead of the GTX 1650, but it is 1.2% behind the Radeon 540.

Specification Differences

• Manufacturer: AMD vs NVIDIA
• Chip: Lexa vs GF114
• Architecture: GCN 4.0 vs Fermi 2.0
• Process Node: 14 nm vs 40 nm
• Foundry: GlobalFoundries vs TSMC
• Transistors: 2,200 million vs 1,950 million
• Die Size: 103 mm² vs 332 mm²
• Transistor Density: 21.4M / mm² vs 5.9M / mm²
• Base Clock: 925 MHz vs not specified
• Boost Clock: 1219 MHz vs not specified
• Memory Clock: 1500 MHz (6 Gbps effective) vs 750 MHz (3 Gbps effective)
• Memory Size: 4 GB vs 1536 MB
• Memory Bus: 128 bit vs 192 bit
• Memory Bandwidth: 96.00 GB/s vs 72.00 GB/s
• Shading Units: 512 vs 336
• TMUs: 32 vs 56
• ROPs: 16 vs 24
• Pixel Rate: 19.50 GPixel/s vs 8.372 GPixel/s
• Texture Rate: 39.01 GTexel/s vs 33.49 GTexel/s
• FP32: 1,248.3 GFLOPS vs 803.7 GFLOPS
• FP16: 1,248.3 GFLOPS (1:1) vs not specified
• TDP: 65 W vs 75 W
• Slot Width: Single-slot vs MXM Module
• Bus Interface: PCIe 3.0 x8 vs MXM-B (3.0)
• Display Outputs: 1x DisplayPort 1.4a, 2x mini-DisplayPort 1.4a vs Portable Device Dependent
• DirectX: 12 (12_0) vs 12 (11_0)
• Vulkan: 1.3 vs not specified
• Dimensions: 168 mm (6.6 inches) by 69 mm (2.7 inches) vs not specified
• Release Date: 2017-06-11 vs 2012-03-21
• Predecessor: Radeon Pro GCN vs GeForce 500M
• Successor: Radeon Pro Vega vs GeForce 700M
• Launch MSRP: The WX 3100 has a launch MSRP of 199 USD; the GTX 670M has no launch MSRP listed.

The Verdict

The data is unambiguous: the AMD Radeon Pro WX 3100 is the superior GPU in every measured category. Its OpenCL score is 12.6% higher, its average benchmark score is 16% higher, and it adds Vulkan support that the GTX 670M simply does not have. The WX 3100 also has over twice the die density, a smaller node, more memory, and higher bandwidth, all while consuming 10 W less power.

Any user choosing between these two should pick the AMD card, provided it fits their system. The GTX 670M is a legacy mobile part with no advantages in measured performance. Its only potential use is in a laptop that requires the MXM form factor, but even there, the performance gap is too large to justify. For a desktop workstation or a system with PCIe 3.0 x8, the WX 3100 is the only sensible choice, based strictly on the recorded benchmark results and specification differences.