AMD Radeon R7 M365X vs NVIDIA GeForce GTX 670MX Comparison

# The Verdict

The NVIDIA GeForce GTX 670MX and AMD Radeon R7 M365X are both end-of-life mobile GPUs, but the benchmark data shows a clear hierarchy between them. In the head-to-head comparison, the GTX 670MX wins both available tests: Geekbench OpenCL and Geekbench Vulkan. The OpenCL margin is modest at 3.1%, while the Vulkan gap widens to 8.6%, indicating that NVIDIA's advantage grows under compute-heavy API workloads.

The GTX 670MX posts an average benchmark score of 5721, placing it in the 33rd percentile of all GPUs. The R7 M365X averages 5416, sitting at the 32nd percentile. While the percentile difference appears small, the absolute gap of roughly 305 points represents a meaningful performance tier separation. The GTX 670MX's nearest rivals include the NVIDIA GeForce GTX 550 Ti (5731, within 0.2%) and the Intel Iris Pro Graphics P6300 (5712, within 0.2%), placing it in a competitive mid-range desktop-class bracket. The R7 M365X, by contrast, competes against integrated-class parts like the AMD Radeon 610M (5444, 0.5% ahead) and the Intel UHD Graphics P630 (5370, 0.9% behind), showing it operates at a lower performance stratum.

Who should pick which? Based strictly on the data, the GTX 670MX is the choice for users prioritizing raw compute and Vulkan API performance. Its 960 shading units and 80 texture mapping units dwarf the R7 M365X's 384 shading units and 24 TMUs, and this hardware advantage translates directly into benchmark wins. The R7 M365X, however, is the newer part by release date (May 2015 versus September 2012) and offers a more modern DirectX 12 feature level (12_1 versus 11_0). Users needing newer API support for specific titles may prefer the R7 M365X, but for pure performance, the data is unambiguous.

# Architecture Differences

The two GPUs come from entirely different architectural generations, and the specifications reflect that divergence. The GTX 670MX uses NVIDIA's Kepler architecture on the GK104 chip, built on a 28 nm TSMC process. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0 million per square millimeter. The R7 M365X employs AMD's GCN 1.0 architecture on the Litho chip, also fabricated on a 28 nm TSMC process, but with just 950 million transistors on a 77 mm² die, resulting in a slightly higher density of 12.3 million per square millimeter.

The compute resources are starkly different. The GTX 670MX features 960 shading units, 80 TMUs, and 24 ROPs, enabling a pixel rate of 12.02 GPixel/s and a texture rate of 48.08 GTexel/s. Its floating-point throughput reaches 1,153.9 GFLOPS in FP32. The R7 M365X, in comparison, has 384 shading units, 24 TMUs, and only 8 ROPs, producing 6.60 GPixel/s and 19.80 GTexel/s, with an FP32 output of 633.6 GFLOPS. The GTX 670MX delivers roughly 82% more FP32 throughput, 143% more texture fill, and 82% higher pixel fill than the R7 M365X.

Memory configurations also differ significantly. The GTX 670MX carries 3 GB of GDDR5 on a 192-bit bus, achieving 67.20 GB/s of bandwidth. The R7 M365X has 1 GB of GDDR5 on a 128-bit bus, providing 64.00 GB/s. Despite the R7's higher memory clock (1000 MHz versus 700 MHz, or 4 Gbps effective versus 2.8 Gbps effective), the narrower bus limits its bandwidth to slightly below the GTX 670MX. The GTX 670MX also enjoys a wider PCIe interface — PCIe 3.0 x16 versus PCIe 3.0 x8 on the R7 M365X — which can matter for data transfer in compute workloads.

The feature sets show generational differences. The GTX 670MX supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The R7 M365X supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The R7's higher DirectX feature level is notable for newer games, but the GTX 670MX's Vulkan version is marginally newer. The R7 M365X draws power without a specified TDP in the data, while the GTX 670MX is rated at 75 W with no power connectors required, suggesting it was designed to run within standard laptop power budgets.

# Where Each One Wins

The head-to-head results show the GTX 670MX winning both benchmark tests, but the nature of those wins reveals different usage scenarios. In Geekbench OpenCL, the GTX 670MX scores 6125 against the R7 M365X's 5939, a 3.1% advantage. This test exercises general-purpose compute across a wide range of workloads, and the smaller margin suggests that the R7 M365X's GCN architecture is more competitive in this area than raw specifications might indicate. The GTX 670MX's advantage here is real but not overwhelming.

The Vulkan test tells a different story. The GTX 670MX scores 5316 versus 4893 for the R7 M365X, an 8.6% margin — nearly triple the OpenCL gap. Vulkan workloads often stress driver efficiency and low-level hardware scheduling, and the data suggests Kepler holds a firmer edge in this API. For users running Vulkan-based games or compute applications, the GTX 670MX is the stronger choice by a comfortable margin.

The R7 M365X's wins are less about performance and more about positioning. Its 32nd percentile placement puts it among integrated-class rivals like the AMD Radeon 610M (5444, 0.5% ahead) and the Intel UHD Graphics P630 (5370, 0.9% behind). This suggests the R7 M365X is best suited for light-duty tasks where its DirectX 12 (11_1) support might enable compatibility with newer titles that the GTX 670MX cannot run at full feature levels. The R7 also has a more recent release date (2015 versus 2012), which could matter for driver maturity in certain software stacks, though the data does not quantify this.

# FAQ

Q: Which GPU wins in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 670MX wins with a score of 6125 versus 5939 for the AMD Radeon R7 M365X, a 3.1% difference.

Q: How large is the Vulkan performance gap?

A: The GTX 670MX scores 5316 in Geekbench Vulkan against 4893 for the R7 M365X, giving NVIDIA an 8.6% advantage.

Q: Which GPU has more shaders?

A: The GTX 670MX has 960 shading units compared to 384 on the R7 M365X, a 2.5x difference in raw shader count.

Q: What is the memory bandwidth difference?

A: The GTX 670MX provides 67.20 GB/s over a 192-bit bus, while the R7 M365X offers 64.00 GB/s on a 128-bit bus, a difference of about 5%.

Q: Which GPU supports a higher DirectX feature level?

A: The AMD Radeon R7 M365X supports DirectX 12 (11_1), while the GTX 670MX supports DirectX 12 (11_0), giving AMD a newer feature level.

Q: How do these GPUs rank against all others?

A: The GTX 670MX sits at the 33rd percentile with an average benchmark score of 5721, while the R7 M365X is at the 32nd percentile with an average score of 5416.

# Head-to-Head Benchmarks

The two available head-to-head tests provide a compact but informative picture of how these GPUs compare. In Geekbench OpenCL, the GTX 670MX posts 6125 points, edging out the R7 M365X's 5939 by 186 points. This 3.1% margin is substantial enough to indicate a real performance advantage, but it is not a landslide. The OpenCL test typically exercises a mix of memory bandwidth, compute throughput, and driver efficiency. Given that the GTX 670MX has 1,153.9 GFLOPS of FP32 performance versus 633.6 GFLOPS for the R7, one might expect a larger gap; the fact that it is only 3.1% suggests the R7's GCN architecture extracts more practical efficiency from its hardware in this workload. The R7's higher memory clock (4 Gbps effective versus 2.8 Gbps) and narrower bus (128-bit versus 192-bit) may also balance out in ways that narrow the theoretical bandwidth gap.

The Geekbench Vulkan test shows a more decisive outcome. The GTX 670MX scores 5316, while the R7 M365X manages 4893 — a 423-point gap translating to 8.6%. This is nearly three times the OpenCL margin, indicating that the GTX 670MX's architectural advantages compound under Vulkan's lower-level API model. Kepler's larger ROP count (24 versus 8) and higher texture fill rate (48.08 GTexel/s versus 19.80) likely contribute to this result, as Vulkan workloads often stress these fixed-function units more heavily. The GTX 670MX's average benchmark score of 5721 further reinforces its position, placing it near the NVIDIA GeForce GTX 550 Ti (5731) and ahead of the Intel Iris Pro Graphics P6300 (5712). The R7 M365X, with its 5416 average, sits closer to the AMD Radeon 610M (5444) and the NVIDIA Quadro M4000 (5467), showing it belongs to a lower performance class.

In terms of raw specifications, the GTX 670MX wins on nearly every measurable front: shading units (960 versus 384), TMUs (80 versus 24), ROPs (24 versus 8), memory capacity (3 GB versus 1 GB), and memory bandwidth (67.20 GB/s versus 64.00 GB/s). The R7 M365X counters only with a higher effective memory clock (4 Gbps versus 2.8 Gbps), a higher DirectX feature level (11_1 versus 11_0), and a slightly newer Vulkan API version (1.2.170 versus 1.2.175, with the GTX actually having the newer Vulkan). The transistor density is nearly identical (12.3M/mm² versus 12.0M/mm²), reflecting that both use the same 28 nm TSMC process, but the GTX 670MX's much larger die (294 mm² versus 77 mm²) shows where the performance investment went.

The benchmark data, when combined with the specification sheet, paints a clear picture. The GTX 670MX is the stronger performer in both available tests, with its advantage growing from 3.1% in OpenCL to 8.6% in Vulkan. The R7 M365X offers newer API support and a more recent release date, but it cannot match the GTX 670MX's compute throughput or memory subsystem. For any application that leverages the tested workloads, the GTX 670MX is the data-backed choice.