AMD Radeon R9 M375 vs NVIDIA GeForce GTX 960 Comparison

AMD Radeon R9 M375 and NVIDIA GeForce GTX 960 represent two very different approaches to mobile and desktop graphics from the same era, and the benchmark data reveals a clear split in their capabilities. The R9 M375, a 28 nm GCN 1.0 part aimed at notebooks, wins the Vulkan test outright, while the GTX 960, a desktop-focused Maxwell 2.0 card, dominates the OpenCL workload by a wide margin. The two GPUs each claim one head-to-head victory, but the magnitude of those wins is far from symmetrical, and the underlying specifications explain why.

Head-to-Head Benchmarks

The single most decisive result in the entire comparison is the Geekbench OpenCL score. Here, the GTX 960 posts 18,925 points against the R9 M375’s 10,457, a delta of -44.7% from the AMD part’s perspective. That means the NVIDIA card delivers roughly 81% higher raw throughput in this compute-heavy API. The gap is not a marginal edge; it is a generational chasm in raw execution resources. The GTX 960’s 2.413 TFLOPS of FP32 performance versus the R9 M375’s 1,299.2 GFLOPS is the primary driver, but the memory subsystem also plays a role. The GTX 960’s 112.2 GB/s of bandwidth is nearly four times the R9 M375’s 28.80 GB/s, which directly impacts OpenCL workloads that stream data through shared memory.

The Vulkan result flips the script, though the margin is far more modest. The R9 M375 scores 9,682 while the GTX 960 scores 9,231, giving the AMD part a 4.9% advantage. This is a meaningful win for the R9 M375, especially considering its much lower theoretical specs. Vulkan’s lower overhead and explicit control can favor architectures with better driver efficiency per compute unit, and the data suggests the GCN 1.0 design handles this API slightly better than Maxwell 2.0 in this specific test. It is worth remembering the GTX 960’s Vulkan score of 9,231 is actually lower than its own OpenCL score of 18,925, while the R9 M375’s Vulkan score of 9,682 is only slightly below its OpenCL score of 10,457. This indicates the NVIDIA part scales better with OpenCL’s more traditional compute model, while the AMD part remains more consistent across APIs.

Looking at overall averages, the picture becomes more nuanced. The R9 M375 has an average benchmark score of 10,070, placing it in the 48th percentile of all GPUs. The GTX 960’s average is 9,273, which is in the 45th percentile. This is counterintuitive given the OpenCL blowout, but it reflects the fact that the GTX 960’s average is dragged down by its Vulkan result and the inclusion of a 3DMark Steel Nomad DX12 score of just 162 points, which is a very low number for a desktop card. The R9 M375, by contrast, only has two benchmark entries, both of which are relatively close to each other. The net effect is that the R9 M375’s average sits 8.6% higher than the GTX 960’s, even though the GTX 960 wins the most demanding test by a landslide.

Architecture Differences

The two GPUs are built on the same 28 nm TSMC process, but their internal designs could hardly be more different. The R9 M375 uses the Tropo chip with GCN 1.0 architecture, packing 1,500 million transistors into a 123 mm² die. The GTX 960 uses the GM206 chip with Maxwell 2.0 architecture, which crams 2,940 million transistors into a 228 mm² die. That is nearly double the transistor count and almost double the die area, which explains the massive compute advantage. Transistor density is similar — 12.2M per mm² for AMD versus 12.9M per mm² for NVIDIA — so the difference is purely about scale, not packing efficiency.

The compute resources tell the story. The R9 M375 has 640 shading units, 40 texture mapping units, and 16 ROPs. The GTX 960 more than doubles the shading units to 1,024, doubles the TMUs to 64, and doubles the ROPs to 32. This translates directly to pixel rate: 16.24 GPixel/s for the AMD part versus 37.70 GPixel/s for the NVIDIA part. Texture rate follows the same pattern, with the GTX 960 hitting 75.39 GTexel/s against the R9 M375’s 40.60 GTexel/s. These are not subtle differences; the GTX 960 has more than twice the fill rate in every category.

Memory is another major divergence. Both cards have 2 GB and a 128-bit bus, but the R9 M375 uses DDR3 at 900 MHz (1,800 Mbps effective), yielding 28.80 GB/s. The GTX 960 uses GDDR5 at 1,753 MHz (7 Gbps effective), yielding 112.2 GB/s. That is a 3.9x bandwidth advantage for NVIDIA, which matters enormously for texture-heavy scenes and compute kernels that access large datasets. Clock speeds also favor the GTX 960: it runs at 1,127 MHz base and 1,178 MHz boost, versus the R9 M375’s 1,000 MHz base and 1,015 MHz boost. The combination of more cores, higher clocks, and far faster memory makes the GTX 960 a categorically stronger part on paper.

API support shows a split as well. The R9 M375 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The GTX 960 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The higher DirectX feature level on the GTX 960 (12_1 versus 11_1) indicates better support for advanced rendering features, while the newer Vulkan version on the NVIDIA card suggests more recent driver optimizations. Both are end-of-life products, but the R9 M375 was released on May 4, 2015, while the GTX 960 came earlier on January 21, 2015.

Where Each One Wins

The R9 M375 wins in Vulkan compute, posting a 4.9% higher score than the GTX 960. This makes it the better choice for applications that rely heavily on Vulkan’s compute shaders, particularly if the workload is not memory-bandwidth bound. The R9 M375 also holds a higher average benchmark score (10,070 versus 9,273) and a better percentile ranking (48th versus 45th), which suggests it is more consistent across the limited set of tests available. For users running older or lighter titles that favor GCN 1.0’s driver profile, the R9 M375 may deliver a smoother experience relative to its specs.

The GTX 960 wins decisively in OpenCL, where its 18,925 score crushes the R9 M375 by 44.7%. This is the kind of result that matters for GPU-accelerated productivity tasks, video encoding, and any compute workload that scales with raw FP32 throughput and memory bandwidth. The GTX 960 also wins on every raw specification: more shading units, more TMUs, more ROPs, higher clocks, and 3.9x the memory bandwidth. In any game or application that is not Vulkan-limited, the GTX 960 should be substantially faster. The 3DMark Steel Nomad DX12 score of 162, while low in absolute terms, is a data point the R9 M375 simply does not have, indicating the NVIDIA card can run modern DX12 workloads that the AMD part cannot even attempt.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R9 M375 has an average benchmark score of 10,070, while the NVIDIA GeForce GTX 960 averages 9,273, a difference of 797 points in AMD’s favor.

Q: How large is the OpenCL performance gap?

A: The GTX 960 scores 18,925 in Geekbench OpenCL versus the R9 M375’s 10,457, giving NVIDIA a 44.7% lead in that test.

Q: Does the R9 M375 win any benchmark?

A: Yes, the R9 M375 wins the Geekbench Vulkan test with 9,682 points against the GTX 960’s 9,231, a 4.9% margin.

Q: What is the memory bandwidth difference?

A: The GTX 960 has 112.2 GB/s of bandwidth, while the R9 M375 has 28.80 GB/s, making the NVIDIA card roughly 3.9 times faster in memory throughput.

Q: Which GPU has more shading units?

A: The GTX 960 has 1,024 shading units, compared to 640 on the R9 M375, a 60% increase.

Q: What is the transistor count for each chip?

A: The R9 M375’s Tropo chip contains 1,500 million transistors, while the GTX 960’s GM206 chip contains 2,940 million transistors.

Specification Differences

The two GPUs differ on nearly every measurable specification. The R9 M375 uses the Tropo chip with GCN 1.0 architecture, while the GTX 960 uses GM206 with Maxwell 2.0. Transistor counts are 1,500 million versus 2,940 million, and die sizes are 123 mm² versus 228 mm². Base clocks are 1,000 MHz versus 1,127 MHz, with boost clocks of 1,015 MHz versus 1,178 MHz. Memory types differ: DDR3 at 900 MHz (1,800 Mbps effective) for AMD, GDDR5 at 1,753 MHz (7 Gbps effective) for NVIDIA. Both have 2 GB and a 128-bit bus, but bandwidth is 28.80 GB/s versus 112.2 GB/s. Shading units are 640 versus 1,024, TMUs are 40 versus 64, and ROPs are 16 versus 32. Pixel rates are 16.24 GPixel/s versus 37.70 GPixel/s, and texture rates are 40.60 GTexel/s versus 75.39 GTexel/s. FP32 performance is 1,299.2 GFLOPS versus 2.413 TFLOPS. The GTX 960 has a TDP of 120 W, a dual-slot cooler, a 1x 6-pin power connector, a 300 W suggested PSU, and a 241 mm (9.5 inches) length. The R9 M375 has no listed TDP, slot width, power connectors, or dimensions. The GTX 960 also has display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2) while the R9 M375 has none listed. DirectX support is 12 (11_1) for AMD versus 12 (12_1) for NVIDIA, and Vulkan versions are 1.2.170 versus 1.4. The GTX 960 has a launch MSRP of 199 USD; the R9 M375 has no launch MSRP listed. The R9 M375’s predecessor is “Solar System” and successor is “Polaris Mobile,” while the GTX 960’s predecessor is “GeForce 700” and successor is “GeForce 10.”

The Verdict

The data points to a clear, if nuanced, recommendation. The NVIDIA GeForce GTX 960 is the superior GPU for anyone prioritizing raw compute performance, particularly in OpenCL applications. Its 44.7% lead in that benchmark, combined with more than double the shading units, double the ROPs, and nearly four times the memory bandwidth, makes it the obvious choice for gaming, video editing, or any workload that can leverage its massive parallelism. The higher DirectX 12 feature level (12_1 versus 11_1) and newer Vulkan 1.4 support also give it a longer software lifespan. The 199 USD launch MSRP, while not directly comparable to the R9 M375’s absent pricing, positions it as a mainstream desktop part.

The AMD Radeon R9 M375 should be selected only if the workload is specifically Vulkan-centric, where it wins by 4.9%, or if the user values consistency across APIs, given its higher average score and percentile ranking. Its lower transistor count, smaller die, and slower memory make it a less capable part in absolute terms, but its Vulkan win shows that architecture and driver efficiency can sometimes overcome hardware deficits. For a mobile GPU, the R9 M375’s lower specs likely translate to lower power consumption, though no TDP is listed. Ultimately, the GTX 960 is the stronger choice for nearly every scenario except the narrow Vulkan case, and even there the margin is slim. The R9 M375 is a respectable part that wins one test, but the GTX 960 wins the tests that matter more.