AMD Radeon R7 M350 vs NVIDIA GeForce GTX 765M Comparison

Head-to-Head Benchmarks

The recorded data shows a clear, but not overwhelming, win for the NVIDIA GeForce GTX 765M in the direct comparisons. Across the two shared benchmark tests, the GTX 765M takes both wins, though the margin varies significantly by workload.

In Geekbench OpenCL, the NVIDIA card scores 7176 against the AMD Radeon R7 M350's 6991. That is a 2.6% lead, a narrow gap that puts both cards in the same general performance class for compute-heavy tasks. The delta is small enough that driver versions, thermal state, or workload specifics could easily flip the result in any given run.

The Vulkan test tells a different story. Here the GTX 765M posts 6714 versus 5662 for the R7 M350, a 15.7% advantage. That is a substantial margin, indicating the NVIDIA architecture handles the lower-level API significantly better than the AMD part in this specific scenario. For any application or game that leans on Vulkan, the GTX 765M is clearly the stronger option.

Looking at the broader database averages, the picture becomes more nuanced. The R7 M350 has an average benchmark score of 6327, placing it at the 36th percentile of all GPUs. The GTX 765M, despite winning both head-to-head tests, has a lower average score of 5501 and sits at the 32nd percentile. This discrepancy is worth noting: the GTX 765M wins the direct matchups, but the R7 M350 has a higher overall standing when measured against the full GPU landscape. The R7 M350's nearest rivals hover very close to its average, with the AMD Radeon Pro WX 4100 at 6330 (0% delta), the NVIDIA Quadro K620 at 6282 (0.7% behind), and the NVIDIA GeForce RTX 5070 Ti SUPER and RTX 4070 Ti SUPER AD102 both at 6270 (0.9% behind). The GTX 765M's nearest rivals are equally tight, with the NVIDIA GeForce MX130 at 5508 (0.1% ahead), the AMD FirePro M4000 at 5537 (0.7% ahead), and the AMD Radeon R7 M440 at 5483 (0.3% behind).

The raw score data suggests that while the GTX 765M is faster in the two direct tests, the R7 M350 has a better profile across the entire benchmark suite in the database. This is likely because the OpenCL result dominates the average, and the R7 M350's OpenCL score is close enough to the GTX 765M's that the higher Vulkan score for NVIDIA does not fully compensate in the aggregate.

Architecture Differences

The two GPUs come from different architectural generations and design philosophies. The AMD Radeon R7 M350 uses the Meso chip built on GCN 3.0 architecture, part of the Gem System generation under the R7 M300 family. The NVIDIA GeForce GTX 765M uses the GK106 chip with Kepler architecture, from the GeForce 700M family. Both are fabricated by TSMC on a 28 nm process node, so the underlying manufacturing technology is identical.

The transistor budgets differ notably. The NVIDIA chip packs 2,540 million transistors on a 221 mm² die, while the AMD chip has 1,550 million transistors on a 125 mm² die. The AMD die is denser at 12.4 million transistors per mm² versus 11.5 million for NVIDIA, but the NVIDIA chip is physically larger and has nearly two-thirds more transistors overall. This scale difference shows up in the execution resources: the GTX 765M has 768 shading units, 64 texture mapping units, and 16 render output units, while the R7 M350 has 384 shading units, 24 TMUs, and 8 ROPs. The NVIDIA part has exactly double the shaders and TMUs, and double the ROPs.

Clock behavior also diverges. The R7 M350 runs at a 1000 MHz base and 1015 MHz boost, while the GTX 765M runs at 797 MHz base and 863 MHz boost. Despite the lower clocks, the GTX 765M achieves higher throughput because of its wider execution resources. The pixel rate for the GTX 765M is 13.81 GPixel/s versus 8.120 GPixel/s for the AMD card, and the texture rate is 55.23 GTexel/s versus 24.36 GTexel/s. FP32 compute follows the same pattern: 1,325.6 GFLOPS for NVIDIA versus 779.5 GFLOPS for AMD. The AMD card does support FP16 at a 1:1 ratio (779.5 GFLOPS), while the NVIDIA part has no recorded FP16 capability.

Memory configurations are another clear split. The R7 M350 comes with 4 GB of DDR3 on a 64-bit bus, yielding 16.00 GB/s of bandwidth. The GTX 765M has 2 GB of GDDR5 on a 128-bit bus, delivering 64.13 GB/s. That is roughly four times the memory bandwidth, which heavily influences the Vulkan result and any bandwidth-sensitive workload. Memory clocks are 1000 MHz (2 Gbps effective) for AMD and 1002 MHz (4 Gbps effective) for NVIDIA.

API support differs in DirectX and Vulkan versions. The AMD card lists DirectX 12 (12_0) while NVIDIA lists DirectX 12 (11_0), so the AMD part supports a higher feature level. Vulkan is close: 1.2.170 for AMD and 1.2.175 for NVIDIA. Both support OpenGL 4.6. The R7 M350 uses a PCIe 3.0 x8 interface, while the GTX 765M uses an MXM-B (3.0) module interface. The GTX 765M has a recorded TDP of 75 W and no power connectors, while the R7 M350 has no TDP listed. Display outputs for the GTX 765M are listed as portable device dependent, and the R7 M350 has no display output data recorded.

Where Each One Wins

The GTX 765M wins in raw compute throughput, memory bandwidth, and Vulkan performance. Its 64.13 GB/s of bandwidth versus 16.00 GB/s is a massive advantage for any workload that streams data, such as high-resolution textures, large compute buffers, or frame-heavy rendering passes. The doubled shader count and doubled ROPs also give it a clear edge in fill-rate-bound scenarios. The 15.7% Vulkan lead is the single largest gap in the head-to-head data, so applications built on Vulkan will favor the NVIDIA part decisively. The 2.6% OpenCL win is smaller but still consistent.

The R7 M350 wins in a few specific areas. It has 4 GB of memory versus 2 GB, which matters for workloads that need larger working sets, even if the bandwidth is much lower. The DirectX 12 (12_0) feature level is ahead of the GTX 765M's 12 (11_0), so games or applications that use DirectX 12 features beyond the 11_0 tier will see better compatibility or performance on the AMD part. Its higher average benchmark score (6327 versus 5501) and better percentile standing (36th versus 32nd) indicate that across the full range of database tests, the R7 M350 is more balanced. The FP16 support at a 1:1 ratio is also a feature the GTX 765M lacks entirely, which can matter for compute tasks that use half-precision arithmetic.

The transistor density advantage (12.4M per mm² versus 11.5M) suggests the AMD die is more efficient in terms of resources per area, though the NVIDIA chip uses its larger die to deliver more absolute performance.

The Verdict

The data points to different buyers for each card. The NVIDIA GeForce GTX 765M is the pick for anyone prioritizing raw performance in the direct benchmarks. It wins both head-to-head tests, with a particularly strong showing in Vulkan where it leads by 15.7%. Its memory bandwidth is four times higher, its shader count is double, and its FP32 throughput is 70% higher. If the workload is gaming, OpenCL compute, or anything that moves large amounts of data, the GTX 765M is the better choice. Its nearest rivals in the database (the MX130 at 0.1% ahead, the FirePro M4000 at 0.7% ahead) show it sits in a well-populated performance tier, so it is not an outlier in either direction.

The AMD Radeon R7 M350 is the pick for users who need more memory capacity (4 GB versus 2 GB), DirectX 12 (12_0) features, or FP16 compute. Its higher average benchmark score (6327) and percentile rank (36th versus 32nd) suggest it holds up better across a wider variety of tests, even though it loses the two direct matchups. The 0% delta to the Radeon Pro WX 4100 and the 0.7% gap to the Quadro K620 place it in a slightly higher overall tier than the GTX 765M's cluster. For a laptop or small-form-factor system where the workload is mixed rather than gaming-focused, the R7 M350 might be the safer bet.

There is no clear overall winner. The GTX 765M wins where it counts for gaming and bandwidth-heavy tasks. The R7 M350 wins on features and average standing. Choose based on the workload, not the brand.

FAQ

Q: Which GPU wins in OpenCL performance?

A: The NVIDIA GeForce GTX 765M wins the Geekbench OpenCL test with a score of 7176 versus 6991 for the AMD Radeon R7 M350, a 2.6% margin.

Q: How large is the Vulkan performance gap?

A: The GTX 765M scores 6714 in Geekbench Vulkan, while the R7 M350 scores 5662. That is a 15.7% lead for the NVIDIA card, the largest difference in any recorded benchmark between the two.

Q: Does the AMD card have more memory?

A: Yes. The R7 M350 has 4 GB of DDR3 memory, while the GTX 765M has 2 GB of GDDR5. However, the GTX 765M's memory bandwidth is 64.13 GB/s versus 16.00 GB/s for the AMD part.

Q: Which card has better DirectX support?

A: The AMD Radeon R7 M350 supports DirectX 12 (12_0), while the NVIDIA GeForce GTX 765M supports DirectX 12 (11_0). The AMD card has the higher feature level.

Q: How do their average benchmark scores compare?

A: The R7 M350 has a higher average benchmark score of 6327, placing it at the 36th percentile of all GPUs. The GTX 765M averages 5501, at the 32nd percentile.

Q: Is the GTX 765M more powerful in raw compute?

A: Yes. The GTX 765M has 768 shading units versus 384, and its FP32 throughput is 1,325.6 GFLOPS versus 779.5 GFLOPS for the R7 M350.

Specification Differences

The following specifications differ between the two GPUs:

• Chip: Meso (AMD) versus GK106 (NVIDIA)
• Architecture: GCN 3.0 versus Kepler
• Generation: Gem System (R7 M300) versus GeForce 700M
• Transistors: 1,550 million versus 2,540 million
• Die Size: 125 mm² versus 221 mm²
• Transistor Density: 12.4M / mm² versus 11.5M / mm²
• Base Clock: 1000 MHz versus 797 MHz
• Boost Clock: 1015 MHz versus 863 MHz
• Memory Clock: 1000 MHz (2 Gbps effective) versus 1002 MHz (4 Gbps effective)
• Memory Size: 4 GB versus 2 GB
• Memory Type: DDR3 versus GDDR5
• Memory Bus Width: 64 bit versus 128 bit
• Memory Bandwidth: 16.00 GB/s versus 64.13 GB/s
• Shading Units: 384 versus 768
• TMUs: 24 versus 64
• ROPs: 8 versus 16
• Pixel Rate: 8.120 GPixel/s versus 13.81 GPixel/s
• Texture Rate: 24.36 GTexel/s versus 55.23 GTexel/s
• FP32: 779.5 GFLOPS versus 1,325.6 GFLOPS
• FP16: 779.5 GFLOPS (1:1) versus null
• TDP: null versus 75 W
• Slot Width: null versus MXM Module
• Power Connectors: null versus None
• Bus Interface: PCIe 3.0 x8 versus MXM-B (3.0)
• Display Outputs: null versus Portable Device Dependent
• DirectX: 12 (12_0) versus 12 (11_0)
• Vulkan: 1.2.170 versus 1.2.175
• Release Date: 2015-05-04 versus 2013-05-29
• Predecessor: Solar System versus GeForce 600M
• Successor: Polaris Mobile versus GeForce 800M