AMD Radeon R7 M350 vs AMD Radeon R7 M365X Comparison

Where Each One Wins

The recorded data splits these two mobile GPUs cleanly. The AMD Radeon R7 M350 wins both head-to-head benchmark comparisons, taking the Geekbench OpenCL test and the Geekbench Vulkan test. The AMD Radeon R7 M365X does not win either recorded benchmark. That gives the M350 a 2-0 record in direct comparison.

Looking at the broader database, the M350 sits at the 36th percentile among all GPUs, while the M365X sits at the 32nd percentile. That four-point gap in percentile placement reflects the average benchmark scores: the M350 averages 6327 points, and the M365X averages 5416 points. The difference is 911 points, or roughly 16.8% of the M365X's average score.

For workloads that lean on OpenCL compute, the M350 is the stronger choice. Its OpenCL score of 6991 beats the M365X's 5939 by 17.7%. For Vulkan-based tasks, the M350 again leads, scoring 5662 against 4893, a 15.7% advantage. The M365X has no recorded test where it pulls ahead.

The M350's nearest rivals in the database include the AMD Radeon Pro WX 4100 (6330 average, 0% delta), the NVIDIA Quadro K620 (6282, 0.7% ahead), and the NVIDIA GeForce RTX 5070 Ti SUPER (6270, 0.9% ahead). The M365X's nearest rivals include the AMD Radeon 610M (5444, 0.5% ahead), the Intel UHD Graphics P630 (5370, 0.9% behind), and the NVIDIA Quadro M4000 (5467, 0.9% ahead). The M350 competes in a higher performance tier than the M365X, as shown by the overlapping rival sets.

Architecture Differences

The two GPUs share the same 28 nm process node and TSMC as the foundry, but their underlying chips differ. The M350 uses the Meso chip with GCN 3.0 architecture. The M365X uses the Litho chip with GCN 1.0 architecture. This is a significant generational gap within the same R7 M300 series.

Transistor counts differ notably. The M350 packs 1,550 million transistors on a 125 mm² die, giving a transistor density of 12.4 million per mm². The M365X has 950 million transistors on a 77 mm² die, with a density of 12.3 million per mm². The M350 has roughly 63% more transistors and a 62% larger die, yet the density is nearly identical, indicating the M350's extra silicon goes toward more compute resources rather than tighter packing.

Both GPUs have the same count of shading units (384), texture mapping units (24), and render output units (8). The architectural advantage for the M350 comes from the newer GCN 3.0 design and its higher clock speeds. The M350 runs at a base clock of 1000 MHz and a boost clock of 1015 MHz. The M365X has no base or boost clock listed in the database, so the comparison relies on the recorded throughput figures.

The M350 supports DirectX 12 (12_0), while the M365X supports DirectX 12 (11_1). Both support OpenGL 4.6 and Vulkan 1.2.170. The FP16 capability also differs: the M350 delivers 779.5 GFLOPS FP16 with a 1:1 ratio to FP32, while the M365X has no FP16 figure recorded. This suggests the M350 has a more complete feature set for modern compute workloads.

Head-to-Head Benchmarks

The Geekbench OpenCL test shows the M350 scoring 6991 against the M365X's 5939. The M350 wins by 17.7%. In absolute terms, that is a 1052-point gap. This is the larger of the two victories.

The Geekbench Vulkan test shows the M350 at 5662 and the M365X at 4893. The M350 wins by 15.7%, a 769-point gap. Both tests confirm the same ordering, with the M350 consistently ahead by roughly 16% to 18%.

The M350's pixel rate is 8.120 GPixel/s versus the M365X's 6.600 GPixel/s, a 23% advantage. The texture rate is 24.36 GTexel/s versus 19.80 GTexel/s, a 23% advantage. FP32 compute is 779.5 GFLOPS versus 633.6 GFLOPS, a 23% advantage. These throughput figures align closely with the benchmark results, though the benchmark deltas are slightly smaller, suggesting real-world workloads do not perfectly scale with raw rates.

The M350's average benchmark score of 6327 places it just 0.7% above the NVIDIA Quadro K620 (6282) and 0.9% above the NVIDIA GeForce RTX 4070 Ti SUPER AD102 (6270). The M365X's average of 5416 places it 0.5% behind the AMD Radeon 610M (5444) and 0.9% behind the NVIDIA Quadro M4000 (5467). In the database ranking, the M350 sits one full tier above the M365X, with no overlap in their nearest rival groups.

Specification Differences

The two GPUs differ in several key specification fields. Memory size: the M350 has 4 GB of DDR3, while the M365X has 1024 MB of GDDR5. Memory bus width: the M350 uses a 64-bit bus, the M365X uses a 128-bit bus. Memory bandwidth: the M350 delivers 16.00 GB/s, the M365X delivers 64.00 GB/s, a 4x advantage for the M365X.

Memory clock: the M350 runs at 1000 MHz with 2 Gbps effective, the M365X runs at 1000 MHz with 4 Gbps effective. The M365X's GDDR5 memory and wider bus give it substantially more bandwidth, yet this does not translate into a benchmark win.

Chip and architecture: Meso with GCN 3.0 versus Litho with GCN 1.0. Transistors: 1,550 million versus 950 million. Die size: 125 mm² versus 77 mm². Transistor density: 12.4M per mm² versus 12.3M per mm², a negligible difference.

Base and boost clocks: the M350 lists 1000 MHz base and 1015 MHz boost, the M365X lists neither. Pixel rate: 8.120 GPixel/s versus 6.600 GPixel/s. Texture rate: 24.36 GTexel/s versus 19.80 GTexel/s. FP32: 779.5 GFLOPS versus 633.6 GFLOPS. FP16: 779.5 GFLOPS (1:1) versus null. DirectX: 12 (12_0) versus 12 (11_1).

Both share the same PCIe 3.0 x8 interface, same 28 nm process, same TSMC foundry, same generation (Gem System R7 M300), same release date, same predecessor (Solar System), same successor (Polaris Mobile), same production status (end-of-life), and no TDP listed for either.

FAQ

Q: Which GPU wins the OpenCL benchmark?

A: The AMD Radeon R7 M350 wins Geekbench OpenCL with a score of 6991 against the M365X's 5939, a 17.7% advantage.

Q: Which GPU wins the Vulkan benchmark?

A: The AMD Radeon R7 M350 wins Geekbench Vulkan with a score of 5662 against the M365X's 4893, a 15.7% advantage.

Q: How much memory does each GPU have?

A: The M350 has 4 GB of DDR3 memory on a 64-bit bus. The M365X has 1024 MB of GDDR5 memory on a 128-bit bus.

Q: Which GPU has higher memory bandwidth?

A: The M365X has 64.00 GB/s of bandwidth, four times the M350's 16.00 GB/s. Despite this, the M365X loses both recorded benchmarks.

Q: What are the architecture differences?

A: The M350 uses the Meso chip with GCN 3.0 architecture, while the M365X uses the Litho chip with GCN 1.0. The M350 has 1,550 million transistors on 125 mm², the M365X has 950 million on 77 mm².

Q: Do both GPUs support the same DirectX version?

A: No. The M350 supports DirectX 12 (12_0), while the M365X supports DirectX 12 (11_1). Both support OpenGL 4.6 and Vulkan 1.2.170.

The Verdict

The data points to one clear choice for compute performance: the AMD Radeon R7 M350. It wins both head-to-head benchmarks, has a higher average score (6327 versus 5416), and sits at a higher percentile (36th versus 32nd). The M350 also has a more modern architecture (GCN 3.0 versus GCN 1.0), more transistors (1,550 million versus 950 million), and higher throughput in pixel rate, texture rate, and FP32 compute.

The M365X does have one notable advantage: memory bandwidth. Its 64.00 GB/s GDDR5 setup with a 128-bit bus is four times the M350's 16.00 GB/s DDR3 with a 64-bit bus. It also has more memory per pin, running at 4 Gbps effective versus 2 Gbps. Yet in the recorded benchmarks, this bandwidth advantage does not overcome the M350's compute lead.

For users choosing between these two end-of-life mobile GPUs, the M350 is the stronger candidate for OpenCL and Vulkan workloads. The M365X may be preferable in scenarios where memory bandwidth is the bottleneck, but no recorded benchmark supports that conclusion. The M350 wins both tests, and the specification sheet explains why: newer architecture, higher clocks, and roughly 23% more compute throughput across the board.

The M350's nearest rivals include professional workstation GPUs like the AMD Radeon Pro WX 4100 and NVIDIA Quadro K620, with deltas of 0% and 0.7% respectively. The M365X's rivals are lower-tier parts like the AMD Radeon 610M and Intel UHD Graphics P630. The database places the M350 in a higher performance class, and the benchmark results confirm that placement. Choose the M350 for better compute performance; choose the M365X only if the GDDR5 memory bandwidth is specifically required for a workload not represented in these tests.