AMD Radeon HD 8790M vs AMD Radeon R7 M350 Comparison

The Verdict

The AMD Radeon R7 M350 and AMD Radeon HD 8790M are both end-of-life mobile GPUs, but they represent different generations of AMD's GCN architecture and deliver different strengths. The recorded data shows a split decision: the R7 M350 wins in OpenCL compute workloads with a 39.3% margin, while the HD 8790M wins in Vulkan graphics workloads by 11%.

For users prioritizing raw compute throughput, especially in OpenCL-accelerated applications, the R7 M350 is the clear choice. Its 779.5 GFLOPS FP32 performance and 8.120 GPixel/s pixel rate give it a mathematical edge in general-purpose GPU tasks. The higher base clock of 1000 MHz versus 850 MHz and boost clock of 1015 MHz versus 900 MHz contributes directly to this advantage.

For users who care about graphics API performance, particularly Vulkan, the HD 8790M is the better pick. Its 64.00 GB/s memory bandwidth, enabled by a 128-bit bus with GDDR5, is four times the bandwidth of the R7 M350's 16.00 GB/s. This explains its Vulkan score of 6365 versus 5662. The HD 8790M also supports MXM-A (3.0) form factor with no power connectors, making it a straightforward drop-in replacement in systems designed for that standard.

The database places the R7 M350 at the 36th percentile among all GPUs, with an average benchmark score of 6327. Its nearest rivals include the AMD Radeon Pro WX 4100 (6330, 0% delta), NVIDIA Quadro K620 (6282, 0.7% ahead), and two NVIDIA GeForce RTX 40/50 series cards (6270, 0.9% ahead). The HD 8790M sits at the 33rd percentile with an average score of 5691, competing closely with Intel Iris Pro Graphics P6300 (5712, 0.4% behind), NVIDIA GeForce GTX 670MX (5721, 0.5% behind), and NVIDIA GeForce GTX 550 Ti (5731, 0.7% behind). The R7 M350 is roughly 11.2% ahead of the HD 8790M in average benchmark score.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R7 M350 has an average benchmark score of 6327, while the AMD Radeon HD 8790M scores 5691. The R7 M350 is ahead by approximately 11.2%.

Q: Which GPU wins in OpenCL performance?

A: The AMD Radeon R7 M350 wins in Geekbench OpenCL with a score of 6991 versus 5017 for the HD 8790M, a delta of 39.3%.

Q: Which GPU wins in Vulkan performance?

A: The AMD Radeon HD 8790M wins in Geekbench Vulkan with a score of 6365 versus 5662 for the R7 M350, a delta of 11% in favor of the HD 8790M.

Q: What are the memory differences between these two GPUs?

A: The R7 M350 has 4 GB of DDR3 memory on a 64-bit bus with 16.00 GB/s bandwidth. The HD 8790M has 2 GB of GDDR5 memory on a 128-bit bus with 64.00 GB/s bandwidth.

Q: Which GPU has higher clock speeds?

A: The R7 M350 has a base clock of 1000 MHz and boost clock of 1015 MHz. The HD 8790M has a base clock of 850 MHz and boost clock of 900 MHz.

Q: What is the form factor difference?

A: The R7 M350 uses a PCIe 3.0 x8 interface, while the HD 8790M uses an MXM-A (3.0) module with no power connectors and portable device dependent display outputs.

Architecture Differences

The two GPUs come from different GCN generations. The R7 M350 uses GCN 3.0 architecture on the Meso chip, a 28 nm TSMC process with 1,550 million transistors on a 125 mm² die. The HD 8790M uses GCN 1.0 architecture on the Mars chip, also 28 nm TSMC, but with 950 million transistors on a 77 mm² die. Transistor density is nearly identical: 12.4M per mm² for the R7 M350 versus 12.3M per mm² for the HD 8790M.

Both GPUs share the same core counts: 384 shading units, 24 texture mapping units, and 8 raster operation units. The R7 M350 supports DirectX 12 (12_0), while the HD 8790M supports DirectX 12 (11_1). Both support OpenGL 4.6 and Vulkan 1.2.170.

The R7 M350 belongs to the Gem System (R7 M300) generation, succeeding the Solar System generation to which the HD 8790M belongs. The HD 8790M's predecessor is London, and its successor is Gem System. The R7 M350's predecessor is Solar System, and its successor is Polaris Mobile.

The R7 M350 has no listed FP16 support ratio, while the HD 8790M has no FP16 data at all. Both have identical FP32 compute per shading unit proportionally, but the R7 M350's higher clocks push its raw FP32 to 779.5 GFLOPS versus 691.2 GFLOPS for the HD 8790M.

Specification Differences

The R7 M350 and HD 8790M differ in several key specification fields:

• Process node: Both are 28 nm TSMC, but transistor counts differ: 1,550 million versus 950 million.
• Die size: The R7 M350 is 125 mm², the HD 8790M is 77 mm².
• Base clock: 1000 MHz versus 850 MHz.
• Boost clock: 1015 MHz versus 900 MHz.
• Memory clock: Both run at 1000 MHz, but the R7 M350 has 2 Gbps effective, while the HD 8790M has 4 Gbps effective.
• Memory size: 4 GB DDR3 versus 2 GB GDDR5.
• Memory bus width: 64-bit versus 128-bit.
• Memory bandwidth: 16.00 GB/s versus 64.00 GB/s.
• Pixel rate: 8.120 GPixel/s versus 7.200 GPixel/s.
• Texture rate: 24.36 GTexel/s versus 21.60 GTexel/s.
• FP32: 779.5 GFLOPS versus 691.2 GFLOPS.
• Bus interface: PCIe 3.0 x8 versus MXM-A (3.0).
• Power connectors: Not listed for the R7 M350, none for the HD 8790M.
• Display outputs: Not listed for the R7 M350, portable device dependent for the HD 8790M.
• Release date: 2015-05-04 versus 2013-03-31.

Head-to-Head Benchmarks

The two recorded benchmarks show a clear trade-off. In Geekbench OpenCL, the R7 M350 scores 6991 against the HD 8790M's 5017, a 39.3% advantage. This is a substantial margin that reflects the R7 M350's higher clocks and newer GCN 3.0 architecture. The FP32 output of 779.5 GFLOPS versus 691.2 GFLOPS, combined with the higher pixel rate of 8.120 GPixel/s versus 7.200 GPixel/s, drives this OpenCL dominance.

In Geekbench Vulkan, the tables turn. The HD 8790M scores 6365 against the R7 M350's 5662, an 11% advantage for the older card. This result correlates strongly with memory bandwidth: the HD 8790M's 64.00 GB/s is four times the R7 M350's 16.00 GB/s. Vulkan workloads often stress memory throughput, and the 128-bit GDDR5 configuration gives the HD 8790M a decisive edge despite lower clocks.

The average benchmark scores reflect this split. The R7 M350 averages 6327 across both tests, while the HD 8790M averages 5691. The R7 M350's nearest rival, the AMD Radeon Pro WX 4100, scores 6330 with a 0% delta, meaning the R7 M350 is essentially tied with that workstation card. The HD 8790M's nearest rival, the Intel Iris Pro Graphics P6300, scores 5712, just 0.4% ahead of the HD 8790M.

Where Each One Wins

The R7 M350 wins in compute-oriented scenarios. Its OpenCL score of 6991 is 39.3% higher, and its average benchmark score of 6327 places it in the 36th percentile versus the HD 8790M's 33rd percentile. Applications that leverage OpenCL for general-purpose computation, such as video encoding, physics simulation, or data processing, will benefit from the R7 M350's higher FP32 throughput of 779.5 GFLOPS and its GCN 3.0 feature set. The R7 M350 also has a higher pixel rate (8.120 GPixel/s) and texture rate (24.36 GTexel/s), which helps in tasks that rasterize geometry.

The HD 8790M wins in graphics API performance, specifically Vulkan. Its score of 6365 is 11% higher, and its memory bandwidth advantage (64.00 GB/s versus 16.00 GB/s) makes it more suitable for texture-heavy workloads, high-resolution rendering, and framebuffer-intensive operations. The HD 8790M's 128-bit bus and GDDR5 memory type are better suited to the random access patterns common in graphics workloads. Its MXM-A (3.0) form factor also makes it compatible with a standardized mobile module slot, which may simplify upgrades in systems that accept MXM modules.

For users choosing between these two end-of-life GPUs, the decision hinges on workload type. If OpenCL compute matters more, the R7 M350 is the superior part. If Vulkan graphics performance is the priority, the HD 8790M holds the advantage. Given that both are end-of-life, availability and system compatibility will likely dictate the practical choice, but the benchmark data clearly separates their strengths.