AMD Radeon HD 8790M vs AMD Radeon R7 M365X Comparison

AMD Radeon HD 8790M and AMD Radeon R7 M365X are both 28 nm GCN 1.0 parts with identical transistor counts and die sizes, but benchmark results show they are not interchangeable. The data reveals a split decision: the R7 M365X wins the OpenCL test by a significant margin, while the HD 8790M dominates Vulkan performance. Each GPU holds a distinct advantage in a different API workload, making the choice between them dependent on the target application.

Head-to-Head Benchmarks

The two GPUs split their benchmark wins exactly one apiece, but the magnitude of each victory is uneven. In the Geekbench OpenCL test, the AMD Radeon R7 M365X scores 5939, which is 15.5% higher than the HD 8790M’s 5017. This is a decisive lead in compute-oriented workloads, suggesting the R7 M365X has a notable edge in general-purpose GPU computing tasks that leverage OpenCL. The HD 8790M, by contrast, trails by nearly a sixth of its own score in this test, a gap that cannot be dismissed as a minor variance.

The reverse is true in the Geekbench Vulkan test, where the HD 8790M turns the tables with a score of 6365 against the R7 M365X’s 4893. That represents a 30.1% advantage for the older part, a massive swing that dwarfs the OpenCL delta. The HD 8790M’s Vulkan score is not merely better; it is categorically superior, indicating that its driver stack or hardware implementation handles the Vulkan API far more efficiently. This 30.1% lead is the largest margin in either direction across both tests, making it the single most important differentiator in this comparison.

Looking at the broader context, the HD 8790M’s average benchmark score of 5691 places it in the 33rd percentile of all GPUs, while the R7 M365X’s average of 5416 lands in the 32nd percentile. The HD 8790M’s higher average is driven entirely by its Vulkan performance, as its OpenCL score drags the mean down. The R7 M365X, conversely, benefits from its OpenCL strength but suffers in Vulkan. The nearest rival data reinforces this split: the HD 8790M’s closest competitor, the Intel Iris Pro Graphics P6300, averages 5712, a delta of -0.4%, while the R7 M365X’s nearest rival, the AMD Radeon 610M, averages 5444, a delta of -0.5%. Both GPUs sit within 1% of their respective nearest rivals, but the HD 8790M’s average is 275 points higher in absolute terms.

Where Each One Wins

The AMD Radeon R7 M365X is the clear winner for OpenCL compute tasks. Its 5939 score in that benchmark is not only higher than the HD 8790M’s but also places it ahead of several competitors in its own rival list. For applications that rely on OpenCL for acceleration — such as certain video encoding, physics simulation, or general-purpose GPU computing — the R7 M365X offers measurably better performance. The 15.5% delta is substantial enough to be felt in real workloads, and the R7 M365X’s position relative to rivals like the Intel UHD Graphics P630 (which it beats by 0.9%) suggests it is competitive within its class.

The AMD Radeon HD 8790M, on the other hand, is the definitive choice for Vulkan-based workloads. Its 6365 Vulkan score is 30.1% ahead of the R7 M365X, a margin that is impossible to ignore for any application built on Vulkan, including modern game engines and compute frameworks that use the API. The HD 8790M’s Vulkan performance is so strong that it pulls its average score above the R7 M365X despite losing OpenCL. For users running Vulkan titles or Vulkan-based compute, the HD 8790M is the only rational pick between these two.

The wins are mutually exclusive in practice. No single workload benefits from both GPUs equally; the R7 M365X excels where OpenCL is the API, and the HD 8790M excels where Vulkan is the API. There is no overlap in their strengths, which means the decision hinges entirely on which API the user’s software stack prioritizes.

Architecture Differences

Both GPUs share the same fundamental architecture: GCN 1.0, fabricated on TSMC’s 28 nm process node. The die size is identical at 77 mm², and both pack 950 million transistors, resulting in the same transistor density of 12.3M per square millimeter. The shading unit count is also identical at 384, with 24 texture mapping units and 8 render output units. Memory configurations match in type and bus width: both use GDDR5 with a 128-bit interface, and both run at 1000 MHz with 4 Gbps effective speed, yielding identical 64.00 GB/s bandwidth.

The critical differences lie in clock speeds and derived rates. The HD 8790M has explicit base and boost clocks of 850 MHz and 900 MHz, respectively, while the R7 M365X lists no base or boost clocks in the data. This clock advantage translates directly to throughput: the HD 8790M achieves 7.200 GPixel/s pixel rate and 21.60 GTexel/s texture rate, whereas the R7 M365X is lower at 6.600 GPixel/s and 19.80 GTexel/s. Floating-point performance follows the same pattern, with the HD 8790M delivering 691.2 GFLOPS FP32 versus the R7 M365X’s 633.6 GFLOPS. The memory clock is the same, but the GPU core clock difference explains why the HD 8790M has higher compute throughput in raw terms.

The R7 M365X differs in memory capacity, offering 1024 MB (1 GB) compared to the HD 8790M’s 2 GB. This is a notable divergence given identical memory type and bus width. The bus interface also differs: the HD 8790M uses MXM-A (3.0) in an MXM Module slot, while the R7 M365X uses PCIe 3.0 x8. The HD 8790M lists display outputs as "Portable Device Dependent," while the R7 M365X has no display output information. Both support the same API levels: DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon HD 8790M has an average benchmark score of 5691, which is higher than the AMD Radeon R7 M365X’s 5416. The HD 8790M also sits at the 33rd percentile of all GPUs, versus the R7 M365X’s 32nd percentile.

Q: How much faster is the R7 M365X in OpenCL?

A: The R7 M365X scores 5939 in Geekbench OpenCL, which is 15.5% higher than the HD 8790M’s 5017. This is the R7 M365X’s only benchmark win.

Q: What is the Vulkan performance gap between the two?

A: The HD 8790M scores 6365 in Geekbench Vulkan, a 30.1% advantage over the R7 M365X’s 4893. This is the largest performance delta in either direction across both tests.

Q: Do both GPUs have the same memory bandwidth?

A: Yes, both have a 128-bit bus with GDDR5 memory at 1000 MHz and 4 Gbps effective speed, resulting in identical 64.00 GB/s bandwidth. However, the HD 8790M has 2 GB while the R7 M365X has 1024 MB.

Q: Are the compute capabilities different despite the same architecture?

A: Yes. The HD 8790M has higher clocks (850 MHz base, 900 MHz boost) leading to 691.2 GFLOPS FP32 and 7.200 GPixel/s, while the R7 M365X has no listed clocks and delivers 633.6 GFLOPS and 6.600 GPixel/s.

Q: Which GPU is better for modern gaming APIs?

A: For Vulkan, the HD 8790M is decisively better with its 30.1% lead. Both support DirectX 12 (11_1), but the Vulkan benchmark is the only modern API test in the data, and it favors the HD 8790M.

Specification Differences

| Specification | AMD Radeon HD 8790M | AMD Radeon R7 M365X |

|---|---|---|

| Chip | Mars | Litho |

| Generation | Solar System (HD 8700M) | Gem System (R7 M300) |

| Base Clock | 850 MHz | Not listed |

| Boost Clock | 900 MHz | Not listed |

| Memory Size | 2 GB | 1024 MB |

| Pixel Rate | 7.200 GPixel/s | 6.600 GPixel/s |

| Texture Rate | 21.60 GTexel/s | 19.80 GTexel/s |

| FP32 | 691.2 GFLOPS | 633.6 GFLOPS |

| Slot Width | MXM Module | Not listed |

| Bus Interface | MXM-A (3.0) | PCIe 3.0 x8 |

| Display Outputs | Portable Device Dependent | Not listed |

| Release Date | 2013-03-31 | 2015-05-04 |

| Predecessor | London | Solar System |

| Successor | Gem System | Polaris Mobile |

The Verdict

The data supports a clear verdict based on API preference. If your workloads rely on Vulkan, the AMD Radeon HD 8790M is the unequivocal choice. Its 6365 Vulkan score is 30.1% higher than the R7 M365X’s, and this single benchmark gives it a higher average score (5691 vs 5416) and a better percentile ranking (33rd vs 32nd). The HD 8790M also has superior raw compute rates — 691.2 GFLOPS FP32, 7.200 GPixel/s, and 21.60 GTexel/s — and double the memory capacity at 2 GB. For any Vulkan-centric application, the HD 8790M is faster in every relevant metric except OpenCL.

Conversely, if OpenCL is the dominant API in your software stack, the AMD Radeon R7 M365X is the better option. Its 5939 OpenCL score beats the HD 8790M by 15.5%, and it does so with half the memory (1024 MB) and lower compute rates, suggesting its driver optimization or architectural tweaks favor OpenCL execution. The R7 M365X also comes from a later generation (Gem System R7 M300, released 2015-05-04) compared to the HD 8790M’s Solar System (released 2013-03-31), which may explain its OpenCL strength.

There is no universal winner. The HD 8790M wins on Vulkan, average score, raw throughput, memory capacity, and percentile. The R7 M365X wins on OpenCL alone. Given that the HD 8790M’s Vulkan lead (30.1%) is nearly double the R7 M365X’s OpenCL lead (15.5%), and that Vulkan is a more modern and widely supported API in the data, the HD 8790M is the more versatile GPU overall. The R7 M365X is a specialized tool for OpenCL compute, but the HD 8790M offers broader strength across the two tested workloads.