AMD Radeon HD 8690M vs AMD Radeon R7 M365X Comparison

The Verdict

The data separates these two mobile GPUs more by generation and configuration than by raw architectural philosophy. The AMD Radeon HD 8690M wins the only shared benchmark, Geekbench OpenCL, with a score of 6137 against the Radeon R7 M365X's 5939, a 3.3% advantage. The R7 M365X counters with a Geekbench Vulkan score of 4893, a test the HD 8690M did not record.

For users prioritizing raw compute in legacy OpenCL workloads, the HD 8690M is the stronger pick, sitting at the 36th percentile of all GPUs. The R7 M365X, at the 32nd percentile, trades that slight compute edge for a wider memory bus and a newer generation label, but its average benchmark score of 5416 trails the HD 8690M's 6137. Neither card is a modern gaming solution, but the HD 8690M holds the higher recorded performance ceiling, while the R7 M365X offers a Vulkan path for newer API-based tasks.

Architecture Differences

Both GPUs share the GCN 1.0 architecture and a 28 nm TSMC process node, but their physical implementations diverge. The HD 8690M uses the Sun chip with 690 million transistors on a 56 mm² die, while the R7 M365X uses the Litho chip with 950 million transistors on a 77 mm² die. Transistor density is identical at 12.3M per mm², indicating the same manufacturing maturity, but the Litho die packs 27.4% more transistors into 37.5% more area.

The shading resources differ meaningfully. The R7 M365X carries 384 shading units and 24 texture mapping units, while the HD 8690M has 320 shading units and 20 TMUs. Both have 8 ROPs. This gives the R7 M365X a 20% advantage in shading units and TMUs, yet its pixel rate is lower at 6.600 GPixel/s versus 7.800 GPixel/s for the HD 8690M, because the HD 8690M's higher clock speeds compensate for fewer units. Texture rates are nearly identical: 19.80 GTexel/s for the R7 M365X versus 19.50 GTexel/s for the HD 8690M.

The memory subsystem is where the R7 M365X pulls ahead structurally. It has a 128-bit bus width and 64.00 GB/s bandwidth, exactly double the HD 8690M's 64-bit bus and 32.00 GB/s bandwidth. Both use GDDR5 memory at 1000 MHz with 4 Gbps effective speed, but the R7 M365X ships with 1024 MB while the HD 8690M has 2 GB. This means the HD 8690M holds a capacity advantage while the R7 M365X has a bandwidth advantage, a classic trade-off between frame buffering and memory throughput.

Feature support is identical on the API front: both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Both use PCIe 3.0 x8. The generations differ, with the HD 8690M in the Solar System (HD 8600M) family and the R7 M365X in the Gem System (R7 M300) family. The HD 8690M's predecessor is London and its successor is Gem System, while the R7 M365X's predecessor is Solar System and its successor is Polaris Mobile. Both are end-of-life products.

Head-to-Head Benchmarks

The database records only one direct comparison: Geekbench OpenCL. The HD 8690M scores 6137 against the R7 M365X's 5939, a 3.3% win. This margin is modest but consistent with the HD 8690M's higher clock rates: its base clock is 925 MHz with a boost of 975 MHz, while the R7 M365X has no recorded base or boost clocks. The HD 8690M's FP32 throughput is 624.0 GFLOPS, slightly below the R7 M365X's 633.6 GFLOPS, yet the HD 8690M still wins the compute test, suggesting that clock speed and memory latency play a larger role than raw shader count.

The R7 M365X's Vulkan score of 4893 is its only other recorded benchmark, and no comparable Vulkan result exists for the HD 8690M. This makes a direct comparison impossible, but the Vulkan result indicates the R7 M365X can execute modern API workloads, a capability the HD 8690M's absence of data cannot confirm. The R7 M365X's average benchmark score of 5416, which includes both OpenCL and Vulkan results, falls below its OpenCL-only score, reflecting the lower Vulkan number pulling the average down.

The nearest rival data contextualizes both GPUs. The HD 8690M sits within 1.2% of the AMD Radeon HD 6950 (which scores 6210) and is 0.3% ahead of the Intel Iris Pro Graphics 6200 (6117). It also edges out the NVIDIA RTX A400 (6078) and GeForce MX230 (6077) by 0.3% and 0.3% respectively. The R7 M365X, by contrast, is 0.5% behind the AMD Radeon 610M (5444) and 0.9% ahead of the Intel UHD Graphics P630 (5370), while trailing the NVIDIA Quadro M4000 (5467) by 0.9% and leading the AMD Radeon R7 M445 (5358) by 1.1%. The HD 8690M competes in a higher performance tier overall, while the R7 M365X sits among much weaker integrated and entry-level parts.

Specification Differences

| Specification | AMD Radeon HD 8690M | AMD Radeon R7 M365X |

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

| Chip | Sun | Litho |

| Generation | Solar System (HD 8600M) | Gem System (R7 M300) |

| Transistors | 690 million | 950 million |

| Die Size | 56 mm² | 77 mm² |

| Base Clock | 925 MHz | Not recorded |

| Boost Clock | 975 MHz | Not recorded |

| Memory Size | 2 GB | 1024 MB |

| Memory Bus Width | 64 bit | 128 bit |

| Memory Bandwidth | 32.00 GB/s | 64.00 GB/s |

| Shading Units | 320 | 384 |

| TMUs | 20 | 24 |

| Pixel Rate | 7.800 GPixel/s | 6.600 GPixel/s |

| Texture Rate | 19.50 GTexel/s | 19.80 GTexel/s |

| FP32 | 624.0 GFLOPS | 633.6 GFLOPS |

| Release Date | 2013-02-28 | 2015-05-04 |

| Predecessor | London | Solar System |

| Successor | Gem System | Polaris Mobile |

| Benchmarks | OpenCL: 6137 | OpenCL: 5939, Vulkan: 4893 |

| Avg Benchmark Score | 6137 | 5416 |

| Percentile | 36 | 32 |

Identical between both: 28 nm process, TSMC foundry, 12.3M / mm² transistor density, GDDR5 memory type, 1000 MHz memory clock, 4 Gbps effective memory speed, 8 ROPs, PCIe 3.0 x8, DirectX 12 (11_1), OpenGL 4.6, Vulkan 1.2.170, and end-of-life production status.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The AMD Radeon HD 8690M scores 6137 in Geekbench OpenCL, while the AMD Radeon R7 M365X scores 5939, giving the HD 8690M a 3.3% advantage.

Q: Does the R7 M365X support Vulkan?

A: Yes, the R7 M365X records a Geekbench Vulkan score of 4893. The HD 8690M has no recorded Vulkan benchmark, though both list Vulkan 1.2.170 in their API support.

Q: Which GPU has more memory bandwidth?

A: The R7 M365X has 64.00 GB/s bandwidth via a 128-bit bus, exactly double the HD 8690M's 32.00 GB/s on a 64-bit bus. Both use GDDR5 at 1000 MHz with 4 Gbps effective speed.

Q: Which GPU has more shading units?

A: The R7 M365X has 384 shading units and 24 TMUs, compared to the HD 8690M's 320 shading units and 20 TMUs. Despite this, the HD 8690M wins the OpenCL test due to its higher clock speeds.

Q: How do these GPUs compare to their nearest rivals?

A: The HD 8690M is 0.3% ahead of the Intel Iris Pro Graphics 6200 and 1% ahead of the NVIDIA RTX A400 and GeForce MX230, while trailing the AMD Radeon HD 6950 by 1.2%. The R7 M365X is 0.5% behind the AMD Radeon 610M, 0.9% ahead of the Intel UHD Graphics P630, 0.9% behind the NVIDIA Quadro M4000, and 1.1% ahead of the AMD Radeon R7 M445.

Q: Which GPU has more memory capacity?

A: The HD 8690M has 2 GB of GDDR5, while the R7 M365X has 1024 MB. This means the HD 8690M can hold larger frame buffers, even though its bus is narrower.

Where Each One Wins

The HD 8690M wins in compute performance, as demonstrated by its 3.3% OpenCL advantage over the R7 M365X. It also holds a higher percentile rank at 36 versus 32, a higher average benchmark score of 6137 versus 5416, and double the memory capacity at 2 GB. Its pixel rate of 7.800 GPixel/s is 18.2% higher than the R7 M365X's 6.600 GPixel/s, making it better suited for fill-rate-bound tasks. The HD 8690M's nearest rivals include desktop-class parts like the Radeon HD 6950, placing it in a more competitive performance bracket.

The R7 M365X wins on memory bandwidth, delivering 64.00 GB/s versus 32.00 GB/s, which is critical for texture-heavy workloads and higher resolutions. It also has more shading units (384 versus 320) and TMUs (24 versus 20), giving it a theoretical edge in shader-bound compute, even though its FP32 output of 633.6 GFLOPS is only 1.5% higher than the HD 8690M's 624.0 GFLOPS. The R7 M365X's Vulkan benchmark of 4893 is a unique data point, suggesting it can handle newer API workloads that the HD 8690M never recorded. Its texture rate of 19.80 GTexel/s slightly exceeds the HD 8690M's 19.50 GTexel/s.

For users with legacy OpenCL applications, the HD 8690M is the clear choice. For those needing Vulkan support or higher memory throughput, the R7 M365X offers capabilities the HD 8690M cannot match in the recorded data. The R7 M365X's later release date of 2015-05-04 versus 2013-02-28 also implies a more recent production window, though both are now end-of-life. The data ultimately shows two aging GCN 1.0 parts with different optimization targets: one tuned for peak clocks and fill rate, the other for bandwidth and shader volume.