AMD Radeon 610M vs AMD Radeon HD 8690M Comparison

Where Each One Wins

The benchmark data splits cleanly between these two mobile graphics solutions. The AMD Radeon HD 8690M wins the only direct head-to-head comparison recorded in the database. In Geekbench OpenCL, the older card posts 6137 points against the Radeon 610M's 4535, a 35.3% advantage. That is a decisive margin, not a narrow scrape. If your workload leans on OpenCL compute, the HD 8690M is the stronger choice based on recorded results.

The AMD Radeon 610M, however, counters with a different set of capabilities. It has a Vulkan score of 6353 in the database, a test the HD 8690M does not appear in. That Vulkan result exceeds the HD 8690M's OpenCL score of 6137, but since the tests differ, direct comparison is not possible. What the data does show is that the 610M is competitive in modern API environments, which matters for current games and applications.

The wins breakdown is one for the HD 8690M, zero for the 610M in head-to-head tests. Yet the 610M's architecture and API support suggest it handles newer workloads better, even if the recorded OpenCL test favors the older GPU. For pure compute throughput as measured by Geekbench OpenCL, the HD 8690M is ahead. For feature support and modern rendering paths, the 610M has the edge.

Architecture Differences

These are two very different designs separated by nearly a decade of GPU evolution. The HD 8690M uses the Sun chip built on GCN 1.0 architecture, a 28 nm TSMC part with 690 million transistors on a 56 mm² die. Its transistor density works out to 12.3 million per square millimeter. The 610M uses the Mendocino chip on RDNA 2.0 architecture, fabricated on TSMC's 6 nm process. Its die size is 100 mm², though the transistor count is not recorded in the database.

The shading hardware tells the story of two design philosophies. The HD 8690M packs 320 shading units, 20 texture mapping units, and 8 ROPs. The 610M has only 128 shading units, 8 TMUs, and 4 ROPs. Fewer shaders but a much newer architecture means the 610M relies on efficiency and instruction-level improvements rather than raw unit count.

Clock speeds favor the 610M substantially. The HD 8690M runs at 925 MHz base and 975 MHz boost. The 610M starts at 1500 MHz and boosts to 1900 MHz. That clock advantage helps close the gap created by fewer shading units. Memory configurations diverge completely. The HD 8690M uses 2 GB of GDDR5 on a 64-bit bus with 32.00 GB/s bandwidth. The 610M uses shared system memory, with bandwidth described as system dependent. That means the 610M's memory performance varies with the laptop's RAM, while the HD 8690M has dedicated VRAM with fixed bandwidth.

Feature support moves in the 610M's favor. It lists ray tracing cores (2 of them) and DirectX 12 Ultimate support (12_2), while the HD 8690M only reaches DirectX 12 (11_1). The 610M also has Vulkan 1.4 support versus 1.2.170 for the older card. Both support OpenGL 4.6.

Rates and throughput numbers reflect the architectural split. The HD 8690M achieves 7.800 GPixel/s pixel rate and 19.50 GTexel/s texture rate. The 610M posts 7.600 GPixel/s and 15.20 GTexel/s. Pixel rates are nearly identical, but the HD 8690M leads in texture fill. Floating point performance goes the other way in FP16: the 610M delivers 972.8 GFLOPS with 2:1 ratio, while the HD 8690M has no recorded FP16 capability. In FP32, the HD 8690M leads with 624.0 GFLOPS versus 486.4 GFLOPS.

Head-to-Head Benchmarks

The only recorded head-to-head test is Geekbench OpenCL. The HD 8690M scores 6137, the 610M scores 4535. That is a 35.3% delta in favor of the older GPU. Context from the nearest rivals puts both in perspective. The HD 8690M sits at the 36th percentile of all GPUs, with neighbors like the Intel Iris Pro Graphics 6200 at 6117 (0.3% behind), NVIDIA RTX A400 at 6078 (1% behind), NVIDIA GeForce MX230 at 6077 (1% behind), and AMD Radeon HD 6950 at 6210 (1.2% ahead). The HD 8690M is right in the middle of a cluster spanning roughly 6077 to 6210.

The 610M's average benchmark score is 5444, which combines its OpenCL 4535 and Vulkan 6353 results. Its percentile sits at 32, slightly below the HD 8690M's 36. The 610M's rivals include NVIDIA Quadro M4000 at 5467 (0.4% ahead), AMD Radeon R7 M365X at 5416 (0.5% behind), AMD Radeon R7 M440 at 5483 (0.7% ahead), and NVIDIA GeForce GTX 765M at 5501 (1% ahead). The 610M's average lands between 5416 and 5501, a tight grouping.

The Vulkan score of 6353 for the 610M deserves attention. It is the single highest recorded score between both GPUs across all tests. That number suggests the 610M responds well to modern low-level APIs, which is consistent with its RDNA 2.0 architecture and newer feature set. The HD 8690M has no Vulkan result in the database, so its Vulkan performance remains unmeasured.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The AMD Radeon HD 8690M scores 6137, which is 35.3% higher than the Radeon 610M's 4535 in the same test.

Q: Does the Radeon 610M support ray tracing?

A: Yes, the database lists 2 ray tracing cores for the 610M. The HD 8690M has no ray tracing cores recorded.

Q: What memory type does each GPU use?

A: The HD 8690M uses 2 GB of GDDR5 on a 64-bit bus with 32.00 GB/s bandwidth. The 610M uses system shared memory with bandwidth described as system dependent.

Q: Which GPU has a higher boost clock?

A: The Radeon 610M boosts to 1900 MHz, while the HD 8690M boosts to 975 MHz. The 610M's base clock of 1500 MHz is also higher than the HD 8690M's base of 925 MHz.

Q: How do their average benchmark scores compare?

A: The HD 8690M has an average benchmark score of 6137. The 610M has an average of 5444, which combines its OpenCL score of 4535 and Vulkan score of 6353.

Q: Which GPU has better DirectX support?

A: The Radeon 610M supports DirectX 12 Ultimate (12_2). The HD 8690M supports DirectX 12 (11_1), which is a lower feature level.

The Verdict

The data points to a clear choice depending on workload. If your priority is OpenCL compute performance, the AMD Radeon HD 8690M wins outright with a 35.3% lead over the 610M. Its dedicated GDDR5 memory with fixed 32.00 GB/s bandwidth provides predictable performance that does not depend on system RAM. The HD 8690M also holds a 36th percentile ranking versus the 610M's 32nd, and its FP32 throughput of 624.0 GFLOPS exceeds the 610M's 486.4 GFLOPS.

If your priority is modern API support and features, the Radeon 610M is the better pick. It has DirectX 12 Ultimate, Vulkan 1.4, ray tracing cores, and FP16 support at 972.8 GFLOPS. Its 1900 MHz boost clock is nearly double the HD 8690M's 975 MHz. The Vulkan score of 6353 shows strong performance in current rendering APIs. The 610M also runs at a 15 W TDP with no power connectors, making it suitable for thin laptops, while the HD 8690M has no TDP recorded in the database.

The HD 8690M is end-of-life hardware from 2013. The 610M is end-of-life from 2022. For a new purchase, the 610M's newer architecture and feature set make more sense despite losing the OpenCL test. For existing hardware or specific compute workloads that rely on OpenCL, the HD 8690M remains competitive, as its 6137 score places it among GPUs like the Intel Iris Pro Graphics 6200 and NVIDIA GeForce MX230.

Specification Differences

| Specification | AMD Radeon HD 8690M | AMD Radeon 610M |

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

| Architecture | GCN 1.0 | RDNA 2.0 |

| Chip | Sun | Mendocino |

| Process Node | 28 nm | 6 nm |

| Die Size | 56 mm² | 100 mm² |

| Transistors | 690 million | Not recorded |

| Base Clock | 925 MHz | 1500 MHz |

| Boost Clock | 975 MHz | 1900 MHz |

| Memory Size | 2 GB | System Shared |

| Memory Type | GDDR5 | System Shared |

| Memory Bus | 64 bit | System Shared |

| Memory Bandwidth | 32.00 GB/s | System Dependent |

| Shading Units | 320 | 128 |

| TMUs | 20 | 8 |

| ROPs | 8 | 4 |

| Ray Tracing Cores | None | 2 |

| Pixel Rate | 7.800 GPixel/s | 7.600 GPixel/s |

| Texture Rate | 19.50 GTexel/s | 15.20 GTexel/s |

| FP32 | 624.0 GFLOPS | 486.4 GFLOPS |

| FP16 | Not recorded | 972.8 GFLOPS (2:1) |

| TDP | Not recorded | 15 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 4.0 x8 |

| DirectX | 12 (11_1) | 12 Ultimate (12_2) |

| Vulkan | 1.2.170 | 1.4 |

| Release Date | 2013-02-28 | 2022-09-19 |

| Production Status | End-of-life | End-of-life |