AMD Radeon HD 8690M vs NVIDIA GeForce MX230 Comparison

The AMD Radeon HD 8690M and NVIDIA GeForce MX230 represent two distinct eras of mobile graphics, with the data showing a narrow but clear overall victory for the older AMD part. In the only shared benchmark test, the Geekbench OpenCL score, the Radeon HD 8690M achieves 6137 points against the MX230’s 5739, a 6.9% advantage. However, the MX230 counters with a Geekbench Vulkan score of 6414, a test the HD 8690M did not run, which complicates any direct narrative of superiority. The average benchmark scores tell a similar story of near-parity: the HD 8690M averages 6137, while the MX230 averages 6077, a difference of exactly 1% that places both cards within the same performance tier.

Head-to-Head Benchmarks

The single directly comparable data point between these two GPUs is the Geekbench OpenCL benchmark, where the AMD Radeon HD 8690M wins decisively. The HD 8690M scores 6137, while the GeForce MX230 scores 5739, yielding a 6.9% delta in favor of the AMD part. This is a meaningful margin in the context of these low-power mobile GPUs, indicating that AMD’s older GCN architecture still delivers competitive raw compute throughput in OpenCL workloads. The 6.9% gap places the HD 8690M in the 36th percentile of all GPUs, while the MX230 sits just below at the 35th percentile, confirming that this head-to-head result is consistent with their broader standing in the benchmark database.

The MX230’s Vulkan score of 6414, while not directly comparable to the HD 8690M’s OpenCL result, suggests the NVIDIA part has a stronger showing in modern graphics APIs. This score is substantially higher than its own OpenCL result, a 11.8% improvement within the same GPU, and indicates that the Pascal architecture’s driver optimizations and API support are more effective in Vulkan contexts. The HD 8690M, with no Vulkan benchmark entry, cannot be evaluated on this axis, leaving a gap in the data that favors the MX230 in modern API scenarios. The delta between the two cards’ average scores is just 1% (6137 vs 6077), and both are within 1% of the NVIDIA RTX A400 (6078), which appears as a rival for both cards — the HD 8690M is 1% ahead of it, while the MX230 is exactly level with it.

Where Each One Wins

The AMD Radeon HD 8690M wins in the OpenCL compute benchmark, where its 624.0 GFLOPS of FP32 throughput and 320 shading units give it a computational edge over the MX230’s 256 shading units and 783.9 GFLOPS. The 6.9% score advantage in OpenCL reflects this raw compute capability, making the HD 8690M the better choice for OpenCL-accelerated applications like certain physics simulations, image processing filters, or compute shaders that rely on this API. The HD 8690M also shows a 0.3% advantage over the Intel Iris Pro Graphics 6200 (6117) and a 1% edge over the NVIDIA RTX A400 (6078), placing it slightly ahead of its nearest rivals in compute tasks.

The NVIDIA GeForce MX230 wins in Vulkan performance, where its 6414 score demonstrates superior optimization for this modern API. This is particularly relevant for gaming and applications that leverage Vulkan’s lower overhead and better multi-threading. The MX230’s 16 ROPs (double the HD 8690M’s 8) contribute to a pixel rate of 24.50 GPixel/s, which is more than three times the HD 8690M’s 7.800 GPixel/s, making the NVIDIA part significantly faster in fill-rate-bound scenarios like high-resolution texturing or certain post-processing effects. The MX230 also achieves a 24.50 GTexel/s texture rate versus 19.50 GTexel/s for the AMD part, giving it a 25.6% advantage in texture-heavy workloads. The MX230’s 1.4 Vulkan API version support, compared to the HD 8690M’s 1.2.170, further solidifies its position in modern graphics environments.

Architecture Differences

The architectural divide between these two GPUs is stark, reflecting six years of semiconductor evolution. The AMD Radeon HD 8690M uses the Sun chip built on GCN 1.0 architecture, fabricated on TSMC’s 28 nm process with 690 million transistors packed into a 56 mm² die, yielding a transistor density of 12.3 million per square millimeter. In contrast, the NVIDIA GeForce MX230 uses the GP108 chip on the Pascal architecture, fabricated on Samsung’s 14 nm process with 1,800 million transistors on a 74 mm² die, achieving a far higher transistor density of 24.3 million per square millimeter — nearly double that of the AMD part. This density advantage allows the MX230 to pack more than 2.6 times the transistors into a 32% larger die, enabling its higher clock speeds and newer feature set.

The clock speeds tell a dramatic story of process node advantage. The MX230 runs at a 1519 MHz base clock and 1531 MHz boost, while the HD 8690M operates at 925 MHz base and 975 MHz boost — the NVIDIA part runs 64% faster at base and 57% faster at boost. Memory clocks also favor the MX230: its 1502 MHz memory (6 Gbps effective) delivers 48.06 GB/s of bandwidth, a 50% improvement over the HD 8690M’s 1000 MHz memory (4 Gbps effective) at 32.00 GB/s. Both cards use 2 GB of GDDR5 on a 64-bit bus, but the MX230’s higher memory clock is the differentiator.

The compute resource allocation differs significantly. The HD 8690M has 320 shading units, 20 TMUs, and 8 ROPs, while the MX230 has 256 shading units, 16 TMUs, and 16 ROPs. This means the AMD card has 25% more shaders and TMUs, but the NVIDIA card has twice the ROPs. The FP32 throughput reflects this: the MX230’s 783.9 GFLOPS is 25.6% higher than the HD 8690M’s 624.0 GFLOPS, despite fewer shaders, due to its much higher clock speed. The MX230 also has FP16 support at 12.25 GFLOPS (1:64 ratio), a feature absent from the HD 8690M’s specifications, though the 1:64 ratio indicates it is not a strong FP16 performer.

Power and interface differences are notable. The MX230 has a TDP of 10 W and is listed as an IGP with no power connectors, while the HD 8690M has no TDP data available. The MX230 uses a PCIe 3.0 x4 interface, while the HD 8690M uses PCIe 3.0 x8, giving the AMD part twice the bus bandwidth for data transfer. The MX230 supports DirectX 12 (12_1) and Vulkan 1.4, while the HD 8690M supports DirectX 12 (11_1) and Vulkan 1.2.170 — the NVIDIA part’s newer API versions reflect its 2019 release date versus the HD 8690M’s 2013 launch.

FAQ

Q: Which GPU has a higher benchmark score in OpenCL?

A: The AMD Radeon HD 8690M scores 6137 in Geekbench OpenCL, which is 6.9% higher than the NVIDIA GeForce MX230’s 5739 in the same test.

Q: Does the NVIDIA GeForce MX230 have any benchmark where it wins?

A: Yes, the MX230 has a Geekbench Vulkan score of 6414, though the HD 8690M has no Vulkan benchmark entry, so this result is not directly comparable between the two.

Q: How do the two GPUs compare in terms of transistor density?

A: The MX230 has a transistor density of 24.3 million per square millimeter on a 14 nm Samsung process, while the HD 8690M has 12.3 million per square millimeter on a 28 nm TSMC process, making the NVIDIA part nearly twice as dense.

Q: What are the memory bandwidth differences between the two?

A: The MX230 provides 48.06 GB/s of bandwidth from its 6 Gbps effective memory clock, while the HD 8690M delivers 32.00 GB/s from a 4 Gbps effective clock, a 50% advantage for the NVIDIA card.

Q: Which GPU has more shading units?

A: The AMD Radeon HD 8690M has 320 shading units, while the NVIDIA GeForce MX230 has 256, giving the AMD part a 25% advantage in shader count.

Q: What is the pixel rate difference between the two cards?

A: The MX230 achieves 24.50 GPixel/s, which is over three times the HD 8690M’s 7.800 GPixel/s, due to the NVIDIA card’s 16 ROPs versus the AMD card’s 8.

Specification Differences

| Specification | AMD Radeon HD 8690M | NVIDIA GeForce MX230 |

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

| Architecture | GCN 1.0 | Pascal |

| Process Node | 28 nm | 14 nm |

| Foundry | TSMC | Samsung |

| Transistors | 690 million | 1,800 million |

| Die Size | 56 mm² | 74 mm² |

| Transistor Density | 12.3M / mm² | 24.3M / mm² |

| Base Clock | 925 MHz | 1519 MHz |

| Boost Clock | 975 MHz | 1531 MHz |

| Memory Clock | 1000 MHz (4 Gbps effective) | 1502 MHz (6 Gbps effective) |

| Memory Bandwidth | 32.00 GB/s | 48.06 GB/s |

| Shading Units | 320 | 256 |

| TMUs | 20 | 16 |

| ROPs | 8 | 16 |

| Pixel Rate | 7.800 GPixel/s | 24.50 GPixel/s |

| Texture Rate | 19.50 GTexel/s | 24.50 GTexel/s |

| FP32 Performance | 624.0 GFLOPS | 783.9 GFLOPS |

| FP16 Performance | Not listed | 12.25 GFLOPS (1:64) |

| TDP | Not listed | 10 W |

| Slot Width | Not listed | IGP |

| Power Connectors | Not listed | None |

| Bus Interface | PCIe 3.0 x8 | PCIe 3.0 x4 |

| Display Outputs | Not listed | Portable Device Dependent |

| DirectX Support | 12 (11_1) | 12 (12_1) |

| Vulkan Support | 1.2.170 | 1.4 |

| Release Date | 2013-02-28 | 2019-02-20 |

| Predecessor | London | Not listed |

| Successor | Gem System | Not listed |

| Geekbench OpenCL Score | 6137 | 5739 |

| Percentile vs All GPUs | 36 | 35 |