AMD Radeon HD 8750M vs NVIDIA GeForce MX230 Comparison

NVIDIA GeForce MX230 and AMD Radeon HD 8750M are both end-of-life mobile graphics processors, but they represent vastly different eras of GPU design. The MX230 is a 2019 Pascal-based part built on a 14nm process, while the HD 8750M is a 2013 GCN 1.0 part on a 28nm node. The benchmark data shows they are remarkably close in overall performance, with the AMD part holding a slight edge in the single available test, despite the NVIDIA chip having a much more modern feature set and significantly higher clock speeds.

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and the results are surprisingly tight. The AMD Radeon HD 8750M scores 5970, while the NVIDIA GeForce MX230 scores 5739. This gives the AMD part a 3.9% lead in this particular test. The delta is small enough that it falls well within the range of typical run-to-run variation, but the data records it as a win for the older AMD silicon.

Looking at the average benchmark scores across all tests each GPU appears in, the picture is similar. The MX230 averages 6077, while the HD 8750M averages 5970. That works out to roughly a 1.8% advantage for the NVIDIA part when considering the broader dataset. These two GPUs are effectively performance peers in raw compute workloads, which is remarkable given the six-year gap in their release dates.

The percentile rankings reinforce this parity. The MX230 sits at the 35th percentile of all GPUs, while the HD 8750M is at the 34th percentile. That one-percentile difference is negligible in real-world terms. When you look at the nearest rivals for each card, the MX230 trades blows with the NVIDIA RTX A400 (6078 avg, 0% delta), NVIDIA Quadro P2000 (6049 avg, 0.5% delta), and Intel Iris Pro Graphics 6200 (6117 avg, -0.7% delta). The HD 8750M is similarly matched against the NVIDIA Quadro K4000 (5982 avg, -0.2% delta), NVIDIA Quadro K620M (5957 avg, 0.2% delta), and AMD Radeon HD 8730M (5955 avg, 0.3% delta). Neither card is decisively ahead of its immediate competition.

Where Each One Wins

The AMD Radeon HD 8750M wins the only head-to-head benchmark, taking Geekbench OpenCL by 231 points. This suggests that in pure compute tasks that leverage OpenCL, the older GCN architecture can still hold its own. The HD 8750M achieves this with a wider 128-bit memory bus and 384 shading units, which help compensate for its lower clocks.

The NVIDIA GeForce MX230 wins on overall average benchmark score, posting 6077 versus 5970 for the AMD card. This indicates that across a broader range of workloads, the Pascal architecture tends to edge out the older GCN design. The MX230 also has a substantial advantage in raw clock speed — its base clock of 1519 MHz is nearly double the HD 8750M's 775 MHz base. The boost clock tells the same story: 1531 MHz versus 825 MHz.

In terms of memory bandwidth, the MX230 pulls ahead with 48.06 GB/s compared to 28.80 GB/s for the HD 8750M. This is despite the NVIDIA part using a narrow 64-bit bus, because it runs GDDR5 at an effective 6 Gbps, whereas the AMD card uses DDR3 at an effective 1800 Mbps on a 128-bit bus. The MX230 also delivers far higher pixel and texture rates: 24.50 GPixel/s and 24.50 GTexel/s versus 6.600 GPixel/s and 19.80 GTexel/s for the AMD part.

Architecture Differences

The architectural gap between these two is substantial. The NVIDIA GeForce MX230 uses the GP108 chip built on Pascal architecture, manufactured by Samsung on a 14nm process. It packs 1,800 million transistors into a 74 mm² die, giving a transistor density of 24.3M per mm². The AMD Radeon HD 8750M uses the Mars chip based on GCN 1.0 architecture, built by TSMC on a 28nm process. It has 950 million transistors on a 77 mm² die, with a transistor density of just 12.3M per mm².

The MX230 has 256 shading units, 16 texture mapping units, and 16 ROPs. The HD 8750M has more shading units at 384, and more TMUs at 24, but only 8 ROPs. This explains why the AMD card loses so badly in pixel rate despite having more compute units. The NVIDIA part also supports FP16 compute at 12.25 GFLOPS (1:64 ratio), while the AMD card has no listed FP16 capability. In FP32, the MX230 outputs 783.9 GFLOPS versus 633.6 GFLOPS for the HD 8750M.

The memory subsystems differ fundamentally. The MX230 uses 2 GB of GDDR5 on a 64-bit bus, while the HD 8750M uses 1 GB of DDR3 on a 128-bit bus. The NVIDIA card's memory runs at 1502 MHz (6 Gbps effective), while the AMD part's memory runs at 900 MHz (1800 Mbps effective). This gives the MX230 the bandwidth advantage despite the narrower bus.

API support also differs. The MX230 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The HD 8750M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. So the NVIDIA part has a slightly higher DirectX feature level and a newer Vulkan version. The MX230 connects via PCIe 3.0 x4, while the HD 8750M uses PCIe 3.0 x8, giving the AMD card twice the bus lanes.

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The AMD Radeon HD 8750M scores 5970 in Geekbench OpenCL, which is 3.9% higher than the NVIDIA GeForce MX230's 5739.

Q: Does the newer NVIDIA card have better memory bandwidth?

A: Yes. The MX230 delivers 48.06 GB/s using 2 GB of GDDR5 on a 64-bit bus, while the HD 8750M provides 28.80 GB/s with 1 GB of DDR3 on a 128-bit bus.

Q: How do their overall average benchmark scores compare?

A: The MX230 averages 6077 across its benchmarks, which is about 1.8% higher than the HD 8750M's average of 5970.

Q: Which GPU has higher clock speeds?

A: The NVIDIA MX230 runs at a 1519 MHz base and 1531 MHz boost, compared to the AMD HD 8750M's 775 MHz base and 825 MHz boost.

Q: What are the DirectX version differences?

A: The MX230 supports DirectX 12 (12_1), while the HD 8750M supports DirectX 12 (11_1). The NVIDIA part has the higher feature level.

Q: Which card is more efficient per transistor?

A: The MX230 has a transistor density of 24.3M per mm² on a 74 mm² die with 1,800 million transistors, while the HD 8750M has 12.3M per mm² on a 77 mm² die with 950 million transistors.

Specification Differences

| Specification | NVIDIA GeForce MX230 | AMD Radeon HD 8750M |

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

| Architecture | Pascal | GCN 1.0 |

| Process Node | 14 nm | 28 nm |

| Foundry | Samsung | TSMC |

| Transistors | 1,800 million | 950 million |

| Die Size | 74 mm² | 77 mm² |

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

| Base Clock | 1519 MHz | 775 MHz |

| Boost Clock | 1531 MHz | 825 MHz |

| Memory Clock | 1502 MHz (6 Gbps effective) | 900 MHz (1800 Mbps effective) |

| Memory Size | 2 GB | 1024 MB |

| Memory Type | GDDR5 | DDR3 |

| Memory Bus Width | 64 bit | 128 bit |

| Memory Bandwidth | 48.06 GB/s | 28.80 GB/s |

| Shading Units | 256 | 384 |

| TMUs | 16 | 24 |

| ROPs | 16 | 8 |

| Pixel Rate | 24.50 GPixel/s | 6.600 GPixel/s |

| Texture Rate | 24.50 GTexel/s | 19.80 GTexel/s |

| FP32 Performance | 783.9 GFLOPS | 633.6 GFLOPS |

| FP16 Performance | 12.25 GFLOPS (1:64) | null |

| TDP | 10 W | null |

| Slot Width | IGP | null |

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

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

| Vulkan | 1.4 | 1.2.170 |

| Release Date | 2019-02-20 | 2013-02-25 |

| Predecessor | null | London |

| Successor | null | Gem System |

The Verdict

The data presents a genuine toss-up, with each card winning a different metric. The AMD Radeon HD 8750M takes the direct head-to-head OpenCL benchmark with a 3.9% lead, proving that its larger shading unit count and wider memory bus still matter in compute workloads. If you only care about that specific test, the AMD card is the pick.

The NVIDIA GeForce MX230 is the better all-rounder. It has a higher average benchmark score (6077 versus 5970), nearly double the memory bandwidth, and dramatically higher pixel and texture rates. It also consumes only 10 W TDP, is a simple IGP solution, and supports newer API versions including DirectX 12_1 and Vulkan 1.4. The HD 8750M has no TDP listed, which makes power planning harder.

For modern workloads, the MX230 is the sensible choice. It has 2 GB of VRAM versus 1 GB, doubles the memory bandwidth, and offers better feature support. The HD 8750M is a relic from 2013 that happens to still be competitive in raw OpenCL compute, but it lacks the memory capacity and API modernity needed for current applications. If you are building a system today, the MX230 is the data-backed recommendation. If you are comparing legacy laptops, the HD 8750M is not the embarrassment its age suggests.