AMD Radeon HD 8790M vs NVIDIA GeForce MX230 Comparison

The NVIDIA GeForce MX230 and AMD Radeon HD 8790M are both end-of-life mobile graphics solutions, but they represent different eras of design. The MX230, built on a modern 14 nm process, and the HD 8790M, a 28 nm part from 2013, are separated by roughly six years of architectural progress. Benchmark data from Geekbench shows the MX230 winning both head-to-head tests, but the margins tell a more nuanced story about where each GPU still holds ground.

Head-to-Head Benchmarks

In the Geekbench OpenCL test, the NVIDIA GeForce MX230 scores 5739 against the AMD Radeon HD 8790M’s 5017. That is a 14.4% advantage for the MX230, a clear and substantial win. The MX230’s higher average benchmark score of 6077 versus 5691 for the HD 8790M reinforces this lead. The OpenCL result aligns with the raw compute specifications: the MX230 delivers 783.9 GFLOPS of FP32 performance, while the HD 8790M manages 691.2 GFLOPS. The NVIDIA part’s 14 nm process also allows for a much higher clock speed, with a base of 1519 MHz boosting to 1531 MHz, compared to the AMD’s 850 MHz base and 900 MHz boost.

The Vulkan test is much closer. The MX230 scores 6414, narrowly edging out the HD 8790M’s 6365, a delta of just 0.8%. This near-tie is notable because the HD 8790M, despite its older GCN 1.0 architecture, has a wider 128-bit memory bus and higher memory bandwidth of 64.00 GB/s, versus the MX230’s 64-bit bus and 48.06 GB/s. The AMD part also has more shading units (384 vs 256) and more texture mapping units (24 vs 16), which helps it stay competitive in API-level workloads that leverage these resources. However, the MX230’s newer architecture supports Vulkan 1.4, while the HD 8790M is limited to Vulkan 1.2.170, which may explain why the NVIDIA chip still wins even in a test where the AMD hardware has theoretical advantages.

Looking at the percentile rankings, the MX230 sits at the 35th percentile of all GPUs, while the HD 8790M is at the 33rd percentile. The MX230’s nearest rivals include the NVIDIA RTX A400 (average score 6078, delta 0%), the NVIDIA Quadro P2000 (6049, +0.5%), and the Intel Iris Pro Graphics 6200 (6117, -0.7%). The HD 8790M’s closest competitors are the Intel Iris Pro Graphics P6300 (5712, -0.4%), the NVIDIA GeForce GTX 670MX (5721, -0.5%), and the NVIDIA GeForce GTX 550 Ti (5731, -0.7%). These figures place both GPUs in a similar performance tier, but the MX230 has a slight edge that is consistent across both tests.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce MX230 has an average benchmark score of 6077, while the AMD Radeon HD 8790M averages 5691. This represents a roughly 6.8% advantage for the MX230.

Q: How do the two GPUs compare in Vulkan performance?

A: The MX230 scores 6414 in Geekbench Vulkan, beating the HD 8790M’s 6365 by a slim 0.8% margin. Despite the close result, the MX230’s support for Vulkan 1.4 versus the HD 8790M’s Vulkan 1.2.170 gives it a software feature advantage.

Q: What is the memory bandwidth difference?

A: The AMD Radeon HD 8790M has significantly higher memory bandwidth at 64.00 GB/s, thanks to its 128-bit bus. The NVIDIA GeForce MX230 is limited to 48.06 GB/s over a 64-bit bus, a 33% deficit for the NVIDIA part.

Q: Which GPU has more shading units?

A: The AMD Radeon HD 8790M has 384 shading units, compared to 256 on the NVIDIA GeForce MX230. The AMD part also has 24 TMUs versus 16 on the NVIDIA chip, though the MX230 has double the ROPs (16 vs 8).

Q: What are the process node differences?

A: The NVIDIA GeForce MX230 is fabricated on a 14 nm process at Samsung, while the AMD Radeon HD 8790M uses a 28 nm process at TSMC. The MX230’s die is 74 mm² with 1,800 million transistors, while the HD 8790M’s die is 77 mm² with 950 million transistors.

Q: Which GPU supports a newer DirectX version?

A: The NVIDIA GeForce MX230 supports DirectX 12 (12_1), while the AMD Radeon HD 8790M supports DirectX 12 (11_1). Both support OpenGL 4.6.

The Verdict

The data points to the NVIDIA GeForce MX230 as the stronger performer overall. It wins both head-to-head benchmarks, has a higher average score, and sits at a higher percentile rank. The 14.4% lead in OpenCL is decisive, and even in Vulkan, where the HD 8790M’s wider memory bus and more shading units should help, the MX230 still comes out ahead. For any workload that relies on raw compute throughput, the MX230 is the clear choice.

However, the AMD Radeon HD 8790M is not without merit. Its 64.00 GB/s memory bandwidth is substantially higher, which could benefit memory-bound tasks in legacy titles or OpenCL workloads that are sensitive to bandwidth rather than shader count. The HD 8790M also has a higher texture fill rate in terms of raw texture units, though the MX230’s higher clocks result in a higher texture rate of 24.50 GTexel/s versus 21.60 GTexel/s. If you are working with older software that favors the GCN architecture’s wider memory interface, the HD 8790M may hold up better than its overall score suggests.

For most users, the MX230 is the safer pick. It is faster, more efficient on paper, and supports newer API features like Vulkan 1.4. The HD 8790M is a legacy part that, while competitive in narrow scenarios, cannot match the MX230’s sustained performance across the benchmark suite.

Specification Differences

The two GPUs diverge on nearly every core specification. The NVIDIA GeForce MX230 uses the GP108 chip on a 14 nm Samsung process, while the AMD Radeon HD 8790M uses the Mars chip on a 28 nm TSMC process. Transistor counts differ sharply: 1,800 million for the MX230 versus 950 million for the HD 8790M. Die sizes are similar (74 mm² vs 77 mm²), but transistor density is far higher on the MX230 at 24.3M / mm² versus 12.3M / mm².

Clock speeds are the most dramatic difference. The MX230 runs at a base of 1519 MHz and boost of 1531 MHz, while the HD 8790M runs at 850 MHz base and 900 MHz boost. Memory clocks also differ: the MX230’s memory runs at 1502 MHz (6 Gbps effective), while the HD 8790M’s runs at 1000 MHz (4 Gbps effective). The HD 8790M compensates with a 128-bit bus versus the MX230’s 64-bit bus, yielding 64.00 GB/s versus 48.06 GB/s.

The AMD part has more shading units (384 vs 256) and TMUs (24 vs 16), but the NVIDIA part has more ROPs (16 vs 8). This results in a pixel rate of 24.50 GPixel/s for the MX230 versus 7.200 GPixel/s for the HD 8790M. FP32 performance favors the MX230 at 783.9 GFLOPS versus 691.2 GFLOPS. The MX230 is an IGP with a PCIe 3.0 x4 interface, while the HD 8790M is an MXM module with an MXM-A (3.0) interface. The MX230 has a TDP of 10 W, while the HD 8790M’s TDP is not listed.

Architecture Differences

The NVIDIA GeForce MX230 is built on the Pascal architecture, which is optimized for power efficiency and modern API support. It features a 14 nm process, allowing for high clocks at low power. The MX230 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Its shading unit count of 256 is arranged in a design that yields 783.9 GFLOPS of FP32 performance. The MX230’s FP16 performance is listed as 12.25 GFLOPS (1:64), indicating a limited half-precision path.

The AMD Radeon HD 8790M uses the GCN 1.0 architecture, which was designed for compute-heavy workloads. It is built on a 28 nm process at TSMC, with a transistor density of just 12.3M / mm². The GCN design provides 384 shading units, which is why the HD 8790M can compete in Vulkan despite lower clocks. The HD 8790M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. It does not list FP16 performance. The architecture differences also manifest in the memory subsystem: the HD 8790M’s wider 128-bit bus is a hallmark of older GCN parts, while the MX230’s narrow 64-bit bus is a cost-saving measure typical of entry-level Pascal products.

Where Each One Wins

The NVIDIA GeForce MX230 wins in all head-to-head benchmarks, but its biggest victory is in OpenCL, where it leads by 14.4%. This makes it the superior choice for general compute tasks, including OpenCL-based rendering, physics simulations, and any workload that leverages FP32 throughput. Its higher pixel rate (24.50 GPixel/s) and texture rate (24.50 GTexel/s) also make it better for rasterization-heavy tasks in older games that rely on fill rates. The MX230’s 10 W TDP and IGP form factor mean it is suited for thin-and-light laptops where power draw is a constraint.

The AMD Radeon HD 8790M, despite losing both tests, has specific advantages. Its 64.00 GB/s memory bandwidth is 33% higher than the MX230’s, which can benefit workloads that are memory-bandwidth bound, such as large texture streaming or certain compute kernels. The HD 8790M’s higher shading unit count (384 vs 256) and TMU count (24 vs 16) give it raw parallel resources that could help in Vulkan applications that are not optimized for NVIDIA’s architecture. Its MXM form factor means it was designed for upgradeable laptop modules, though it is now end-of-life. For users with legacy software that favors GCN’s wider memory interface, the HD 8790M may still offer acceptable performance, but the data shows it is generally outclassed by the MX230.