AMD Radeon HD 8730M vs AMD Radeon HD 8790M Comparison

AMD Radeon HD 8730M and AMD Radeon HD 8790M are two mobile graphics solutions from the same Solar System (HD 8700M) generation, both built on the GCN 1.0 architecture and the Mars chip. Despite sharing the same fundamental silicon, the data reveals a more complex relationship than a simple tiered hierarchy, with the lower-numbered HD 8730M actually taking the single head-to-head benchmark victory, while the HD 8790M counters with a feature set that includes a faster memory standard and an additional benchmark score. This analysis breaks down the benchmark data, architectural differences, and use-case scenarios to determine which GPU holds the edge in specific workloads.

Head-to-Head Benchmarks

The direct comparison between these two GPUs is remarkably one-sided in the available data. The head-to-head benchmark suite contains a single test: Geekbench OpenCL. In this test, the AMD Radeon HD 8730M scores 5955 points, while the AMD Radeon HD 8790M scores 5017 points. This results in a decisive 18.7% advantage for the HD 8730M, marking it as the clear winner in the only shared benchmark metric.

This outcome is counterintuitive given the HD 8790M's higher clock speeds and superior memory type, but the data is unambiguous. The HD 8730M's OpenCL score of 5955 places it in the 34th percentile of all GPUs, while the HD 8790M's OpenCL score of 5017 contributes to its lower average. When looking at the average benchmark score across all available tests, the HD 8730M maintains a lead with an average of 5955, compared to the HD 8790M's average of 5691. This difference is driven entirely by the HD 8790M's weaker OpenCL showing, despite its additional Vulkan benchmark.

The HD 8790M does post a significantly higher score in its Geekbench Vulkan test, achieving 6365 points. However, since the HD 8730M lacks a Vulkan benchmark in its profile, this cannot be used as a direct head-to-head comparison. The closest the HD 8790M comes to a direct win is through its average score relative to the HD 8730M's nearest rivals; the HD 8790M sits 0.4% ahead of Intel Iris Pro Graphics P6300 and 0.5% ahead of NVIDIA GeForce GTX 670MX, but these are not direct comparisons to the HD 8730M.

Architecture Differences

Both GPUs are built on the same Mars chip using the GCN 1.0 architecture, manufactured by TSMC on a 28 nm process node. This means they share identical core configurations: 384 shading units, 24 texture mapping units (TMUs), and 8 raster operation units (ROPs). The transistor count is also identical at 950 million, packed into the same 77 mm² die size, resulting in a transistor density of 12.3M per mm². Both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170, and were released on the same date, with the same predecessor (London) and successor (Gem System).

The key architectural differences lie in clock speeds and memory subsystems. The HD 8730M operates at a base clock of 650 MHz with a boost clock of 700 MHz, while the HD 8790M runs significantly higher at 850 MHz base and 900 MHz boost. This clock advantage translates directly to compute throughput: the HD 8730M delivers 537.6 GFLOPS of FP32 performance, whereas the HD 8790M achieves 691.2 GFLOPS. Similarly, pixel and texture fill rates are higher on the HD 8790M, with 7.200 GPixel/s and 21.60 GTexel/s respectively, compared to 5.600 GPixel/s and 16.80 GTexel/s on the HD 8730M.

Memory configuration is where the cards diverge most starkly. The HD 8730M uses 2 GB of DDR3 memory on a 128-bit bus, running at 900 MHz (1800 Mbps effective), yielding a bandwidth of 28.80 GB/s. The HD 8790M also has 2 GB on a 128-bit bus, but it uses GDDR5 memory clocked at 1000 MHz (4 Gbps effective), more than doubling the bandwidth to 64.00 GB/s. This is a substantial difference that should theoretically favor the HD 8790M in memory-bound workloads, yet the benchmark data does not reflect this advantage in OpenCL.

The physical form factors also differ. The HD 8730M uses a PCIe 3.0 x8 bus interface, while the HD 8790M is designed as an MXM Module with a MXM-A (3.0) interface. The HD 8790M lists "Portable Device Dependent" for display outputs and has no power connectors, whereas the HD 8730M does not specify these details.

Where Each One Wins

Based strictly on the benchmark data, the AMD Radeon HD 8730M wins in OpenCL compute workloads. Its Geekbench OpenCL score of 5955 is 18.7% higher than the HD 8790M's 5017, making it the superior choice for applications that rely on OpenCL acceleration, such as certain video encoding, physics simulation, and general-purpose GPU computing tasks. The HD 8730M also holds a higher percentile ranking at 34th versus the HD 8790M's 33rd, and its average benchmark score of 5955 exceeds the HD 8790M's 5691 by a meaningful margin.

The AMD Radeon HD 8790M wins in raw theoretical performance metrics. Its higher clock speeds and GDDR5 memory give it clear advantages in pure compute throughput (691.2 GFLOPS vs 537.6 GFLOPS), pixel fill rate (7.200 GPixel/s vs 5.600 GPixel/s), texture fill rate (21.60 GTexel/s vs 16.80 GTexel/s), and memory bandwidth (64.00 GB/s vs 28.80 GB/s). For workloads that are sensitive to memory bandwidth, such as high-resolution texture streaming or large data set processing, the HD 8790M's architecture is objectively more capable on paper.

The HD 8790M also demonstrates its capability in its Vulkan benchmark, scoring 6365 points. While this cannot be directly compared to the HD 8730M, it indicates strong performance in Vulkan-based applications and games. The HD 8790M's nearest rivals in average score include the NVIDIA GeForce GTX 670MX (5721, deltaPct -0.5) and the NVIDIA GeForce GTX 550 Ti (5731, deltaPct -0.7), both of which are desktop or high-end mobile parts, suggesting the HD 8790M holds its own in that performance bracket.

The Verdict

The data presents a paradox: the HD 8790M is the faster card on paper, with higher clocks, more memory bandwidth, and better fill rates, yet it loses decisively in the only directly comparable benchmark. The HD 8730M's 18.7% lead in Geekbench OpenCL is too large to ignore, and its higher average benchmark score (5955 vs 5691) reinforces this result.

For users prioritizing OpenCL compute performance, the AMD Radeon HD 8730M is the clear choice. Its benchmark score of 5955 places it right alongside the NVIDIA Quadro K620M (5957, deltaPct 0) and just ahead of the Intel UHD Graphics 730 (5929, deltaPct 0.4), indicating competitive OpenCL capability. The HD 8730M's nearest rival, the AMD Radeon HD 8750M, scores 5970 (deltaPct -0.3), meaning the HD 8730M is nearly identical to its direct sibling in this metric.

For users who prioritize Vulkan performance or rely on memory bandwidth-intensive workloads, the AMD Radeon HD 8790M is the better option, provided they are not dependent on OpenCL. Its Vulkan score of 6365 is substantial, and its theoretical memory bandwidth of 64.00 GB/s is more than double that of the HD 8730M. The HD 8790M's average score of 5691, while lower than the HD 8730M's, is still respectable, sitting 1.6% ahead of the NVIDIA Quadro M500M (5604, deltaPct 1.6).

The production status of both cards is end-of-life, so these are legacy parts. The choice between them should be dictated by the specific application requirements: OpenCL compute favors the HD 8730M, while Vulkan and memory bandwidth favor the HD 8790M. There is no universal winner based on the data; the correct pick depends entirely on the workload.

FAQ

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

A: The AMD Radeon HD 8730M scores 5955 in Geekbench OpenCL, which is 18.7% higher than the AMD Radeon HD 8790M's score of 5017 in the same test.

Q: Does the AMD Radeon HD 8790M have any benchmark advantage?

A: Yes, the HD 8790M scores 6365 in Geekbench Vulkan, a test not available for the HD 8730M. Its average benchmark score across all tests is 5691, which is lower than the HD 8730M's 5955 average.

Q: How do the memory subsystems differ between the two GPUs?

A: The HD 8730M uses 2 GB of DDR3 memory on a 128-bit bus with a bandwidth of 28.80 GB/s. The HD 8790M uses 2 GB of GDDR5 memory on the same 128-bit bus but achieves a bandwidth of 64.00 GB/s.

Q: Are the core configurations identical?

A: Yes, both GPUs feature 384 shading units, 24 TMUs, and 8 ROPs. They are both built on the Mars chip with GCN 1.0 architecture and 950 million transistors.

Q: What is the difference in clock speeds?

A: The HD 8730M has a base clock of 650 MHz and a boost clock of 700 MHz. The HD 8790M operates at a base clock of 850 MHz and a boost clock of 900 MHz.

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

A: The HD 8730M is nearly tied with the NVIDIA Quadro K620M (5957, deltaPct 0) and trails the AMD Radeon HD 8750M by 0.3%. The HD 8790M sits 0.4% ahead of the Intel Iris Pro Graphics P6300 and 0.5% ahead of the NVIDIA GeForce GTX 670MX.