AMD Radeon HD 8730A

The AMD Radeon HD 8730A is a 28 nm GCN 1.0 part fabricated by TSMC, using the Mars chip. It carries 950 million transistors on a 77 mm² die, yielding a transistor density of 12.3M per mm². The specification sheet places it at the 50th percentile of all GPUs in the database, indicating a median performance standing. With 384 shading units, 24 texture mapping units, and 8 render output units, it operates at a base clock of 650 MHz and a boost clock of 700 MHz. The memory runs at 1000 MHz, providing an effective data rate of 2 Gbps. The release date is March 31, 2013, and the production status is end-of-life. This GPU belongs to the All-In-One (HD 8000) generation, suggesting its intended use in integrated all-in-one systems. The average benchmark score in the database is 0, and the benchmarks array is empty, meaning no direct performance measurements are recorded.

Benchmark Performance

The database does not list any synthetic benchmark scores for the HD 8730A; the benchmarks array is empty. Consequently, the only performance metric available is the 50th percentile ranking across all GPUs tracked in the database. This percentile implies that the GPU performs at the exact median of the entire GPU population, outperforming half of the GPUs and trailing the other half. The nearestRivals array is also empty, so no direct percentage deltas against competing GPUs can be provided. The raw compute throughput is 537.6 GFLOPS, derived from 384 shading units operating at a 700 MHz boost clock. This FP32 figure is modest for a 28 nm part, and it places the GPU in the entry-level segment of its era. The pixel fill rate is 5.600 GPixel/s, and the texture fill rate is 16.80 GTexel/s. These rates indicate that the GPU can handle 1080p rendering at low to medium settings, but it will struggle with high-resolution or high-detail workloads. The boost clock of 700 MHz is only slightly above the base 650 MHz, suggesting a narrow power envelope and limited thermal headroom. The transistor density of 12.3M per mm² is a product of the 950 million transistors on a 77 mm² die, which is typical for a mid-range 28 nm GPU. The 8 ROPs limit the pixel throughput, while the 24 TMUs handle texture filtering, and the 384 shaders execute vertex and pixel operations. Without rivals to compare against, the 50th percentile is the only anchor, and it suggests a balanced but unremarkable performance profile.

Ray Tracing and Feature Set

The HD 8730A does not include dedicated ray tracing cores or tensor cores; both fields are null in the specification. Consequently, hardware-accelerated ray tracing is not available on this GPU. The feature set relies on the GCN 1.0 architecture, which provides a baseline for DirectX 12 (11_1) support, along with OpenGL 4.6 and Vulkan 1.2.170. The DirectX 12 support is limited to feature level 11_1, which means it does not support the full DirectX 12 feature set, such as mesh shaders or variable rate shading. This limits the GPU to games that can run on the 11_1 feature level, which includes many titles from the DirectX 11 era. Vulkan 1.2.170 support allows modern Vulkan titles to run, though the lack of dedicated RT cores means any ray tracing effects must be computed via software or fallback paths. The absence of tensor cores also means no hardware-accelerated machine learning features are available. The API list is the only software feature set provided; display outputs and other connectivity options are not specified. The architecture is GCN 1.0, which is the first generation of AMD's Graphics Core Next, known for its compute-oriented design. This architecture supports asynchronous compute, though the implementation is early-generation and may not be as efficient as later GCN iterations. The OpenGL 4.6 support ensures compatibility with a wide range of professional and legacy applications, while Vulkan 1.2.170 offers a modern low-overhead API for game developers. However, the lack of RT and tensor cores places this GPU firmly in the rasterization-only category.

Memory Subsystem

The memory subsystem is a key bottleneck for this GPU. It is equipped with 1024 MB of DDR3 memory on a 128-bit bus, resulting in a memory bandwidth of 32.00 GB/s. The memory clock is 1000 MHz, which translates to 2 Gbps effective data rate. For a GPU with a pixel rate of 5.600 GPixel/s, the 32.00 GB/s bandwidth is sufficient for low-resolution rendering but becomes a constraint at higher resolutions. The 1024 MB frame buffer is modest by modern standards, limiting the ability to load high-resolution textures without exceeding capacity. The 128-bit bus width is narrow compared to higher-tier GPUs, which typically use wider buses. The bandwidth of 32.00 GB/s is a fraction of what high-end GPUs offer, reinforcing the entry-level positioning. At 1080p with medium textures, the memory capacity may suffice for older titles, but modern games with large texture packs will likely exceed the 1024 MB limit, causing stuttering or texture thrashing. The memory type is DDR3, and the effective data rate of 2 Gbps is low, and the 128-bit bus limits the total bandwidth to 32.00 GB/s. This bandwidth must feed 384 shading units and 24 TMUs, which can create a bottleneck in texture-heavy scenes. The 8 ROPs are responsible for pixel output, and with a pixel rate of 5.600 GPixel/s, the GPU is limited to that rate, which is adequate for 1080p but not for higher refresh rates. The memory clock of 1000 MHz is the base for the 2 Gbps effective rate, and the 32.00 GB/s bandwidth is the ceiling for data transfer between the GPU and VRAM.

Who Should Consider It

Given the 50th percentile ranking and the memory constraints, this GPU is best suited for users who prioritize low-resolution gaming or legacy applications. The 1024 MB frame buffer and 32.00 GB/s bandwidth indicate that 720p or 1080p with low-to-medium settings is the realistic ceiling. The 537.6 GFLOPS FP32 performance is adequate for older DirectX 11 titles but will struggle with recent releases. The bus interface is PCIe 3.0 x8, which provides a reduced lane count compared to x16 slots, but the bandwidth is unlikely to be a limiting factor given the GPU's modest compute capabilities. The 28 nm process node and 77 mm² die size suggest a low-power design, though TDP is not listed. The generation label "All-In-One (HD 8000)" indicates it was designed for all-in-one systems, so it may be found in pre-built machines rather than as a discrete upgrade. Users with legacy systems that require a basic display output or light 2D/3D acceleration may find it acceptable, but for any serious gaming, the 50th percentile standing means it is below the average modern GPU. The 8 ROPs and 5.600 GPixel/s pixel rate limit the output to 1080p in simple scenes. The texture rate of 16.80 GTexel/s is similarly limited, so games with heavy texture filtering will see performance drops. The GPU's 384 shading units can handle basic vertex and pixel shaders, but complex shaders with many instructions will reduce frame rates. For users who only need a basic display output for office work or media playback, the HD 8730A can serve adequately, but it is not a gaming GPU by modern standards. The lack of dedicated RT and tensor cores further narrows its use case to rasterized graphics only. The 50th percentile ranking places it exactly at the median, meaning it is neither a low-end outlier nor a high-performance part, but rather a middle-of-the-road solution for undemanding workloads.

FAQ

Q: What is the memory size of the AMD Radeon HD 8730A?

A: The GPU is equipped with 1024 MB of DDR3 memory.

Q: What is the memory bus width?

A: The memory bus width is 128 bits.

Q: What is the FP32 compute performance?

A: The FP32 performance is 537.6 GFLOPS.

Q: Does it support hardware ray tracing?

A: No, the GPU has no dedicated ray tracing cores; the RT cores field is null.

Q: What DirectX version does it support?

A: It supports DirectX 12 (11_1), which corresponds to feature level 11_1.

Q: What is the process node used?

A: The process node is 28 nm, fabricated by TSMC.