AMD Radeon HD 8750M vs NVIDIA GeForce 930A Comparison

FAQ

Q: How do the two GPUs compare in overall benchmark performance?

A: The recorded data shows the AMD Radeon HD 8750M scores 5970 in Geekbench OpenCL, while the NVIDIA GeForce 930A scores 5317. The AMD part leads by 12.3% in this single head-to-head test.

Q: Where does each GPU sit relative to the broader GPU landscape?

A: The AMD Radeon HD 8750M lands in the 34th percentile of all GPUs, while the NVIDIA GeForce 930A sits in the 31st percentile. Both are entry-level mobile parts, but the AMD chip ranks slightly higher overall.

Q: What are the closest rivals to the AMD Radeon HD 8750M?

A: The nearest competitors by average score are the NVIDIA Quadro K4000M (5986, 0.3% higher), NVIDIA Quadro K4000 (5982, 0.2% higher), NVIDIA Quadro K620M (5957, 0.2% lower), and AMD Radeon HD 8730M (5955, 0.3% lower). The AMD HD 8750M essentially trades blows with all of them.

Q: What are the closest rivals to the NVIDIA GeForce 930A?

A: The nearest competitors are the AMD Radeon R7 M445 (5358, 0.8% higher), NVIDIA GeForce 840M (5322, 0.1% higher), NVIDIA GeForce GTX 980M (5308, 0.2% lower), and NVIDIA GeForce 940M (5284, 0.6% lower). The 930A sits in a tight cluster around the 5300 mark.

Q: Do both GPUs support the same DirectX version?

A: No. The AMD Radeon HD 8750M lists DirectX 12 (11_1), while the NVIDIA GeForce 930A lists DirectX 12 (11_0). The AMD part has a slightly higher feature level.

Q: Which GPU has more memory bandwidth?

A: The AMD Radeon HD 8750M has 28.80 GB/s of bandwidth, while the NVIDIA GeForce 930A has 16.02 GB/s. The AMD part delivers nearly 80% more bandwidth despite having a 128-bit bus versus the NVIDIA's 64-bit bus.

Where Each One Wins

The AMD Radeon HD 8750M wins the only recorded head-to-head benchmark, the Geekbench OpenCL test, with a score of 5970 versus 5317 for the NVIDIA GeForce 930A, a 12.3% advantage. This indicates the AMD part has a clear edge in compute-oriented workloads that stress raw shader throughput and memory bandwidth. The AMD chip also carries a wider 128-bit memory bus and significantly more bandwidth (28.80 GB/s versus 16.02 GB/s), which can benefit memory-heavy tasks such as texture streaming or large buffer operations.

The NVIDIA GeForce 930A, by contrast, wins in the efficiency and feature-set domain rather than raw performance. It has a higher base clock (928 MHz versus 775 MHz) and boost clock (941 MHz versus 825 MHz), and it draws only 33 W with no power connectors required, making it a more power-lean option for thin portable devices. The NVIDIA part also supports Vulkan 1.4, whereas the AMD part lists Vulkan 1.2.170, so software that leverages the newest Vulkan features may favor the NVIDIA solution. Additionally, the NVIDIA chip comes with 2 GB of memory versus the AMD's 1 GB, which can matter for workloads that exceed the 1 GB capacity, even if the AMD bus is faster.

Architecture Differences

The AMD Radeon HD 8750M is built on Graphics Core Next 1.0 (GCN 1.0), using the Mars chip. The NVIDIA GeForce 930A is built on Maxwell, using the GM108 chip. Both are manufactured on a 28 nm process at TSMC, and both have a die size of 77 mm². However, the transistor counts differ slightly: the AMD chip packs 950 million transistors, while the NVIDIA chip has 1,020 million. This results in a transistor density of 12.3M per mm² for AMD and 13.2M per mm² for NVIDIA, meaning the Maxwell architecture crams a bit more logic into the same area.

The AMD part belongs to the Solar System generation (HD 8700M family), released in early 2013, with a predecessor named London and a successor named Gem System. The NVIDIA part belongs to the GeForce 900A generation, released in early 2015, with a predecessor named GeForce 800A and no listed successor. The two-year gap shows in their API support: AMD lists DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170, while NVIDIA lists DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4. Both support PCIe 3.0 x8, so the host interface is identical.

A notable architectural difference is the memory subsystem design. The AMD chip uses a 128-bit bus with DDR3 memory at 1800 Mbps effective, yielding 28.80 GB/s. The NVIDIA chip uses a 64-bit bus with DDR3 at 2 Gbps effective, yielding only 16.02 GB/s. The NVIDIA part compensates with a smaller bus but higher effective memory clock. The core configuration, however, is identical: both have 384 shading units, 24 texture mapping units, and 8 raster operation pipelines. The differences in raw throughput therefore come from clock speeds rather than raw ALU counts.

Specification Differences

The two GPUs differ in several key specification fields. The AMD Radeon HD 8750M has a base clock of 775 MHz and a boost clock of 825 MHz, while the NVIDIA GeForce 930A has a base clock of 928 MHz and a boost clock of 941 MHz, making the NVIDIA part roughly 20% faster on paper at the core level. The memory configuration diverges sharply: AMD offers 1024 MB of DDR3 on a 128-bit bus with 28.80 GB/s bandwidth, while NVIDIA offers 2048 MB of DDR3 on a 64-bit bus with 16.02 GB/s bandwidth. The AMD part has 950 million transistors, the NVIDIA part 1,020 million, on the same 77 mm² die size.

Pixel and texture rates also favor the NVIDIA chip. The NVIDIA part achieves 7.528 GPixel/s and 22.58 GTexel/s, versus 6.600 GPixel/s and 19.80 GTexel/s for AMD. Floating-point throughput follows the same pattern: NVIDIA lists 722.7 GFLOPS of FP32, while AMD lists 633.6 GFLOPS, a 14% advantage for NVIDIA. The NVIDIA part also carries a TDP of 33 W, whereas the AMD part has no TDP listed in the database, and it is classified as an IGP with no power connectors. The AMD part, by comparison, has no slot width or power connector data recorded.

The production status for both is end-of-life. The AMD part was released on February 25, 2013, and the NVIDIA part on March 12, 2015. The NVIDIA part supports Vulkan 1.4, which is a newer API revision than the AMD's Vulkan 1.2.170. Neither part has a launch MSRP recorded in the database.

Head-to-Head Benchmarks

The only head-to-head benchmark recorded is Geekbench OpenCL. In that test, the AMD Radeon HD 8750M scores 5970, and the NVIDIA GeForce 930A scores 5317. The delta is 12.3% in favor of AMD. This is a substantial margin for two parts that are architecturally different but share the same shading unit count (384), TMU count (24), and ROP count (8). The AMD part wins despite having lower clocks (775 MHz base versus 928 MHz base), which strongly suggests that the wider 128-bit memory bus and higher bandwidth (28.80 GB/s versus 16.02 GB/s) play a decisive role in OpenCL compute workloads. Memory-bound kernels often scale with bandwidth, and the AMD part has an 80% bandwidth advantage.

The NVIDIA part, however, wins the theoretical throughput comparison in pixel rate (7.528 GPixel/s versus 6.600 GPixel/s), texture rate (22.58 GTexel/s versus 19.80 GTexel/s), and FP32 throughput (722.7 GFLOPS versus 633.6 GFLOPS). These numbers indicate that in pure shading or texture-heavy scenarios, the NVIDIA chip might outperform the AMD part, yet the actual OpenCL score tells the opposite story. The implication is that the AMD GCN architecture extracts more useful compute performance per clock, or that the benchmark is sensitive to memory bandwidth rather than raw ALU throughput.

The nearest rival data reinforces this interpretation. The AMD HD 8750M sits within a 0.3% band of the Quadro K4000, Quadro K620M, HD 8730M, and Quadro K4000M, all of which score between 5955 and 5986. The NVIDIA 930A sits within a 0.8% band of the R7 M445, GeForce 840M, GTX 980M, and GeForce 940M, all of which score between 5284 and 5358. The two clusters do not overlap: the AMD cluster is roughly 600 points higher. This suggests that the AMD chip competes in a higher performance tier than the NVIDIA chip, at least in OpenCL compute.

The Verdict

The data points to a clear compute-performance winner: the AMD Radeon HD 8750M beats the NVIDIA GeForce 930A by 12.3% in Geekbench OpenCL, scoring 5970 versus 5317. The AMD part also holds a significant bandwidth advantage (28.80 GB/s versus 16.02 GB/s) and a wider memory bus (128-bit versus 64-bit), which likely explains its OpenCL lead. For users whose workloads involve general-purpose GPU compute, OpenCL acceleration, or memory-intensive tasks, the AMD part is the stronger choice based on the recorded measurements.

The NVIDIA GeForce 930A, however, has its own advantages that matter in different contexts. It offers 2 GB of memory versus 1 GB, which can be the deciding factor for applications that need more capacity than the AMD part can provide. It also has higher clocks (928 MHz base, 941 MHz boost), higher pixel rate (7.528 GPixel/s), higher texture rate (22.58 GTexel/s), higher FP32 throughput (722.7 GFLOPS), a lower TDP of 33 W, and newer Vulkan support (1.4 versus 1.2.170). In a thin-and-light laptop where power draw and API compatibility matter more than raw compute, the NVIDIA part is the more sensible option.

The percentile rankings confirm the gap: the AMD part sits at the 34th percentile of all GPUs, while the NVIDIA part sits at the 31st. Both are entry-level mobile parts, but the AMD chip sits slightly higher in the overall distribution. The nearest rival clusters also tell a story: the AMD HD 8750M trades closely with professional Quadro parts like the K4000 and K4000M, while the NVIDIA 930A trades closely with consumer GeForce parts like the 840M and 940M. The AMD part therefore punches slightly above its class in compute terms, while the NVIDIA part is better positioned for power-constrained designs. Pick the AMD part for raw OpenCL performance and bandwidth; pick the NVIDIA part for capacity, efficiency, and newer API support.