AMD Radeon HD 8750M vs NVIDIA GeForce GTX 460 SE Comparison

The NVIDIA GeForce GTX 460 SE and the AMD Radeon HD 8750M represent two distinct approaches to GPU design from different eras, with the former being a desktop part built on Fermi and the latter a mobile chip based on GCN 1.0. In the single benchmark test available, the GeForce GTX 460 SE posts a Geekbench OpenCL score of 6389, which places it 7% ahead of the Radeon HD 8750M’s score of 5970. The data shows a narrow but definitive win for the NVIDIA card, driven by its older architecture and significantly wider memory interface, despite the AMD part being built on a much newer process node.

Where Each One Wins

The benchmark results indicate a clear, albeit singular, advantage for the NVIDIA GeForce GTX 460 SE. It wins the only head-to-head test, Geekbench OpenCL, with a score of 6389 against the AMD Radeon HD 8750M’s 5970. This 7% delta is the sole measure of performance provided, meaning the GTX 460 SE is the outright winner in raw compute output. Its 748.8 GFLOPS of FP32 compute power, combined with a texture rate of 31.20 GTexel/s, gives it a mathematical edge in workloads that rely on these specific throughput metrics.

The AMD Radeon HD 8750M, while losing the overall benchmark, shows strengths in architectural efficiency and feature support rather than raw score. It has a higher shading unit count (384 vs. 288), which indicates a design intended for a different class of workload distribution. Its GCN 1.0 architecture also supports Vulkan 1.2.170, a feature the GTX 460 SE lacks entirely, making it the more modern choice for API compatibility. The HD 8750M’s wins are qualitative—it is a later product with a smaller die and a lower transistor count, suggesting it achieves its performance with less physical resources.

Neither part wins on memory capacity, as both have 1024 MB, but the GTX 460 SE dominates in memory bandwidth with 108.8 GB/s versus the HD 8750M’s 28.80 GB/s. This is a critical differentiator for bandwidth-sensitive tasks. The data shows that the NVIDIA card wins the performance crown, while the AMD card wins on architectural modernity and API feature set, though neither advantage translates into a benchmark victory for the latter.

Architecture Differences

The two GPUs are built on fundamentally different foundations. The NVIDIA GeForce GTX 460 SE uses the GF104 chip on a 40 nm TSMC process, packing 1,950 million transistors into a 332 mm² die. This results in a transistor density of 5.9M per mm². In contrast, the AMD Radeon HD 8750M uses the Mars chip on a 28 nm TSMC process, with 950 million transistors on a much smaller 77 mm² die, achieving a higher density of 12.3M per mm². The AMD part is physically more efficient, but the NVIDIA part has over twice the transistor count.

The compute configurations diverge significantly. The GTX 460 SE has 288 shading units, 48 TMUs, and 32 ROPs, while the HD 8750M has 384 shading units, 24 TMUs, and only 8 ROPs. This means the NVIDIA card has double the texture units and four times the ROPs, which explains its higher pixel rate of 7.800 GPixel/s versus 6.600 GPixel/s. The AMD card compensates with more shading units, but its texture rate of 19.80 GTexel/s is well below the GTX 460 SE’s 31.20 GTexel/s. The FP32 throughput also favors NVIDIA at 748.8 GFLOPS versus 633.6 GFLOPS.

Memory architecture is the starkest contrast. The GTX 460 SE uses a 256-bit GDDR5 bus running at 850 MHz (3.4 Gbps effective), yielding 108.8 GB/s of bandwidth. The HD 8750M uses a 128-bit DDR3 bus at 900 MHz (1800 Mbps effective), yielding only 28.80 GB/s. This 3.8x bandwidth difference is a massive factor in real-world performance. The GTX 460 SE also has a dual-slot cooler and requires 2x 6-pin power connectors with a 150 W TDP, while the HD 8750M has no listed TDP, slot width, or power connectors, indicating it is designed for low-power mobile integration. The NVIDIA card uses PCIe 2.0 x16, while the AMD card uses PCIe 3.0 x8.

Head-to-Head Benchmarks

The only available benchmark, Geekbench OpenCL, provides a straightforward comparison. The NVIDIA GeForce GTX 460 SE scores 6389, while the AMD Radeon HD 8750M scores 5970. This results in a 7% delta in favor of the NVIDIA card. This is a meaningful lead, but it is not overwhelming, and it reflects the compute-oriented nature of the OpenCL test, which likely favors the GTX 460 SE’s higher memory bandwidth and pixel throughput.

Looking at the nearest rivals for each card adds context to the scores. The GTX 460 SE is exactly tied with the NVIDIA GeForce GTX 580M at 6389, and it is only 0.3% behind the NVIDIA RTX PRO 5000 72 GB Blackwell. It is 0.9% ahead of the AMD Radeon Pro WX 4100 and 1% ahead of the AMD Radeon R7 M350. This places the GTX 460 SE in a competitive tier where its performance is essentially equivalent to a high-end mobile part from a later generation.

The HD 8750M’s score of 5970 is similarly competitive within its own peer group. It is 0.2% behind the NVIDIA Quadro K4000 (5982) and 0.3% behind the NVIDIA Quadro K4000M (5986). It is 0.2% ahead of the NVIDIA Quadro K620M (5957) and 0.3% ahead of the AMD Radeon HD 8730M (5955). The delta between the two cards is larger than the differences within their respective rival groups, reinforcing that the GTX 460 SE holds a genuine performance edge in this test. The data shows no benchmark where the AMD card wins, as the win count is 1 for NVIDIA and 0 for AMD.

The Verdict

Based strictly on the data, the NVIDIA GeForce GTX 460 SE is the superior performer. It wins the only benchmark by 7%, and its underlying specifications—specifically the 108.8 GB/s memory bandwidth and 32 ROPs—suggest it will handle memory-intensive and pixel-heavy workloads more effectively. For users prioritizing raw compute in OpenCL, the choice is clear. The GTX 460 SE also has a launch MSRP of 160 USD, which provides a reference point for its original market position.

The AMD Radeon HD 8750M is the better choice for scenarios where architectural efficiency and modern API support matter more than raw benchmark scores. Its support for Vulkan 1.2.170 and DirectX 12 (11_1) versus the GTX 460 SE’s DirectX 12 (11_0) and lack of Vulkan gives it a compatibility advantage for newer software. Its 28 nm process and 77 mm² die make it a far more compact and power-efficient design, which is critical for mobile implementations. However, its 28.80 GB/s memory bandwidth is a severe bottleneck that limits its practical performance.

The data shows that the GTX 460 SE is a desktop-class card with a 150 W TDP and dual-slot cooler, while the HD 8750M is a mobile-class part with no listed power draw. They were designed for different form factors, and the benchmark reflects the GTX 460 SE’s advantage in raw throughput. For a desktop user with access to a 450 W PSU and 2x 6-pin connectors, the GTX 460 SE is the clear winner. For a laptop user needing a low-power part with Vulkan support, the HD 8750M is the only sensible option despite its lower score.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The NVIDIA GeForce GTX 460 SE scores 6389, which is 7% higher than the AMD Radeon HD 8750M’s score of 5970.

Q: How does the memory bandwidth compare between the two cards?

A: The GTX 460 SE has a 256-bit GDDR5 interface providing 108.8 GB/s, while the HD 8750M has a 128-bit DDR3 interface providing 28.80 GB/s.

Q: Does the AMD Radeon HD 8750M support Vulkan?

A: Yes, the HD 8750M supports Vulkan 1.2.170, whereas the GTX 460 SE has no Vulkan support listed.

Q: What is the difference in transistor count and die size?

A: The GTX 460 SE has 1,950 million transistors on a 332 mm² die, while the HD 8750M has 950 million transistors on a 77 mm² die.

Q: Which card is tied with the NVIDIA GeForce GTX 580M in the benchmark?

A: The NVIDIA GeForce GTX 460 SE is exactly tied with the GTX 580M at a score of 6389.

Q: What is the pixel rate difference between the two GPUs?

A: The GTX 460 SE achieves 7.800 GPixel/s, which is higher than the HD 8750M’s 6.600 GPixel/s, due to its 32 ROPs versus 8 ROPs.