AMD Radeon HD 8750M vs NVIDIA GeForce GTX 765M Comparison

The NVIDIA GeForce GTX 765M and AMD Radeon HD 8750M are both end-of-life mobile graphics solutions from 2013, but they represent fundamentally different design philosophies. The GTX 765M is a large, fully-featured Kepler chip, while the HD 8750M is a compact GCN 1.0 part. Benchmark data shows a clear performance hierarchy, with the NVIDIA part leading in compute workloads, but the AMD card holds its own in relative standings due to the broader landscape of competing mobile GPUs.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon HD 8750M reports a higher average benchmark score of 5970, while the NVIDIA GeForce GTX 765M averages 5501. This difference is notable because it contradicts the head-to-head result, which is explained by the fact that the AMD card has only one benchmark score (OpenCL) contributing to its average, whereas the NVIDIA card has three (Metal, OpenCL, and Vulkan).

Q: In the direct head-to-head OpenCL test, what is the performance gap?

A: The NVIDIA GeForce GTX 765M wins the Geekbench OpenCL test with a score of 7176 versus the AMD Radeon HD 8750M's 5970. This represents a 20.2% advantage for the NVIDIA part, making it the decisive winner in the only directly comparable benchmark.

Q: How do their transistor counts and die sizes compare?

A: The NVIDIA GeForce GTX 765M integrates 2,540 million transistors on a 221 mm² die, while the AMD Radeon HD 8750M packs 950 million transistors on a much smaller 77 mm² die. Despite the smaller size, the AMD chip achieves a slightly higher transistor density at 12.3M / mm² versus 11.5M / mm² for NVIDIA.

Q: What are the memory specifications for each card?

A: The NVIDIA GeForce GTX 765M comes with 2 GB of GDDR5 memory on a 128-bit bus, delivering 64.13 GB/s of bandwidth. The AMD Radeon HD 8750M is equipped with 1024 MB of DDR3 memory, also on a 128-bit bus, but offers significantly lower bandwidth at 28.80 GB/s.

Q: Which GPU has a higher pixel and texture fill rate?

A: The NVIDIA GeForce GTX 765M is substantially faster, with a pixel rate of 13.81 GPixel/s and a texture rate of 55.23 GTexel/s. The AMD Radeon HD 8750M trails considerably, with 6.600 GPixel/s and 19.80 GTexel/s respectively.

Q: How does the NVIDIA card's percentile ranking compare to the AMD card's?

A: The AMD Radeon HD 8750M sits at the 34th percentile of all GPUs, which is slightly higher than the NVIDIA GeForce GTX 765M's 32nd percentile. This indicates that while the NVIDIA part wins its direct comparison, the AMD card is positioned marginally better relative to the entire GPU market.

Architecture Differences

The core architectural split here is between NVIDIA's Kepler generation and AMD's Graphics Core Next (GCN) 1.0. The NVIDIA GeForce GTX 765M is built on the GK106 chip, which is a derivative of the larger Kepler designs, manufactured on a 28 nm process at TSMC. The AMD Radeon HD 8750M uses the smaller "Mars" chip, also on TSMC's 28 nm node, but with a fundamentally different compute architecture optimized for parallel workloads.

The most striking difference is physical scale. The NVIDIA chip contains 2,540 million transistors across a 221 mm² die, making it nearly three times larger than the AMD chip, which packs 950 million transistors into just 77 mm². This size disparity directly translates into raw resource counts: the GTX 765M has 768 shading units, 64 texture mapping units (TMUs), and 16 raster operation units (ROPs). The HD 8750M, by contrast, is configured with 384 shading units, 24 TMUs, and 8 ROPs—exactly half the shaders, 37.5% of the TMUs, and half the ROPs.

Clock behavior also differs. The NVIDIA part operates at a base clock of 797 MHz with a boost up to 863 MHz, while the AMD card runs at 775 MHz base and 825 MHz boost. Although the clocks are similar, the NVIDIA card's higher shader count and wider texture pipeline produce far greater theoretical throughput. Memory architecture is another major divergence: the GTX 765M uses 2 GB of GDDR5 at 1002 MHz (4 Gbps effective), yielding 64.13 GB/s of bandwidth. The HD 8750M relies on 1024 MB of DDR3 at 900 MHz (1800 Mbps effective), which caps bandwidth at 28.80 GB/s—less than half of NVIDIA's figure.

API support shows subtle differences. Both support DirectX 12 (with NVIDIA at 11_0 feature level and AMD at 11_1) and OpenGL 4.6. Vulkan support is present on both, with NVIDIA at version 1.2.175 and AMD at 1.2.170. The bus interface also differs: NVIDIA uses an MXM-B (3.0) connector, while AMD utilizes PCIe 3.0 x8. The NVIDIA card is specified with a 75 W TDP and no power connectors, indicating an MXM module design; AMD does not list a TDP or slot width. The GTX 765M has a later release date of May 2013, while the HD 8750M launched in February 2013.

Head-to-Head Benchmarks

The only directly comparable benchmark between these two GPUs is Geekbench OpenCL, and the results are unambiguous. The NVIDIA GeForce GTX 765M scores 7176, while the AMD Radeon HD 8750M reaches 5970. This yields a 20.2% delta in favor of NVIDIA, a substantial margin that aligns with the raw specifications. The GTX 765M's advantage in shading units (768 vs 384), TMUs (64 vs 24), and memory bandwidth (64.13 GB/s vs 28.80 GB/s) all contribute to this OpenCL performance gap. The higher boost clock of 863 MHz versus 825 MHz adds a small additional edge.

However, the aggregate benchmark picture is more nuanced. The NVIDIA card's average benchmark score of 5501 is derived from three tests: Geekbench Metal (2612), OpenCL (7176), and Vulkan (6714). The AMD card's average of 5970 is based solely on its OpenCL result. If one were to compare only the OpenCL scores, NVIDIA is clearly superior. But the AMD card's higher overall percentile (34th vs 32nd) and higher average score suggest that in a broader context, the HD 8750M performs competitively against a wider range of GPUs. This is reflected in their nearest rivals: the GTX 765M sits near the NVIDIA GeForce MX130 (deltaPct -0.1) and AMD Radeon R7 M440 (deltaPct 0.3), while the HD 8750M is closely matched with the NVIDIA Quadro K4000 (deltaPct -0.2) and Quadro K620M (deltaPct 0.2).

The wins tally is 1-0 in favor of NVIDIA. The single head-to-head test represents the only direct comparison available, and NVIDIA dominates it. The 20.2% delta is significant enough to suggest that in compute-heavy applications—such as OpenCL-accelerated rendering or physics simulations—the GTX 765M is the more capable part. The AMD card's absence of Metal and Vulkan benchmark scores limits its comparative scope, but it does demonstrate respectable OpenCL performance given its much smaller die and lower power envelope (though TDP is not listed for AMD).

The Verdict

Based strictly on the data, the NVIDIA GeForce GTX 765M is the superior performer in direct comparison. Its 20.2% lead in OpenCL, combined with roughly double the shading units, TMUs, and ROPs, plus more than twice the memory bandwidth, positions it as the stronger choice for raw compute and texture-heavy workloads. The 2 GB GDDR5 frame buffer also offers more capacity than the 1 GB DDR3 on the AMD card, which matters for larger textures or higher-resolution framebuffers.

The AMD Radeon HD 8750M, however, is not without merit. Its higher average benchmark score (5970 vs 5501) and better percentile ranking (34 vs 32) indicate that it performs well relative to the entire GPU market, even if it loses the direct head-to-head. The smaller die size (77 mm² vs 221 mm²) and lower transistor count (950 million vs 2,540 million) suggest it is a more power-efficient design, though no TDP is listed for AMD. The HD 8750M's nearest rivals include the NVIDIA Quadro K4000 and K620M, with deltas under 0.3%, meaning it is tightly clustered with mid-range professional GPUs.

For users prioritizing maximum compute performance, the GTX 765M is the clear pick. Its higher clock speeds, larger memory pool, and superior fill rates make it the better part for gaming at higher settings or running GPU-accelerated applications. The AMD card is better suited for scenarios where the smaller footprint and lower power draw are advantageous, such as ultrathin notebooks. The data shows the GTX 765M wins the only direct benchmark, but the HD 8750M's stronger market position suggests it is a more balanced product for its intended segment. Neither card is current—both are end-of-life—so the choice depends entirely on the specific workload and the system they are installed in.

Specification Differences

| Specification | NVIDIA GeForce GTX 765M | AMD Radeon HD 8750M |

|---|---|---|

| Chip | GK106 | Mars |

| Architecture | Kepler | GCN 1.0 |

| Transistors | 2,540 million | 950 million |

| Die Size | 221 mm² | 77 mm² |

| Transistor Density | 11.5M / mm² | 12.3M / mm² |

| Base Clock | 797 MHz | 775 MHz |

| Boost Clock | 863 MHz | 825 MHz |

| Memory Size | 2 GB | 1024 MB |

| Memory Type | GDDR5 | DDR3 |

| Memory Clock | 1002 MHz (4 Gbps effective) | 900 MHz (1800 Mbps effective) |

| Memory Bandwidth | 64.13 GB/s | 28.80 GB/s |

| Shading Units | 768 | 384 |

| TMUs | 64 | 24 |

| ROPs | 16 | 8 |

| Pixel Rate | 13.81 GPixel/s | 6.600 GPixel/s |

| Texture Rate | 55.23 GTexel/s | 19.80 GTexel/s |

| FP32 | 1,325.6 GFLOPS | 633.6 GFLOPS |

| TDP | 75 W | Not listed |

| Bus Interface | MXM-B (3.0) | PCIe 3.0 x8 |

| Vulkan Version | 1.2.175 | 1.2.170 |

| Release Date | 2013-05-29 | 2013-02-25 |

| Predecessor | GeForce 600M | London |

| Successor | GeForce 800M | Gem System |