AMD Radeon HD 8690M vs NVIDIA GeForce GTX 765M Comparison

FAQ

Q: Which GPU is faster in the Geekbench OpenCL benchmark?

A: The NVIDIA GeForce GTX 765M scores 7176, while the AMD Radeon HD 8690M scores 6137. That puts the GTX 765M ahead by 14.5%, the only direct head-to-head benchmark recorded.

Q: How do these two GPUs rank against all other GPUs in the database?

A: The AMD Radeon HD 8690M sits at the 36th percentile, while the NVIDIA GeForce GTX 765M sits at the 32nd percentile. Despite the GTX 765M winning the direct comparison, its average benchmark score is lower because it includes additional tests that drag its average down.

Q: What is the average benchmark score for each card?

A: The AMD Radeon HD 8690M averages 6137, which comes from a single OpenCL test. The NVIDIA GeForce GTX 765M averages 5501, which is the mean of three tests: OpenCL at 7176, Vulkan at 6714, and Metal at 2612.

Q: Which GPU has the larger memory bus?

A: The NVIDIA GeForce GTX 765M has a 128-bit memory bus, double the 64-bit bus of the AMD Radeon HD 8690M. This gives the GTX 765M a memory bandwidth of 64.13 GB/s versus 32.00 GB/s for the AMD card.

Q: Do both GPUs support modern graphics APIs?

A: Both support DirectX 12 and OpenGL 4.6. The AMD card supports Vulkan 1.2.170, while the NVIDIA card supports a slightly newer Vulkan 1.2.175. The AMD card's DirectX support is listed as 12 (11_1), meaning it targets the 11_1 feature level.

Q: What are the release dates for these two mobile GPUs?

A: The AMD Radeon HD 8690M was released on February 28, 2013. The NVIDIA GeForce GTX 765M followed three months later on May 29, 2013. Both are now end-of-life products.

Architecture Differences

The AMD Radeon HD 8690M uses the Sun chip built on GCN 1.0 architecture, while the NVIDIA GeForce GTX 765M uses the GK106 chip on Kepler architecture. Both are fabricated by TSMC on a 28 nm process, but the similarities end there.

The transistor counts differ dramatically. AMD packs 690 million transistors into a 56 mm² die, resulting in a transistor density of 12.3 million per mm². NVIDIA uses 2,540 million transistors across a 221 mm² die, which works out to 11.5 million per mm². The NVIDIA chip is nearly four times larger by die area and has roughly 3.7 times more transistors.

The compute configurations tell the story of two different design philosophies. The AMD card has 320 shading units, 20 texture mapping units, and 8 raster output units. The NVIDIA card fields 768 shading units, 64 TMUs, and 16 ROPs. In every category, NVIDIA has more than double the hardware resources.

Clock speeds go the other way. The AMD card runs at a 925 MHz base clock and boosts to 975 MHz. The NVIDIA card starts lower at 797 MHz and boosts to 863 MHz. The higher clocks help AMD close some of the gap, but the raw throughput numbers still favor NVIDIA heavily.

Memory configurations are a major split. Both cards use 2 GB of GDDR5, but the AMD card has a 64-bit bus delivering 32.00 GB/s, while the NVIDIA card has a 128-bit bus delivering 64.13 GB/s. The memory clock is nearly identical at 1000 MHz versus 1002 MHz, so the bandwidth difference comes entirely from the bus width.

The AMD card connects via PCIe 3.0 x8, while the NVIDIA card uses an MXM-B (3.0) interface. The NVIDIA card is a full MXM module with a 75 W TDP and no additional power connectors. The AMD card has no listed TDP or power connector information.

Head-to-Head Benchmarks

The database records a single direct comparison between these two GPUs: the Geekbench OpenCL test. The NVIDIA GeForce GTX 765M scores 7176, beating the AMD Radeon HD 8690M's 6137 by 14.5%. This is a decisive win for NVIDIA in the one test where both cards were measured under identical conditions.

The OpenCL result aligns with the raw specifications. NVIDIA's 1,325.6 GFLOPS of FP32 performance is more than double AMD's 624.0 GFLOPS. The pixel rate of 13.81 GPixel/s versus 7.800 GPixel/s and the texture rate of 55.23 GTexel/s versus 19.50 GTexel/s all point in the same direction. The GTX 765M has roughly 2.1 times the compute throughput, 1.8 times the pixel throughput, and 2.8 times the texture throughput.

Looking at the nearest rival data provides context for both cards. The AMD Radeon HD 8690M sits within 1.2% of the AMD Radeon HD 6950, which scores 6210. It is also close to the Intel Iris Pro Graphics 6200 at 6117 (0.3% delta), the NVIDIA RTX A400 at 6078 (1% delta), and the NVIDIA GeForce MX230 at 6077 (1% delta). These rivals cluster tightly around the 6077 to 6210 range, showing the AMD card is competitive with a specific tier of GPUs.

The NVIDIA GeForce GTX 765M's nearest rivals include the NVIDIA GeForce MX130 at 5508 (-0.1% delta), the AMD Radeon R7 M440 at 5483 (0.3% delta), and the NVIDIA Quadro M4000 at 5467 (0.6% delta). The AMD FirePro M4000 scores 5537, which is 0.7% ahead of the GTX 765M. This cluster sits in the 5467 to 5537 range, meaning the GTX 765M's average score is lower than the AMD HD 8690M's average, even though the direct comparison favors NVIDIA.

The GTX 765M's Vulkan score of 6714 and Metal score of 2612 show significant variation depending on the API. The OpenCL result is its strongest, while the Metal result is far weaker. The AMD card only has an OpenCL score, so no direct comparison is possible on Vulkan or Metal.

The Verdict

The data tells a clear story for raw compute performance: the NVIDIA GeForce GTX 765M wins the only direct head-to-head benchmark by 14.5%. Its higher shading unit count, wider memory bus, and greater FP32 throughput give it a substantial advantage in OpenCL workloads. Anyone choosing between these two for compute-heavy tasks should favor the GTX 765M based on the recorded measurements.

However, the percentile rankings present a more nuanced picture. The AMD card sits at the 36th percentile, which is actually higher than the GTX 765M's 32nd percentile. This happens because the GTX 765M's average benchmark score of 5501 is dragged down by its weak Metal performance of 2612. The AMD card's average of 6137 comes from a single strong OpenCL result.

The GTX 765M's Vulkan score of 6714 is close to its OpenCL score of 7176, within about 6%. This suggests consistent performance across Vulkan and OpenCL. The Metal score is an outlier, though, and anyone planning to use Metal workloads should be cautious with this GPU.

Both cards are end-of-life products from 2013. The AMD card has no listed TDP, while the NVIDIA card draws 75 W. The NVIDIA card is an MXM module, which means it is designed for laptops with replaceable graphics modules. The AMD card uses a standard PCIe connection.

For gaming and general GPU compute, the GTX 765M is the stronger choice. Its 128-bit memory bus doubles the bandwidth of the AMD card, and its 768 shading units provide ample parallel processing power. The AMD card's higher clock speeds and smaller die do not compensate for NVIDIA's hardware advantage.

The AMD Radeon HD 8690M is not without merit. Its GCN 1.0 architecture supports DirectX 12 (11_1) feature level, and its Vulkan support is nearly as recent as NVIDIA's. The smaller die and lower transistor count also suggest it may be more power-efficient, though the database does not record a TDP for it.

Specification Differences

| Specification | AMD Radeon HD 8690M | NVIDIA GeForce GTX 765M |

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

| Chip | Sun | GK106 |

| Architecture | GCN 1.0 | Kepler |

| Transistors | 690 million | 2,540 million |

| Die Size | 56 mm² | 221 mm² |

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

| Base Clock | 925 MHz | 797 MHz |

| Boost Clock | 975 MHz | 863 MHz |

| Memory Clock | 1000 MHz (4 Gbps) | 1002 MHz (4 Gbps) |

| Memory Bus Width | 64 bit | 128 bit |

| Memory Bandwidth | 32.00 GB/s | 64.13 GB/s |

| Shading Units | 320 | 768 |

| TMUs | 20 | 64 |

| ROPs | 8 | 16 |

| Pixel Rate | 7.800 GPixel/s | 13.81 GPixel/s |

| Texture Rate | 19.50 GTexel/s | 55.23 GTexel/s |

| FP32 | 624.0 GFLOPS | 1,325.6 GFLOPS |

| TDP | Not listed | 75 W |

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

| Vulkan | 1.2.170 | 1.2.175 |

| DirectX | 12 (11_1) | 12 (11_0) |

| Release Date | 2013-02-28 | 2013-05-29 |

| Predecessor | London | GeForce 600M |

| Successor | Gem System | GeForce 800M |

Where Each One Wins

The NVIDIA GeForce GTX 765M wins in every recorded compute benchmark. Its OpenCL score of 7176 beats the AMD card's 6137 by 14.5%. It also delivers higher pixel rate, texture rate, and FP32 throughput. The 128-bit memory bus provides twice the bandwidth, which benefits memory-intensive workloads. The Vulkan score of 6714 indicates strong performance in Vulkan applications. This GPU is the clear choice for OpenCL compute, Vulkan workloads, and any task that relies on raw shader throughput.

The AMD Radeon HD 8690M wins on the metric of average benchmark consistency. Its single OpenCL score of 6137 produces an average of 6137, which beats the GTX 765M's average of 5501. The higher percentile ranking of 36 versus 32 reflects this. The AMD card also has a higher boost clock of 975 MHz versus 863 MHz, which can help in lightly threaded workloads that depend on clock speed. Its smaller die and lower transistor count suggest better transistor density at 12.3M per mm², though this does not translate into higher performance.

The GTX 765M is the better choice for gaming laptops with MXM slots, given its 75 W TDP and modular design. The AMD card, with its PCIe 3.0 x8 interface, fits into more conventional laptop designs. Neither card supports ray tracing or tensor cores, so those features are absent from both.

For the specific use case of OpenCL compute, the GTX 765M wins outright. For a balanced assessment across multiple APIs, the AMD card has a higher average score and percentile. The decision ultimately depends on whether the user prioritizes peak compute performance or consistent average results. The database records one direct win for NVIDIA, and that carries the most weight in a head-to-head comparison.