AMD Radeon HD 8690M vs NVIDIA GeForce GTX 760M Comparison

Where Each One Wins

The recorded data splits cleanly between the two mobile GPUs based on workload type. The AMD Radeon HD 8690M takes the single measured OpenCL benchmark, scoring 6,137 points against the NVIDIA GeForce GTX 760M's 5,604 points, a 9.5% advantage. This is the only head-to-head benchmark result in the database, so the AMD part wins the only direct comparison available.

The NVIDIA GeForce GTX 760M, however, has an additional recorded benchmark in Vulkan, scoring 4,868 points. That test has no corresponding AMD result, so it cannot be compared directly. The AMD card's 36th percentile ranking among all GPUs places it ahead of the NVIDIA card's 31st percentile, suggesting that the AMD part sits slightly higher in the overall performance distribution despite having fewer shading units and lower theoretical throughput numbers.

For compute-heavy OpenCL workloads, the AMD Radeon HD 8690M is the stronger choice based on the recorded scores. For users prioritizing Vulkan performance, the GTX 760M has at least a recorded result, but without a matching AMD Vulkan score, the database cannot establish which card wins that API. The practical takeaway: the AMD card wins the measurable compute comparison, while the NVIDIA card offers broader API coverage in the recorded data.

Architecture Differences

The two GPUs come from different design philosophies. The AMD Radeon HD 8690M uses the Sun chip built on GCN 1.0 architecture, part of the Solar System generation (HD 8600M series). The NVIDIA GeForce GTX 760M uses the GK106S chip on Kepler architecture, part of the GeForce 700M generation. Both are fabricated on a 28 nm process at TSMC, but the similarities end there.

The transistor counts differ substantially. The AMD chip packs 690 million transistors on a 56 mm² die, giving a transistor density of 12.3 million per mm². The NVIDIA chip uses 2,540 million transistors on a 221 mm² die, yielding 11.5 million per mm². The NVIDIA die is nearly four times larger in area and holds roughly 3.7 times more transistors, but the AMD die achieves a slightly higher packing density.

The compute configuration favors NVIDIA numerically. The GTX 760M has 768 shading units, 64 texture mapping units, and 16 raster operation pipelines. The HD 8690M has 320 shading units, 20 TMUs, and 8 ROPs. Despite this 2.4x gap in shading units, the AMD card still wins the OpenCL benchmark, which points to architectural efficiency differences rather than raw unit counts.

Clock speeds tell a different story. The AMD card runs at a 925 MHz base and 975 MHz boost, while the NVIDIA card runs at a 628 MHz base and 719 MHz boost. The AMD part's higher clocks partially compensate for its smaller compute configuration. Memory clocks are nearly identical at 1000 MHz (AMD) and 1002 MHz (NVIDIA), both yielding 4 Gbps effective.

The memory subsystem favors NVIDIA. Both cards have 2 GB of GDDR5, but the GTX 760M uses a 128-bit bus delivering 64.13 GB/s of bandwidth, double the HD 8690M's 64-bit bus and 32.00 GB/s. This bandwidth advantage is significant for texture-heavy workloads, though the OpenCL result suggests it did not translate into a compute win.

API support differs slightly. The AMD card lists DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The NVIDIA card lists DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The AMD part supports a slightly newer DirectX feature level, while the NVIDIA part has a marginally newer Vulkan version.

The bus interface also differs: the HD 8690M uses PCIe 3.0 x8, while the GTX 760M uses PCIe 3.0 x16. The NVIDIA card is a 55 W MXM module with no power connectors, while the AMD card's TDP is not recorded in the database. Both are end-of-life products, with the AMD card released on 2013-02-28 and the NVIDIA card on 2013-05-29.

The Verdict

The data supports a clear but narrow verdict. The AMD Radeon HD 8690M wins the only direct benchmark comparison, beating the NVIDIA GeForce GTX 760M by 9.5% in Geekbench OpenCL. Its 36th percentile ranking among all GPUs also exceeds the NVIDIA card's 31st percentile. For compute-oriented tasks using OpenCL, the AMD card is the better performer according to the recorded measurements.

The NVIDIA GeForce GTX 760M, however, offers advantages that the benchmark data does not capture. It has double the memory bandwidth (64.13 GB/s versus 32.00 GB/s), more than double the shading units (768 versus 320), and a 128-bit memory bus versus 64-bit. These specifications suggest it would handle memory-intensive workloads better, even though the OpenCL score does not reflect that. The GTX 760M also has a recorded Vulkan benchmark, which the AMD card lacks, making it the only option with verified Vulkan performance in the database.

Users who prioritize compute benchmarks and OpenCL performance should choose the AMD Radeon HD 8690M. Users who need verified Vulkan support or who value the larger memory interface and higher theoretical throughput should choose the NVIDIA GeForce GTX 760M. The recorded data does not support a universal winner, but it does support a workload-specific recommendation.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The AMD Radeon HD 8690M scores 6,137, which is 9.5% higher than the NVIDIA GeForce GTX 760M's 5,604.

Q: Does the NVIDIA GeForce GTX 760M have any benchmark advantage?

A: The GTX 760M has a recorded Geekbench Vulkan score of 4,868, which is the only Vulkan result in the database for either card. The AMD Radeon HD 8690M has no Vulkan benchmark recorded.

Q: How do their memory configurations compare?

A: Both cards have 2 GB of GDDR5 memory, but the GTX 760M uses a 128-bit bus with 64.13 GB/s bandwidth, while the HD 8690M uses a 64-bit bus with 32.00 GB/s bandwidth.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce GTX 760M has 768 shading units, compared to 320 on the AMD Radeon HD 8690M.

Q: What is the clock speed difference?

A: The AMD card runs at a 925 MHz base and 975 MHz boost, while the NVIDIA card runs at a 628 MHz base and 719 MHz boost.

Q: How do their overall rankings compare?

A: The AMD Radeon HD 8690M sits at the 36th percentile among all GPUs, while the NVIDIA GeForce GTX 760M sits at the 31st percentile.

Head-to-Head Benchmarks

The database contains exactly one head-to-head benchmark result between these two GPUs: Geekbench OpenCL. The AMD Radeon HD 8690M scores 6,137, and the NVIDIA GeForce GTX 760M scores 5,604. The AMD card wins by 533 points, a 9.5% margin.

Context from the nearest rivals helps interpret this result. The HD 8690M's closest competitors are the Intel Iris Pro Graphics 6200 at 6,117 (0.3% behind), the NVIDIA RTX A400 at 6,078 (1% behind), and the NVIDIA GeForce MX230 at 6,077 (1% behind). The only rival that beats it is the AMD Radeon HD 6950 at 6,210, which is 1.2% ahead. This clustering suggests the HD 8690M sits in a tight performance band around 6,100 to 6,200 points.

The GTX 760M's nearest rivals form a similar cluster around 5,200 to 5,300 points. The NVIDIA Quadro 4000M scores 5,211 (0.5% behind), the AMD Radeon R7 M260X scores 5,161 (1.5% ahead), and the GeForce 940M scores 5,284 (0.9% ahead). The GTX 980M scores 5,308, which is 1.3% ahead. The GTX 760M's 5,604 OpenCL score actually places it above all of these listed rivals, but it still trails the HD 8690M.

The 9.5% delta between the two cards is meaningful given how tightly clustered their respective rival groups are. The HD 8690M outperforms most of its nearest rivals by only 0.3% to 1%, and the GTX 760M trades places with its rivals within a 1.5% band. A 9.5% gap between the two cards is therefore a substantial difference in this performance neighborhood.

The Vulkan benchmark for the GTX 760M, scoring 4,868, sits roughly 13% below its own OpenCL score. This gap between APIs on the same card is larger than the 9.5% difference between the two cards in OpenCL, which suggests the GTX 760M's Vulkan performance is comparatively weaker than its OpenCL performance. Without a corresponding AMD Vulkan score, this remains a single-card observation.

The average benchmark score in the database reflects the same pattern. The HD 8690M has an average of 6,137 (matching its single OpenCL score), while the GTX 760M has an average of 5,236, pulled down by the lower Vulkan result. The HD 8690M's average is 17.2% higher, though this partly reflects the different benchmark mixes available for each card.

Specification Differences

The two GPUs differ across nearly every recorded specification field.

The process node is identical: both use 28 nm at TSMC. The transistor counts diverge sharply, with the NVIDIA GK106S using 2,540 million transistors versus 690 million on the AMD Sun chip. Die size follows suit at 221 mm² versus 56 mm². Transistor density slightly favors AMD at 12.3M per mm² versus 11.5M per mm².

Clock speeds favor AMD. The HD 8690M runs at 925 MHz base and 975 MHz boost, while the GTX 760M runs at 628 MHz base and 719 MHz boost. Memory clocks are effectively tied at 1000 MHz versus 1002 MHz, both delivering 4 Gbps effective.

Memory capacity matches at 2 GB of GDDR5, but bandwidth favors NVIDIA at 64.13 GB/s versus 32.00 GB/s due to the 128-bit bus versus 64-bit bus. The compute configuration favors NVIDIA in unit counts: 768 shading units, 64 TMUs, and 16 ROPs versus 320 shading units, 20 TMUs, and 8 ROPs. Pixel rate favors NVIDIA at 11.50 GPixel/s versus 7.800 GPixel/s, and texture rate favors NVIDIA at 46.02 GTexel/s versus 19.50 GTexel/s. FP32 throughput favors NVIDIA at 1,104.4 GFLOPS versus 624.0 GFLOPS.

Power characteristics differ, with the GTX 760M recorded at 55 W TDP as an MXM Module with no power connectors, while the HD 8690M has no TDP or slot width recorded. The bus interface favors NVIDIA at PCIe 3.0 x16 versus PCIe 3.0 x8. Display outputs are listed as portable device dependent for NVIDIA, with no display output data for AMD.

API support differs in DirectX and Vulkan versions. AMD lists DirectX 12 (11_1), while NVIDIA lists DirectX 12 (11_0). Vulkan versions are 1.2.170 for AMD and 1.2.175 for NVIDIA. OpenGL is identical at 4.6.

Release dates differ by three months, with AMD on 2013-02-28 and NVIDIA on 2013-05-29. Both are end-of-life. The AMD predecessor is London with successor Gem System, while the NVIDIA predecessor is GeForce 600M with successor GeForce 800M. Neither card has a recorded launch MSRP.