AMD Radeon HD 8690M vs NVIDIA GeForce GTX 550 Ti Comparison

The NVIDIA GeForce GTX 550 Ti and the AMD Radeon HD 8690M represent two distinct generations of GPU design, and the benchmark data highlights a clear, albeit moderate, performance gap. In the single Geekbench OpenCL test available, the AMD Radeon HD 8690M scores 6137, which is 6.6% higher than the NVIDIA GeForce GTX 550 Ti's score of 5731. This result places the AMD part in the 36th percentile of all GPUs, while the NVIDIA card sits in the 33rd percentile, confirming the AMD’s advantage is consistent with its higher overall ranking. The data indicates that in raw compute performance, the mobile-oriented AMD chip outperforms the older desktop NVIDIA card by a meaningful margin.

Head-to-Head Benchmarks

The sole head-to-head benchmark comparison available is Geekbench OpenCL, a test that measures general-purpose compute performance across a wide range of workloads. The AMD Radeon HD 8690M emerges as the winner with a score of 6137, while the NVIDIA GeForce GTX 550 Ti trails at 5731. This translates to a 6.6% performance advantage for the AMD solution. This delta is not trivial, as it positions the HD 8690M a full 406 points ahead of the GTX 550 Ti. When looking at the nearest rival scores for context, the GTX 550 Ti's score of 5731 is within 0.2% of the NVIDIA GeForce GTX 670MX (5721) and 0.3% of the Intel Iris Pro Graphics P6300 (5712), placing it in a tightly contested performance band. Conversely, the HD 8690M's score of 6137 is 0.3% above the Intel Iris Pro Graphics 6200 (6117) and 1% above both the NVIDIA RTX A400 (6078) and NVIDIA GeForce MX230 (6077). The data shows that the HD 8690M sits in a slightly higher performance tier, as it competes with more recent and specialized parts, while the GTX 550 Ti is more aligned with older integrated and mid-range mobile solutions.

The performance spread between the two cards is confirmed by their nearest rival deltas. The GTX 550 Ti is only 1.9% behind the AMD Radeon R7 M465 (5841), which is the fastest of its listed rivals. In contrast, the HD 8690M is only 1.2% behind the AMD Radeon HD 6950 (6210), a much more powerful high-end desktop card from an earlier generation. This suggests the HD 8690M's lead over the GTX 550 Ti is more significant than the raw 6.6% delta might initially suggest, as the AMD part is capable of approaching the performance of a previous flagship, whereas the NVIDIA card is firmly in the mid-range segment. No other benchmark results are provided, so the analysis must rely on this single, decisive data point.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process technologies, which explains their generational gap. The NVIDIA GeForce GTX 550 Ti is based on the GF116 chip, using the Fermi 2.0 architecture, and is manufactured on a 40 nm process at TSMC. This older process results in a relatively large die size of 238 mm², housing 1,170 million transistors, which yields a transistor density of 4.9 million per square millimeter. In contrast, the AMD Radeon HD 8690M uses the Sun chip with the GCN 1.0 architecture, built on a much more advanced 28 nm process. This allows for a significantly smaller die size of just 56 mm² while containing 690 million transistors, resulting in a much higher transistor density of 12.3 million per square millimeter. The architectural shift is stark: Fermi 2.0 is a dated design, while GCN 1.0 introduces a modern compute-oriented architecture.

The core configurations differ substantially. The GTX 550 Ti has 192 shading units, 32 texture mapping units (TMUs), and 24 raster operation units (ROPs). The HD 8690M, despite being a smaller chip, packs 320 shading units, which is 66% more than the NVIDIA card. However, it has fewer TMUs (20) and significantly fewer ROPs (8). This configuration suggests the AMD card is designed for high compute throughput but lower fill-rate tasks. The clock speeds also differ, with the HD 8690M running at a base clock of 925 MHz and a boost clock of 975 MHz, while the GTX 550 Ti's core clock is not specified in the data. In terms of raw compute, the GTX 550 Ti achieves a higher FP32 rate of 691.2 GFLOPS compared to the HD 8690M's 624.0 GFLOPS, thanks to its higher clock speeds and larger number of ROPs. Conversely, the HD 8690M's higher pixel rate of 7.800 GPixel/s edges out the GTX 550 Ti's 7.200 GPixel/s.

Memory subsystems also present a major divergence. The GTX 550 Ti uses a 192-bit memory bus with 1024 MB of GDDR5, achieving a bandwidth of 98.50 GB/s. The HD 8690M, on the other hand, has a narrow 64-bit bus but doubles the capacity to 2 GB of GDDR5, resulting in a much lower bandwidth of 32.00 GB/s. This indicates the NVIDIA card is better suited for high-resolution textures and bandwidth-intensive tasks, while the AMD card's larger framebuffer is more useful for storing data. The bus interfaces differ as well, with the GTX 550 Ti using PCIe 2.0 x16 and the HD 8690M using the newer PCIe 3.0 x8. API support shows the HD 8690M has a slight edge, supporting DirectX 12 (11_1) and Vulkan 1.2.170, while the GTX 550 Ti only supports DirectX 12 (11_0) and no Vulkan. Both support OpenGL 4.6.

Where Each One Wins

Based on the available benchmark results, the AMD Radeon HD 8690M wins decisively in general-purpose compute performance. Its 6.6% lead in Geekbench OpenCL indicates that applications leveraging GPGPU features, such as video encoding, physics simulations, and certain productivity tasks, will run noticeably faster on the AMD card. The higher shading unit count (320 vs 192) is the primary driver of this advantage, as it allows for more parallel threads to be executed simultaneously. This makes the HD 8690M the better choice for compute-oriented workloads, despite its lower raw FP32 rating, which is offset by its more modern architecture and boost clock capabilities.

The NVIDIA GeForce GTX 550 Ti, while losing the compute test, retains strengths in other areas that are not directly measured in the provided data. Its significantly higher memory bandwidth of 98.50 GB/s, compared to 32.00 GB/s for the HD 8690M, suggests it would be superior in traditional 3D gaming scenarios where high texture fill rates are critical. The GTX 550 Ti also has a higher texture rate of 28.80 GTexel/s versus 19.50 GTexel/s, and a higher FP32 rate of 691.2 GFLOPS, indicating that its pixel and texture processing pipelines are more robust. For older games or applications that are not heavily parallelized, the NVIDIA card's higher clock speeds and wider memory bus could translate to better performance. Furthermore, the GTX 550 Ti is a desktop card with a dual-slot design and a 116 W TDP, while the HD 8690M's TDP is not specified, but it is clearly a mobile-oriented chip, meaning the desktop NVIDIA card may have more sustained power headroom for extended workloads.

The wins are not evenly distributed across all use cases. The HD 8690M wins on compute and modern API compatibility, making it more future-proof for newer software that leverages DirectX 12 (11_1) and Vulkan. The GTX 550 Ti, with its older architecture and lack of Vulkan support, is more of a legacy part that will rely on raw bandwidth for its performance. In a direct comparison for a user seeking a GPU for modern, multi-threaded applications, the AMD card is the clear winner. For a user running older, bandwidth-sensitive applications, the NVIDIA card may still hold its own, but the benchmark data does not provide evidence for this.

FAQ

Q: Which GPU has a higher Geekbench OpenCL score?

A: The AMD Radeon HD 8690M has a higher score of 6137, compared to the NVIDIA GeForce GTX 550 Ti's score of 5731. This gives the AMD card a 6.6% performance lead in this specific benchmark.

Q: How does the GTX 550 Ti compare to its closest rivals?

A: The GTX 550 Ti's score of 5731 is within a very tight margin of its nearest competitors. It is 0.2% ahead of the NVIDIA GeForce GTX 670MX (5721) and 0.3% ahead of the Intel Iris Pro Graphics P6300 (5712), but 1.9% behind the AMD Radeon R7 M465 (5841).

Q: What are the key architectural differences between the two cards?

A: The GTX 550 Ti uses the Fermi 2.0 architecture on a 40 nm process with a 238 mm² die size, while the HD 8690M uses the GCN 1.0 architecture on a 28 nm process with a much smaller 56 mm² die. The AMD card has 320 shading units, while the NVIDIA card has 192, but the NVIDIA card has a wider 192-bit memory bus.

Q: Does the GTX 550 Ti have any advantages in raw throughput?

A: Yes, the GTX 550 Ti achieves a higher FP32 rate of 691.2 GFLOPS and a higher texture rate of 28.80 GTexel/s. It also has more ROPs (24 vs 8) and a significantly higher memory bandwidth of 98.50 GB/s compared to 32.00 GB/s for the HD 8690M.

Q: Which GPU supports more modern graphics APIs?

A: The AMD Radeon HD 8690M supports both DirectX 12 (11_1) and Vulkan 1.2.170, while the NVIDIA GeForce GTX 550 Ti only supports DirectX 12 (11_0) and does not list Vulkan support. Both support OpenGL 4.6.

Q: What is the memory configuration difference?

A: The GTX 550 Ti has 1024 MB of GDDR5 on a 192-bit bus, while the HD 8690M has 2 GB of GDDR5 on a 64-bit bus. This results in the NVIDIA card having a much higher bandwidth of 98.50 GB/s, while the AMD card has lower bandwidth but double the capacity.

The Verdict

The benchmark data unequivocally favors the AMD Radeon HD 8690M in compute performance. With a 6.6% higher score in Geekbench OpenCL and a better overall percentile ranking (36th vs 33rd), the AMD part is the superior choice for any workload that leverages general-purpose GPU computing. Its modern GCN 1.0 architecture, higher shading unit count, and support for Vulkan and DirectX 12 (11_1) make it a more capable and future-proof solution. The data shows that the HD 8690M is competitive with more recent parts like the NVIDIA GeForce MX230 and RTX A400, while the GTX 550 Ti is aligned with older integrated graphics and mid-range mobile chips.

However, the verdict is not a complete sweep. The NVIDIA GeForce GTX 550 Ti's strengths lie in its memory subsystem and raw pixel-pushing capabilities. Its 98.50 GB/s bandwidth is more than triple that of the HD 8690M, and its higher FP32 and texture rates suggest it could outperform the AMD card in traditional, fill-rate-limited 3D rendering. The GTX 550 Ti is a desktop card with a 116 W TDP and a 300 W suggested PSU, indicating it is designed for more sustained, high-power workloads, while the HD 8690M is a mobile chip with no specified TDP. For a user building a legacy desktop system that prioritizes raw bandwidth for older games, the GTX 550 Ti remains a relevant option. But for anyone dealing with modern software, compute tasks, or seeking a more balanced and efficient part, the AMD Radeon HD 8690M is the definitive winner based on the provided benchmark results. The single data point is clear, and it points to the AMD card as the stronger performer.