AMD Radeon RX 6650 XT vs NVIDIA GeForce GTX 690 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon RX 6650 XT has an average benchmark score of 19,765, while the NVIDIA GeForce GTX 690 scores 17,037. The AMD card sits in the 65th percentile of all GPUs, while the GTX 690 sits in the 60th percentile.

Q: How do the two compare in OpenCL performance?

A: The NVIDIA GeForce GTX 690 is the winner in the Geekbench OpenCL test, scoring 17,399 versus the RX 6650 XT's 10,111. That is a 41.9% deficit for the AMD card in this specific workload.

Q: What about Vulkan performance?

A: The AMD Radeon RX 6650 XT dominates in Geekbench Vulkan, scoring 81,306 versus the GTX 690's 16,675. The AMD card leads by 387.6% in this test, which is a massive margin.

Q: Which GPU has more memory?

A: The AMD Radeon RX 6650 XT has 8 GB of GDDR6 memory on a 128-bit bus, while the NVIDIA GeForce GTX 690 has 2 GB of GDDR5 memory on a 256-bit bus. The RX 6650 XT also has higher memory bandwidth at 280.3 GB/s versus 192.3 GB/s.

Q: What are the power requirements for each card?

A: The AMD Radeon RX 6650 XT has a TDP of 176 W and requires a 450 W power supply with a single 8-pin connector. The NVIDIA GeForce GTX 690 has a TDP of 300 W and requires a 700 W power supply with two 8-pin connectors.

Q: Which GPU was released more recently?

A: The AMD Radeon RX 6650 XT was released on May 9, 2022, while the NVIDIA GeForce GTX 690 was released on May 2, 2012, a full decade earlier.

Architecture Differences

The AMD Radeon RX 6650 XT and NVIDIA GeForce GTX 690 come from completely different eras and design philosophies. The RX 6650 XT is built on the Navi 23 chip using the RDNA 2.0 architecture, which is part of the Navi II generation (RX 6000 series). It is manufactured on a 7 nm process at TSMC, with 11,060 million transistors packed into a 237 mm² die, resulting in a transistor density of 46.7 million per square millimeter.

The GTX 690, by contrast, uses the GK104 chip with the Kepler architecture from the GeForce 600 generation. It is built on a 28 nm process, also at TSMC, but contains only 3,540 million transistors spread across a larger 294 mm² die. Its transistor density is just 12.0 million per square millimeter, reflecting the older manufacturing node.

The RX 6650 XT features 2048 shading units, 128 texture mapping units, 64 raster operation units, and 32 dedicated ray tracing cores. The GTX 690 has 1536 shading units, 128 texture mapping units, and 32 raster operation units, but no ray tracing cores at all. This means the AMD card supports hardware-accelerated ray tracing, while the NVIDIA card predates that technology entirely.

The RX 6650 XT also supports DirectX 12 Ultimate (12_2), while the GTX 690 only supports DirectX 12 (11_0). Both cards support OpenGL 4.6, but the Vulkan API support differs: the RX 6650 XT supports Vulkan 1.4, while the GTX 690 supports Vulkan 1.2.175. The AMD card uses a PCIe 4.0 x8 interface, whereas the GTX 690 uses the older PCIe 3.0 x16 interface.

Memory configuration is another major architectural difference. The RX 6650 XT has 8 GB of GDDR6 memory running at 17.5 Gbps effective, while the GTX 690 has 2 GB of GDDR5 memory running at 6 Gbps effective. Despite the GTX 690's wider 256-bit bus versus the RX 6650 XT's 128-bit bus, the AMD card achieves higher bandwidth at 280.3 GB/s compared to 192.3 GB/s.

Head-to-Head Benchmarks

The database records two shared benchmark tests between these GPUs, and the results are starkly split. In Geekbench OpenCL, the NVIDIA GeForce GTX 690 takes a decisive win with a score of 17,399, while the AMD Radeon RX 6650 XT manages only 10,111. This represents a 41.9% advantage for the older NVIDIA card, which is a surprising result given the generational gap.

The story flips completely in Geekbench Vulkan. Here, the RX 6650 XT scores 81,306, while the GTX 690 scores just 16,675. The AMD card leads by an enormous 387.6% margin, making this one of the most lopsided results in the comparison. The Vulkan performance difference reflects not just newer architecture but also the substantial improvements in API efficiency and driver maturity that the RDNA 2.0 platform brings.

It is important to note that the GTX 690's OpenCL score is likely boosted by its dual-GPU design. The card was built with two GK104 chips, which explains why its raw compute throughput in certain legacy APIs can still compete with modern single-GPU cards. However, the Vulkan test shows the limits of that aging architecture, as the modern API implementation on the RX 6650 XT produces dramatically better results.

The overall wins are split evenly at one apiece, but the magnitude of the Vulkan victory heavily favors the AMD card in modern workloads. The average benchmark score of 19,765 for the RX 6650 XT versus 17,037 for the GTX 690 reinforces this conclusion, as does the percentile ranking of 65 versus 60.

Specification Differences

The table below highlights the key fields where the two GPUs differ based on the database records.

| Specification | AMD Radeon RX 6650 XT | NVIDIA GeForce GTX 690 |

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

| Architecture | RDNA 2.0 | Kepler |

| Process Node | 7 nm | 28 nm |

| Transistors | 11,060 million | 3,540 million |

| Die Size | 237 mm² | 294 mm² |

| Transistor Density | 46.7M / mm² | 12.0M / mm² |

| Base Clock | 2055 MHz | 915 MHz |

| Boost Clock | 2635 MHz | 1019 MHz |

| Memory Clock | 17.5 Gbps effective | 6 Gbps effective |

| Memory Size | 8 GB | 2 GB |

| Memory Type | GDDR6 | GDDR5 |

| Memory Bus Width | 128 bit | 256 bit |

| Memory Bandwidth | 280.3 GB/s | 192.3 GB/s |

| Shading Units | 2048 | 1536 |

| ROPs | 64 | 32 |

| RT Cores | 32 | None |

| Pixel Rate | 168.6 GPixel/s | 32.61 GPixel/s |

| Texture Rate | 337.3 GTexel/s | 130.4 GTexel/s |

| FP32 | 10.79 TFLOPS | 3.130 TFLOPS |

| FP16 | 21.59 TFLOPS (2:1) | None |

| TDP | 176 W | 300 W |

| Power Connectors | 1x 8-pin | 2x 8-pin |

| Suggested PSU | 450 W | 700 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 3.0 x16 |

| Display Outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a | 3x DVI, 1x mini-DisplayPort 1.2 |

| DirectX Support | 12 Ultimate (12_2) | 12 (11_0) |

| Vulkan Support | 1.4 | 1.2.175 |

| Release Date | 2022-05-09 | 2012-05-02 |

| Launch MSRP | 399 USD | 999 USD |

The RX 6650 XT also has physical dimensions listed as null for length, height, and width, while the GTX 690 measures 279 mm in length, 111 mm in height, and 38 mm in width. Both cards are dual-slot designs.

Where Each One Wins

The AMD Radeon RX 6650 XT is the clear choice for modern workloads and API compatibility. Its Vulkan performance is in a completely different league, with a 387.6% lead over the GTX 690. The RX 6650 XT also offers ray tracing support, which the GTX 690 completely lacks. For any user running current games or applications that leverage DirectX 12 Ultimate or Vulkan 1.4, the AMD card is the only sensible option.

The RX 6650 XT also wins on raw compute metrics that matter for newer software. Its FP32 throughput of 10.79 TFLOPS is more than three times the GTX 690's 3.130 TFLOPS. The pixel rate of 168.6 GPixel/s and texture rate of 337.3 GTexel/s dwarf the GTX 690's 32.61 GPixel/s and 130.4 GTexel/s. The 8 GB memory capacity is also a significant advantage for modern games and high-resolution textures.

The NVIDIA GeForce GTX 690 wins in one specific area: OpenCL performance. Its score of 17,399 in the Geekbench OpenCL test beats the RX 6650 XT by 41.9%. This suggests that in legacy compute workloads that rely heavily on OpenCL, the dual-GPU design of the GTX 690 can still deliver respectable results. However, this is a narrow victory in an aging API.

The GTX 690 also has a wider memory bus at 256-bit versus 128-bit, which historically helped with certain bandwidth-sensitive workloads. But the RX 6650 XT compensates with faster GDDR6 memory and higher effective bandwidth, so this advantage is largely neutralized in practice.

The Verdict

The data makes the choice clear for most users. The AMD Radeon RX 6650 XT is the superior GPU for anyone building a system today or upgrading from older hardware. It offers dramatically better Vulkan performance, modern API support, ray tracing capability, and more than triple the FP32 compute throughput of the GTX 690. Its 65th percentile ranking versus 60th for the GTX 690, combined with an average benchmark score of 19,765 versus 17,037, confirms its overall advantage.

The RX 6650 XT also wins on efficiency. Its TDP of 176 W is significantly lower than the GTX 690's 300 W, and it requires a smaller 450 W power supply versus 700 W. The single 8-pin connector is simpler than the GTX 690's dual 8-pin setup. Despite being more efficient, the AMD card delivers higher performance in nearly every measured metric.

The NVIDIA GeForce GTX 690 is a historical curiosity at this point. Its OpenCL win is notable, but it is confined to a single legacy benchmark. For any practical purpose in 2025 and beyond, the GTX 690's 2 GB memory capacity, lack of ray tracing, and older DirectX 12 (11_0) support make it unsuitable for modern gaming or compute tasks. Its launch MSRP of 999 USD was high in 2012, and the hardware has not aged well in terms of feature support.

The verdict is straightforward: the AMD Radeon RX 6650 XT is the better investment for anyone who needs a working GPU today. The GTX 690 should only be considered by collectors or those running very specific legacy OpenCL workloads where its dual-chip design still shines. For everyone else, the RX 6650 XT's modern architecture, superior API support, and massive Vulkan advantage make it the definitive choice.