AMD Radeon RX 9060 vs NVIDIA GeForce GTX 690 Comparison

# The Verdict

The data presents a decisive generational shift. The AMD Radeon RX 9060 outperforms the NVIDIA GeForce GTX 690 in every shared benchmark, with the largest margin appearing in OpenCL compute workloads. In the Geekbench OpenCL test, the RX 9060 scores 88,183 against the GTX 690’s 17,399, a delta of -80.3% from the perspective of the older card, meaning the RX 9060 is roughly five times faster. The Vulkan result is similarly lopsided: 39,476 for the RX 9060 versus 16,675 for the GTX 690, a -57.8% delta. The head-to-head record is 2 wins for the RX 9060 and 0 for the GTX 690.

However, the average benchmark scores tell a more nuanced story. The RX 9060’s average across all its tests is 16,014, while the GTX 690’s average is 17,037. This discrepancy highlights that the RX 9060’s lead is concentrated in modern, compute-heavy APIs, while legacy or specific workloads may favor the older card’s architecture. The GTX 690 sits in the 60th percentile of all GPUs, while the RX 9060 sits in the 59th, indicating that on a broad aggregate basis, they are roughly comparable, despite the RX 9060’s dominance in the two shared tests.

For a user selecting between these two, the choice is clear based on the data: the RX 9060 is the only sensible pick for any modern workload that leverages OpenCL or Vulkan. Its 8 GB of memory, 288.0 GB/s bandwidth, and 21.43 TFLOPS FP32 performance place it in a different performance class. The GTX 690, with its 2 GB memory and 3.130 TFLOPS FP32, is a historical artifact. The RX 9060 is actively produced, while the GTX 690 is end-of-life. If the task involves current software, the RX 9060 is the definitive choice. If the task is purely archival or targets ancient APIs, the GTX 690’s average score is marginally higher, but the lack of modern feature support makes it a poor recommendation.

FAQ

Q: Which GPU is faster in compute-heavy applications?

A: The AMD Radeon RX 9060 is decisively faster. In the Geekbench OpenCL test, it scores 88,183 versus 17,399 for the NVIDIA GeForce GTX 690, a -80.3% delta. This indicates the RX 9060 delivers roughly five times the compute throughput in this workload.

Q: How do the two compare in Vulkan graphics performance?

A: The RX 9060 wins by a wide margin. It scores 39,476 in Geekbench Vulkan, while the GTX 690 scores 16,675, a -57.8% delta. This suggests the RX 9060 has a significant advantage in modern graphics APIs.

Q: Is the GTX 690 competitive in any benchmark?

A: In the two head-to-head benchmarks available, the GTX 690 does not win either. However, its average benchmark score across all tests (17,037) is higher than the RX 9060’s average (16,014). This is likely due to the RX 9060’s lower scores in legacy DirectX 9, 10, and 11 tests, where it scores 280, 104, and 182 respectively on PassMark.

Q: What are the memory specifications of each card?

A: The GTX 690 has 2 GB of GDDR5 memory on a 256-bit bus, providing 192.3 GB/s of bandwidth. The RX 9060 has 8 GB of GDDR6 memory on a 128-bit bus, providing 288.0 GB/s of bandwidth. The RX 9060 has four times the capacity and roughly 50% more bandwidth.

Q: Which card is more power-efficient?

A: The RX 9060 has a TDP of 132 W and requires a 300 W power supply, while the GTX 690 has a TDP of 300 W and requires a 700 W power supply. The RX 9060 uses less than half the power while delivering significantly higher performance in modern benchmarks.

Q: What is the architectural generation gap?

A: The GTX 690 is based on the Kepler architecture (GK104 chip) on a 28 nm process, released in 2012. The RX 9060 is based on RDNA 4.0 (Navi 44 chip) on a 4 nm process, released in 2025. This represents a 13-year generational leap in process technology and architecture design.

Architecture Differences

The architectural divide between these two GPUs is vast, reflecting over a decade of development. The NVIDIA GeForce GTX 690 employs the Kepler architecture, built on the GK104 chip using a 28 nm process at TSMC. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0M per mm². This is a dual-GPU design, essentially two GK104 dies on a single card, which explains its 300 W TDP and dual-slot cooler with two 8-pin power connectors.

In contrast, the AMD Radeon RX 9060 uses the RDNA 4.0 architecture with the Navi 44 chip, codenamed Strix Point, on a 4 nm process, also from TSMC. It contains 29,700 million transistors on a smaller 199 mm² die, achieving a remarkable transistor density of 149.2M per mm² — over 12 times denser than the GTX 690. This is a single-die solution with a much more efficient design, evidenced by its 132 W TDP and single 8-pin connector.

The feature sets diverge sharply. The RX 9060 includes 28 dedicated ray tracing cores, a feature entirely absent from the GTX 690. The RX 9060 also supports DirectX 12 Ultimate (12_2), while the GTX 690 is limited to DirectX 12 (11_0). Vulkan support differs as well: the RX 9060 supports Vulkan 1.4, while the GTX 690 supports Vulkan 1.2.175. Both cards support OpenGL 4.6. The RX 9060 offers FP16 performance at a 1:1 ratio with FP32 (21.43 TFLOPS), while the GTX 690 has no listed FP16 capability. The RX 9060 supports PCIe 5.0 x16, whereas the GTX 690 uses PCIe 3.0 x16.

Specification Differences

The specification sheet reveals the full extent of the generational gap. The RX 9060 has 1,792 shading units, 112 TMUs, and 64 ROPs, compared to the GTX 690’s 1,536 shading units, 128 TMUs, and 32 ROPs. While the GTX 690 has more TMUs, the RX 9060 doubles the ROP count and adds 28 ray tracing cores.

Clock speeds are dramatically different. The GTX 690 runs at a base clock of 915 MHz with a boost of 1019 MHz. The RX 9060 has a base clock of 1700 MHz, a game clock of 2400 MHz, and a boost clock of 2990 MHz. Memory clocks also differ: the GTX 690 uses 1502 MHz (6 Gbps effective) GDDR5, while the RX 9060 uses 2250 MHz (18 Gbps effective) GDDR6.

The resulting compute rates are stark. The RX 9060 achieves 21.43 TFLOPS FP32, 334.9 GTexel/s texture rate, and 191.4 GPixel/s pixel rate. The GTX 690 achieves 3.130 TFLOPS FP32, 130.4 GTexel/s, and 32.61 GPixel/s. The RX 9060 is approximately 6.8 times faster in FP32, 2.6 times faster in texture fill, and 5.9 times faster in pixel fill.

Memory capacity and bandwidth favor the RX 9060: 8 GB GDDR6 on a 128-bit bus with 288.0 GB/s bandwidth, versus 2 GB GDDR5 on a 256-bit bus with 192.3 GB/s. The RX 9060 uses a single 8-pin power connector and a 300 W suggested PSU, while the GTX 690 requires two 8-pin connectors and a 700 W PSU. Display outputs differ: the GTX 690 offers 3x DVI and 1x mini-DisplayPort 1.2, while the RX 9060 offers 1x HDMI 2.1b and 2x DisplayPort 2.1a.

Head-to-Head Benchmarks

The two shared benchmarks paint a clear picture of the RX 9060’s dominance. In Geekbench OpenCL, the RX 9060 scores 88,183 against the GTX 690’s 17,399. This is a -80.3% delta, meaning the GTX 690 scores roughly 20% of the RX 9060’s result. This massive gap reflects the RX 9060’s superior compute throughput, enabled by its higher clock speeds, more shading units, and modern architecture.

In Geekbench Vulkan, the RX 9060 scores 39,476 versus 16,675 for the GTX 690, a -57.8% delta. While not as extreme as the OpenCL gap, this still represents a lead of over 2.3 times for the RX 9060. This suggests that even in graphics-centric workloads, the modern architecture and higher fill rates of the RX 9060 provide a substantial advantage.

The GTX 690’s nearest rivals in the aggregate benchmark database include the AMD Radeon HD 7970M (average score 17,019, delta 0.1%), the AMD Radeon RX 7600 XT (17,083, delta -0.3%), and the NVIDIA GeForce RTX 3070 (17,208, delta -1%). This places the GTX 690 in a performance tier that, on average, matches modern mid-range cards, though its individual benchmark wins are limited to the two tests where it has data.

The RX 9060’s nearest rivals include the NVIDIA GeForce RTX 3060 Ti (average score 16,129, delta -0.7%), the AMD Radeon R9 370X (15,862, delta 1%), and the AMD Radeon RX 7700 (15,852, delta 1%). This indicates that the RX 9060’s aggregate performance is competitive with mid-range GPUs from the previous generation, despite its massive lead in the specific OpenCL and Vulkan tests shared with the GTX 690.

The data shows that the RX 9060’s high scores in compute-heavy benchmarks are offset by lower scores in legacy DirectX tests (PassMark DirectX 9: 280, DirectX 10: 104, DirectX 11: 182, DirectX 12: 44). These legacy scores drag down its average, explaining why its percentile rank (59th) is nearly identical to the GTX 690’s (60th). For any modern workload, the RX 9060 is the clear winner; for purely legacy applications, the GTX 690’s average score is marginally higher, but this is a moot point given its end-of-life status and lack of modern feature support.