AMD Radeon RX 460 vs NVIDIA GeForce GTX 690 Comparison

Head-to-Head Benchmarks

The benchmark data shows a clear victory for the AMD Radeon RX 460 in the two compute tests where both cards were measured. The most decisive margin comes in the Geekbench Vulkan test, where the RX 460 scores 20,198 against the GTX 690’s 16,675. That is a 21.1% advantage for the AMD card, a substantial gap that reflects the modern API efficiency of the GCN 4.0 architecture compared to the older Kepler design. In practical terms, this means the RX 460 is not merely faster in Vulkan workloads; it is dramatically ahead, a result that would be visible in any Vulkan-based game or compute application.

The second head-to-head test, Geekbench OpenCL, is much closer. The RX 460 posts 17,855, while the GTX 690 manages 17,399. The delta here is only 2.6% in favor of the AMD card. This is a statistical near-tie, well within the margin of error for such benchmarks. However, the data still awards the win to the RX 460, giving it a clean sweep with 2 wins and 0 losses against the GTX 690. For a card released four years later, that sweep is telling, but the OpenCL result also reveals that the GTX 690’s raw compute horsepower remains competitive in legacy APIs.

Looking at the broader benchmark landscape, the RX 460’s average score across all tests is 18,373, placing it at the 62nd percentile of all GPUs. The GTX 690’s average is 17,037, at the 60th percentile. The RX 460’s nearest rival is the Intel Arc A770M, which scores 18,383, just 0.1% higher. The AMD FirePro D500 is 0.9% ahead at 18,533, while the Radeon Pro 5700 trails by 1% at 18,189. The GTX 690’s closest competitor is the AMD Radeon HD 7970M at 17,019, a mere 0.1% behind, and the RX 7600 XT is 0.3% behind at 17,083. Notably, the GTX 690 sits just 1% below the RTX 3070’s score of 17,208, suggesting that despite its age, its compute performance in these specific tests is not entirely obsolete.

Where Each One Wins

The AMD Radeon RX 460 secures its wins in modern, low-level APIs. The Vulkan score of 20,198 is not just a head-to-head victory; it is the card’s single best benchmark result across all tests, surpassing its own OpenCL score by over 13%. This indicates that the RX 460’s architecture is highly optimized for contemporary driver models and explicit multi-threading. The 21.1% lead over the GTX 690 in Vulkan is the dominant story here. Anyone running Vulkan-based titles or compute workloads would see a decisive advantage on the RX 460.

The OpenCL result is a narrower victory, but it still belongs to the RX 460. With a 2.6% edge, the AMD card is ahead, but the GTX 690 is clearly not outclassed in this older API. The GTX 690’s 17,399 OpenCL score is respectable, and its 60th percentile ranking overall shows it can still hold its own against much newer hardware. However, the GTX 690 has no benchmark wins at all in this comparison. Its weakness is particularly evident in Vulkan, where it scores 16,675, its lowest result. This suggests that the Kepler architecture, released in 2012, lacks the hardware features or driver optimizations needed to accelerate Vulkan workloads effectively.

For use cases, the RX 460 is the clear choice for modern compute and gaming with Vulkan. Its 2 GB of memory and 112.0 GB/s bandwidth are modest, but the architecture’s efficiency compensates. The GTX 690, with its dual-GPU design and 3.130 TFLOPS of FP32 performance, has more raw theoretical throughput, but the benchmark data shows that this does not translate into practical wins in the tested workloads. The GTX 690’s 192.3 GB/s memory bandwidth and 256-bit bus are superior on paper, but they do not overcome the RX 460’s API-level advantages.

FAQ

Q: Which card has the higher average benchmark score?

A: The AMD Radeon RX 460 has an average score of 18,373, which is higher than the NVIDIA GeForce GTX 690’s average of 17,037. The RX 460 also ranks at the 62nd percentile of all GPUs, while the GTX 690 ranks at the 60th.

Q: How large is the Vulkan performance gap between the two cards?

A: The RX 460 scores 20,198 in Geekbench Vulkan, which is 21.1% higher than the GTX 690’s 16,675. This is the largest margin in any head-to-head test between the two cards.

Q: Does the GTX 690 win any benchmark comparison?

A: No. The GTX 690 has 0 wins in the head-to-head benchmarks. The RX 460 wins both the OpenCL test (17,855 vs 17,399) and the Vulkan test (20,198 vs 16,675).

Q: What is the difference in transistor density between the two GPUs?

A: The RX 460 has a transistor density of 24.4M per mm², while the GTX 690 has a density of 12.0M per mm². This is due to the RX 460’s smaller 14 nm process node compared to the GTX 690’s 28 nm node.

Q: Which card has a higher pixel rate?

A: The NVIDIA GeForce GTX 690 has a pixel rate of 32.61 GPixel/s, which is higher than the AMD Radeon RX 460’s 19.20 GPixel/s. Similarly, the GTX 690’s texture rate of 130.4 GTexel/s outpaces the RX 460’s 67.20 GTexel/s.

Q: How do the two cards compare in terms of shading units?

A: The GTX 690 has 1,536 shading units, while the RX 460 has 896. The GTX 690 also has 128 texture mapping units and 32 ROPs, compared to the RX 460’s 56 TMUs and 16 ROPs.

Specification Differences

The two cards differ significantly across nearly every hardware specification. The AMD Radeon RX 460 uses a 14 nm process node manufactured by GlobalFoundries, while the NVIDIA GeForce GTX 690 uses a 28 nm node from TSMC. The RX 460’s die size is 123 mm², containing 3,000 million transistors, whereas the GTX 690’s die is much larger at 294 mm² with 3,540 million transistors. This results in a transistor density of 24.4M per mm² for the RX 460 versus 12.0M per mm² for the GTX 690.

Clock speeds also differ. The RX 460 has a base clock of 1090 MHz and a boost clock of 1200 MHz, while the GTX 690 runs at 915 MHz base and 1019 MHz boost. Memory configurations are similar in size but differ in bandwidth: both cards have 2 GB of GDDR5, but the RX 460 uses a 128-bit bus with 112.0 GB/s bandwidth, while the GTX 690 uses a 256-bit bus with 192.3 GB/s bandwidth. The memory clock is 1750 MHz (7 Gbps effective) for the RX 460 and 1502 MHz (6 Gbps effective) for the GTX 690.

Compute metrics show the GTX 690 with higher raw throughput. It delivers 3.130 TFLOPS of FP32 performance, 32.61 GPixel/s pixel rate, and 130.4 GTexel/s texture rate. The RX 460 produces 2.150 TFLOPS FP32, 19.20 GPixel/s, and 67.20 GTexel/s. The GTX 690 has no listed FP16 performance, while the RX 460 lists FP16 at 2.150 TFLOPS (1:1). Power requirements are vastly different: the RX 460 has a TDP of 75 W and requires no power connectors, while the GTX 690 has a 300 W TDP and needs two 8-pin connectors. The suggested PSU is 250 W for the RX 460 and 700 W for the GTX 690.

Physical dimensions differ as well. The RX 460 is 170 mm long, while the GTX 690 is 279 mm long, 111 mm tall, and 38 mm wide. Both are dual-slot cards. The bus interface is PCIe 3.0 x8 for the RX 460 and PCIe 3.0 x16 for the GTX 690. Display outputs also vary: the RX 460 offers one DVI, one HDMI 2.0b, and one DisplayPort 1.4a, while the GTX 690 offers three DVI ports and one mini-DisplayPort 1.2. API support differs, with the RX 460 supporting DirectX 12 (12_0) and Vulkan 1.3, while the GTX 690 supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Architecture Differences

The architectural divide between these two GPUs is generational. The AMD Radeon RX 460 is built on the GCN 4.0 architecture, part of the Arctic Islands (RX 400) generation, and uses the Baffin chip. The NVIDIA GeForce GTX 690 is based on the Kepler architecture, part of the GeForce 600 generation, and uses the GK104 chip. The RX 460’s process node is 14 nm, produced by GlobalFoundries, while the GTX 690’s is 28 nm from TSMC. This process advantage allows the RX 460 to pack more transistors per area despite having fewer total transistors.

The RX 460’s GCN 4.0 architecture natively supports FP16 compute at a 1:1 ratio with FP32, a feature absent from the GTX 690’s Kepler design, which has no listed FP16 capability. This makes the RX 460 better suited for workloads that leverage half-precision arithmetic. The RX 460 also supports Vulkan 1.3, while the GTX 690 is limited to Vulkan 1.2.175. Both cards lack ray tracing and tensor cores, but their respective architectures handle traditional rasterization differently. The RX 460’s 896 shading units, 56 TMUs, and 16 ROPs represent a balanced, efficiency-focused design. The GTX 690’s 1,536 shading units, 128 TMUs, and 32 ROPs indicate a dual-GPU configuration aimed at maximum throughput.

The GTX 690’s Kepler architecture was a pioneer for its time, but the benchmark results show it struggles with modern APIs. The RX 460’s GCN 4.0, released in the Arctic Islands generation, was designed with contemporary compute workloads in mind. The GTX 690’s release date predates the RX 460 by over four years, and its predecessor is the GeForce 500 series, while the RX 460’s predecessor is Pirate Islands. Both cards are end-of-life products, but the RX 460’s successor is Polaris, while the GTX 690’s successor is GeForce 700. The GTX 690 had a launch MSRP of 999 USD.