AMD Radeon RX 9060 vs NVIDIA GeForce GTX 680 Comparison

Head-to-Head Benchmarks

The benchmark data reveals a generational chasm between the AMD Radeon RX 9060 and the NVIDIA GeForce GTX 680. In the two directly comparable head-to-head tests, the RX 9060 wins decisively on both occasions. The most lopsided result comes in Geekbench OpenCL, where the AMD card scores 88,183 against the GTX 680's 16,906. That represents a 421.6% advantage, a margin that dwarfs any other comparison in this matchup. Even the more modern API test, Geekbench Vulkan, shows a 123.7% gap, with the RX 9060 posting 39,476 versus the GTX 680's 17,646.

These are not close contests. The RX 9060's average benchmark score across all recorded tests sits at 16,014, while the GTX 680 manages 14,150. That overall average places the AMD card in the 59th percentile of all GPUs in the database, compared to the 55th percentile for the NVIDIA card. The delta between their aggregate scores is roughly 13%, but the head-to-head results paint a far more dramatic picture because the older card simply lacks the API support and compute throughput to compete in modern workloads.

The RX 9060 also holds a commanding lead in raw specification-driven metrics. Its pixel rate of 191.4 GPixel/s is more than five times the GTX 680's 33.86 GPixel/s. Texture rate follows a similar pattern: 334.9 GTexel/s versus 135.4 GTexel/s, a 2.5x advantage. The FP32 compute figure is the most striking: 21.43 TFLOPS for the AMD part against 3.250 TFLOPS for the NVIDIA part, a 6.6x difference. None of these numbers appear in the head-to-head table, but they explain why the benchmark scores diverge so sharply.

There are no benchmark categories where the GTX 680 wins. The recorded data shows winsA at 2 and winsB at 0. The NVIDIA card does have a Geekbench Metal score of 7,897, but the RX 9060 has no Metal result in the database, so that test cannot be compared directly. Even if it could, the OpenCL and Vulkan results suggest the AMD card would likely dominate there too, though the database does not provide that comparison.

Where Each One Wins

The RX 9060 is the clear choice for any modern compute workload. Its OpenCL score of 88,183 indicates exceptional general-purpose GPU compute performance, suitable for tasks like rendering, physics simulation, and data processing. The Vulkan score of 39,476 shows strong low-level API performance, which translates directly to modern game engines and compute-heavy applications that leverage Vulkan's explicit control. The card's 21.43 TFLOPS FP32 throughput and 21.43 TFLOPS FP16 (at a 1:1 ratio) make it a capable platform for both gaming and professional compute tasks.

The GTX 680's strengths are entirely historical. Its best recorded result is the Geekbench Vulkan score of 17,646, which is still less than half of the RX 9060's Vulkan result. The OpenCL score of 16,906 is similarly modest. The card does have a Metal benchmark result, 7,897, which the RX 9060 lacks, but that only matters for macOS environments and does not offset the massive deficits elsewhere. For any user considering these two cards today, the GTX 680 is only relevant for legacy systems where its 2012-era feature set is sufficient. The RX 9060 wins in every measurable modern scenario.

Architecture Differences

The architectural gap between these two GPUs spans three generations of design philosophy. The RX 9060 is built on RDNA 4.0 architecture, using the Navi 44 chip with the codename Strix Point. It belongs to the Navi IV generation, part of the Radeon RX 9000 series. The GTX 680, in contrast, uses the Kepler architecture with the GK104 chip and belongs to the GeForce 600 generation. The process node difference is stark: the AMD card is fabricated on a 4 nm process at TSMC, while the NVIDIA card uses a 28 nm process, also at TSMC. This process shrink allows the RX 9060 to pack 29,700 million transistors into a 199 mm² die, yielding a transistor density of 149.2M per mm². The GTX 680 has just 3,540 million transistors on a 294 mm² die, for a density of 12.0M per mm².

The RX 9060's feature set reflects its modern origins. It includes 28 dedicated ray tracing cores, something the GTX 680 lacks entirely. The AMD card supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6. The GTX 680 supports DirectX 12 (11_0), Vulkan 1.2.175, and OpenGL 4.6. The DirectX version difference is particularly meaningful: 12_2 enables hardware ray tracing and mesh shaders, while 11_0 is a feature level from the DirectX 11 era. The RX 9060 also supports PCIe 5.0 x16, while the GTX 680 is limited to PCIe 3.0 x16. Display output capabilities differ as well: the AMD card offers 1x HDMI 2.1b and 2x DisplayPort 2.1a, whereas the NVIDIA card has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2.

Specification Differences

The specification sheet shows a complete overhaul between generations. The RX 9060 has 1,792 shading units, 112 texture mapping units, and 64 ROPs. The GTX 680 has 1,536 shading units, 128 TMUs, and 32 ROPs. Despite having fewer TMUs, the RX 9060's texture rate is more than double because of its much higher clock speeds. The AMD card runs at a base clock of 1700 MHz, a game clock of 2400 MHz, and a boost clock of 2990 MHz. The NVIDIA card's base clock is 1006 MHz with a boost of 1058 MHz. Memory configuration also diverges sharply: the RX 9060 has 8 GB of GDDR6 on a 128-bit bus, delivering 288.0 GB/s of bandwidth. The GTX 680 has 2 GB of GDDR5 on a 256-bit bus, providing 192.3 GB/s. The wider bus cannot compensate for the older memory type and lower clock speed.

Power requirements reflect the performance gap. The RX 9060 has a TDP of 132 W and requires a single 8-pin power connector, with a suggested power supply of 300 W. The GTX 680 draws 195 W and needs two 6-pin connectors, with a suggested power supply of 450 W. The RX 9060 achieves far higher performance while consuming 63 W less power. Both cards are dual-slot designs. The GTX 680 has recorded physical dimensions of 256 mm in length, 111 mm in height, and 38 mm in width; the RX 9060's dimensions are not recorded in the database. The production status also differs: the RX 9060 is marked as active, while the GTX 680 is end-of-life.

FAQ

Q: How much faster is the AMD Radeon RX 9060 than the NVIDIA GeForce GTX 680 in OpenCL?

A: In the Geekbench OpenCL test, the RX 9060 scores 88,183 compared to the GTX 680's 16,906, which is a 421.6% advantage.

Q: Does the GTX 680 win any benchmark in the head-to-head comparison?

A: No. The recorded head-to-head data shows the RX 9060 winning 2 tests and the GTX 680 winning 0 tests.

Q: What are the power consumption figures for each card?

A: The RX 9060 has a TDP of 132 W and requires a 300 W power supply. The GTX 680 has a TDP of 195 W and requires a 450 W power supply.

Q: Which card has more memory bandwidth?

A: The RX 9060 delivers 288.0 GB/s from 8 GB of GDDR6 on a 128-bit bus. The GTX 680 provides 192.3 GB/s from 2 GB of GDDR5 on a 256-bit bus.

Q: Does the GTX 680 support ray tracing?

A: No. The GTX 680 has no ray tracing cores. The RX 9060 includes 28 dedicated ray tracing cores.

Q: What is the launch MSRP of the GTX 680?

A: The GTX 680 had a launch MSRP of 499 USD. The RX 9060 has no recorded launch MSRP in the database.