AMD Radeon RX 7700 vs NVIDIA GeForce GTX 690 Comparison

# Head-to-Head Benchmarks

The benchmark data for these two GPUs tells a starkly one-sided story, with the AMD Radeon RX 7700 dominating the only directly comparable test. In Geekbench OpenCL, the RX 7700 scores 106,028 points, while the NVIDIA GeForce GTX 690 manages just 17,399. That represents an 83.6% advantage for the AMD card, a margin that essentially places these products in different performance tiers altogether.

The GTX 690's average benchmark score of 17,037 across all tested workloads places it at the 60th percentile of all GPUs in the database. Its nearest rivals include the AMD Radeon HD 7970M, which scores 17,019 (a 0.1% difference), and the NVIDIA Tesla M4 at 16,932 (0.6% behind). The GTX 690 also sits within 1% of the AMD Radeon RX 7600 XT, which scores 17,083, and trails the NVIDIA GeForce RTX 3070 by 1% (17,208 vs. 17,399). These tight margins suggest the GTX 690, despite its age, remains competitive with much newer entry-level and mid-range offerings in synthetic compute workloads.

The RX 7700, by contrast, posts an average benchmark score of 15,852, which places it at the 58th percentile of all GPUs. Its nearest rivals include the AMD Radeon R9 370X at 15,862 (a 0.1% difference), the AMD Radeon Pro W5500 at 15,679 (1.1% behind), the AMD Radeon RX 9060 at 16,014 (1% ahead), and the NVIDIA GeForce RTX 3060 Ti at 16,129 (1.7% ahead). Interestingly, the RX 7700's average score is lower than the GTX 690's average, despite the RX 7700's overwhelming victory in the single head-to-head OpenCL test. This discrepancy arises because the RX 7700's benchmark suite includes additional tests—PassMark DirectX 9, 10, 11, and 12, plus PassMark G2D and G3D—that drag down its average. The GTX 690 only has two benchmarks in its profile, both Geekbench-based.

Looking at the RX 7700's individual benchmark results, it achieves 2,865 in 3DMark Steel Nomad DX12, 20974 in PassMark G3D, and 10,963 in PassMark GPU Compute. The PassMark DirectX variants show scores of 261 for DX9, 92 for DX10, 186 for DX11, and 96 for DX12, while the G2D test yields 1,206. These results demonstrate a wide-ranging capability across API generations, from legacy DirectX 9 to modern DX12 workloads.

FAQ

Q: Which GPU wins the only direct head-to-head benchmark comparison?

A: The AMD Radeon RX 7700 wins decisively in Geekbench OpenCL, scoring 106,028 versus the NVIDIA GeForce GTX 690's 17,399, an 83.6% advantage.

Q: How do these GPUs compare in terms of overall database percentile ranking?

A: The GTX 690 sits at the 60th percentile of all GPUs, while the RX 7700 ranks at the 58th percentile, despite the RX 7700's much higher raw scores in individual tests.

Q: What is the average benchmark score for each card?

A: The GTX 690 has an average benchmark score of 17,037, while the RX 7700 averages 15,852. The RX 7700's average is pulled down by its broader test suite, which includes multiple PassMark workloads.

Q: Which GPUs are nearest rivals to the GTX 690 in the database?

A: The GTX 690's nearest rivals include the AMD Radeon HD 7970M (0.1% ahead), AMD Radeon RX 7600 XT (0.3% behind), NVIDIA Tesla M4 (0.6% ahead), and NVIDIA GeForce RTX 3070 (1% ahead).

Q: Which GPUs are nearest rivals to the RX 7700 in the database?

A: The RX 7700's nearest rivals include the AMD Radeon R9 370X (0.1% ahead), AMD Radeon Pro W5500 (1.1% behind), AMD Radeon RX 9060 (1% ahead), and NVIDIA GeForce RTX 3060 Ti (1.7% ahead).

Q: Does the GTX 690 have any benchmark wins over the RX 7700?

A: No. The head-to-head data shows zero wins for the GTX 690 and one win for the RX 7700 in the Geekbench OpenCL test.

Architecture Differences

The architectural gap between these two GPUs spans over a decade of semiconductor evolution. The NVIDIA GeForce GTX 690 is built on the GK104 chip using the Kepler architecture, manufactured on TSMC's 28 nm process. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0 million transistors per square millimeter. The RX 7700, by contrast, uses AMD's Navi 32 chip built on the RDNA 3.0 architecture, fabricated on TSMC's 5 nm process. This newer node allows AMD to pack 28,100 million transistors into a 346 mm² die, achieving a density of 81.2 million transistors per square millimeter—nearly seven times denser than the Kepler chip.

The RX 7700 also introduces hardware features that did not exist during the GTX 690's era. It includes 40 ray tracing cores, enabling hardware-accelerated ray tracing, a capability the Kepler architecture lacks entirely. The RX 7700 supports DirectX 12 Ultimate (12_2), while the GTX 690 is limited to DirectX 12 (11_0). Both cards support OpenGL 4.6, but the RX 7700 features Vulkan 1.4 support compared to the GTX 690's Vulkan 1.2.175.

The GTX 690 is a dual-GPU card at its core, combining two Kepler dies on a single board, which explains its 300 W TDP and dual-slot design with two 8-pin power connectors. The RX 7700, despite being a single-die solution, is more power-efficient at 200 W TDP, requiring a 550 W suggested PSU compared to the GTX 690's 700 W recommendation. The RX 7700 also supports modern display outputs including HDMI 2.1a, dual DisplayPort 2.1, and USB Type-C, while the GTX 690 offers three DVI ports and a single mini-DisplayPort 1.2.

Specification Differences

The specification sheets reveal dramatic generational differences. The RX 7700 features 2,560 shading units, 160 texture mapping units, and 96 render output units, compared to the GTX 690's 1,536 shading units, 128 TMUs, and 32 ROPs. The RX 7700 also includes 40 ray tracing cores, a feature absent from the GTX 690.

Memory configurations differ substantially. The GTX 690 offers 2 GB of GDDR5 memory on a 256-bit bus with 192.3 GB/s bandwidth, while the RX 7700 provides 16 GB of GDDR6 memory on a 256-bit bus with 624.1 GB/s bandwidth—more than three times the bandwidth and eight times the capacity.

Clock speeds show the RX 7700's process advantage: it runs at 1900 MHz base and 2459 MHz boost (2041 MHz game clock), while the GTX 690 operates at 915 MHz base and 1019 MHz boost. Memory clocks are similarly lopsided: 2438 MHz (19.5 Gbps effective) for the RX 7700 versus 1502 MHz (6 Gbps effective) for the GTX 690.

Compute throughput is dramatically higher on the RX 7700, with 25.18 TFLOPS FP32 performance and matching 25.18 TFLOPS FP16 (1:1 ratio). The GTX 690 manages 3.130 TFLOPS FP32 with no FP16 capability listed. Pixel and texture rates follow suit: the RX 7700 achieves 236.1 GPixel/s and 393.4 GTexel/s, versus 32.61 GPixel/s and 130.4 GTexel/s for the GTX 690.

The bus interface also differs: PCIe 4.0 x16 on the RX 7700 versus PCIe 3.0 x16 on the GTX 690. Physical dimensions vary slightly, with the RX 7700 measuring 267 mm in length, 135 mm in height, and 50 mm in width, while the GTX 690 is 279 mm long, 111 mm tall, and 38 mm wide. Both are dual-slot cards.

Where Each One Wins

The RX 7700 is the clear winner for modern gaming and compute workloads. Its 83.6% advantage in OpenCL performance indicates vastly superior compute throughput, while its 16 GB memory capacity and 624.1 GB/s bandwidth make it suitable for high-resolution textures and memory-intensive applications. The inclusion of ray tracing cores and DirectX 12 Ultimate support means it can handle contemporary games with advanced graphics features. Its 200 W TDP and 550 W suggested PSU also make it more practical for modern system builds.

The RX 7700's nearest rival comparisons show it competing well against cards like the RTX 3060 Ti and RX 9060, suggesting it occupies a solid mid-range position. Its PassMark G3D score of 20,974 and 3DMark Steel Nomad score of 2,865 indicate strong rasterization performance across both legacy and modern APIs.

The GTX 690's advantages are limited to its higher percentile ranking (60th vs. 58th) and its slightly higher average benchmark score (17,037 vs. 15,852). However, these metrics are misleading—the GTX 690's benchmark profile is much narrower, and its average is not diluted by the additional PassMark tests that the RX 7700 undergoes. The GTX 690's nearest rival comparisons show it within 1% of the RTX 3070 in average score, which is noteworthy for a 2012 product. For users with legacy software that relies on older API implementations, or for those seeking a historical comparison point, the GTX 690 demonstrates surprising longevity. However, its 2 GB memory capacity and lack of ray tracing support severely limit its relevance for modern workloads. The data clearly positions the RX 7700 as the superior choice for essentially any current use case, from gaming to compute tasks, while the GTX 690 remains a fascinating relic of the Kepler era that still holds its own in synthetic benchmarks against much newer hardware.