AMD Radeon RX 7600 vs NVIDIA GeForce GTX 690 Comparison

Head-to-Head Benchmarks

The recorded data contains two direct comparison points between the NVIDIA GeForce GTX 690 and the AMD Radeon RX 7600. Both are compute-oriented tests, and the results show a decisive advantage for the AMD card. In the Geekbench OpenCL test, the GTX 690 scores 17,399, while the RX 7600 posts 88,051. That is a delta of -80.2%, meaning the RX 7600 is roughly five times faster in raw compute throughput. The Geekbench Vulkan test narrows the gap somewhat but still favors AMD overwhelmingly: 16,675 for the GTX 690 versus 34,401 for the RX 7600, a delta of -51.5%. The RX 7600 wins both head-to-head tests, giving it a clean 2-0 record in the direct comparison set.

These deltas are not marginal. In OpenCL, the RX 7600’s score is 70,652 points higher, a margin that dwarfs any difference seen among the GTX 690’s nearest rivals. For context, the GTX 690’s nearest rivals in the database include the AMD Radeon HD 7970M (17,019, +0.1% delta), the AMD Radeon RX 7600 XT (17,083, -0.3%), the NVIDIA Tesla M4 (16,932, +0.6%), and the NVIDIA GeForce RTX 3070 (17,208, -1%). The GTX 690 sits within 1% of all four, meaning its 17,037 average benchmark score places it in a tightly packed cluster of similarly performing GPUs. The RX 7600, by contrast, has an average benchmark score of 15,171, which places it near the NVIDIA GeForce RTX 3050 OEM (15,199, -0.2%), the AMD Radeon Pro 560X (15,082, +0.6%), the AMD Radeon 680M (15,270, -0.7%), and the NVIDIA GeForce GTX 660 Ti (15,063, +0.7%). The two cards do not even inhabit the same performance tier in the database’s aggregate metrics.

The Vulkan delta of -51.5% is also substantial. The GTX 690’s 16,675 is only about 48% of the RX 7600’s 34,401. This is not a case where one card edges out the other by a few percentage points; the RX 7600 nearly doubles the older NVIDIA card in this API. Across both tests, the RX 7600 demonstrates a generational leap in compute capability that is consistent with its much higher FP32 throughput, though the benchmark scores themselves are the primary evidence.

Where Each One Wins

The RX 7600 wins every direct head-to-head benchmark in the dataset. That includes both OpenCL and Vulkan, so in any workload that leverages these APIs, the AMD card is the clear choice. The OpenCL advantage is particularly stark: 88,051 versus 17,399. This suggests that compute-heavy tasks such as GPGPU processing, rendering, or scientific workloads would see massive gains on the RX 7600. The Vulkan result, 34,401 versus 16,675, also points to strong performance in modern graphics APIs, which is relevant for current games and applications that use Vulkan for rendering.

The GTX 690, despite losing both head-to-head tests, still holds its own in the absolute sense. Its average benchmark score of 17,037 places it in the 60th percentile among all GPUs in the database. That is not a poor showing for a card from the GeForce 600 generation. However, in a direct comparison, it cannot match the RX 7600’s compute output. The RX 7600’s percentile rank is 57, slightly lower than the GTX 690’s 60, but this is a reflection of the wider pool of modern GPUs that the RX 7600 competes against, not a sign that it is slower. The average benchmark score of the RX 7600 is 15,171, which is lower than the GTX 690’s 17,037, yet the head-to-head tests tell a different story. This discrepancy arises because the GTX 690’s average is pulled up by its two Geekbench scores, while the RX 7600’s average includes additional tests such as 3DMark and Passmark entries that drag its mean down. In the two tests where both cards were measured under identical conditions, the RX 7600 dominates.

For gaming specifically, the RX 7600 offers 32 ray accelerators, a feature the GTX 690 lacks entirely. The RX 7600 also supports DirectX 12 Ultimate (12_2), while the GTX 690 is limited to DirectX 12 (11_0). That API gap means the RX 7600 can run the latest DirectX 12 Ultimate features, while the GTX 690 cannot. The GTX 690 does have a higher memory bus width (256 bit versus 128 bit), but the RX 7600 compensates with faster GDDR6 memory and a higher bandwidth of 288.0 GB/s versus 192.3 GB/s.

Architecture Differences

The two GPUs come from different eras and entirely different architectures. The GTX 690 is built on NVIDIA’s Kepler architecture, using the GK104 chip, and was released in May 2012. It is manufactured on a 28 nm process at TSMC, with 3,540 million transistors on a 294 mm² die. The RX 7600, by contrast, uses AMD’s RDNA 3.0 architecture with the Navi 33 chip, released in May 2023. It is built on a 6 nm process, also at TSMC, but packs 13,300 million transistors into a smaller 204 mm² die. The transistor density difference is stark: the RX 7600 achieves 65.2 million transistors per square millimeter, versus 12.0 million for the GTX 690. That is more than a fivefold increase in density, reflecting over a decade of process node improvements.

Clock speeds also differ dramatically. The GTX 690 has a base clock of 915 MHz and a boost clock of 1019 MHz. The RX 7600 runs at a base clock of 1720 MHz, a game clock of 2250 MHz, and a boost clock of 2655 MHz. The boost clock alone is 1636 MHz higher than the GTX 690’s boost. Memory clocks show a similar gap: the GTX 690’s memory runs at 1502 MHz (6 Gbps effective), while the RX 7600’s memory runs at 2250 MHz (18 Gbps effective). The RX 7600 has 8 GB of GDDR6 memory on a 128-bit bus, versus 2 GB of GDDR5 on a 256-bit bus for the GTX 690. Despite the narrower bus, the RX 7600’s higher memory clock yields 288.0 GB/s of bandwidth, which is 95.7 GB/s more than the GTX 690’s 192.3 GB/s.

Compute resources also favor the RX 7600. It has 2048 shading units, 128 texture mapping units, and 64 raster operation units. The GTX 690 has 1536 shading units, 128 TMUs, and 32 ROPs. The pixel rate of the RX 7600 is 169.9 GPixel/s versus 32.61 GPixel/s for the GTX 690, and the texture rate is 339.8 GTexel/s versus 130.4 GTexel/s. FP32 performance is 21.75 TFLOPS for the RX 7600 versus 3.130 TFLOPS for the GTX 690. The RX 7600 also supports FP16 at 21.75 TFLOPS (1:1 ratio), while the GTX 690 has no listed FP16 capability. The RX 7600 includes 32 ray accelerators, which the GTX 690 does not have.

Power and interface specifications also differ. The GTX 690 has a TDP of 300 W and requires two 8-pin power connectors, with a suggested power supply of 700 W. The RX 7600 has a TDP of 165 W, needs a single 8-pin connector, and suggests a 450 W power supply. The RX 7600 uses PCIe 4.0 x8, while the GTX 690 uses PCIe 3.0 x16. Display outputs also differ: the GTX 690 offers 3x DVI and 1x mini-DisplayPort 1.2, while the RX 7600 offers 1x HDMI 2.1a and 3x DisplayPort 2.1. The RX 7600 supports Vulkan 1.4 and DirectX 12 Ultimate, while the GTX 690 supports Vulkan 1.2.175 and DirectX 12 (11_0). Both support OpenGL 4.6.

FAQ

Q: Which card has better compute performance in OpenCL?

A: The AMD Radeon RX 7600 is significantly faster, scoring 88,051 in Geekbench OpenCL versus 17,399 for the GTX 690, a delta of -80.2% in favor of AMD.

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

A: The RX 7600 scores 34,401 in Geekbench Vulkan, while the GTX 690 scores 16,675. The RX 7600 leads by 17,726 points, corresponding to a delta of -51.5%.

Q: Does the GTX 690 support ray tracing?

A: No. The GTX 690 has no ray accelerator cores. The RX 7600 includes 32 ray accelerators.

Q: What is the memory configuration difference between the two GPUs?

A: The GTX 690 has 2 GB of GDDR5 memory on a 256-bit bus with 192.3 GB/s bandwidth. The RX 7600 has 8 GB of GDDR6 memory on a 128-bit bus with 288.0 GB/s bandwidth.

Q: Which card has a higher FP32 performance?

A: The RX 7600 delivers 21.75 TFLOPS of FP32 performance, compared to 3.130 TFLOPS for the GTX 690.

Q: How do the power requirements differ?

A: The GTX 690 has a TDP of 300 W and needs two 8-pin connectors, while the RX 7600 has a TDP of 165 W and requires one 8-pin connector. The suggested power supply is 700 W for the GTX 690 and 450 W for the RX 7600.

The Verdict

The benchmark data is unambiguous. The AMD Radeon RX 7600 wins both head-to-head tests decisively, with an 80.2% advantage in OpenCL and a 51.5% advantage in Vulkan. For anyone choosing between these two cards based on compute performance, the RX 7600 is the only rational option. Its newer architecture, higher clocks, larger memory capacity, and support for modern APIs such as DirectX 12 Ultimate and Vulkan 1.4 give it capabilities that the GTX 690 simply cannot match.

The GTX 690, while still occupying a respectable 60th percentile position in the database, is an end-of-life product from 2012. Its 2 GB memory capacity is a fraction of the RX 7600’s 8 GB, and its lack of ray accelerators means it cannot handle hardware-accelerated ray tracing. The RX 7600 also runs cooler in terms of power draw, with a 165 W TDP versus 300 W, and requires a less demanding power supply.

The RX 7600’s average benchmark score of 15,171 is lower than the GTX 690’s 17,037, but this is misleading because the RX 7600’s average includes a broader set of tests, including Passmark entries where it scores relatively low (e.g., 84 in DirectX 10, 58 in DirectX 12). In the head-to-head tests, the RX 7600 is far faster. The GTX 690’s nearest rivals cluster within 1% of its average score, indicating that it performs similarly to cards like the RTX 3070 in the database’s aggregate metrics, but that aggregate includes only two Geekbench tests for the GTX 690. The RX 7600’s nearest rivals include the GTX 660 Ti, a much older card, which suggests that in the broader database context, the RX 7600 is being compared against a wider range of GPUs, some of which drag its average down.

For a user prioritizing raw compute in OpenCL or Vulkan, the RX 7600 is the clear pick. For legacy compatibility or scenarios where the GTX 690’s specific feature set (such as its 256-bit memory bus or dual-DVI outputs) is required, the GTX 690 remains functional, but the data shows no performance scenario where it beats the RX 7600. The RX 7600 is the superior card in every direct measurement recorded, and its modern feature set makes it the appropriate recommendation for any new build or upgrade.