NVIDIA GeForce GTX 690 vs NVIDIA GeForce RTX 3060 Ti Comparison

The GeForce GTX 690 and the GeForce RTX 3060 Ti represent two very different eras of NVIDIA engineering. The GTX 690, a dual-GPU flagship from 2012, was built for raw multi-card scaling, while the RTX 3060 Ti is a modern single-chip Ampere design focused on efficiency and feature support. The benchmark data reveals a decisive generational shift, but each card retains a distinct identity for specific use cases.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The GTX 690 has a higher average benchmark score of 17037, compared to the RTX 3060 Ti's 16129. However, this average is pulled from only two legacy tests (Geekbench OpenCL and Vulkan) for the GTX 690, while the RTX 3060 Ti's average includes a broader range of modern workloads.

Q: How does the RTX 3060 Ti perform in raw compute tests?

A: In Geekbench OpenCL, the RTX 3060 Ti scores 78927, which is 78% higher than the GTX 690's 17399. The RTX 3060 Ti also leads in Geekbench Vulkan with 47784 points versus 16675, a 65.1% advantage.

Q: What is the memory configuration difference?

A: The GTX 690 has 2 GB of GDDR5 memory on a 256-bit bus, delivering 192.3 GB/s bandwidth. The RTX 3060 Ti doubles capacity to 8 GB of GDDR6 on the same 256-bit bus, but more than doubles bandwidth to 448.0 GB/s.

Q: Do both cards support the same DirectX version?

A: No. The GTX 690 supports DirectX 12 (11_0), while the RTX 3060 Ti supports DirectX 12 Ultimate (12_2). The newer card also supports a newer Vulkan version (1.4 vs 1.2.175).

Q: Which card has hardware ray tracing and tensor cores?

A: Only the RTX 3060 Ti has dedicated hardware for these features, with 38 RT cores and 152 tensor cores. The GTX 690 has no RT or tensor cores listed.

Q: How do their transistor counts compare?

A: The RTX 3060 Ti packs 17,400 million transistors on a 392 mm² die (44.4M / mm²), while the GTX 690 has 3,540 million transistors on a 294 mm² die (12.0M / mm²).

Architecture Differences

The architectural gap between these two cards is massive and goes far beyond simple clock speeds. The GTX 690 is built on the Kepler architecture using the GK104 chip, fabricated on a 28 nm process at TSMC. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0M / mm². The RTX 3060 Ti uses the Ampere architecture with the GA104 chip, manufactured on Samsung's 8 nm process. This newer node allows 17,400 million transistors in a 392 mm² die, achieving a density of 44.4M / mm² — nearly four times the density of the older chip.

The shading unit count tells a similar story. The GTX 690 features 1536 shading units, 128 texture mapping units (TMUs), and 32 ROPs. The RTX 3060 Ti dramatically increases these numbers to 4864 shading units, 152 TMUs, and 80 ROPs. This expansion is complemented by the addition of 38 RT cores and 152 tensor cores, which are entirely absent from the GTX 690. These dedicated units enable hardware-accelerated ray tracing and AI-driven features like DLSS, fundamentally changing what the card can do.

Memory architecture also diverges. The GTX 690 uses 2 GB of GDDR5 at 1502 MHz (6 Gbps effective), while the RTX 3060 Ti uses 8 GB of GDDR6 at 1750 MHz (14 Gbps effective). Both use a 256-bit bus, but the newer memory type and higher clock speed push bandwidth from 192.3 GB/s to 448.0 GB/s. The GTX 690's dual-GPU design means its 2 GB pool is split across two GPUs, effectively limiting usable memory per GPU to 1 GB in many scenarios — a critical limitation for modern games.

The API support also reflects their eras. The GTX 690 lists DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The RTX 3060 Ti supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The newer Vulkan version and DirectX 12 Ultimate feature set enable advanced rendering techniques that the older card simply cannot handle.

Head-to-Head Benchmarks

The head-to-head data is limited to two Geekbench tests, but the results are unambiguous. In Geekbench OpenCL, the RTX 3060 Ti scores 78927 against the GTX 690's 17399. This is a 78% delta in favor of the newer card. The RTX 3060 Ti also wins Geekbench Vulkan with 47784 points versus 16675, a 65.1% advantage. In both tests, the RTX 3060 Ti more than doubles the GTX 690's raw output.

These scores align with the theoretical compute figures. The GTX 690 delivers 3.130 TFLOPS of FP32 performance, while the RTX 3060 Ti delivers 16.20 TFLOPS. Texture fill rates follow suit: 130.4 GTexel/s for the GTX 690 versus 253.1 GTexel/s for the RTX 3060 Ti. Pixel rates show an even wider gap, with the RTX 3060 Ti hitting 133.2 GPixel/s compared to the GTX 690's 32.61 GPixel/s.

The GTX 690's average benchmark score of 17037 is actually higher than the RTX 3060 Ti's 16129, but this is misleading. The GTX 690's score is based on only two tests, both of which are older API workloads. The RTX 3060 Ti's score includes modern tests like PassMark G3D (20349) and 3DMark Steel Nomad DX12 (2626), which stress features the GTX 690 cannot run efficiently. In the two shared tests, the RTX 3060 Ti wins decisively.

Looking at the nearest rivals provides additional context. The GTX 690 sits at the 60th percentile of all GPUs, with its closest competitor being the NVIDIA GeForce RTX 3070, which scores 17208 (a 1% delta). The RTX 3060 Ti sits at the 59th percentile, with the AMD Radeon RX 5700 XT (16361) as a close reference point (1.4% delta). Both cards occupy similar overall performance tiers, but the RTX 3060 Ti does so with far more modern features and efficiency.

Specification Differences

The specification sheets reveal stark contrasts across nearly every field. The GTX 690 has a base clock of 915 MHz and a boost of 1019 MHz, while the RTX 3060 Ti runs at 1410 MHz base and 1665 MHz boost. Memory clocks are 1502 MHz (6 Gbps effective) for the GTX 690 versus 1750 MHz (14 Gbps effective) for the RTX 3060 Ti.

Power requirements differ significantly. The GTX 690 draws 300 W TDP and requires a 700 W suggested PSU, with two 8-pin power connectors. The RTX 3060 Ti is more efficient at 200 W TDP, needs only a 550 W PSU, and uses a single 12-pin connector. Physical dimensions also change: the GTX 690 is 279 mm long, 111 mm tall, and 38 mm wide, while the RTX 3060 Ti is shorter at 242 mm, similar at 112 mm tall, with no width listed.

Display outputs shift from the GTX 690's 3x DVI and 1x mini-DisplayPort 1.2 to the RTX 3060 Ti's 1x HDMI 2.1 and 3x DisplayPort 1.4a. The bus interface upgrades from PCIe 3.0 x16 to PCIe 4.0 x16. The RTX 3060 Ti also lists FP16 performance at 16.20 TFLOPS (1:1 ratio), a feature entirely absent from the GTX 690's specifications.

The release dates are nearly a decade apart: May 2, 2012 for the GTX 690 and November 30, 2020 for the RTX 3060 Ti. Both are now end-of-life products. The GTX 690's successor is the GeForce 700 series, while the RTX 3060 Ti's successor is the GeForce 40 series.

Where Each One Wins

The RTX 3060 Ti wins every measurable performance category in the head-to-head data. It is 78% faster in OpenCL and 65.1% faster in Vulkan. The newer card also dominates in shader count (4864 vs 1536), texture rate (253.1 vs 130.4 GTexel/s), and pixel rate (133.2 vs 32.61 GPixel/s). Its 8 GB memory capacity versus 2 GB makes it viable for modern games and applications that require larger frame buffers. The 448.0 GB/s bandwidth is more than double the GTX 690's 192.3 GB/s, directly benefiting high-resolution textures and compute workloads.

The GTX 690's only "win" is its higher average benchmark score of 17037 versus 16129, but this is a statistical artifact of the limited test set. It also has a higher percentile ranking at 60 versus 59, a negligible difference. The GTX 690's launch MSRP of 999 USD reflects its original dual-GPU flagship positioning, but the data shows no real-world advantage in modern workloads.

For use cases, the RTX 3060 Ti is the clear choice for any current gaming or productivity workload. Its DirectX 12 Ultimate support, ray tracing cores, and tensor cores unlock features that the GTX 690 cannot access. The GTX 690 might be interesting only for retro benchmarking or testing legacy software from its 2012 era, but even then, the RTX 3060 Ti's superior driver support and modern APIs make it more practical.

The Verdict

The data is unambiguous: the RTX 3060 Ti is the superior GPU by every meaningful performance metric. It delivers 4.5x the FP32 compute (16.20 vs 3.130 TFLOPS), 2.3x the memory bandwidth (448.0 vs 192.3 GB/s), and 3.2x the shading units (4864 vs 1536). In head-to-head tests, it wins by 78% and 65.1% margins. This is not a close competition — it is a generational leap.

Anyone building a system today should choose the RTX 3060 Ti if they want to play modern games, use ray tracing, or run compute workloads that leverage tensor cores. The 8 GB frame buffer is essential for current titles, and the lower 200 W TDP means less strain on the power supply. The GTX 690's 300 W draw and dual-GPU complexity offer no tangible benefit in any scenario covered by the data.

The only reason to pick the GTX 690 is historical interest or a very specific legacy application from its 2012 release window. Its 2 GB memory and lack of modern API support make it unsuitable for anything beyond light, older tasks. The RTX 3060 Ti is the practical choice for virtually every buyer, offering more performance, more features, and better efficiency in a smaller physical footprint. The verdict is simple: the RTX 3060 Ti wins decisively, and the GTX 690 belongs in a museum, not a PC case.