NVIDIA GeForce RTX 2060

The NVIDIA GeForce RTX 2060, built on the Turing architecture with the TU106 chip at TSMC’s 12 nm process, holds a distinct position in the benchmark database. With an average benchmark score of 16074, it sits near the 58th percentile of all GPUs tracked, indicating a solid mid-range standing rather than a top-tier performer. The card integrates 1920 shading units, 120 texture mapping units, and 48 raster operation pipelines, alongside 30 RT cores and 240 tensor cores, which underpin its ray tracing and AI-assisted features. Its production status is end-of-life, and it was released in early 2019, succeeding the GeForce 10 series and preceding the GeForce 30 series. The data reveals a capable card for its era, though the benchmarks show a mix of strengths across different API generations and workloads.

Benchmark Performance

The RTX 2060’s performance profile is best understood through its varied benchmark results. In 3DMark Steel Nomad (DX12), the card scores 1242, a raw metric that places it in a moderate performance tier for modern DirectX 12 titles. Geekbench results show a split between compute-oriented and graphics-oriented loads: OpenCL scores 71278, while Vulkan trails slightly at 67425, suggesting the card handles OpenCL compute tasks more efficiently than Vulkan graphics workloads in synthetic tests. PassMark results offer a broader view — the GPU scores 14111 in the G3D test, which aggregates overall 3D performance, and 5488 in GPU compute, indicating a more modest compute throughput relative to its rasterization capabilities.

Comparing to nearest rivals, the average benchmark score of 16074 places the RTX 2060 marginally ahead of the AMD Radeon RX 7600S, which averages 15996, a delta of 0.5%. This is a razor-thin margin, effectively a statistical tie in aggregate scores. Against the AMD Radeon Pro 5600M and the AMD Radeon Pro 5700, both averaging 16351, the RTX 2060 trails by 1.7%, a small but consistent deficit. The AMD Radeon Pro W5500 averages 15786, meaning the RTX 2060 leads it by 1.8%. These deltas highlight that the RTX 2060 sits in a tightly contested performance band, where no rival holds a dominant edge in overall average scores.

Delving into specific benchmark deltas, the PassMark DirectX 11 score of 110 versus the DirectX 12 score of 53 shows a notable regression in newer API overhead, while DirectX 9 scores 190 and DirectX 10 scores 98. This pattern suggests the card’s driver optimizations favor legacy APIs, but modern titles using DX12 may not extract full potential. The pixel rate of 80.64 GPixel/s and texture rate of 201.6 GTexel/s, derived from the 48 ROPs and 120 TMUs at boost clocks, provide theoretical ceilings that align with its mid-range positioning. FP32 compute sits at 6.451 TFLOPS, with FP16 at 12.90 TFLOPS (2:1 ratio), indicating the tensor cores enable accelerated half-precision workloads.

Power and Cooling

The RTX 2060 carries a TDP of 160 W, a figure that dictates its thermal and power delivery requirements. The suggested PSU rating is 450 W, which accommodates the card’s draw along with a typical system configuration, though higher-wattage units are never a drawback for headroom. Power is supplied via a single 8-pin connector, simplifying installation compared to cards requiring multiple connectors. The card is dual-slot in width, measuring 229 mm in length (9 inches), 113 mm in height (4.4 inches), and 35 mm in width (1.4 inches). These dimensions allow it to fit in most mid-tower cases, but compact builds should verify clearance.

Cooling is handled by a dual-slot design, which provides adequate thermal dissipation for the 160 W TDP, but the data does not specify fan configurations or heat sink sizes. The 12 nm process node from TSMC, with 10,800 million transistors on a 445 mm² die, yields a transistor density of 24.3M per mm², which is moderate by modern standards but was efficient for its generation. The card’s boost clock of 1680 MHz, up from a base of 1365 MHz, means the GPU will dynamically increase frequency under load, raising power draw toward the TDP ceiling. For users with older or lower-capacity PSUs, the 450 W recommendation serves as a baseline, but the single 8-pin connector limits peak power input to a known threshold.

Memory Subsystem

Memory configuration is a defining characteristic: 6 GB of GDDR6 on a 192-bit bus, yielding a bandwidth of 336.0 GB/s. The memory clock runs at 1750 MHz, translating to 14 Gbps effective data rate. This bandwidth is sufficient for 1080p and 1440p gaming, but the 6 GB capacity becomes a constraint at higher resolutions or with texture-heavy titles. At 4K, the combination of 6 GB VRAM and 336.0 GB/s bandwidth may cause frame pacing issues when textures exceed the available memory, leading to stutters or reduced texture quality settings.

The 192-bit bus width is narrower than higher-tier cards, but the GDDR6 type mitigates this with higher per-pin data rates. For ray tracing workloads, the 30 RT cores and 240 tensor cores rely on memory bandwidth to fetch acceleration structures and neural network data; the 336.0 GB/s figure is adequate for the RTX 2060’s intended performance level, but it does not provide the headroom seen in larger-memory rivals. The FP16 throughput of 12.90 TFLOPS suggests the tensor cores can process mixed-precision workloads efficiently, but memory bandwidth remains the bottleneck in scenarios with large datasets. Overall, the memory subsystem supports smooth gameplay at 1080p with high settings, while 1440p users may need to adjust settings for VRAM-intensive titles.

How It Compares

Against the AMD Radeon RX 7600S, the RTX 2060 holds a 0.5% average score advantage (16074 vs 15996). This delta is negligible in real-world terms, meaning both cards deliver nearly identical aggregate performance. The RX 7600S, being a mobile-oriented variant, may have different power characteristics, but the benchmark data treats them as equivalent in compute output.

The AMD Radeon Pro 5600M and AMD Radeon Pro 5700 both outscore the RTX 2060 by 1.7% (16351 vs 16074). These professional-grade cards, despite their workstation focus, exhibit a slight performance lead in the average benchmark suite. For gaming, this margin is unlikely to produce noticeable differences, but in compute-heavy professional tasks, the Pro cards may edge ahead.

The AMD Radeon Pro W5500 trails the RTX 2060 by 1.8% (15786 vs 16074). This is the largest delta among the nearest rivals, yet still a small gap. The RTX 2060’s advantage here suggests its Turing architecture offers a modest performance uplift over the W5500 in the tested workloads. Overall, the RTX 2060 occupies a narrow performance band where rival GPUs cluster within ±2% of its average score, making architectural features and driver support more decisive than raw benchmark differences.

Who Should Consider It

Benchmark results indicate the RTX 2060 is best suited for 1080p gaming at high to ultra settings. The PassMark G3D score of 14111 and 3DMark Steel Nomad score of 1242 suggest it can handle modern DirectX 12 titles at this resolution with consistent frame rates. For 1440p, the card remains viable, but users should expect to dial back some settings to maintain smooth performance, given the 6 GB VRAM and 336.0 GB/s bandwidth. The 58th percentile ranking means it outperforms roughly 58% of all GPUs in the database, placing it above entry-level cards but below high-end models.

Users who prioritize ray tracing will find the 30 RT cores functional, but the 160 W TDP and mid-range compute power (6.451 TFLOPS FP32) limit ray-traced effects to lighter scenes or lower resolutions. The end-of-life status means no new driver optimizations are expected, so longevity depends on existing software support. The Geekbench Vulkan score of 67425 versus OpenCL’s 71278 suggests Vulkan-based games may run slightly slower than OpenCL compute tasks, but this is a minor consideration. For budget-conscious builders who find this card at a reduced price, its performance is competitive with the nearest rivals, though the lack of modern features like DLSS 3 or larger VRAM buffers is a drawback.

FAQ

Q: What is the average benchmark score of the RTX 2060?

A: The average benchmark score is 16074, placing it at the 58th percentile of all GPUs tracked.

Q: How does the RTX 2060 compare to the AMD Radeon RX 7600S?

A: The RTX 2060 scores 0.5% higher on average (16074 vs 15996), making the performance difference negligible.

Q: What is the TDP of the RTX 2060 and what PSU is recommended?

A: The TDP is 160 W, and the suggested PSU rating is 450 W, with power supplied via a single 8-pin connector.

Q: How much VRAM does the RTX 2060 have, and what is its bandwidth?

A: It has 6 GB of GDDR6 memory on a 192-bit bus, providing 336.0 GB/s of bandwidth at 14 Gbps effective.

Q: Does the RTX 2060 support ray tracing?

A: Yes, it includes 30 RT cores and 240 tensor cores, enabling ray tracing and AI-accelerated features, though performance is limited by its mid-range compute power.

Q: What is the launch MSRP of the RTX 2060?

A: The launch MSRP is 349 USD, which was set at its release in early 2019.