NVIDIA GeForce RTX 5060

The NVIDIA GeForce RTX 5060 is an active GeForce 50-series GPU from NVIDIA, based on the GB206 chip and Blackwell 2.0 architecture. It is manufactured by TSMC on a 5 nm process. The die contains 21,900 million transistors on 181 mm², for a transistor density of 121.0M / mm². Its predecessor is GeForce 40. The base clock is 2280 MHz and the boost clock is 2497 MHz. The release date is 2025-05-18, and the launch MSRP is 299 USD. The average benchmark score is 30109, placing it at the 73rd percentile of all GPUs. In the nearestRivals data, the AMD Radeon RX 7700 XT has an avgScore of 30097, the AMD Radeon RX 570 has 30158, the NVIDIA Tesla M60 has 29969, and the NVIDIA GeForce RTX 5070 Mobile has 29928. The corresponding deltaPct values are 0, -0.2, 0.5, and 0.6.

Who Should Consider It

Benchmark results indicate a card with strong compute and cross-API rendering. Geekbench OpenCL reports 120785, while Geekbench Vulkan reports 116626. Passmark G3D is 20891, and Passmark GPU compute is 10899. The Passmark G2D score is 1154. The Passmark DirectX 9 score is 225, DirectX 10 is 127, DirectX 11 is 200, and DirectX 12 is 77. The DirectX 12 result is the weakest Passmark DirectX score, which matters for users running modern feature-heavy workloads. The DirectX 9 and 11 scores are higher, so older API workloads fare better. At high resolutions, the 8 GB GDDR7 frame buffer and 128-bit bus define the practical settings ceiling. The 448.0 GB/s bandwidth and 19.18 TFLOPS FP32 rate are strong, but capacity is the primary constraint. The 3840 shading units and 120 TMUs need the memory subsystem to stay within 8 GB to avoid a capacity ceiling. The 73rd percentile ranking places it in the middle of the full GPU database. Users whose workloads fit inside the memory budget will see the best results. Users who push beyond that budget should not rely on the 19.18 TFLOPS figure to compensate.

Memory Subsystem

The RTX 5060 uses 8 GB of GDDR7 memory. The bus width is 128 bits, and memory bandwidth is 448.0 GB/s. The memory clock is 1750 MHz, with an effective data rate of 28 Gbps. The memory type is GDDR7, and 28 Gbps is the effective per-pin rate. With a 128-bit bus, total bandwidth is 448.0 GB/s. For high resolutions, the 8 GB capacity is the primary working-set limit. The 128-bit bus then determines the rate at which that working set can be refreshed. The 1750 MHz clock is the memory clock reference; the 28 Gbps effective figure is the higher-speed signaling rate. A 448.0 GB/s bandwidth is the aggregate figure for texture streaming demands. The 120 TMUs and 48 ROPs can generate 299.6 GTexel/s and 119.9 GPixel/s, respectively, but sustained output requires the memory subsystem to keep up. The combination of 8 GB capacity and 448.0 GB/s bandwidth is the memory envelope for all high-resolution work.

Power and Cooling

The TDP is 145 W. The suggested power supply is 300 W. The power connector requirement is 1x 8-pin. The card occupies a dual-slot width. Physical dimensions are 241 mm length, 111 mm height, and 40 mm width. The 145 W TDP is the card's thermal design power. A 300 W PSU recommendation is the system-level guide. The 1x 8-pin connector is the only power input listed. Builders need clearance for a 241 mm long card, 111 mm height, and 40 mm width. Dual-slot cooling is the specified cooler footprint. The 5 nm TSMC process and 21,900 million transistor count are the manufacturing context, while the power figures are the planning numbers. This is not a multi-connector power card.

How It Compares

The RTX 5060's average benchmark score is 30109. The nearestRivals data shows deltaPct values of 0, -0.2, 0.5, and 0.6, so all listed rivals sit within under a percentage point of the RTX 5060 in aggregate average score.

The AMD Radeon RX 7700 XT has an avgScore of 30097 and a deltaPct of 0. This puts the RX 7700 XT at the same aggregate average performance as the RTX 5060. A deltaPct of 0 is a precise tie in the benchmark data, not a near miss.

The AMD Radeon RX 570 has an avgScore of 30158 and a deltaPct of -0.2. The RTX 5060 is 0.2% behind the RX 570 in average score. A gap of 0.2% is too small to separate the cards in practical ranking.

The NVIDIA Tesla M60 has an avgScore of 29969 and a deltaPct of 0.5. The RTX 5060 is 0.5% ahead of the Tesla M60. The Tesla M60 average score of 29969 is lower than the RTX 5060's 30109 but still places both in the same band.

The NVIDIA GeForce RTX 5070 Mobile has an avgScore of 29928 and a deltaPct of 0.6. The RTX 5060 is 0.6% ahead of this mobile variant. This is the largest deltaPct in the nearestRivals list, yet it is still under a percentage point. The RTX 5060's 73rd percentile rank therefore comes from the broader database, not from decisively beating these rivals.

Ray Tracing and Feature Set

The RTX 5060 has 30 RT cores and 120 tensor cores. It also has 3840 shading units, 120 TMUs, and 48 ROPs. The RT cores are the ray tracing hardware block. The tensor cores are the tensor hardware block. The architecture is Blackwell 2.0. The FP32 compute rate is 19.18 TFLOPS, and the FP16 rate is also 19.18 TFLOPS at a 1:1 ratio. The pixel rate is 119.9 GPixel/s, and the texture rate is 299.6 GTexel/s. The API list is DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The bus interface is PCIe 5.0 x8. Display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b. Geekbench OpenCL and Vulkan scores of 120785 and 116626 further characterize the feature set in compute terms. The 120 tensor cores and 19.18 TFLOPS FP16 rate are the tensor-related throughput figures, though the FACT PACK does not provide a dedicated tensor benchmark. The 30 RT cores provide the ray tracing capability, and DirectX 12 Ultimate (12_2) is the feature level that software can target.

FAQ

Q: What are the core specifications of the RTX 5060?

A: It uses the GB206 chip on Blackwell 2.0 architecture with 3840 shading units, 120 TMUs, 48 ROPs, 30 RT cores, and 120 tensor cores. It is built on TSMC's 5 nm process with 21,900 million transistors.

Q: What is the memory configuration?

A: The card has 8 GB of GDDR7 memory on a 128-bit bus, with a 1750 MHz memory clock and a 28 Gbps effective data rate, for 448.0 GB/s of bandwidth.

Q: What power supply and connector are specified?

A: The TDP is 145 W, the suggested PSU is 300 W, and the power connector is 1x 8-pin.

Q: What are the average benchmark score and percentile?

A: The average benchmark score is 30109, and the percentile vs all GPUs is 73.

Q: How does the RTX 5060 compare to the AMD Radeon RX 7700 XT?

A: The RX 7700 XT has an avgScore of 30097 and a deltaPct of 0, so the cards are tied in aggregate average benchmark score.

Q: What APIs and display outputs are supported?

A: It supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b.