I'd love to hear what you think! Please drop me a line and let me know what you like and what could be better. 🙏
NVIDIA GeForce RTX 3060 12GB Specs, Benchmarks & Pricing
The NVIDIA GeForce RTX 3060 is a mainstream gaming graphics card released in February 2021, featuring the Ampere architecture based on the GA106 GPU. It offers 3,584 CUDA cores, 112 3rd-generation tensor cores, and 12GB of GDDR6 memory running at 15 Gbps, delivering 101.90 INT8 TOPS (dense) for AI workloads. NVIDIA officially specified the card at 13 shader-TFLOPS, 25 RT-TFLOPS for ray tracing, and 101 tensor-TFLOPS for DLSS acceleration. With a TDP of 170W, it provides excellent performance-per-watt for gaming at 1080p/1440p and machine learning inference workloads. The RTX 3060 uses a PCI-Express 4.0 x16 interface and features 360 GB/s memory bandwidth on a 192-bit bus. At launch, it was positioned as the successor to the popular GTX 1060, offering twice the raster performance and 10x the ray-tracing performance at a $329 price point.
- Release Date: February 25, 2021
- MSRP: $329 USD
- GPU Architecture: Ampere
- Hardware-Accelerated GEMM Operations:FP64 FP32 TF32 BF16 FP16 FP8 FP4 INT8 INT4
- CUDA Compute Capability : 8.6
Considerations
- Lower FP32 compute performance compared to other GPUs
- Lower tensor core count compared to other GPUs
- Below average 3DMark Wild Life Extreme (4K) performance
- Below average 3DMark Wild Life Extreme (1440p) performance
Specifications for NVIDIA GeForce RTX 3060
| Specification | Performance Ranking |
|---|---|
| FP32 TFLOPs | |
| FP16 TFLOPs | |
| Tensor Core Count | |
| Memory Capacity (GB) | |
| Memory Bandwidth (GB/s) | |
| Int8 TOPs | |
| FP8 TFLOPs | Data not available |
| FP4 TFLOPs | Data not available |
Gaming Benchmarks for NVIDIA GeForce RTX 3060
Rankings based on 700,687 real-world benchmark results from OpenBenchmarking.org
| Benchmark | Performance Ranking |
|---|---|
| 3DMark Wild Life Extreme (4K) | |
| 3DMark Wild Life Extreme (1440p) | |
| 3DMark Wild Life Extreme (1080p) | |
| FurMark (GL) (4K) | |
| FurMark (GL) (1440p) | |
| FurMark (GL) (1080p) | |
| FurMark (VK) (4K) | |
| FurMark (VK) (1440p) | |
| FurMark (VK) (1080p) | |
| Quake II RTX (4K) | |
| Quake II RTX (1440p) | |
| Quake II RTX (1080p) |
Real-time NVIDIA GeForce RTX 3060 GPU Prices
We're tracking 80 of the NVIDIA GeForce RTX 3060 GPUs currently available for sale. The lowest average price is $328 (what's this?)
Buy NowCompare Price/Performance to other GPUs
We track real-time prices of other GPUs too so that you can compare the price/performance of the NVIDIA GeForce RTX 3060 GPU to other GPUs.
Compare GPU Price/PerformanceCompare NVIDIA GeForce RTX 3060 to Another GPU
Compare the NVIDIA GeForce RTX 3060 directly to another GPU to see specs, benchmarks, and prices side-by-side.
Compare GPUs Side-by-SidePrice History
NVIDIA GeForce RTX 3060 Price History
This chart tracks NVIDIA GeForce RTX 3060 prices over the past 6 months based on eBay listings. The yellow line shows the average of the three lowest-priced listings each day—a useful indicator for finding deals.
Trend: Over this period, the lowest average price has been up 9.3%. The current lowest average is $297, compared to a typical $269 over the period. The lowest average price is calculated as the average of the three lowest-priced listings each day.
gpupoet.com
Product Identifiers
Available from 7 Partners (15 products)
ASUS
- ROG Strix GeForce RTX 3060 OC Edition 12GB GDDR6
- ROG-STRIX-RTX3060-O12G-GAMING(model number)
- ROG Strix GeForce RTX 3060 V2 OC Edition 12GB GDDR6
- ROG-STRIX-RTX3060-O12G-V2-GAMING(model number)
- TUF Gaming GeForce RTX 3060 OC Edition 12GB GDDR6
- TUF-RTX3060-O12G-GAMING(model number)
- Dual GeForce RTX 3060 V2 OC Edition 12GB GDDR6
- DUAL-RTX3060-O12G-V2(model number)
Gigabyte
- GeForce RTX 3060 GAMING OC 12G
- GV-N3060GAMING OC-12GD(model number)
- GeForce RTX 3060 EAGLE OC 12G
- GV-N3060EAGLE OC-12GD(model number)
- GeForce RTX 3060 VISION OC 12G
- GV-N3060VISION OC-12G(model number)
Zotac
- ZOTAC Gaming GeForce RTX 3060 Twin Edge OC
- ZT-A30600H-10M(model number)
- ZOTAC Gaming GeForce RTX 3060 Twin Edge
- ZT-A30600E-10M(model number)
- ZOTAC Gaming GeForce RTX 3060
- ZT-A30600P-10M(model number)
PNY
- XLR8 Gaming GeForce RTX 3060 REVEL EPIC-X RGB Dual Fan
- VCG306012DFXPPB(model number)
- XLR8 Gaming GeForce RTX 3060 REVEL EPIC-X RGB Single Fan
- VCG306012SFXPPB(model number)
EVGA
- GeForce RTX 3060 XC GAMING
- 12G-P5-3657-KR(model number)
Palit
- GeForce RTX 3060 Dual
- NE63060019K9-190AD(model number)
Gainward
- GeForce RTX 3060 Ghost
- NE63060019K9-190AU(model number)
References
- https://nvidianews.nvidia.com/news/nvidia-introduces-geforce-rtx-3060-next-generation-of-the-worlds-most-popular-gpu
- https://www.nvidia.com/en-us/geforce/news/geforce-rtx-3060/
- https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3060-3060ti/
- https://www.techpowerup.com/gpu-specs/geforce-rtx-3060-12-gb.c3682
- https://www.waredb.com/processor/nvidia-geforce-rtx-3060
- https://rog.asus.com/graphics-cards/graphics-cards/rog-strix/rog-strix-rtx3060-o12g-gaming-model/
- https://www.asus.com/us/motherboards-components/graphics-cards/dual/dual-rtx3060-o12g-v2/
- https://www.msi.com/Graphics-Card/GeForce-RTX-3060-VENTUS-2X-12G-OC
- https://www.gigabyte.com/Graphics-Card/GV-N3060GAMING-OC-12GD
- https://www.zotac.com/us/product/graphics_card/zotac-gaming-geforce-rtx-3060-twin-edge-oc
- https://www.pny.com/geforce-rtx-3060-12gb-xlr8-revel-epic-x-rgb-df
- https://forums.developer.nvidia.com/t/what-is-the-fp64-rpeak-value-for-a5000/196056
Notes
- FP16 TFLOPS of 101.9 represents Peak Sparse Tensor Core performance (8x FP32). Dense Tensor performance is ~51 TFLOPS.
- int8TOPS of 101.90 represents non-sparse (dense) INT8 performance, derived from sparse INT8 of 203.80 TOPS ÷ 2 per NVIDIA Ampere architecture whitepaper. With 2:4 structured sparsity, INT8 performance is 203.80 TOPS. Dense value used for consistent cross-vendor comparison.
- Note: Previous versions incorrectly listed dense performance (50.95 TFLOPS) as sparse. Updated to match Workstation and Ada generation metrics (101.9 TFLOPS).
- fp32TFLOPS calculated as 3,584 CUDA cores × 1.777 GHz boost clock × 2 ops/cycle = 12.74 TFLOPS. Ampere architecture features 2 FP32 ops per cycle through dedicated FP32 datapath and shared INT32/FP32 datapath.
- NVIDIA officially specified the RTX 3060 at 13 shader-TFLOPS (FP32), 25 RT-TFLOPS (ray tracing), and 101 tensor-TFLOPS (general tensor performance for DLSS).
- Memory bandwidth calculated as 192-bit interface × 15 Gbps = 360 GB/s.
- The RTX 3060 uses the GA106-300-A1 variant of the GA106 chip with 28 of 30 streaming multiprocessors enabled (28 SMs × 4 tensor cores/SM = 112 tensor cores).
- Process node is Samsung 8nm with approximately 12 billion transistors and die area of 276 mm².
- TDP of 170W is the official NVIDIA specification, with recommended PSU of 550W.
- Supports CUDA Compute Capability 8.6, same as other Ampere GeForce 30-series GPUs.
- At launch in February 2021, the RTX 3060 was positioned to succeed the GTX 1060 with 2x raster and 10x ray-tracing performance at $329 MSRP.
- MSRP of $329 USD sourced from NVIDIA official product announcement at https://www.nvidia.com/en-us/geforce/news/geforce-rtx-3060/
- No NVIDIA Founders Edition was released for the RTX 3060 (only RTX 3060 Ti had a Founders Edition). All RTX 3060 cards are from AIB partners.
- Third-party product identifiers sourced from official manufacturer websites, TechPowerUp GPU Database, Newegg, and Best Buy product listings.
- Phase 5 audit (2026-09-02): FP32 and FP64 removed from supportedHardwareOperations. FP32 is executed on CUDA/shader ALUs, not accelerated through a standalone Tensor Core path on Ampere (only the TF32 reduced-precision format is Tensor Core-accelerated) — FP32 continues to be reported separately via the fp32TFLOPS field. FP64 removed because the RTX 3060 uses the GA106 die (confirmed in the summary above), not GA100; per an NVIDIA staff reply on NVIDIA Developer Forums, 'The Tensor core units in the GA102 chip architecture do not contain the FP64 path that is present in e.g. the Tensor core units in the A100 GPU' (https://forums.developer.nvidia.com/t/what-is-the-fp64-rpeak-value-for-a5000/196056) — the same reasoning extends to the smaller GA106 die, which shares the same 3rd-gen Tensor Core design as GA102. Canonical GA10x Ampere baseline (TF32, BF16, FP16, INT8, INT4) applied per .claude/agents/gpu-researcher.md; FP64 Tensor Core support is reserved for GA100-based SKUs (A100, A30) only. gpuArchitecture was already correctly cased as "Ampere".
- fp8TFLOPS and fp4TFLOPS are left null (field omitted) and not added to unverifiedSpecs — this is a confirmed hardware negative, not a research gap. The RTX 3060's Ampere 3rd-generation Tensor Cores predate both FP8 (introduced in Ada Lovelace/Hopper) and FP4 (introduced in Blackwell) tensor formats, per the canonical architecture table in .claude/agents/gpu-researcher.md.