The NVIDIA RTX 3060 Ti and RTX 2080 Ti graphics cards represent two powerful but distinct generations of GPU tech.
The RTX 2080 Ti was the very best Turing card NVIDIA offered. It came at an extremely high price point but delivered unmatched performance for over two years after launch.
Meanwhile, the more affordable RTX 3060 Ti showcases major leaps with NVIDIA‘s new Ampere architecture. It stands as a fantastic 1440p gaming solution.
How do these GPUs stack up against each other though? Let‘s dive into a full comparison of specs, features, and real-world game benchmarks.
RTX 2080 Ti vs. 3060 Ti Specs
First, let‘s directly compare the key specifications between the RTX 2080 Ti and RTX 3060 Ti:
| **Specification** | **RTX 2080 Ti** | **RTX 3060 Ti** |
| CUDA Cores | 4,352 | 4,864 |
| Tensor Cores | 544 | 152 |
| RT Cores | 68 | 38 |
| GPU Base Clock | 1,350 MHz | 1,410 MHz |
| GPU Boost Clock | 1,545 MHz | 1,665 MHz |
| Memory Capacity | 11 GB GDDR6 | 8 GB GDDR6 |
| Memory Speed | 14 Gbps | 14 Gbps |
| Memory Interface | 352-bit | 256-bit |
| TDP | 250W | 200W |
| Manufacturing Process | 12 nm | 8 nm |
| Architecture | Turing | Ampere |
| Launch Date | Q3 2018 | Q4 2020 |
Right away the generational differences become obvious.
The RTX 3060 Ti is built on a smaller, more efficient 8nm manufacturing process. And it utilizes NVIDIA‘s latest graphics architecture in Ampere, bringing both architectural and process refinements.
As far as raw specifications, the 3060 Ti actually wins out with a higher CUDA core count at 4,864 vs 4,352 on the 2080 Ti. These cores handle most shader workloads in games and creative apps.
However, the 2080 Ti still packs more dedicated ray tracing and tensor cores. These accelerate advanced workloads like ray traced lighting, DLSS, and AI-powered rendering features. The 3060 Ti closes much of the feature gap though with its 2nd-gen RT and 3rd-gen tensor cores.
The 3060 Ti also clocks slightly higher, further boosting its throughput advantage. But as we‘ll see soon, architectural differences have just as much impact on real-world speed.
First, let‘s better understand what makes these GPUs tick…
Turing vs. Ampere Architecture
The RTX 2080 Ti utilizes NVIDIA‘s Turing architecture first introduced alongside the RTX 20-series in 2018. Turing debuted ray tracing cores and tensor cores to hardware accelerate these advanced graphics features.
Turing also optimized the traditional CUDA shader pipeline. It achieves high clocks and introduces concurrent floating point and integer execution for improved throughput. The 2080 Ti pushes Turing to its limits as the fully unlocked TU102 GPU.
Ampere then builds on Turing‘s foundation with its RTX 30-series. 2nd-gen ray tracing cores double performance while also accelerating motion blur and mesh shading. The 3rd-gen tensor cores bring much faster FP32 and new FP16 data types for AI.
But the biggest change is a redesigned CUDA core. Ampere doubles FP32 throughput while greatly improving FP16 and INT32 performance too. Throughput scales much better with higher clocks as a result.
Higher clocks and improved efficiency also allow the 3060 Ti to achieve similar or faster speeds while drawing much less power at just 200W. That‘s 20% lower than the 2080 Ti and a big advantage.
Let‘s now put these specs to the test in games!
Performance Benchmarks: 1080p and 1440p
To represent real gaming performance, we‘ll look at benchmark results at both 1080p and 1440p resolutions. 1080p stresses the GPU more while 1440p better matches what these cards are designed for.
1080p benchmark data courtesy of Tom‘s Hardware
At 1080p, the RTX 2080 Ti still leads in average frame rates. Its advantage ranges from around 5% in Assassin‘s Creed Valhalla up to 34% in Rainbow Six Siege. Much of this comes thanks to its faster memory and more RT/tensor hardware.
However, the 3060 Ti is no slouch, trailing by less than 10% in most games tested. Considering it costs literally hundreds less, the performance-per-dollar value is outstanding. Plus in several games tested, the two GPUs are dead even.
Now at 1440p:
1440p results data from Guru3D
When pushing up to 1440p, most game engines become more bound by shader performance rather than memory bandwidth. Here the 3060 Ti and its extra CUDA cores shine.
The two GPUs are now evenly matched, trading blows in different games. The 3060 Ti wins in Horizon Zero Dawn, Assassin‘s Creed Valhalla, and Borderlands 3 while the 2080 Ti leads again in Rainbow Six.
For gaming, the RTX 3060 Ti effectively matches the former $1200 GPU king for $399. That‘s incredibly impressive.
Now what about content creation performance?
Content Creation Benchmarks
The RTX 2080 Ti and 3060 Ti are both capable creative workstation cards as well. NVIDIA‘s CUDA platform accelerates everything from video editing to 3D rendering and AI tools.
Here the 2080 Ti still holds a slight performance advantage thanks to its more built-out tensor core design:
Benchmark data courtesy of Puget Systems
The 2080 Ti is around 13% faster in CUDA GPU compute workloads on average. For video editing in Premiere Pro, that gap widens to 20% faster 4K render times.
Still, the 3060 Ti isn‘t slow by any means. And it‘s much more affordable. For burgeoning 3D modelers, video editors, or graphics professionals, it remains a fantastic choice that won‘t break the bank.
Now for a breakdown of other important considerations:
Ray Tracing and DLSS Support
The GeForce RTX series introduced dedicated hardware acceleration for real-time ray traced lighting and deep learning super sampling (DLSS).
While the RTX 2080 Ti was the first card with these features, its capabilities are now dated. The RTX 3060 Ti sports improved ray tracing (RT) cores alongside 3rd-gen tensor cores for DLSS:
Ray tracing performance sees the biggest uplift. The 3060 Ti‘s 2nd-gen RT cores double throughput to keep up with the demands of modern games. DLSS also gets better, using AI and tensor cores for smarter upscaling algorithms.
Ultimately, the 3060 Ti handles both ray traced games and DLSS much better in spite of some spec disadvantages. Performance is just as fast if not faster for these advanced effects.
Power Draw, Thermals, and Noise
We‘ve noted the RTX 3060 Ti‘s much lower 200W power rating compared to the 2080 Ti‘s 250W TDP. And that lower power draw directly leads to advantages in other areas:
Tester data from both Tom‘s Hardware and Guru3D show the 3060 Ti running a good 18-20°C cooler on average. That‘s with stock fan profiles.
And the 3060 Ti manages that while producing just 40 dBA of noise. That‘s very quiet, while the 2080 Ti hits 51-52 dBA under load. Your ears will thank you with the newer Ampere card.
For small form factor builds or quiet gaming rigs, power and thermals become especially critical. Here the 3060 Ti is much better suited.
Availability and Longevity
While this comparison is strictly technical, there‘s an important practical consideration around availability that impacts buying decisions:
The RTX 2080 Ti is now discontinued as NVIDIA shifts production fully to newer RTX 30-series cards. That means diminishing stock that won‘t ever replenish.
Of course the flip side is heavily discounted pricing while 2080 Ti inventory sells out. Prices now sit around $550 to $700 for models once over $1000.
Meanwhile the 3060 Ti just launched at the end of 2020. It‘s readily in stock around its $399 MSRP with prices even dropping below that at times.
And unlike the outgoing 2080 Ti, the current-gen 3060 Ti will see better long term support too. Game developers will optimize for Ampere over the next couple years. And NVIDIA continues releasing improved drivers and features for RTX 30-series owners.
If you don‘t need a GPU right away, the 3060 Ti is the smarter buy that will pay dividends down the road.
Performance Overview: Frame Rates and FPS Per Dollar
Let‘s summarize gaming performance with a couple more charts. First, average frame rates across our tested games:
Next, value calculated as FPS per dollar spent. Higher is better:
This perfectly captures the overall story. While the RTX 2080 Ti is marginally faster, the RTX 3060 Ti provides practically the same level of 1080p and 1440p gaming performance. All while costing literally hundreds less.
Dollar for dollar, you‘re getting over twice the frame rate value with Ampere. And with RTX 30-series prices continuing to improve, that value gap may widen further.
When to Choose the RTX 2080 Ti Over the 3060 Ti
We‘ve clearly demonstrated why the RTX 3060 Ti makes so much sense for the money versus even the former flagship 2080 Ti. But there are still a couple cases where it can pay to find a discounted 2080 Ti:
1. High Refresh Rate 4K Gaming – For gaming at 4K above 60 FPS, the 2080 Ti maintains enough of a throughput advantage to push higher frame rates. Especially in titles using its extra ray tracing and tensor hardware. If your primary goal is maxing graphics at high refresh 4K, a 2080 Ti still gets you closer right now.
2. Budget Professional Workstations – The 2080 Ti retains a performance advantage in GPU compute and creative applications. Someone building out a lower cost workstation for 3D modeling, video production, code compiling etc will benefit from more CUDA power today. Just make sure to find a decent discount.
Outside of those two scenarios, I generally recommend the much more affordable RTX 3060 Ti that narrowly trails the former $1200 king. It really does stand as an incredible value gaming card with better efficiency and future prospects too.
Final Thoughts
When first announced alongside Turing back in 2018, the RTX 2080 Ti stood completely unchallenged as the world‘s fastest gaming GPU. It carried an equally intimidating $999 price tag.
Two years later, NVIDIA sets a new standard for high-end value with the RTX 3060 Ti. It nearly measures up to the once mighty 2080 Ti across the board at less than half the cost.
Built on superior Ampere architecture, the 3060 Ti marks a new era where advanced graphics horsepower finally reaches mainstream buyers. This battle ultimately signals the passing of the torch to a more efficient, affordable generation of gaming hardware.
