Conclusion
In today’s performance comparison GeForce RTX 4060 Ti vs GeForce RTX 2060 SUPER review, we have pit head-to-head NVIDIA’s midrange 2019 GPU vs NVIDIA’s midrange 2023 GPU at the same $399 MSRP.
The GeForce RTX 2060 SUPER was launched in the summer of 2019 as a refresh to the GeForce RTX 20 Turing series of GPUs. It is based on the 12nm process, Turing architecture, has 2,176 CUDA Cores, 32 RT Cores, and 272 Tensor Cores with 64 ROPs, and runs at a Boost Clock of 1650MHz. It has 8GB of GDDR6 on a 256-bit memory bus at 448GB/s of bandwidth with PCIe 3.0 x16 interface and 175W TDP.
By contrast, the new GeForce RTX 4060 Ti was launched in May of 2023 filling out the GeForce RTX 40 Ada Lovelace series of GPUs. It is based on the TSMC 4N process and has 4,352 CUDA Cores, 34 3rd Gen RT Cores, and 136 4th Gen Tensor Cores with 64 ROPs. It also has 8GB of GDDR6, but on a 128-bit memory bus at 288GB/s of bandwidth and a PCIe 4.0 x8 interface and 160W TDP.
Performance Summary
At 1080p, in Returnal the GeForce RTX 4060 Ti was 43% faster than the GeForce RTX 2060 SUPER. In Forspoken the GeForce RTX 4060 Ti was 53% faster than the GeForce RTX 2060 SUPER. In The Callisto Protocol the GeForce RTX 4060 Ti was 58% faster than the GeForce RTX 2060 SUPER. In Call of Duty Modern Warfare II the GeForce RTX 4060 Ti was 81% faster than the GeForce RTX 2060 SUPER. In Chernobylite Enhanced Edition the GeForce RTX 4060 Ti was 60% faster than the GeForce RTX 2060 SUPER. In Dying Light 2 the GeForce RTX 4060 Ti was 55% faster than the GeForce RTX 2060 SUPER. In Cyberpunk 2077 the GeForce RTX 4060 Ti was 47% faster than the GeForce RTX 2060 SUPER.
At 1080p with Ray Tracing enabled, in Returnal the GeForce RTX 4060 Ti was 55% faster than the GeForce RTX 2060 SUPER. In Forspoken the GeForce RTX 4060 Ti was 36% faster than the GeForce RTX 2060 SUPER. In The Callisto Protocol the GeForce RTX 4060 Ti was 95% faster than the GeForce RTX 2060 SUPER. In Chernobylite Enhanced Edition the GeForce RTX 4060 Ti was 126% faster than the GeForce RTX 2060 SUPER. In Dying Light 2 the GeForce RTX 4060 Ti was 82% faster than the GeForce RTX 2060 SUPER. In Cyberpunk 2077 the GeForce RTX 4060 Ti was 92% faster than the GeForce RTX 2060 SUPER.
At 1440p, in Returnal the GeForce RTX 4060 Ti was 46% faster than the GeForce RTX 2060 SUPER. In Forspoken the GeForce RTX 4060 Ti was 39% faster than the GeForce RTX 2060 SUPER. In The Callisto Protocol the GeForce RTX 4060 Ti was 58% faster than the GeForce RTX 2060 SUPER. In Call of Duty Modern Warfare II the GeForce RTX 4060 Ti was 84% faster than the GeForce RTX 2060 SUPER. In Chernobylite Enhanced Edition the GeForce RTX 4060 Ti was 53% faster than the GeForce RTX 2060 SUPER. In Dying Light 2 the GeForce RTX 4060 Ti was 52% faster than the GeForce RTX 2060 SUPER. In Cyberpunk 2077 the GeForce RTX 4060 Ti was 48% faster than the GeForce RTX 2060 SUPER.
With DLSS performance, we did find that the GeForce RTX 4060 Ti gained more performance versus the GeForce RTX 2060 SUPER with DLSS. It is clear that the DLSS performance gains are higher with the GeForce RTX 4060 Ti versus the older GeForce RTX 2060 SUPER in games.
The Pros and the Cons
PCI-Express
The GeForce RTX 4060 Ti utilizes a PCI-Express 4.0 interface, while the GeForce RTX 2060 SUPER utilizes a PCI-Express 3.0 interface. However, the GeForce RTX 2060 SUPER uses a full 16-lanes (x16) interface, while the GeForce RTX 4060 Ti uses an 8-lanes (x8) interface. This technically gives both video cards the exact same PCI-Express bandwidth of 16 GB/s. However, this is only true if the GeForce RTX 4060 Ti is running on a PCI-Express 4.0 capable motherboard. Whereas the GeForce RTX 2060 SUPER will run at 16 GB/s on either a PCIe 3.0 or 4.0 motherboard, the GeForce RTX 4060 Ti will not.
On a PCI-Express 3.0-only capable motherboard, the GeForce RTX 4060 Ti will run at a slower 8 GB/s of bandwidth due to the PCIe 3.0 bus at x8. This matters because it is more likely that gamers already running a GeForce RTX 2060 SUPER that want to upgrade, are more than likely also only running a PCI-Express 3.0 only motherboard, and not a PCI-Express 4.0 capable one. If this is the case, the GeForce RTX 4060 Ti will actually be a downgrade in specific terms of PCI-Express bandwidth on those motherboards. However, if you have a PCI-Express 4.0 capable motherboard and graphics slot, then this is not an issue at all, and the PCI-Express is equal to the RTX 2060 SUPER, it is no better, but at least it isn’t worse, on PCIe 4.0 motherboards.
Overall, even with a downgrade in PCI-Express bandwidth, this shouldn’t affect gaming performance to any noticeable degree on this level of performance. It’s just a bit underwhelming that after 4 years, and upgrading two generations, the PCI-Express bandwidth is only equal to a four-year-old video card.
VRAM
The GeForce RTX 2060 SUPER has 8GB of GDDR6 VRAM on board. This amount of VRAM capacity was perfectly fine for this 2019, four-year-old video card from two generations past. While this was perfectly fine for its time, and nothing odd at all about it, the fact that after two generations and four years, the replacement at the same price point in the GeForce RTX 4060 Ti also has the same 8GB of GDDR6 VRAM capacity.
You will not be gaining any VRAM capacity after two generations, and 4 years as an upgrade at this same price point. This is another underwhelming factor about the GeForce RTX 4060 Ti when you compare it as an upgrade coming from a video card from 2019. In 2023, a VRAM capacity of 12GB at the $399 price point, would be a more desirable upgrade path. Therefore, if you upgrade from the GeForce RTX 2060 SUPER, you are not getting a VRAM upgrade.
The new GeForce RTX 4060 Ti does provide an upgrade in L2 cache, versus the GeForce RTX 2060 SUPER thanks to the new Ada Lovelace architecture. The GeForce RTX 2060 SUPER had a small 4MB of L2 cache on board. The new GeForce RTX 4060 Ti upgrades that cache by quite a large leap, with 32MB of L2 cache on board. This is a much larger leap in L2 cache than we are used to seeing from generation to generation. Even the GeForce RTX 3060 Ti only had 4MB as well, the same as the GeForce RTX 2060 SUPER.
This much larger L2 cache does help with gaming performance, allows the GPU to have fewer cache misses, and gives it room to expand more data before needing to hit the VRAM. However, it can only go so far, and ultimately VRAM is still very important for game smoothness, texture loading, pop-in, and geometry.
Ray Tracing
The new GeForce RTX 4060 Ti provides a very substantial gain in ray tracing performance, and tensor core performance, and accuracy, despite the actual number of cores remaining the same or lower. The GeForce RTX 2060 SUPER and GeForce RTX 4060 Ti both have 34 Ray Tracing cores, so there is no upgrade in the number of cores. However, what makes them differ is in the generations. The GeForce RTX 2060 SUPER uses the first generation of RT cores ever introduced in the RTX lineup. The GeForce RTX 4060 Ti uses newer third-generation RT cores, with much improved efficiency and performance. While the core count is the same and not an upgrade, the sheer performance you get with each core has gone up.
Tensor Cores
The same is also true for the Tensor Cores. The GeForce RTX 2060 SUPER actually had more Tensor Cores at 272 compared to the GeForce RTX 4060 Ti which has 136. The GeForce RTX 2060 SUPER uses the second generation Tensor Core, while the GeForce RTX 4060 Ti uses the fourth generation Tensor Cores. In addition to higher Tensor Core performance, per core, accuracy has also improved greatly on the fourth generation Tensor Cores which allow the GeForce RTX 4060 Ti to have usable Optical Flow acceleration which allows DLSS 3 Frame Generation, a feature the RTX 2060 SUPER cannot do.
Architectures
The architectures between the GeForce RTX 2060 SUPER (Turing) and GeForce RTX 4060 Ti (Ada Lovelace) are also much upgraded. The newer Ada Lovelace architecture on the GeForce RTX 4060 Ti is much more efficient, faster, and power friendly. The GeForce RTX 2060 SUPER has 2,176 CUDA Cores (shading units), and the GeForce RTX 4060 Ti upgrades that quite a bit by literally doubling it to 4,352 CUDA Cores or shading units.
The GeForce RTX 4060 Ti also adds AV1 decoding and encoding support, something the RTX 2060 SUPER did not have. The GeForce RTX 4060 Ti also has a power advantage, with a 160W TDP compared to the higher 175W TDP on the GeForce RTX 2060 SUPER. The GeForce RTX 4060 Ti is, therefore, faster, but also uses less power, so that is an upgrade.
Final Points
Overall, if we take an average of performance at 1080p without Ray Tracing, the GeForce RTX 4060 Ti (out of the games we tested) is on average 57% of a performance uplift versus the GeForce RTX 2060 SUPER. At 1080p with Ray Tracing enabled the GeForce RTX 4060 Ti is on average 81% of a performance uplift versus the GeForce RTX 2060 SUPER. At 1440p the GeForce RTX 4060 Ti is on average 54% of an uplift versus the GeForce RTX 2060 SUPER.
With this testing, we can conclude that you will see the most benefit at 1080p with the GeForce RTX 4060 Ti providing a performance uplift versus the GeForce RTX 2060 SUPER, with generally higher performance differences. We can also conclude that Ray Tracing performance is a big gain over the GeForce RTX 2060 SUPER, and you will see usable and playable gameplay experiences with the GeForce RTX 4060 Ti versus the GeForce RTX 2060 SUPER. At 1440p, the GeForce RTX 4060 Ti does allow some games to be playable whereas the GeForce RTX 2060 SUPER is not. We can also say that DLSS performance is overall faster on the GeForce RTX 4060 Ti, and you will see bigger gains by enabling DLSS on the GeForce RTX 4060 Ti versus the GeForce RTX 2060 SUPER.
Overall, if you are upgrading from the GeForce RTX 2060 SUPER, you should have a better gameplay experience in 2023.
Discussion (6 replies)
Join Discussion →Love the review and thank you.
Doing some numbers something stood out to me and it REALLY shows when you compare the target max MHZ between the two cards.
the 4060 is pushing 2535 and the 2070 is pushing 1650, This is a raw difference clock per clock of 53.6.
When we are NOT doing RT work the % of improvement for games at 1080p averages out to 45.3%. This leads me to believe most actual improvement to performance and efficiency is through process node efficiency and not redesign or IPC improvement of the actual rasterization cores. But when you take into account that they ALSO doubled the cuda cores to reach that number... is even more interesting.
This is weird... They doubled the cuda core count
Increased peak MHZ by 53.6%
Nicely brought down the power envelope.
But for DOUBLE the cuda cores running at just over 50% the mhz rating... if the IPC remained the same I would have expected a more linear progression in performance.
So it seems like they DID make some changes to the cuda cores themselves, for efficiency but had to claw back performance through more cores and more mhz taking away some of that efficiency gain. Actually theoretically 100% of the potential gain considering the real performance improvement is more in line with the MHZ as opposed to the DOUBLED cuda core count.
Something does not smell right here with these numbers. We're missing a part of the story. The end user is getting lesser performant cores core to core... if the had left the core counts equal and counted on only generational improvements (not including DLSS I didn't crunch those numbers) a 2070 to 4060 would have been a rasterization downgrade.
I need to run similar numbers for AMD's cards in a generational comparison.
Would have been interesting to include these 3 cards also in the comparison & make it a 5 way battle
2060 12gb
3060 12gb
4060 ti 16gb
Idea is to identify bottlenecks with 8gb vram
For example in Ratchet & Clank in 1080p ultra (RT off) the 2060 12gb almost matches the 8gb 4060 ti !!!
Search engine optimization rules the day - will probably focus on more 1:1 matchups for now. The aim here is to look at upgrade scenarios....
Great writeup!
Grim brings up a good point - the 4060 shows about a 50% improvement in performance, but that 50% increase in clock speed ~and~ core count - the math doesn't seem to add up without drawing the conclusion that you just need to pack more slower (yet more efficient) cores to make Turing work. It's an interesting thing to note, if nothing else. Maybe that's why JSH is saying Moore's law is dead...?
Also - for ~most~ people (not enthusiasts here who chase the absolute top of the performance curve) - I usually recommend they upgrade to something that doubles their existing GPU performance. In the past that was every 2-3 generations, at roughly the same tier/price they had previously paid. Now, starting with RTX, that has been thrown out the window.
nVidia obviously wants you to pay a premium for Raytracing performance. I've never thought it was worth it, but to those that do - it's your money, buy what you like. But for folks that have cards from Pascal and newer, it's been hard to recommend anything to upgrade to without utterly blowing through budgets or accepting much smaller (rasterizing) performance gains.
Thanks, @Brent_Justice for the great comparative write-up. It's bummer about the PCIe and VRAM details but otherwise a solid upgrade in keeping at the exact same price point.
I had gotten a 2080 SUPER back then as a slight upgrade to a 1080 Ti and also so I could enjoy using G-Sync with an LG C9. While mostly a sideways move it had some impressive improvements for the day with RT and DLSS when gaming at 1440p. It even cost me somewhere around $50-100 less than what I paid for the 1080 Ti and only needed 2x 8-pin connectors vs 3x 8-pin for the 1080 Ti. I always cringed at the "SUPER" name but otherwise really had a good experience with that card.
Very nice review, thank you. I've been looking to get a lower tier card at some point for a back up rig again, and this review does help.