Entering 2021, NVIDIA remains the king of graphics cards. While its competitors are coming close to matching both NVIDIA’s performance and name recall, the hype surrounding the company’s new releases are still second to none when it comes to GPUs. With the 3000 series, NVIDIA’s newest RTX release, customers might find that slightly older cards have become both more affordable and more available in online stores.
For those who want to have a PC setup strong enough to handle the highest settings on the highest resolutions without a big drop in framerate, the RTX series is the only choice. This quality, combined with the fact that it’s hard to get RTX 3000 series cards for a fair price at the moment, makes the 2000 series a very attractive option for consumers building a new PC. The question, then, is which one should you get? The RTX 2060 and its seemingly more powerful counterpart the RTX 2070 are some
NVIDIA 2060 vs 2070 Comparison Chart
|Model||GeForce RTX 2060||GeForce RTX 2070|
|Price||Check Price||Check Price|
|Release Date||Q1 2019||Q4 2018|
|Memory||6GB DDR6||8GB DDR6|
|GPU||Turing TU106||Turing TU106|
Both cards run on NVIDIA’s Turing microarchitecture.
Both of these chips are based on the TU106, meaning that both are using Turing microarchitecture developed by NVIDIA. This enables real-time hardware ray tracing in both chips, making either chip already one of the better choices for those who are serious about graphics. Both chips are built on the 12nm manufacturing process, with a total 10.8 billion transistors each and a die size of 445 sq. mm. Power consumption also does not differ much between the two, given that they run the same intense processes. The differences between the two lying mostly in how many cores of each kind there are in each model.
In terms of numbers, the 2070 predictably beats its lesser counterpart. It has more shader processors, texture mapping and render output units, ray tracer cores, and tensor cores. It also has more compute unified device architecture (or CUDA) cores, with around 400 more than the 2060’s 1920 CUDA cores. Quite noticeably, though, the 2060 records a higher boost clock. This does seem like an advantage on paper; however, the large number of conditions you have to consider for overclocking to be possible on the 2060 makes it almost impossible to unlock its full potential.
Ray Tracing and DLSS
The RTX 2000 series is a pioneer of real-time ray tracing technology.
A landmark of any RTX chip that differentiates it from NVIDIA’s other GPU lines is the ability for real-time hardware-accelerated ray tracing and deep learning supersampling (or DLSS). To clarify, all Turing-based chips, like the Ti variant of the GTX 1660, are able to deliver ray tracing. However, the RT cores present on these RTX chips accelerate it, making for a smoother ray tracing performance. Ray-traced lighting effects are becoming more and more common in video games and other related media, and as the demand grows for more photorealistic graphics, the more necessary these RT cores will become.
The tensor cores, on the other hand, are used for DLSS. This is a method that NVIDIA-powered computers use to create images that appear to be high-resolution, without having to push the graphics card’s performance too much. The tensor cores achieve this by relying on an algorithm trained to do so. DLSS also enables a computer to have better anti-aliasing, among other graphics features, resulting in an image that’s less rough and more realistic, while not being as resource-hungry as it would be without tensor cores. It is worth noting, though, that more than 2 years after the introduction of DLSS, it still remains more of a novelty than a necessity. Only a small number of games (for example, Minecraft) even have options for it to be used.
Minimal differences are seen between the two, except when higher resolutions are involved.
The most important point of comparison in any GPU is how it performs when put to intense work, like big production AAA games. Surprisingly enough, though there is a difference in performance between the two, the disparity isn’t large enough to definitively say that the 2060 is the worse choice. Multiple sources online have compared the two using several video games set at maximum settings and at 1440p. The average difference between the two chips is only 10%, meaning that the 2060 performed only 10% worse than its more expensive counterpart. It’s not unnoticeable by any means, as videos show obvious differences between the two. However, it’s not significant enough for the 2060 to be inferior in every aspect.
The difference is even smaller between the two chips if you’re a customer that isn’t super picky about having only the best graphics on the biggest of screens. While running on even higher graphics settings using smaller 1080p screens, the difference between the two chips is small enough to be unnoticeable, even when RTX is enabled.
Price and Availability
The 2060 is generally more affordable, and can offer you a bigger bang for your buck.
After considering that performance differences may not be as large as expected, the last big factor that distinguishes the two is at what price points can you find these chips and how available they are. Availability, thankfully, isn’t a huge issue, as both chips are available for purchase on major online platforms like Amazon. Interested customers may have a bit more trouble looking for an RTX 2070, since it has been superseded by the 2070 Super since Q3 2019, but it’s not exactly a rare find either.
The biggest difference between the two models actually lies in their price. The RTX 2070 is generally around a hundred dollars more expensive than the 2060. While recent discounts may make prices for both chips significantly lower, the difference in price remains and the 2060 still stands out as the more affordable option. Considering all the information given in the earlier sections, the 2070’s 10% increase in performance is accompanied by an increase in price of around 20 to 30 percent. For those who have some cash to spend but don’t want to burn it unwittingly, this comparison point could make or break a deal.
The marginal improvements make the 2060 the better choice for most customers.
The RTX 2060 stands out as the better choice between the two for most consumers. A lot of the technical improvements made to its more expensive counterpart only really benefit those whose lifestyles and livelihood revolve around smooth state-of-the-art graphics, while being mostly unnoticeable or even unusable on more humble machines. The large difference in price, then, makes the 2060 the better choice in terms of corresponding price-performance comparisons.
However, if you want better graphics than the 2060 can provide and have the equipment that can make the difference obvious, go for the RTX 2070. Its larger RAM and higher number of tensor and RT cores could make all the difference, depending on what type of content you create. Between the two chips, though, you really can’t go wrong with either one. It’s just a matter of choosing which one suits both your needs and your budget.
Depending on your situation, it could be. If you need ray tracing capabilities better than the 2060 can provide, then a 2070 or a 2080 will be worth the additional cash.
It’s always been NVIDIA’s strategy to occupy different price points. While it is cheaper than other RTX options, its quality doesn’t suffer. This means that it’s a solid choice for anyone who wants an RTX chip but isn’t willing to break the bank for one.
It all depends on what your budget and needs are. Both are good choices, but the 2060 is more economical and could provide more bang for your buck if you don’t really need the heavy lifting that the 2070 can do.
The RTX 2070 is definitely powerful enough to not need upgrading for at least 3 more years from the time of writing.
Last update on 2020-12-23 / Affiliate links / Images from Amazon Product Advertising API