If you’re considering some of the new Ada options, you might want to see how it compares to the old. So we look at the RTX 4060 Ti vs RTX 3080 Ti vs RTX 3080 and how some of these best graphics cards compare to each other.

These Nvidia GeForce RTX GPUs range in a variety of ways and are on different levels of models. But the generational improvement might diminish that difference. So first we see when they launched. And there we have the RTX 3080 first, coming with the Lovelace generation in September 2020. With the upgraded 12GB model coming out in January 2022. Whilst the 3080 Ti came in between in June 2021. Most recently, the newer 40-series has been developing, with the RTX 4060 Ti now available to buy since May 2023.

There is a lot of variation, but only the new 4060 Ti is on a different generation that will provide the biggest changes. But with that, we look at what those entail and what you might expect from them.

RTX 4060 Ti vs RTX 3080 Ti vs RTX 3080 architecture

There is just one generational gap between the 40 series and the 30 series cards. There are some features limited to the new option, like Nvidia DLSS 3 and frame generation, so the performance boost available can improve how much the 40 cards can run and possibly bring it closer to what the others can do. But it is still limited by the games that support this boost and what specs it features.

For the original card, the RTX 3080 is built on the GA102-200 GPU variant. Made with Samsungs 8nm process, it has 28.3 billion transistors on the 628mm² die size. Whereas the 3080 12GB, is on the GA102-220 and then the RTX 3080 Ti has the GA102-225 GPU variant and both have the same die size and transistor count. And finally, the 4060 Ti has the AD106-350/351 processor based on the Ada Lovelace architecture, with the variation only by name for the 8/16Gb models. Made with TSMCs 5nm process with a die size of 190mm² with a yet unknown number of transistors.

Specifications

	RTX 4060 Ti (8/16GB)	RTX 3080 (10/12GB)	RTX 3080 Ti
GPU	AD106-350/351	GA102-200/220	GA102-225
GPU process	TSMC 5nm	Samsung 8nm	Samsung 8nm
CUDA cores	4,352	8,704/8,960	10,240
Base clock	2.31 GHz	1.44/1.26 GHz	1.37 GHz
Boost clock	2.54 GHz	1.71/1.71 GHz	1.67 GHz
VRAM	8/16GB GDDR6	10/12GB GDDR6X	12GB GDDR6X
Memory clock/bandwidth	18 Gbps / 288 GB/s	19Gbps / 760.3/912.4GB/s	19 Gbps / 912.4 GB/s
Memory interface	128-bit	320/384-bit	384-bit
TDP/TGP	160/165W	320/350W	350W

These aren’t the whole specifications, but the key ones for the difference found between the cards. Offering up the reason for the difference found between the performance of the cards and what they can achieve.

Frequency

Clock speed or frequency is the rate at which the GPU processor works at. The silicon die is able to process those instructions at different speeds. And you can see that with the new generation, the speeds have increased. With the boost clocks reaching over 800 MHz faster than the previous offerings were capable of. Custom models are also capable of reaching higher speeds, as well as the ability to overclock the cards yourself and achieve those greater boost clock speeds.

VRAM

The video memory of a graphics card is the buffer for the pixels of graphics. This processing allows the higher resolutions to work faster and more efficiently. As the higher memory bandwidth and capacity allow the memory configuration to thrive when it comes to the larger pixel counts of the larger resolutions. So if you’re after 4k gaming, you do need one able to achieve it and so the lower bandwidth of the 4060 Ti will hold back what it can do compared to the 3080s since they are over 900 GB/s each compared to 288 of the 60 model.

Processor cores

now when it comes to what the GPU processor holds within it, the transistors will be arranged in a certain way to groups of core counts. Such as the CUDA cores that work on the processing, RT cores for ray tracing processing, or the rest like TMUs, ROPS and such. The count of these will affect what these cards can achieve.

Between the RTX 4060 Ti vs RTX 3080 Ti vs RTX 3080, the generational difference does mean the efficiency and speed of the difference do vary. But the 80 models are over double in CUDA count over the 4060 Ti option. This means they will be much better at getting through those instructions and processing them better.

Power

What used to be TDP is now also called TGP, the thermal design power or total graphics power. Which gives you an indication of the power that the GPU will be using and how hot it may run. And between these options, we see the 80 models over double the 4060 Ti. As such they require more power to run and a larger power supply as well as running hotter in general. So running the lower-powered card will run cooler and less expensive.

How much cheaper is the 4060 Ti than the 3080 or 3080 Ti?

The 4060 Ti has the same launch price tag as its predecessor the RTX 3060 Ti, at least on the 8GB model. So it has an MSRP of $399 and $499 respectively. Which is $300/200 and $400/300 cheaper than the RTX 3080 10/12GB model MSRP. Whilst compared to the Ti, it is $800/700 cheaper. But you’re likely losing out on a lot of performance from the cheaper option.

Since the 30-series is the older series though, you can also shop around or even consider second-hand if you want to save some money. However, the RTX 4060 Ti will still deliver decent value for money, though not worth an upgrade if you already have a 30-series GPU.

RTX 4060 Ti vs RTX 3080 Ti vs RTX 3080 performance

The RTX 4060 Ti has arrived, which means some important benchmarks regarding its performance. Prior to release, Nvidia released a bunch of results, but this was focused around the DLSS 3.0 and ray tracing improvements, making it harder to gauge the raw performance on board. Now we can see it performs similar to the 3070 as expected, so less powerful than both the 3080 Ti and 3080. The TechSpot review shows that the 8GB RTX 4060 Ti 15 game average FPS is 111 at 1080p, 78 at 1440p, and 38 at 4K in modern titles.

For the 3080’s performance, we again look at the performance as seen in the TechSpot review. There we see the RTX 3080 Ti achieve an average of 200 FPS at 1080p, 158 at 1440p, and 95 at 4k. The 3080 10/12GB run at 190/197, 146/153, and 86/91 FPS. So, the 3080 Ti enjoys up to double the frames of the 4060 Ti, and the base 3080 is not too far behind. This is more clearly evident at 4K.

Is the 4060 Ti worth it over the 3080 or 3080 Ti?

Although it is possible the 4060 Ti can reach 60 at 4K in certain titles, or fully utilizing DLSS 3, it’s unlikely to reach that same level as the 30 cards. But it does come at a much lower price and so if you don’t need that higher performance and it is sufficient for your needs when it comes out, it will be likely worth it with what it can do on its own. At lower resolutions like 1440p, and especially 1080p, the 4060 Ti shows its worth. If you’re only going to be gaming at 1080p, there is no need for a RTX 3080 / 3080 Ti, as the 4060 Ti will be great for what you need, even in demanding moderns titles.

Additionally, you need to consider the rest of your current build and power consumption. The RTX 4060 Ti is the vastly more efficient GPU, with a TDP of more than half the high-end options. This means you can afford to upgrade with a less powerful PSU for example, which is great for those that can’t afford to upgrade more than one component.

ASUS ROG Strix LC RTX 3080 Ti OC

Pros

Excellent cooling solution for performance and longevity

Top performance out of choices

Best option for overclocking

Cons

Most expensive choice

More complicated to install, requiring more space

Shop on Amazon

CHECK PRICE

EVGA GeForce RTX 3080 XC3 Ultra Gaming

Pros

Very good build quality

Attractive design

Displays excellent value

Tasteful RGB