Prices, Specs, and Everything Else We Know About Nvidia’s RTX 4060 and RTX 4070 GPUs

In what it’s claiming to be its biggest generation leap ever, Nvidia has already launched its flagship RTX 4090 and 4080 GPUs. However, with a price tag starting at $1,199, it’s safe to say most people are highly anticipating the budget 40-Series options: the RTX 4070 and RTX 4060.

With RTX 4070 and RTX 4060 releases slated for Q2 and Q3 2023, there are rumors, leaks, and speculation about the two new GPUs. How well do these new GPUs stack up against their predecessors? How much will the RTX 4070 and RTX 4060 cost? And are they a good deal? Here’s what we know about the Nvidia RTX 4070 and RTX 4060​​​​​​.

1. RTX 4070 and RTX 4060: Release Date, Pricing, and Availability

According to hardware leaker hongxing2020, the RTX 4070 will reportedly hit on-shelf availability on April 13, 2023. Furthermore, the RTX 4070 launch will come in two parts, with Nvidia expected to officially announce the card on March 12, which is also the review embargo for all MSRP RTX 4070 cards. Non-MSRP model tests will be released one month later, on April 13.

As for the RTX 4060, we’ll likely see a 4060 Ti first before the more budget-oriented non-Ti version comes around. While there’s no word on the release date of the RTX 4060 Ti yet, you can expect the official announcement in a couple of weeks, meaning we should have both the RTX 4060 models on the market by the end of May.

Considering the launch prices for RTX 3070 ($499) and RTX 3060 ($329), we expected the newer-gen RTX 4070 to be just under $500. However, according to Moore’s Law Is Dead, a prolific leaker in the GPU space, Nvidia’s recommended MSRP for the RTX 4070 is $750, which means third-party cards from Nvidia’s partners could easily start at $800.

This pricing doesn’t make sense at all, especially considering that the RTX 4070 Ti is available for $799 with clearly better performance. It also indicates that the RTX 4060 Ti and non-Ti variants could also see major price hikes compared to their predecessors.

Nvidia does seem to be careful about the price hike. By splitting the embargo dates, the company is basically promoting cheaper custom models instead of factory-overclocked versions, which are generally priced higher. However, with such small differences in price between the RTX 4070 and 4070 Ti, the latter does seem better value for money.

2. RTX 4070 and RTX 4060: Specifications

The Nvidia RTX 4070 and RTX 4060 will use Nvidia’s next-generation Ada architecture, so you can expect some serious specifications, at least on paper. The RTX 4070 is expected to come with the AD104-250/251 GPU core.

This means we could see a card with 5,888 CUDA cores with 12 GB GDDR6X VRAM running at 21 Gpbs across a 192-bit bus. The card is reportedly clocked at 1,920 MHz and can go as high as 2,475 MHz when boosted. You can expect around 29 TFLOPs of computing power. There’s also a bigger L2 cache coming in at 36 MB, all for a memory bandwidth of 504.2 GB/s

As for the RTX 4060, the GPU will use the AD106-350-A1 GPU core. That means 3,840 CUDA cores and 8 GB GDDR6 memory running at 18 Gpbs across a 128-bit interface. So you can expect nearly 288 GB/s and an L2 memory cache of 24 MB. The base and boost clock speeds should be around 2,505MHz and 2,640MHz, respectively. Overall, you’re most likely looking at 20 TFLOPs of compute power.

As mentioned above, we will likely see an RTX 4060 Ti launch before the RTX 4060. If that’s the case, you’d be looking at a GPU with 4,352 CUDA cores running on the same AD106-350 GPU core. The base and boost clocks should be 2,100 MHz and 2,600 MHz, respectively, powering 8GB GDDR6 VRAM running at 18Gbps across a 128-bit bus with a total memory bandwidth of 288GB/s

The difference in computer power between the RTX 4060 and RTX 4060 Ti is only expected to be about 2 TFLOPs. These are very similar-looking GPUs on paper, but the AD106 core will be used to a higher potential in the RTX 4060 Ti, meaning better real-world performance.

3. RTX 4070 and RTX 4060: Power Requirements

Both the RTX 4070 and RTX 4060 will be using the new PCIe Gen5 12VHPWR connector as part of Nvidia’s move to standardize the new lineup. They’re not particularly power-hungry cards either, cutting down on roughly 20% of their predecessors’ TGP at 200W and 160W, respectively. However, it’s likely that the RTX 4060 could also demand up to 200W, according to some leaks.

That doesn’t mean you’ll need a new PSU for your shiny new RTX GPU, as old PSUs will work fine as long as you have enough power headroom. Ideally, 750W or above PSUs should be just fine. Similar to their more powerful siblings, these cards squeeze more performance while also being more efficient thanks to the Ada architecture.

4. RTX 4070 and RTX 4060: Performance

Leaks suggest that the RTX 4070 could compete directly with the RTX 3080 Ti and even the RTX 3090. Compared to its predecessor, the RTX 3070, the card managed a Time Spy Extreme score of 10,000, a jump of nearly 30% considering the RTX 3070 was hovering around 7,000.

The RTX 4060 could compete with the RTX 3070 and RTX 3060 Ti with an expected performance boost of around 20% compared to the RTX 3060.

Although that might not be as big of a leap as you’d expect, especially considering just how powerful the higher-end 40-series cards are, it’s still no big feat considering the card is actually running on less power and comes at more or less the same price.

5. RTX 4070 and RTX 4060: Software Support

Similar to their more expensive counterparts, the RTX 4070 and RTX 4060 will be getting all the software benefits coming with the next generation. This means DLSS 3, Nvidia Reflex, Encoder, Broadcast, and Studio will all be accessible on both cards.


Infographic showing Optical flow accelerator on NVIDIA
Image credit: NVIDIA

While DLSS and Reflex are mostly gaming-focused, the creative tools from Nvidia that these GPUs will support will be massively beneficial for creative professionals. It makes the two GPUs a good choice for not just gamers but photographers, filmmakers, digital artists, and just about anyone who needs to crush visual workloads without their denting their wallet too much.

Outside of the raw hardware performance, these software features, especially DLSS 3, should give the GPUs a good performance boost thanks to AI frame generation coupled with lower latency and better responsiveness, not to mention the typically excellent drivers Nvidia releases.

RTX 4070 and RTX 4060: Time to Get a New GPU?

If you’ve been holding off on a GPU upgrade and are scared by the eye-watering prices of new Nvidia and AMD GPUs, the RTX 4070 and RTX 4060 are likely rather appealing options. The performance they may offer can be the best bang for your buck, especially if you have other components to upgrade as well.

That said, if you’re willing to shell out the cash, the RTX 4090 and RTX 4080 will still be one heck of an upgrade. If you want the best performance possible, these are the GPUs for you.


🧪 |Medical Laboratory Scientist 🥇 | Mindset over Everything. 
 🤝 | Let's Grow Together.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button