Skip to main content

Nvidia may be working on a surprising new budget GPU

Nvidia GeForce RTX 3050 graphics card.
Jacob Roach / Digital Trends

This is certainly unexpected. Hints of an upcoming Nvidia graphics card showed up in the PCI-ID database, and it’s definitely not what you think. Instead of working on the RTX 50-series or refreshing a 40-series GPU, Nvidia is bringing back the slowest RTX 30-series card — the RTX 3050 — but with an Ada Lovelace chip. In theory, that could make this the first RTX 30-series GPU to have access to Deep Learning Super Sampling 3 (DLSS 3).

The GPU in question is a laptop version of the RTX 3050. It’s hard to say why Nvidia would choose to release new versions of that particular card in 2024 (or beyond), but the RTX 3050 A exists — VideoCardz found traces of it in the latest Nvidia drivers. Just the fact that there’s an RTX 3050 in the works is surprising, but that it’s based on the AD106 chip is what baffles me.

Recommended Videos

We’re most likely dealing with a cut-down version of the AD106 GPU, which has a maximum of 4,608 CUDA cores. It’s not odd for Nvidia to be using cut-down versions of this chip for another GPU, but seeing it show up in the RTX 3050 is pretty strange. The RTX 3050 appears in several versions, but the mobile card has a mere 2,048 CUDA cores. That’s quite a leap.

The AD106M chip could have many uses within the Lovelace generation, from the RTX 4050 to the RTX 4060 Ti, so the RTX 3050 is … a choice. The most interesting implication of using Ada architecture in an Ampere card lies in DLSS 3 and Nvidia’s frame generation technology.

Nvidia's Ada Lovelace GPU.
Nvidia

DLSS 3 provides a significant improvement over DLSS 2, so it’s no wonder that Nvidia locked it to the RTX 40-series. For many, DLSS 3 provided an incentive to upgrade, and it was much-needed, as some RTX 40-series cards barely moved the needle when compared to their predecessors. It’s not just a fancy on Nvidia’s part, though. DLSS 3 needs the Ada architecture in order to run, as it uses fourth-generation Tensor cores and Nvidia’s optical flow accelerator, so everyone without an RTX 40-series card is out of luck.

This makes me wonder — could this strange, Frankensteined RTX 3050 with an Ada chip actually have access to DLSS 3? It’d be the only card outside of the RTX 40-series lineup to offer frame generation if that were the case. This would also give what would likely end up in a fairly cheap gaming laptop a lot of extra power to handle newer titles.

Technically, with an AD106 chip, it seems like it could be possible for the RTX 3050 to feature DLSS 3 — at least on the hardware side. However, I still think it’s unlikely. We don’t know when, or if, this card will appear in laptops, but chances are that it’ll be a fairly niche product, and there’s no reason for Nvidia to introduce DLSS 3 in a single older card when it has a whole lineup that features it. Although it’s impossible to say with certainty, the RTX 3050 A is most likely just Nvidia’s way to get rid of some AD106 chips before it moves on to the next generation.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
Laptop GPU names feel like a scam
Gamer playing Overwatch on GIGABYTE G6X gaming laptop from GIGABYTE gaming laptop deals.

Despite the top-tier graphics cards from the past few generations being absolute power hogs, drawing hundreds upon hundreds of watts to deliver what feels like increasingly-modest performance gains, their laptop counterparts have taken enormous leaps in capabilities. AMD, Nvidia, and Intel have made great strides in what onboard graphics and dedicated graphics chips can do with relatively limited power and cooling options.

But even so, mobile GPU naming feels like a scam. The latest example is the flagship Nvidia RTX 50 GPU of this generation, the RTX 5090. On desktop it's about 30% faster than an RTX 4090, but with boatloads more memory, support for the latest multi frame generation technology, and a near-600W TDP to go with it.

Read more
Would you pay less for a defective GPU?
The RTX 5090 sitting on top of the RTX 4080.

The graphics card market is an absolute mess at the moment. Stock issues persist despite Nvidia's early claims that it would solve early issues with the 50-series, and AMD has promised to get its exceedingly popular RX 9000-series GPUs back in stock as soon as possible. That's meant pricing is even more ridiculous than ever, with top cards going for hundreds or even thousands of dollars more than they should, and last-generation options priced just as crazily.

But there are alternatives. Alongside buying older and second-hand cards, there's a new brand of GPU available at discounted prices for anyone willing to take a loss on maximum performance: Defective GPUs.

Read more
Nvidia’s new rule could help you pick the right gaming laptop. Here’s why
A Razer Blade 14 gaming laptop on a coffee table.

Buying a laptop might seem easy, but if you want the best gaming laptop within your budget, you'll need to dig beyond surface-level information. With that said, every single spec should be readily available so that you know what you're buying, but that has not always been the case.

Fortunately, Nvidia says that it's working with its partners to make sure that one key spec is always listed: The total graphics power (TGP). Here's why it matters.
Why does TGP matter in laptops?

Read more