Skip to main content

The RTX 50-series may fix the biggest weakness of its predecessor

Nvidia Blackwell chips.
Notebook Check / Nvidia

Nvidia’s RTX 50-series is still largely a mystery, but the occasional leak keeps the hype alive. And this time, it’s gaming laptop enthusiasts who have reason to be happy. According to Moore’s Law Is Dead, Nvidia’s Blackwell laptop GPUs are receiving a much-needed boost to the memory interface, implying that Nvidia may have learned from its past mistakes.

Although rumor has it that Nvidia may be aiming to launch the first RTX 50-series desktop GPUs in the final quarter of 2024, we’ve heard very little about these graphics cards that didn’t come from leaks. We’ve heard even less about their laptop counterparts, but it’s possible that those will come later, as NotebookCheck reports that Nvidia may be delaying those models.

Now, Moore’s Law Is Dead seems to corroborate that theory by claiming that Nvidia decided to make some much-needed tweaks to the mobile versions of the RTX 5080, the RTX 5070, and the RTX 5070 Ti.

Get your weekly teardown of the tech behind PC gaming
Check your inbox!

As per the YouTuber’s latest video, Nvidia is planning to deck out the RTX 5080 with the same GB203 GPU as the RTX 5090, which would put these two graphics cards on a much more even playing field. Not only that, but Nvidia is said to be tweaking the VRAM and the memory bus width in these cards, which, as we now know based on GPUs like the RTX 4060 Ti, is increasingly becoming necessary.

If the leak turns out to be true, the laptop version of the RTX 5080 will come with the GB203 GPU, but also the same 16GB of memory across a 256-bit bus. That’s the same configuration as in the current-gen RTX 4090 (mobile version) and, reportedly, the same as the RTX 5090. That would be huge, as in this generation, the laptop RTX 4080 is limited to 12GB VRAM and a 192-bit bus. The RTX 5070/RTX 5070 Ti is also said to enjoy similar perks, receiving an extra 4GB of video memory for a total of 12GB VRAM across a 192-bit bus. Unfortunately, this doesn’t apply to every GPU across the stack.

Nvidia introducing its Blackwell GPU architecture at GTC 2024.

These gains are as needed as they are welcome. No matter laptop or desktop, VRAM is becoming a hot topic as of late — and with good reason. If Nvidia tunes up the RTX 5080 and the RTX 5070, we might be able to get solid gaming laptops at a more affordable price — bar any price increases.

Unfortunately, the RTX 5060 is still said to be limited to 8GB VRAM and a 128-bit bus in its mobile iteration. If proven true, gamers on a tighter budget who want a top-notch GPU will be faced with a hard choice. As 8GB of VRAM is not ideal even now, it’ll only become less optimal over time, so these laptops may need to be priced competitively to make up for not being able to keep up with the rising requirements in games.

As always, these are just rumors, so we’ll have to wait for Nvidia to make an official announcement before we truly learn the specs of any Blackwell GPUs. If they end up being true, though, it certainly seems like Nvidia has a big upgrade in store for gaming laptops.

Editors' Recommendations

Monica J. White
Monica is a UK-based freelance writer and self-proclaimed geek. A firm believer in the "PC building is just like expensive…
All RTX GPUs now come with a local AI chatbot. Is it any good?
A window showing Nvidia's Chat with RTX.

It's been difficult to justify packing dedicated AI hardware in a PC. Nvidia is trying to change that with Chat with RTX, which is a local AI chatbot that leverages the hardware on your Nvidia GPU to run an AI model.

It provides a few unique advantages over something like ChatGPT, but the tool still has some strange problems. There are the typical quirks you get with any AI chatbot here, but also larger issues that prove Chat with RTX needs some work.
Meet Chat with RTX
Here's the most obvious question about Chat with RTX: How is this different from ChatGPT? Chat with RTX is a local large language model (LLM). It's using TensorRT-LLM compatible models -- Mistral and Llama 2 are included by default -- and applying them to your local data. In addition, the actual computation is happening locally on your graphics card, rather than in the cloud. Chat with RTX requires an Nvidia RTX 30-series or 40-series GPU and at least 8GB of VRAM.

Read more
Why I’m feeling hopeful about Nvidia’s RTX 50-series GPUs
The RTX 4070 Super on a pink background.

I won't lie -- I was pretty scared of Nvidia's RTX 50-series, and I stand by the opinion that those fears were valid. They didn't come out of thin air; they were fueled by Nvidia's approach to GPU pricing and value for the money.

However, the RTX 40 Super refresh is a step in the right direction, and it's one I never expected to happen. Nvidia's most recent choices show that it may have learned an important lesson, and that's good news for future generations of graphics cards.
The price of performance
Nvidia really didn't hold back in the RTX 40 series. It introduced some of the best graphics cards we've seen in a while, but raw performance isn't the only thing to consider when estimating the value of a GPU. The price is the second major factor and weighing it against performance can often tip the scales from "great" to "disappointing." That was the case with several GPUs in the Ada generation.

Read more
The RTX 4080 Super doesn’t move the needle — and it doesn’t need to
The RTX 4080 Super graphics card sitting on a pink background.

I suspect the RTX 4080 Super will be disappointing to some people. That doesn't make much sense, considering the RTX 4080 Super will replace the base RTX 4080, but for a market fixated on generational improvements and small percentage wins, it looks like a disappointing GPU.

Nvidia set the bar low, claiming performance improvements in the low single digits. The tests I ran for our RTX 4080 Super review show even smaller gains -- around 1% on average, often rounded up from less than a single percentage point. So, what's so super about this supposed Super refresh?

Read more