Skip to main content

Nvidia might finally fix its VRAM problem — but it will take time

The Razer Blade 14 and 18 on a table.
Luke Larsen / Digital Trends

It’s no secret that some of Nvidia’s best graphics cards could use a little more VRAM. According to a new leak, Nvidia may be addressing that problem in a big way — at least in laptops. The RTX 5090 laptop GPU is now reported to come with 24GB VRAM across a 256-bit memory bus. The downside? These new laptops might not make it to market as soon as we’d hoped.

The information comes from Moore’s Law Is Dead, who cites his own industry sources as he spills the beans on RTX 50-series laptop specs. Up until now, we’ve not heard much about Nvidia’s plans for RTX 50 laptops, indicating that they might be a few months away. The YouTuber agrees with this, saying that Nvidia might be targeting a launch window in the first or second quarter of 2025. This might not affect the entire lineup, though.

Recommended Videos

The reason why Nvidia might be planning to increase the VRAM on the laptop RTX 5090 is that it’s planning to use newer GDDR7 memory modules for it, with 3GB, up from 2GB. This boosts the memory capacity from 16GB (which the company used in this generation) to 24GB. However, as Moore’s Law Is Dead speculates, that also might be why Nvidia is holding off on the RTX 5090 laptop GPU in particular.

Speculation about the RTX 5090 laptop configuration.
Moore's Law Is Dead

According to the YouTuber, Nvidia’s current plan is to first launch the RTX 5090 and the RTX 5080 in their desktop versions this year. It’s worth noting that most leakers disagree with this timeline — the GPUs aren’t expected to be announced until CES 2025 at this point.

Following the reported desktop launch, Nvidia is said to be unveiling some RTX 50-series laptops at CES 2025 in January, but those will all use the memory modules that are ready now, meaning the 2GB versions. As per the source cited by Moore’s Law Is Dead, the RTX 5090M doesn’t come in a 16GB configuration at all, which is why it won’t be launched alongside the rest of the lineup. Laptop cards like the RTX 5080, RTX 5070, and RTX 5060 might all still make an appearance at CES and launch shortly after with 16GB, 12GB, and 8GB of memory, respectively. They’ll use the 2GB modules, though.

One interesting thing to note is that, while Nvidia will likely use the GB203 die for the desktop RTX 5080 and then use that same die for the RTX 5090 (laptop version), the desktop card will still be limited to just 16GB of memory. In the grand scheme of things, this kind of adds up — if Nvidia is launching desktop GPUs first, the RTX 5080 wouldn’t have been ready with more memory assuming that these 3GB modules aren’t readily available yet. But all of this is still just speculation, and the final specs are unknown to the public.

More VRAM on laptop GPUs is certainly good news for gamers, but it’s hard to know how much of this will turn out to be true. As things stand now, we might not have any concrete information until CES 2025.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
Android 16 might give its own spin to iPhone’s Dynamic Island alerts
The DynamicSpot Dynamic Island at the top of the Pixel 7 Pro.

Over the past few weeks, we’ve come across some interesting details about the next major build of Android. Currently in development under the apparent codename of Baklava, Android 16 will reportedly bring a cool new feature called Priority modes for notifications.

If that sounds familiar, that’s because Apple already offers a bunch of focus modes toward the same goal and bolsters the system with AI-assisted priority notifications in iOS 18. It seems Google doesn’t want to be left behind, and in doing so, could very well lift from a popular iPhone trick.

Read more
It’s finally time to stop buying Nvidia’s RTX 30-series GPUs
RTX 3080 Ti in front of a window.

If you're looking for a budget GPU, the general advice is usually to buy from the previous generation of graphics cards. After all, as the new cards take over the market, the older ones are still waiting to be sold -- and while they're no longer among the best graphics cards, they're still perfectly acceptable alternatives.

We've now reached the point in the current generation of Nvidia GPUs where that advice no longer applies. If you want to get the best bang for your buck, it's time to stop buying Nvidia's RTX 30-series and look for other options.
The RTX 30-series arrived at the worst possible time

Read more
Future GPUs might have up to 48GB VRAM
Samsung's new GDDR7 VRAM.

Samsung has just announced the development of the first 24Gb GDDR7 memory that will one day be found in some of the best graphics cards. An industry first, the new DRAM brings huge improvements to both capacity and performance. With these new memory modules, we might finally stop seeing GPUs with just 8GB VRAM in the near future, with the maximum capacity reaching a massive 48GB.

Samsung's newest DRAM will allow GPU makers to keep the same number of memory modules while reaching higher capacities. Right now, GPUs use either 8Gb (1GB) or 16Gb (2GB) memory modules. At 24Gb (3GB), the capacity will go up considerably, reaching up to 48GB. That's unlikely to happen in the upcoming RTX 5090, but we might see that kind of number in a future RTX 6090.

Read more