When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.
It’s official: Nvidia’s next generation of gaming graphics is here and, my friends, the GeForce RTX 50 series looks to be priced pretty attractively on paper. Nvidia CEO Jensen Huang announced another 8 GeForce RTX 50 series graphics cards at his CES 2025 keynote: four for desktops and four for laptops, all supporting a next-generation fourArray DLSS.
The same GPUs were announced for both form factors: The GeForce RTX 5090, 5080, 5070 Ti, and 5070. It was a big difference from AMD’s keynote, where RDNA 4 and the Radeon RX 9070 weren’t even mentioned despite press receiving a high-level briefing.
Let’s start with Nvidia’s highly anticipated desktop graphics cards, powered by Nvidia’s new Blackwell architecture. They’re less expensive than expected, with lower costs than their predecessors. . . unless it’s the RTX 5090. But there’s an explanation for that.
I was convinced the GeForce RTX 5090 would cost $2,500 or more, based on leaked specs. Well, Jensen didn’t actually move on to the product specifications, however, at $1,999, it’s a smart deal for AI researchers, if not necessarily gamers. With 32GB of VRAM and a rugged 512-bit bus, along with Nvidia’s flagship Blackwell GPU, this device will be fierce in gaming, but unheard of in the device’s learning responsibilities if you can’t have a gaming card. Dear company Nvidia.
The issue goes further: at $999, the RTX 5080 costs $200 less than the 4080 at launch (and the same as the RTX 4080 Super). At $749 and $549, the RTX 5070 Ti and RTX 5070 charge $50 less than their predecessors (although, to be fair, I called the RTX 4070 Ti “limp and wildly expensive”). Better yet, Jensen said the RTX 5070 offers RTX 4090-class functionality for about a third of the price.
Inside the new dual-fan “double flow through” Nvidia Founders Edition design, with an absolutely miniscule PCB.
NVIDIA
If this is true in gaming and not just AI workloads or benchmarks, it’s an exciting start to the next generation; However, keep in mind that those starting values for Nvidia’s revised Founders Edition may not reflect the same value seen in the traditional third generation. -products. party cards. Nvidia’s announcement message for the 50 series, released after the keynote, includes key points of high-level functionality for each desktop GPU, but only in full ray tracing and DLSS situations. Since we don’t have a tangible idea of the functionality benefits that DLSS 4 brings, take them with a bit of a grain of salt for now. However, it shows what is possible.
Nvidia claims the RTX 5090 is “2x faster than the GeForce RTX 4090”; the RTX 5080 is “twice as fast as the GeForce RTX 4080”; the RTX 5070 is “2x faster than the GeForce RTX 4070 Ti”; and the RTX 5070 is as fast as the RTX 4090.
Again, expect to hit those peak levels only in fast-paced games and scenarios that feature Nvidia’s full suite of ray tracing and DLSS features. The RTX 5090 offers many more CUDA cores than the 4090, so it deserves to be a beast no matter what, while the lower-tier RTX 50 cards have more modest CUDA upgrades, so Blackwell’s architectural tweaks (and DLSS 4) possibly wants to be modified. do the heavy lifting if Nvidia plans to deliver big functionality improvements. We’ll see!
Look for RTX 50 series graphics cards to start hitting the market in some form at the end of January. Nvidia specified which cards would be released and when.
Usually, Nvidia releases desktop GPUs much earlier than their computer counterparts, but with the RTX 50 series!Well, a little. They’ll be released in March, while the desk cards will be released this month, but Jensen revealed either on stage.
Details were scarce beyond what is seen in the screenshot above. The GeForce RTX 5090 can be purchased in laptops starting at $2,899; the RTX 5080 in laptops starting at $2,199; the RTX 5070 Ti in laptops starting at $1,599; and the RTX 5070 in laptops starting at $1,299.
However, pay close attention to the TOP AIs indexed under the style numbers for computer and desktop GPUs. Laptop graphics tend to be scaled down compared to their desktop cousins, and AI TOPS suggests that’s the case with the RTX 50 series as well. The RTX 5090 desktop card offers 3400 AI TOPS, while the desktop edition it hits 1850, about the same as the desktop RTX 5080. (Not surprising, since the 4090 computer is necessarily a 4080 desktop built into a computer. ) The same trend continues. in the 50 series line of computers.
Update: Nvidia’s GeForce website provided more specs after the keynote and yep, my initial suspicions were correct. Be sure to also pay attention to memory capacity (especially for the RTX 5070) and bus width size, in addition to CUDA core counts. That laptop 5090 is nowhere near the firepower of the desktop 5090 (though Nvidia wisely increased the memory capacity of the laptop 5090 over the desktop 5080), and the laptop 5070 offers just 8GB of VRAM over a 128-bit bus — a skimpy total that frankly doesn’t cut it at even 1080p resolution in 2025’s latest games, though Nvidia may argue that DLSS 4’s AI benefits reduce memory demands.
That’s all well and good, but how do these puppies perform? Jensen didn’t get into details on-stream, instead relaying some high-level Blackwell specs that matter more for AI researchers than gamers – AI TOPS, RT TFLOPS, “AI management processor,” and the like. You can see it all in the screenshot above.
On the plus side, at least some of Nvidia’s RTX 50 series will come with next-gen GDDR7 memory, which will help deliver up to 1. 8TB/s of memory bandwidth, which is double what you could imagine with its predecessor. It will also offer dual shaders for INT and floating-point calculations (the two most important for classic game graphics), the ability to combine GPU and AI workloads as needed, and programmable GPU shaders that can handle neural network tasks.
If it turns out that Nvidia is betting on AI, well, that’s been the case for years. Haven’t you noticed DLSS and DLSS 3 Frame Gen? On that note, Jensen also teased DLSS 4, which features “neural texture compression” and “neural materials” that further reduce the need for classic GPU rendering.
Nvidia says that while DLSS 3 could inject AI-generated frames between every GPU-rendered frame, DLSS 4 can infer three full frames off of a single traditional frame. Hey, AI upscaling is killing native graphics, after all. Jensen said that a total of about 33 million pixels are generated for four frames of 4K imagery; with DLSS 4, that’s still true, but the GPU is only rendering about two million of those, with AI doing the rest of the heavy lifting. You can read much more about DLSS 4 and multi-frame generation on Nvidia’s website if you’re interested.
As DLSS 3 users who forgot to allow Nvidia Reflex can tell you, injecting AI-generated frames also injects latency – Reflex can recover it. This is a much more pronounced challenge when AI is used to create multiple images. Enter Nvidia Reflex 2, which uses “warp frame generation” to reduce latency by up to 75%. It’s complicated, so watch the video and click this link for a detailed explanation of the generation if you need more details.
Bonkers – and potentially very, very game changing if it works as advertised. That’s a big if, though, and there are plenty of questions still swirling around the RTX 50-series’ capabilities after Nvidia’s detail-light keynote. Stay tuned to PCWorld (and PCWorld’s YouTube channel!) for all the latest updates and news coming from CES 2025.
Editor’s Note: This article has been updated to include videos of MSI and Gigabyte 50 series graphics cards.
Brad Chacos spends his days researching desktop computers and tweeting too much. It specializes in graphics cards and gaming, but covers everything from security to Windows tips to all kinds of PC hardware.
Business
Laptop
Mobile
PC Hardware
Storage
Offers
Tech Hive
Digital Magazine – Subscribe
Digital Magazine – Information
Gift Subscription
Newsletters