The Consumer Electronics Show – better known as CES – is upon us, and that means it’s time for Nvidia to lift the lid on its latest and greatest line of desktop and laptop graphics cards. The company held its annual presentation at CES today, and with it came the first official details for its RTX 50 series graphics cards, including their price, release date, and all the new features.
The RTX 50 series of graphics cards from Nvidia – made up of the RTX 5090, 5080, 5070, and 5070 Ti – were revealed during the presentation, and promise a marked upgrade over their 40 series counterparts. The new cards are built on Nvidia’s new Blackwell architecture, with fifth-generation Tensor cores and fourth-generation RT cores on board, which Nvidia says will bring “breakthroughs in AI-driven rendering, including neural shaders, digital human technologies, geometry and lighting.”
“Blackwell, the engine of AI, has arrived for PC gamers, developers and creatives,” Jensen Huang, founder and CEO of Nvidia, said. “Fusing AI-driven neural rendering and ray tracing, Blackwell is the most significant computer graphics innovation since we introduced programmable shading 25 years ago.”
Nvidia says the 5090 is the fastest RTX GPU to date, with 92 billion transistors, capable of performing over 3342 trillion AI operations per second. What does that mean, exactly? Well, if the numbers are to be believed, the 5090 should provide performance up to double that of the RTX 4090 before it. The rest of the series is looking pretty solid too, with Nvidia claiming that the 5070 should see performance equivalent to the 4090, despite being priced much lower.
Part of this performance boost comes courtesy of DLSS 4, which on 50 series cards has a new feature called Multi Frame Generation. This is an expansion of the Frame Generation seen in DLSS 3 on the 40 series cards, and is capable of using AI to generate up to three additional frames per rendered frame, compared to just one frame in prior generations of the technology. This, combined with an enhanced Nvidia Reflex 2, should allow for much higher frame rates while – hopefully – still keeping input latency low. We’ll have to see how that pans out when the cards get into players’ hands.
The release date for the RTX 5090 and 5080 is set for January 30, 2025, while the 5070 and the 5070 Ti will be released in February 2025. Laptops featuring the 5090, 5080, and 5070 Ti will be available starting in March 2025, while laptops featuring the regular 5070 will be available in April.
In terms of pricing, those interested in the top of the line will have to pay a whopping $1999 USD for the RTX 5090, while other cards in the series are much more reasonably priced. The RTX 5080 will cost $999 USD, the 5070 Ti will cost $749, and the 5070 will cost $549. Granted, these prices are for the reference versions of the cards, so cards from other manufacturers could (and likely will) cost more.
Here are the full specs for the RTX 50 series cards, courtesy of Nvidia:
RTX 5090
RTX 5080
RTX 5070 Ti
RTX 5070
AI TOPS
3352
1801
1406
988
NVENC
3x 9th gen
2x 9th
2x 9th gen
1x 9th gen
NVDEC
2x 6th gen
2x 6th gen
1x 6th gen
1x 6th gen
Memory
32GB GDDR7
16GB GDDR7
16GB GDDR7
12GB GDDR7
Memory bandwidth
1792 GB/s
960 GB/s
896 GB/s
692 GB/s
Price
$1999
$999
$749
$549
