After months of hypothesis and anticipation, Nvidia, despite everything, raised the policy of its new diversity of admission graphics cards, the Nvidia GeForce RTX 5000 series, starting with the flagship RTX 5090.
The CEO of Nvidia, Jensen Huang, made this announcement in a stadium crowded with Mandalay Bay Resort and Casino in Las Vegas, a prevailing occasion that ended a day of main rivals AMD and Intel in 2025.
However, the NVIDIA RTX 5000 series graphics cards were still going to borrow the series, no matter what else it announced, so Huang naturally had the highlight on Monday night when he revealed our first final review of the RTX 5090, among other GPUs and GPUs and Fantasy Ai features.
Possibly the NVIDIA Titan RTX is not called, however, the RTX 5090 will also be, given the specifications of the offer and its scandalous completely of $ 1,999 (£ 1,999 / au $ 4,039) MSRP, not a generational increase in RTX 4090, Tuned, granted, but even a ridiculous amount of cash for a seemingly “conventional” graphics card.
With an amazing 92 billion transistors, the new generation tensioning centers and the shelves tracking nuclei, and more than twice the speed of the AI of 4090, the RTX 5090 will undoubtedly be the admission graphics card harder on the planet, and it will not even be close.
Combine the GPU specifications with 32 GB of new and brilliant Vram GDDR7 in a huge reminiscence bus of 512 bits and a PCIe 5. 0 x16 interface, it has an amazing bandwidth of 1792 GB/s, with an ascending reminiscence speed of 23. 8 GBP.
Given its specs, not only will this graphics card blow through local 4K games (without ramping up) at the highest settings (including ray tracing), but it’s arguably the first true 8K graphics card given the amount of VRAM it has and its reminiscent bandwidth, two key specs that allow a graphics card to proceed with the substantially larger 8k texture files needed for gaming in this resolution.
Of course, few games even support 8K resolution, much less have developers and artists effectively wasting their time on texture files so large that only a rare few will ever see them as intended. But there’s no doubt that if 8K gaming ever becomes a thing, the RTX 5090 will be more than ready to meet the challenge.
Of course, this does not deal with the fact that it is no longer a game of the game, more, and not at this price. And if (finally, when) the resellers get involved, it will be much worse.
After all, we called the $1,199 value on the flagship RTX 2080 TI “almost obscene” In our review, there are 3 generations of GPUs, there are 3. In the absence of a generational drop in value on RTX 4090, it’s not expensive, it’s not Delacta to say that RTX 5090 is purely a professional GPU, meant to deal with raw 4K video streams or make long 3-D sequences generated at Pixar or some other animation studio. As a laugh as possible, it’s not a graphics card meant to play Wolfenstein 3-D.
The Nvidia GeForce RTX 5090 is set to go on sale for $1,999 on January 30.
In a move that has been telegraphed for a while, the Nvidia GeForce RTX 5080 will also be part of the first wave of Nvidia’s new graphics cards to hit store shelves later this month.
The RTX 5080 looks to be almost exactly half of the RTX 5090 in terms of specs. Although Nvidia has yet to release a comprehensive spec sheet for the new GPUs – a strange move, though likely one designed to avoid distracting from all the new AI features – we know that the 5080 has 1,801 AI TOPS (trillion operations per second), a little under half the RTX 5090’s 3,352.
It also has a new GDDR7 VRAM, with a 16GB cluster on a 256-bit reminiscence bus for 960GB/s of reminiscence bandwidth, again, necessarily part of the specs of the 1950s VRAM. Its reminiscence speed is a bit fast at 30 Gbps, which is helping to compensate for the narrower reminiscence bus.
The card will pass on January 21 with an MSRP of $999 (£999/AU$2,019), which coincides with the launch of the last RTX 4080.
The Nvidia RTX 5070 and 5070 Ti also debuted on Monday night, an effort by Nvidia to reassure gamers who haven’t forgotten the mid-range market, specifically with its big rival AMD to target budget and mid-range exclusively with its new rival GPUS.
Once again, we do not have the scope of the specifications that we expected to see at this stage, however, we know that the RTX 5070 TI gives 1,406 peaks of AI, while the RTX 5070 has 988 maximums. As an approximate comparison point, the new Apple M4 chip ends in around 38 peaks: a compromised GPU is undoubtedly a need for the workloads of the serious AI.
On the VRAM front, the RTX 5070 Ti’s reminiscent profile is almost the same for the RTX 5080, with 16GB GDDR7, a 28GBPM reminiscence speed that’s in a position for 4K gaming. Higher, the RTX 5070 has 12GB of GDDR7 at 672GB/sec, still faster than the RTX 4070, albeit the same base amount of VRAM.
The Nvidia GeForce RTX 5070 and 5070 Ti will go on sale in February – date to be confirmed – with respective retail prices of $549 (£549 / AU$1,509) and $749 (£749 / AU$1,109). This is actually quite pleasing to see, since the RTX 4070 retailed at $599 – meaning we finally have a generational price drop from Nvidia.
In addition to the new hardware, Nvidia also showcased a selection of upgraded AI features debuting with the ‘Blackwell’ RTX 5000 generation of graphics cards.
The leader among these, of course, DLSS Four, the last update of the Nvidia solution software, which allows greater fractures in the game by making the game weaker and reinforcing it to an objective solution (for example, 1080p to Fourk) from Ai.
DLSS 4 will also feature ‘Multi Frame Generation’, an improved version of the Frame Generation tech seen in the RTX 4000 generation, which uses AI to extrapolate and produce additional frames and ‘insert’ them between ordinary rendered frames to boost framerate. Unfortunately for users on older GPUs, only the regular DLSS 4 upscaling will be available on older cards; Multi Frame-Gen will be exclusive to RTX 5000 cards.
We also get NVIDIA REFLECTE 2, a new edition of the SLR software for access latency in the games. Reflex 2 will come with “Warp of Table”, which aims to proactively insert the fixed frames through the reading of the Front to the Mouse before the demonstration reaches, this can only the front latency to 75%.
RTX 5000 also brings powers AI A Shader Tech with new RTX neuronal shaders. This uses small AI networks in GPU programmable shaders sets to supply shadows and “movie quality” in the game. “RTX Neural Faces”, as well as the new RTX generation for representation and animation of hair and skin, promises to offer more realistic humans than ever.
Finally, we can expect to see more of the ACE Nvidia with this generation of RTX GPUs: the improved AI features of the Blackwell generation mean that projects like NVIDIA’s (slightly creepy) NPC AI generation can be implemented on a wider level, with a (slightly creepy) Ace Force Fouard characters planned to appear in a handful of titles adding PUBG: Battlegrounds and Inzoi.
Christian is the IT writer based at Techradar in the UK. He came here from PC Maximum magazine, where he fell in love with creating PC and PC hardware. It was normal among our independent review team before switching to TechRadar, and you can find yourself drooling over the new high-end graphics card or gaming computer before looking for your bank account balance and crying.
Christian is a keen campaigner for LGBTQ+ rights and the owner of a charming rescue dog named Lucy, having adopted her after he beat cancer in 2021. She keeps him fit and healthy through a combination of face-licking and long walks, and only occasionally barks at him to demand treats when he’s trying to work from home.
Forget the RTX 5090 – The RTX 5070 is the gift that NVIDIA has given to the PC players at ages
Nvidia’s DLSS Four Is: Here’s What AMD’s FSR Four Will Have To Do To Take It
United Kingdom to the best friend with the EU for horizon projects