Monday, January 6th 2025
NVIDIA Blackwell GeForce RTX 50 Series Opens New World of AI Computer Graphics
NVIDIA today unveiled the most advanced consumer GPUs for gamers, creators and developers—the GeForce RTX 50 Series Desktop and Laptop GPUs. Powered by the NVIDIA Blackwell architecture, fifth-generation Tensor Cores and fourth-generation RT Cores, the GeForce RTX 50 Series delivers 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," said Jensen Huang, founder and CEO of NVIDIA. "Fusing AI-driven neural rendering and ray tracing, Blackwell is the most significant computer graphics innovation since we introduced programmable shading 25 years ago." The GeForce RTX 5090 GPU—the fastest GeForce RTX GPU to date—features 92 billion transistors, providing over 3,352 trillion AI operations per second (TOPS) of computing power. Blackwell architecture innovations and DLSS 4 mean the GeForce RTX 5090 GPU outperforms the GeForce RTX 4090 GPU by up to 2x.
GeForce Blackwell comes to laptops with all the features of desktop models, bringing a considerable upgrade to portable computing, including extraordinary graphics capabilities and remarkable efficiency. The Blackwell generation of NVIDIA Max-Q technology extends battery life by up to 40%, and includes thin and light laptops that maintain their sleek design without sacrificing power or performance.
NVIDIA DLSS 4 Boosts Performance by Up to 8x
DLSS 4 debuts Multi Frame Generation to boost frame rates by using AI to generate up to three frames per rendered frame. It works in unison with the suite of DLSS technologies to increase performance by up to 8x over traditional rendering, while maintaining responsiveness with NVIDIA Reflex technology.
DLSS 4 also introduces the graphics industry's first real-time application of the transformer model architecture. Transformer-based DLSS Ray Reconstruction and Super Resolution models use 2x more parameters and 4x more compute to provide greater stability, reduced ghosting, higher details and enhanced anti-aliasing in game scenes. DLSS 4 will be supported on GeForce RTX 50 Series GPUs in over 75 games and applications the day of launch.
NVIDIA Reflex 2 introduces Frame Warp, an innovative technique to reduce latency in games by updating a rendered frame based on the latest mouse input just before it is sent to the display. Reflex 2 can reduce latency by up to 75%. This gives gamers a competitive edge in multiplayer games and makes single-player titles more responsive.
Blackwell Brings AI to Shaders
Twenty-five years ago, NVIDIA introduced GeForce 3 and programmable shaders, which set the stage for two decades of graphics innovation, from pixel shading to compute shading to real-time ray tracing. Alongside GeForce RTX 50 Series GPUs, NVIDIA is introducing RTX Neural Shaders, which brings small AI networks into programmable shaders, unlocking film-quality materials, lighting and more in real-time games.
Rendering game characters is one of the most challenging tasks in real-time graphics, as people are prone to notice the smallest errors or artifacts in digital humans. RTX Neural Faces takes a simple rasterized face and 3D pose data as input, and uses generative AI to render a temporally stable, high-quality digital face in real time.
RTX Neural Faces is complemented by new RTX technologies for ray-traced hair and skin. Along with the new RTX Mega Geometry, which enables up to 100x more ray-traced triangles in a scene, these advancements are poised to deliver a massive leap in realism for game characters and environments.
The power of neural rendering, DLSS 4 and the new DLSS transformer model is showcased on GeForce RTX 50 Series GPUs with Zorah, a groundbreaking new technology demo from NVIDIA.
Autonomous Game Characters
GeForce RTX 50 Series GPUs bring industry-leading AI TOPS to power autonomous game characters in parallel with game rendering.
NVIDIA is introducing a suite of new NVIDIA ACE technologies that enable game characters to perceive, plan and act like human players. ACE-powered autonomous characters are being integrated into KRAFTON's PUBG: BATTLEGROUNDS and InZOI, the publisher's upcoming life simulation game, as well as Wemade Next's MIR5.
In PUBG, companions powered by NVIDIA ACE plan and execute strategic actions, dynamically working with human players to ensure survival. InZOI features Smart Zoi characters that autonomously adjust behaviors based on life goals and in-game events. In MIR5, large language model (LLM)-driven raid bosses adapt tactics based on player behavior, creating more dynamic, challenging encounters.
AI Foundation Models for RTX AI PCs
Showcasing how RTX enthusiasts and developers can use NVIDIA NIM microservices to build AI agents and assistants, NVIDIA will release a pipeline of NIM microservices and AI Blueprints for RTX AI PCs from top model developers such as Black Forest Labs, Meta, Mistral and Stability AI.
Use cases span LLMs, vision language models, image generation, speech, embedding models for retrieval-augmented generation, PDF extraction and computer vision. The NIM microservices include all the necessary components for running AI on PCs and are optimized for deployment across all NVIDIA GPUs.
To demonstrate how enthusiasts and developers can use NIM to build AI agents and assistants, NVIDIA today previewed Project R2X, a vision-enabled PC avatar that can put information at a user's fingertips, assist with desktop apps and video conference calls, read and summarize documents, and more.
AI-Powered Tools for Creators
The GeForce RTX 50 Series GPUs supercharge creative workflows. RTX 50 Series GPUs are the first consumer GPUs to support FP4 precision, boosting AI image generation performance for models such as FLUX by 2x and enabling generative AI models to run locally in a smaller memory footprint, compared with previous-generation hardware.
The NVIDIA Broadcast app gains two AI-powered beta features for livestreamers: Studio Voice, which upgrades microphone audio, and Virtual Key Light, which relights faces for polished streams. Streamlabs is introducing the Intelligent Streaming Assistant, powered by NVIDIA ACE and Inworld AI, which acts as a cohost, producer and technical assistant to enhance livestreams.
Availability
For desktop users, the GeForce RTX 5090 GPU with 3,352 AI TOPS and the GeForce RTX 5080 GPU with 1,801 AI TOPS will be available on Jan. 30 at $1,999 and $999, respectively.
The GeForce RTX 5070 Ti GPU with 1,406 AI TOPS and GeForce RTX 5070 GPU with 988 AI TOPS will be available starting in February at $749 and $549, respectively.
The NVIDIA Founders Editions of the GeForce RTX 5090, RTX 5080 and RTX 5070 GPUs will be available directly from NVIDIA and select retailers worldwide.
"Blackwell, the engine of AI, has arrived for PC gamers, developers and creatives," said Jensen Huang, founder and CEO of NVIDIA. "Fusing AI-driven neural rendering and ray tracing, Blackwell is the most significant computer graphics innovation since we introduced programmable shading 25 years ago." The GeForce RTX 5090 GPU—the fastest GeForce RTX GPU to date—features 92 billion transistors, providing over 3,352 trillion AI operations per second (TOPS) of computing power. Blackwell architecture innovations and DLSS 4 mean the GeForce RTX 5090 GPU outperforms the GeForce RTX 4090 GPU by up to 2x.
NVIDIA DLSS 4 Boosts Performance by Up to 8x
DLSS 4 debuts Multi Frame Generation to boost frame rates by using AI to generate up to three frames per rendered frame. It works in unison with the suite of DLSS technologies to increase performance by up to 8x over traditional rendering, while maintaining responsiveness with NVIDIA Reflex technology.
DLSS 4 also introduces the graphics industry's first real-time application of the transformer model architecture. Transformer-based DLSS Ray Reconstruction and Super Resolution models use 2x more parameters and 4x more compute to provide greater stability, reduced ghosting, higher details and enhanced anti-aliasing in game scenes. DLSS 4 will be supported on GeForce RTX 50 Series GPUs in over 75 games and applications the day of launch.
NVIDIA Reflex 2 introduces Frame Warp, an innovative technique to reduce latency in games by updating a rendered frame based on the latest mouse input just before it is sent to the display. Reflex 2 can reduce latency by up to 75%. This gives gamers a competitive edge in multiplayer games and makes single-player titles more responsive.
Blackwell Brings AI to Shaders
Twenty-five years ago, NVIDIA introduced GeForce 3 and programmable shaders, which set the stage for two decades of graphics innovation, from pixel shading to compute shading to real-time ray tracing. Alongside GeForce RTX 50 Series GPUs, NVIDIA is introducing RTX Neural Shaders, which brings small AI networks into programmable shaders, unlocking film-quality materials, lighting and more in real-time games.
Rendering game characters is one of the most challenging tasks in real-time graphics, as people are prone to notice the smallest errors or artifacts in digital humans. RTX Neural Faces takes a simple rasterized face and 3D pose data as input, and uses generative AI to render a temporally stable, high-quality digital face in real time.
RTX Neural Faces is complemented by new RTX technologies for ray-traced hair and skin. Along with the new RTX Mega Geometry, which enables up to 100x more ray-traced triangles in a scene, these advancements are poised to deliver a massive leap in realism for game characters and environments.
The power of neural rendering, DLSS 4 and the new DLSS transformer model is showcased on GeForce RTX 50 Series GPUs with Zorah, a groundbreaking new technology demo from NVIDIA.
Autonomous Game Characters
GeForce RTX 50 Series GPUs bring industry-leading AI TOPS to power autonomous game characters in parallel with game rendering.
NVIDIA is introducing a suite of new NVIDIA ACE technologies that enable game characters to perceive, plan and act like human players. ACE-powered autonomous characters are being integrated into KRAFTON's PUBG: BATTLEGROUNDS and InZOI, the publisher's upcoming life simulation game, as well as Wemade Next's MIR5.
In PUBG, companions powered by NVIDIA ACE plan and execute strategic actions, dynamically working with human players to ensure survival. InZOI features Smart Zoi characters that autonomously adjust behaviors based on life goals and in-game events. In MIR5, large language model (LLM)-driven raid bosses adapt tactics based on player behavior, creating more dynamic, challenging encounters.
AI Foundation Models for RTX AI PCs
Showcasing how RTX enthusiasts and developers can use NVIDIA NIM microservices to build AI agents and assistants, NVIDIA will release a pipeline of NIM microservices and AI Blueprints for RTX AI PCs from top model developers such as Black Forest Labs, Meta, Mistral and Stability AI.
Use cases span LLMs, vision language models, image generation, speech, embedding models for retrieval-augmented generation, PDF extraction and computer vision. The NIM microservices include all the necessary components for running AI on PCs and are optimized for deployment across all NVIDIA GPUs.
To demonstrate how enthusiasts and developers can use NIM to build AI agents and assistants, NVIDIA today previewed Project R2X, a vision-enabled PC avatar that can put information at a user's fingertips, assist with desktop apps and video conference calls, read and summarize documents, and more.
AI-Powered Tools for Creators
The GeForce RTX 50 Series GPUs supercharge creative workflows. RTX 50 Series GPUs are the first consumer GPUs to support FP4 precision, boosting AI image generation performance for models such as FLUX by 2x and enabling generative AI models to run locally in a smaller memory footprint, compared with previous-generation hardware.
The NVIDIA Broadcast app gains two AI-powered beta features for livestreamers: Studio Voice, which upgrades microphone audio, and Virtual Key Light, which relights faces for polished streams. Streamlabs is introducing the Intelligent Streaming Assistant, powered by NVIDIA ACE and Inworld AI, which acts as a cohost, producer and technical assistant to enhance livestreams.
Availability
For desktop users, the GeForce RTX 5090 GPU with 3,352 AI TOPS and the GeForce RTX 5080 GPU with 1,801 AI TOPS will be available on Jan. 30 at $1,999 and $999, respectively.
The GeForce RTX 5070 Ti GPU with 1,406 AI TOPS and GeForce RTX 5070 GPU with 988 AI TOPS will be available starting in February at $749 and $549, respectively.
The NVIDIA Founders Editions of the GeForce RTX 5090, RTX 5080 and RTX 5070 GPUs will be available directly from NVIDIA and select retailers worldwide.
38 Comments on NVIDIA Blackwell GeForce RTX 50 Series Opens New World of AI Computer Graphics
Now I need a 4K 240hz screen to enjoy DLSS4 :p
What about productivity? Media engines "NVENC"
What about performance numbers in games, I can't believe I'm asking about games!
I think this is it. Game set and match. Nvidia is at a totally different level in computer graphics. I mean, AMD and Intel still using fast typists, Nvidia is already at photocopying. AMD and Intel will have to split the sub $500 market. With the power Nvidia has in the press even if Intel or AMD come up with something close to what Nvidia will be offering, some Tim out there will be using the magnifying glass to conclude that it is totally junk and last gen anyway. Developers will be making games for PCs optimized to use those next gen graphics, meaning games on AMD and Intel cards wouldn't be just slower and unoptimized but also worst looking. Something that Nvidia tried with PhysX but failed.
also that 5070 looks like a 5060 in disguise, a 106 die has never before been used for 70 series card
Remember these claims?
But the MSRPs are, apart from RTX 5090, actually decent, with no generational price uplift, we only get performance rise - unless there is a plan to offload the price increases to AIB partners and blame the market demand?
Most of the world doesn't have access to Nvidia's own Founder's Edition cards.
Below 30% raster performance increase would be lowest since RTX 20x0 - a generation that mostly focussed on raytracing and DLSS - but we are now in similar change of focus, with Nvidia mostly caring about AI acceleration increase, and using that tensor performance increase in Gaming for DLSS upscaling.
I imagine reviewers will be under great pressure to show all the benefits of DLSS improvements, talk about all the upcoming AI gimmicks, and not focus on raster performance, which could be really underwhelming. Or simply not mention that the average raster increase was about 50% per generation, people have short memory span.
I really hoped they would double RT perf, as they are being rapidly caught up by both AMD and Intel in this department. DLSS 4 looks good though, I can see much more detail and far less fringing in the sample image. Neural rendering looks like a feature that's not going to be used in the lifetime of these cards though and will probably be too slow by the time games come out using it, just like RT on the 2x cards.
But hey, at least we got double the "A.I." perf, whatever that really means to most people... but only because of FP4 support...
Looks like Blackwell is quite the lazy upgrade though, basically a few nice software features, a minor NVENC update, and FP4 support. I hope the 6x series of cards does better than just a +-30% perf uptick on average.