Semiconductor big Nvidia’s CEO Jensen Huang has unveiled the corporate’s new cutting-edge RTX 50-series graphics playing cards on the Client Electronics Present (CES) in Las Vegas, Nevada, on Tuesday, January 7, 2025.
Jensen Huang, co-founder and chief govt officer of Nvidia Corp., speaks whereas holding the corporate’s new GeForce RTX 50 sequence graphics playing cards and a Thor Blackwell robotics processor throughout the 2025 CES occasion in Las Vegas, Nevada, US, on Monday, Jan. 6, 2025. Huang introduced a raft of recent chips, software program and providers, aiming to remain on the forefront of synthetic intelligence computing. (Bridget Bennett/Bloomberg)
Promising a big improve by way of efficiency, effectivity, and innovation, these GPUs are prone to develop into extremely wanted, particularly for PC players, in accordance with a HT Tech report.
Additionally Learn: Who’s Vinay Hiremath, Indian-origin co-founder who offered startup for $975 million?
Design and options of the brand new Nvidia RTX 50-series GPUs
The RTX 50-series will get a smooth redesign of Nvidia’s Founders Version, which can be accompanied by various technical upgrades.
Nvidia claims the highest of the road RTX 5090 presents twice the efficiency of its predecessor, the RTX 4090, whereas additionally being compact, making it appropriate for smaller PC circumstances.
Specs of the Nvidia RTX 5090
The RTX 5090, powered by NVIDIA’s Blackwell structure and the most recent DLSS 4 expertise comes with 32GB GDDR7 reminiscence, 21,760 CUDA cores, and has a bandwidth of 1,792GB/sec.
This interprets to the GPU having the ability to run Cyberpunk 2077 at 238fps with ray tracing enabled.
The caveat is that it requires a 1,000-watt PSU and may eat as much as 575 watts.
Additionally Learn: Amit Banerji, co-founder of workspace operator Desk House, dies of coronary heart assault at 44
Specs of the Nvidia RTX 5080
The RTX 5080 however, will get 16GB GDDR7 reminiscence, 10,752 CUDA cores, and a bandwidth of 960GB/sec, giving it double the velocity of the older RTX 4080.
It nevertheless, requires a 850-watt PSU.
Specs of the mid-range RTX 5070 Ti and RTX 5070
The RTX 5070 Ti will get 16GB GDDR7 reminiscence, 8,960 CUDA cores, and a bandwidth of 896GB/sec, whereas the RTX 5070 will get 12GB GDDR7 reminiscence, 6,144 CUDA cores, and bandwidth of 672GB/sec.
These are the extra reasonably priced, mid-range choices, with double the efficiency of their equal older counterparts, the RTX 40-series.
Additionally Learn: CES 2025: From Nvidia Cosmos to Qualcomm Snapdragon Cockpit, AI is making a case
Particulars of the Nvidia RTX 50-series for laptops
Nvidia has additionally introduced an RTX 50-series line-up for laptops, which is able to launch in March 2025, that includes as much as 24GB of GDDR7 reminiscence.
AI options and DLSS 4 for the Nvidia RTX 50-series
The RTX 50-series can be about supporting synthetic intelligence (AI) options.
DLSS 4: Body charges are boosted by as much as eight instances, which boosts picture high quality whereas additionally decreasing visible artefacts.
Neural Shaders: These improve recreation textures for richer and extra practical visible results.
Neural Faces: Utilising generative AI for lifelike character expressions.
Worth and availability of the brand new Nvidia RTX 50-series GPUs
The RTX 50-series begins at $549 for the RTX 5070 and goes as excessive as an eye-watering $1,999 for the flagship RTX 5090.
The RTX 5080 and 5090 will probably be out there from January 30, 2025, whereas the the RTX 5070 Ti and RTX 5070 will probably be launched someday in February.