#NVIDIA
133 Stories
Nvidia Announces RTX 5060 Ti: Starting at $379 with DLSS 4 and Extreme Performance Gains
Upanishad Sharma
View quick summary
NVIDIA has announced the RTX 5060 Ti desktop GPUs with DLSS 4 and Blackwell architecture, starting at $379. Offering up to 2X frame rates and lower latency, RTX 5060 laptops are also launching in May, starting at $1,099.
Nvidia G-Assist Is Your Personal AI Assistant That Optimizes Games
Ishan Adhikary
View quick summary
The Nvidia G-Assist feature is live, and it helps you optimize your visuals, measure frame rates, and fine-tune the graphics settings. Moreover, it lets you control the lighting if you have supportive devices like Logitech, Corsair, MSI, and Nanoleaf. G-Assist is only available for RTX 30, 40, or 50 series desktop GPUs and requires approximately 10GB of storage.
Nvidia today announced its RTX PRO 6000 Blackwell series. The RTX PRO 6000 for workstations has 96GB of GDDR7 memory with a 600-watt power requirement. RTX PRO 6000 advances in AI and ray tracing, offering 125 TFLOPS FP32 and 4000 AI TOPS. The full RTX PRO lineup will arrive this summer as confirmed by Nvidia in the press release.
Nvidia Rebrands Project Digits as ‘DGX Spark’ and Unveils a Larger DGX Station
Arjun Sha
View quick summary
Nvidia has renamed Project Digits to 'DGX Spark' and opened it for reservations, starting at $3,000. The company has also announced a beefy, server-class DGX Station that comes with the latest Blackwell Ultra GPU.
Nvidia’s Rumored N1X ARM Chip for Windows Laptops Shows Up on Geekbench
Arjun Sha
View quick summary
Nvidia's rumored N1X Arm chipset for Windows laptops is taking shape. It has debuted on Geekbench with an early test run. Check out our article for the Geekbench score.
NVIDIA RTX 5090 & 5090D GPUs Are Getting Bricked and Nobody Knows Why
Upanishad Sharma
View quick summary
Batches of 5090 and 5090D GPUS are failing worldwide with no clear reason why. The problem first started occurring after installing the latest NVIDIA driver update. Keep reading to learn everything about it.
Nvidia is "planning" to enter the Windows PC market, and it may happen by the end of 2025 itself. Leaks reveal that Nvidia is working on "N1X" and "N1" chipsets for Windows PCs.
Nvidia boldly claimed that the affordable RTX 5070 GPU can match RTX 4090 performance. However, RTX 4090 houses a large number of CUDA cores, SMs, Ray Tracing units, and Tensor cores, compared to RTX 5070. It delivers 2x to 3x better performance than the RTX 5070.
Nvidia has finally confirmed that GeForce Now is coming to India in the first half of this year. Nvidia is launching its own RTX 4080-powered data center in India.
Nvidia’s Project Digits is a Mini AI Supercomputer That You Can Actually Buy
Arjun Sha
View quick summary
Nvidia has announced a tiny supercomputer that can process massive AI workloads right on your desk. It costs $3000 and comes with the powerful Blackwell GPU with the latest Tensor cores. On the CPU side, it features Arm's flagship Cortex cores.
Nvidia's RTX 5090 GPU redraws the boundary of graphics and AI performance. It's nearly 2x faster than RTX 4090, thanks to the Blackwell architecture and ultra-fast GDDR7 memory.
Nvidia's latest GeForce RTX 5090 GPU is here and it's based on the latest Blackwell architecture. It comes with 32GB of GDDR7 memory and packs 21,760 CUDA cores. The highest TDP is rated at 575W.
Nvidia RTX 5090 Reveal Showcases ‘Blackwell’ Series With Triple the AI Performance
Ishan Adhikary
View quick summary
Nvidia CEO Jensen Huang revealed the RTX 'Blackwell' series cards, showing off the performance for 5090. The Nvidia RTX 5090 card will come with 3352 AI TOPS and 32 GB GDDR7 memory at a price of $1,999. The other cards from the 'Blackwell' series will be priced based on their VRAM, AI TOPS, and performance: RTX 5080 at $999, RTX 5070 Ti at $749, and RTX 5070 at $549.
First reported by VideoCardz, a latest custom RTX 5090 card leak confirmed the next-gen Nvidia card to have a 32GB GDD7 memory. Inno3D RTX 5090 iChill X3, the custom card will also come with a 3.5-slot cooler. Rumors suggest that the Nvidia RTX 5090 will come with a TDP of 575W.
First Nvidia GeForce RTX 5080 Custom Model Leak Confirms 16 GB GDDR7 Memory
Ishan Adhikary
View quick summary
Latest leaks have revealed the first look at the Nvidia's GeForce RTX 5080 custom model. MSI's GeForce RTX 5080 Gaming Trio OC Edition is the leaked custom card that also confirms 16GB of GDDR7 memory. The leaks also suggest that the Nvidia RTX 5080 will have a TDP of 360W.
Nvidia Releases Nemotron 70B Model; Claims to Beat GPT-4o and Claude 3.5 Sonnet
Arjun Sha
View quick summary
Nvidia's latest Nemotron 70B model claims to beat GPT-4o and Claude 3.5 Sonnet models. It's trained on top of Llama 3.1 70B model using RLHF. Nvidia says it can correctly answer the strawberry question without using reasoning tokens.
RTX 5090 Prices Won’t Be Significantly Higher than 4090: Says Leaker
Ishan Adhikary
View quick summary
YouTuber Moore's Law Is Dead shared a report claiming Nvidia RTX 5090 price to be around $1,999 to $2,499. This is a massive price bump compared to RTX 4090 that launched at an overpriced tag of $1,599. However, another trusted leaker kopite7kimi has said the leaked prices are totally fake and the price bump will not be such high.
Nvidia CEO Jensen Huang to Deliver CES 2025 Keynote: RTX 5090 on the Way?
Upanishad Sharma
View quick summary
Nvidia's CEO Jensen Huang will hold an official keynote at CES 2025. The event will take place on January 6. While it is expected that the company will focus on AI, rumors also point towards a possible RTX 5090 reveal.
Nvidia RTX 5090 to Come with 32GB VRAM and 600W Power Requirement; Leaks Claim
Ishan Adhikary
View quick summary
Thanks to a renowned leaker Kopite7kimi's, we have more definitive specs about the upcoming Nvidia RTX 5090 and 5080. According to the leaks, the RTX 5090 consumes 600W power, where the RTX 5080 consumes 400W power. Rumors also suggests that the RTX 5090 will run on 21,760 CUDA cores with 32GB GDDR7 VRAM and RTX 5080 will run on 10,752 CUDA cores with 16GB of GDDR7 VRAM.
Nvidia is the dominant player in the AI hardware space. However, there are other players developing powerful custom silicon, specifically for AI workloads. From Google to Amazon, Microsoft, Groq, and Cerebras, the race to build highly-performant AI accelerators is heating up. Read our piece to understand what makes Nvidia a unique company and what stops developers from using other hardware for AI training and inferencing.
At Gamescom 2024, Nvidia shows off ACE, the first digital human NPC through a demo inside the Mechabreak game. It uses a new SLM called NVIDIA Nemotron-4 4B Instruct alongside Nvidia Audio2Face-3D NIM. For voice recognition and its realistic feedback, third part Whisper and Elevenlabs are used. The Nvidia ACE digital human NPC can be a companion in the game with genuine knowledge and is beyond a basic generative AI.
Nvidia and MediaTek Partner to Make G-Sync Monitors More Affordable
Ishan Adhikary
View quick summary
At Gamescom 2024, Nvidia has revealed a new partnership to bring G-Sync to more monitors. According to the reveal, Nvidia is partnering with MediaTek, the leading scalers in the world. With the MediaTek 9810 and three new monitors, Nvidia will start their MediaTek partnership to add G-Sync to more monitors. Nvidia further says the G-Sync support will increase for more MediaTek scalers in the future.
GeForce NOW Brings Automatic Xbox Sign-in to Make Cloud Gaming Effortless
Ishan Adhikary
View quick summary
Nvidia GeForce NOW will now support automatic sign-in for Xbox. Players will have to link their Xbox account to GeForce NOW account just like Ubisoft or Epic Games account to access the games seamlessly. GeForce NOW also announces some latest AAA games arriving on the cloud platform including Black Myth: Wukong and Final Fantasy XVI demo.
Nvidia Project G-Assist: Your First-Ever AI Gaming Assistant Is Here!
Ishan Adhikary
View quick summary
Nvidia has announced the first ever AI powered gaming assistant named Project G-Assist after posting about it in 2017 that became an April Fools' Day joke. Project G-Assist will help players progress easily in the game by detecting content on screen or from app API. You can also ask the AI assistant about the best visuals, latency, performance, or other graphics related questions for an easy solution in seconds. The G-Assist uses the power of RTX AI PC and Nvidia cloud.
The best budget graphics card for $350 is the AMD RX 7600 XT. Then, we have options like the AMD RX 7600 and Nvidia RTX 4060. Going down to around $200, the Intel Arc A750 is perfect and gives good performance for the price albeit suffering from stability issues in certain games. Between $150 to $200, there is the AMD RX 6600 and newly launched Nvidia RTX 3050 6GB. At the lower end, Intel Arc A380 costs around $100 and the most affordable graphics card.
Beebom Staff
AMD Radeon RX 7800 XT has 16GB of GDDR6 and Nvidia RTX 4070 Super has 12GB of GDDR6X VRAM. Both offer various advantages, such as AFMF with AMD. Nvidia has a suite of nice-to-have apps such as Chat with RTX. In raw performance, both GPUs stand tall against one another. However, Nvidia has better performance in ray-tracing. Overall, it is up to the user if they prefer the reasons we listed to get RX 7800 XT or RTX 4070 Super. AMD is more value for money, while Nvidia consumes less power.
Sampad Banerjee
As per a recent report, NVIDIA board partners expect the RTX 5080 and RTX 5090 GPUS to launch in the fourth quarter of 2024. While not officially confirmed, the leak has been backed up by a prominent hardware leaker.
The Death of Nvidia GTX Series: Yes, It Has Finally Been Discontinued!
Satyam Kumar
View quick summary
Nvidia GeForce GTX GPUs have served us well for the last 19 years. Now, they are being discontinued. The last production of GTX 16 Series GPUs has been reportedly stopped, and it is expected that they will be gone from the market in the next few months. The RTX 2050 mobile GPU based on Ampere architecture for laptops and the newly released RTX 3050 with 6GB VRAM for desktops will be the new entry-level graphics cards from Nvidia.
Nvidia Unveils New RTX 500 & RTX 1000 Laptop GPUs for AI Workstations
Satyam Kumar
View quick summary
Nvidia's new RTX 1000 and RTX 500 laptop GPUs are meant for the AI era of computing. They bring the Ada Lovelace architecture with 3rd Gen RT cores and 4th Gen Tensor Cores. Suited for professionals such as 3D artists, engineers, scientists, and more these new GPUs will be used in upcoming RTX professional laptops released by various manufacturers such as HP, Dell, Lenovo, and MSI. The RTX 1000 has 6GB GDDR6 VRAM and an AI performance measurement of 154 TOPS, with the RTX 500 having 4GB GDDR6 VRAM with 193 TOPS.
The New NVIDIA App is Here to Help You Break up with GeForce Experience
Upanishad Sharma
View quick summary
It's no secret that the GeForce Experience app was annoyingly bad. However, we don't have to worry about that anymore as NVIDIA has just released a brand new app. Simply called the NVIDIA app, this new experience combines GeForce Experience, NVIDIA Control Panel and RTX Experience apps. Some new features include redesigned UI, improved driver panel screen, and a new AI Filters.
Nvidia Launches Chat with RTX, an AI Chatbot You Can Run Locally on Windows
Satyam Kumar
View quick summary
Nvidia has released a new tool called Chat with RTX. This lets Windows PCs with Nvidia RTX 30 series or RTX 40 series GPU run an AI chatbot locally on their PC. Custom data can be given to the AI including documents, PDFs, and even YouTube videos. It includes the Mistral and Llama-2 open-source LLMs in the download. 16GB of RAM is also required and the GPU must have at least 8GB of VRAM capacity in order to work.
Nvidia has launched the new RTX 3050 6GB budget graphics card. It features a lower 6GB GDDR6 VRAM capacity, lower CUDA core count of 2304, and lower boost clo ck of 1.47 GHz. The total graphics power requirement is also lesser at 70W. The GPU does not have pricing and availability details yet but it has already been listed for sale. Nvidia's website has been updated with this GPU's existence.
Nvidia has released the RTX 4080 Super and now it is on sale. Various brands like MSI, Gigabyte, ASUS, PNY, and more have launched their lineup of new RTX 4080 Super graphics cards. The GPU costs $999 and is 2 to 5 percent more performant as compared to the non-Super RTX 4080.
One Redditor wants to make his RTX 3080 fanless. According to his post, the modder has previously been able to make something that works at 125W cooling capacity and was tested on an i7-10700K. His new project is quite aspirational, and uses several CPU coolers mounted on a copper block in a currently on-going attempt to cool down an RTX 3080 without any fans at all.
Nvidia has Tensor cores on its RTX graphics cards. These are used for various features apart from DLSS 3.5 and the AI frame generation tech released by the company in several games. As part of Nvidia's RTX Video enhancements, they had released RTX VSR before to upscale content. Now, with RTX Video HDR, users of any RTX graphics cards (20 series and above) can also make their content look like HDR. Nvidia touts it as an "AI-based HDR enhancement". Enabling RTX Video HDR is as easy as going into the Nvidia control panel.
Laptop-Based GeForce RTX 4090M & RTX 4080M Desktop GPUs Selling in China
Satyam Kumar
View quick summary
Nvidia RTX 4090M and RTX 4080M have been spotted on the website of China-based retailer Goofish. These are laptop GPUs, repurposed into desktop cards. Specialized companies, who are not official brand partners of Nvidia, have made the RTX 4080M and RTX 4090M GPUs with reduced VRAM and CUDA core count. The price for an RTX 4090M is quite high too, even higher than the desktop variant.
RTX 3050 6GB Specs and Release Date Leaked; New Budget GPU from Nvidia!
Satyam Kumar
View quick summary
Nvidia is expected to launch a new RTX 3050 graphics card with watered-down specifications this month. Basically, this RTX 3050 will cut-down the 8GB VRAM capacity to 6GB and come with other nerfs such as lower CUDA core count, memory bandwidth, GPU clock speeds, and power consumption. The new card is expected to launch in January 2024 at a sub-$200 price point, possibly aiming to become the next big budget-centric GPU.
At CES 2024, Nvidia launched new GeForce RTX 4080 Super, RTX 4070 Super, and RTX 4070 Ti Super graphics cards. The RTX 4070 Ti gets upgraded VRAM capacity, at 16GB GDDR6X. The other cards still feature the same amount of VRAM. The performance has been upgraded on all new Super cards, as compared to their non-Super variants. Nvidia has put more CUDA cores in all these GPUs. The new Super series starts from $599 and will be available starting Jan 17, 2024.
Nvidia is expected to release a cut-down version of the previously launched RTX 3050 graphics card. This new entry-level card will feature 6GB VRAM capacity instead of 8GB, as revealed in a new leak. Moreover, a lower TDP and clock speed for the RTX 3050 6GB has also been spotted in this leak.
China Gets an Exclusive Nvidia RTX 4090 D; Here’s How It’s Different
Satyam Kumar
View quick summary
RTX 4090 D has been launched in China. The GPU costs the same, but has lower amount of CUDA Cores. It also has slightly lower power at 425W, instead of 450W on the normal RTX 4090 which is now banned in the country. Nvidia has stated that RTX 4090 D is 5% less performant than the previous GPU in gaming and content creation. Also, the RTX 4090 D is not overclockable.
12 different RTX 40 Super GPUs have leaked by Videocardz. The new leak shows upcoming MSI GPU models including Suprim X, VENTUS 2X/3X, and Gaming X variants. The VRAM specification of RTX 4070 Ti, RTX 4070, and RTX 4080 Super has also been shown here. The RTX 4070 Ti Super will enjoy 16GB VRAM, while others remain at the same VRAM capacity as their 'non-Super' variants.
Nvidia RTX 4070 Super is an upcoming graphics card which is expected to be revealed at CES 2024, at NVIDIA's 'Special Address' event. However, one particular GPU model known as ASUS Dual RTX 4070 Super has leaked prior to its launch. The leak reveals a dual-fan design for the graphics card along with the same 12GB GDDR6X memory. This is also speculation on the 16-pin power connector. Will it be made of 12VHPWR or the safer, newer 12V-2X6 standard?