All articles tagged with #ada lovelace
NVIDIA confirms DLSS 5 will be ported to the RTX 40-Series after optimizing performance for the RTX 50-Series; DLSS 5 officially launches with NBA 2K27 and is currently supported on RTX 50-series GPUs and GeForce NOW, while Ada Lovelace support is being optimized and a broader RTX 40-Series port will follow once tuning is complete.
A community patch swaps in Ada‑friendly binaries to run NVIDIA's DLSS 5 on RTX 4000 series GPUs despite no official support. The patch, developed after a leaked DLSS 5 tool and reported by Tom's Hardware, reportedly works on an RTX 4080 and is being tested by others in Discord. NVIDIA has not announced formal support for 4000‑series DLSS 5, and compatibility with older architectures (like Ampere) remains uncertain as the project is in early stages.
Nvidia has launched the RTX 500 and 1000 Ada Generation laptop GPUs, based on the Ada Lovelace architecture, to provide AI acceleration and graphics performance for professionals working on the go. These GPUs are designed to meet the increasing demand for potent mobile solutions, incorporating generative AI and graphics performance for tasks such as video conferencing, streaming, and content creation. The new GPUs will be featured in upcoming mobile workstations from global manufacturing partners and are set to be available this spring.
NVIDIA's next-generation gaming GPUs, known as Ada Lovelace-Next, are expected to launch in 2025 according to a recently revealed roadmap. The successor to the Ada Lovelace Gaming GPUs, labeled as Ada Lovelace-Next, will come a year after the Hopper-Next GPUs, which are set to launch in 2024. In the meantime, NVIDIA may offer refreshes to its current product lineup to keep things interesting for gamers. Additionally, NVIDIA's Grace Next CPU architecture utilizing Arm cores is also expected to be announced in 2025.
NVIDIA has confirmed that its GeForce RTX 4060 8 GB graphics card will be launched on June 29th for $299 US. The card offers a 20% improvement over the GeForce RTX 3060 12 GB in traditional rasterization and bigger gains when RT and DLSS are factored in. It features 3072 CUDA cores and 8 GB of GDDR6 memory running at 17 Gbps memory speeds across a 128-bit bus interface. The card will be available in various custom models at launch but there will be no Founders Edition variant for it.
NVIDIA's smallest Ada Lovelace GPU, the AD107-400, has been pictured and will be powering the entry-level and mainstream GeForce RTX 4060 graphics card. The chip measures 146mm2 and is 27% smaller than the GA107 and TU117 GPUs. The GeForce RTX 4060 is expected to feature 3072 CUDA cores, 8 GB of GDDR6 memory, and a 128-bit bus interface. It will have a 35% lower TGP than its predecessor, the RTX 3060 12 GB, and will be priced at $299 US.
Nvidia's stock rose 5% to reach its highest price since December 2021, as the company announced a new line of high-performance gaming cards aimed at budget shoppers. The "Ada Lovelace"-architecture RTX 4060 Ti 8GB gaming card will be available for $399 on May 24, while the RTX 4060 Ti 16GB version will be available in July starting at $499. The plain RTX 4060, starting at $299, will also be available in July. Nvidia's shares have more than doubled in the past eight months and are 117% year-to-date, outperforming the S&P 500 and Nasdaq Composite Index.
The NVIDIA RTX 4070 graphics card, featuring 5,888 FP32 cores and a bandwidth of 504 GB/s, offers up to 40% more performance than the RTX 3070 despite having the same core counts. The larger L2 cache and higher memory bandwidth are the only differences between the two 70-class graphics cards. In real-life scenarios, the RTX 4070 offers roughly RTX 3080 level of performance, demonstrating the generational uplift that NVIDIA was able to bring without changing the shader strength.
NVIDIA showcased the Opacity Micro Maps feature of the Ada Lovelace architecture, which can boost ray tracing performance by reducing the number of shader interactions with rays. Opacity Micro Maps can be used to speed up ray tracing performance whenever rays hit irregular objects, such as particles with empty spaces. The feature can also be used to turn a simple triangle into a leaf, resulting in a large gain in ray-traced performance. Game developers can check out the Opacity Micro Maps SDK available on GitHub to implement this feature.
NVIDIA has announced a wholesale update of their professional visualization product stack, including five new SKUs for professional laptops and a low-profile, small form factor optimized desktop RTX 4000 Ada Generation card. The laptop lineup includes the RTX 5000 Ada Generation Laptop GPU, which is analogous to the consumer RTX 4090 for Laptops, and the RTX 2000 Ada Generation for laptops, which is analogous to the consumer RTX 4060 for Laptops. The desktop RTX 4000 SFF Ada Generation is based on NVIDIA’s AD104 GPU and has a TDP of 70W.
Nvidia has released the RTX 4000 SFF graphics card, which uses the Ada Lovelace architecture and is designed for professional applications such as CAD, graphics design, AI applications, and software development. It has 20GB of GDDR6 memory, 6,144 CUDA cores, 192 Tensor cores, and 48 ray tracing cores, but is limited to 70 watts of power due to its small form factor. The card is not recommended for gamers and will be sold for $1,250 by PNY and Ryoyo Electro starting in April. Nvidia is also bringing the Ada Lovelace architecture to laptop creators.