It is no surprise that the performance of modern AI accelerators is constantly increasing. Nor is it a surprise that they are consuming ever more absurd amounts of energy and mercilessly pushing the thermal limits of the hardware. The solution? For NVIDIA, it’s “liquid cooling or nothing”. With the Blackwell Ultra GB300 AI [...]
AMD has finally thrown down the gauntlet with the Radeon RX 9070 and 9070 XT, taking direct aim at the recently launched GeForce RTX 5070 from NVIDIA. While the RTX 5070 has to make do with 12 GB of GDDR7 memory, AMD consistently relies on 16 GB of GDDR6 and prices both cards at a competitive 629 and 689 euros RRP respectively, although the prices [...]
Yes, what have we here? With the GeForce RTX 5070 FE, NVIDIA has now launched a new graphics card on the market that leaves me somewhat perplexed in many respects. I’m deliberately putting this up front, because you have to read today’s review carefully and also get involved in the subtleties, which I will highlight more visually this [...]
The MSI RTX 5070 Ti Gaming Trio presented today offers a balanced mix of performance, cooling and design and positions itself as the ideal middle ground between the simple Ventus series and the high-end Vanguard variant. Thanks to the powerful Tri-Frozr 3 cooling system with three fans, the card ensures low temperatures and low noise levels [...]
With the RTX 5070 Ti, NVIDIA has now launched a new graphics card on the market that is intended to position itself in the mid-range and promises both performance and modern features. The MSI RTX 5070 Ti Ventus 16 GB is available for me to test today (the better OC models will be available tomorrow). Incidentally, the artificial name [...]
The fact that NVIDIA likes to establish its own hardware standards is nothing new. Now the company also appears to be taking on the memory market for compact AI supercomputers. According to South Korean reports, Team Green is working with Samsung, SK Hynix and Micron on a new memory module called SOCAMM – smaller, more efficient and of [...]
DeepSeek has caused quite a stir in recent months. From an ambitious side project, the company has developed into a serious competitor in the field of Large Language Models (LLMs). But that is apparently not enough. According to DigiTimes, the company is now eyeing its own AI chips. That sounds like a lot of work – and an even bigger [...]
With the latest patches, the first details of the upcoming AMD Instinct MI400 have emerged. The new architecture is based on a modular design with two separate interposer dies, each of which can accommodate four XCDs (Accelerated Compute Dies). Structural innovations: Dual interposer and dedicated multimedia I/O tiles As can be seen from the [...]
Today’s article offers a comprehensive review of the GeForce RTX 5080 and builds on the launch article for the RTX 5090, which I will of course also link to below. But I’ll save myself redundant content and concentrate more on the card. Various performance values have already been reported and leaked and it will certainly be more [...]
It seems that the Chinese AI industry is continuing to make great strides. Currently, AI company DeepSeek is making headlines as its latest language model, the R1, is reportedly running inference workloads on Huawei’s Ascend 910C AI chips. This news not only highlights China’s progress in the AI sector, but also raises interesting [...]
According to the latest shipping documents, NVIDIA is apparently preparing a new generation of workstation graphics cards under the name “RTX Blackwell”. One of the variants is said to be equipped with a whopping 96 GB of GDDR7 memory and a 512-bit memory connection. Double the memory capacity for professional applications Until now [...]
Today’s article offers a comprehensive review of the GeForce RTX 5090 and the underlying NVIDIA Blackwell architecture. As it is a completely new architecture, it is divided into a theoretical analysis of the new technical developments and a practical review that focuses on benchmarks, real applications and user experiences. In the first [...]
New rumors suggest that AMD will rely on TSMC’s advanced N3E manufacturing process for its upcoming Ryzen “Zen 6” CPUs and Radeon “UDNA” GPUs. In addition, AMD is expected to use 3D stacking technology in the next Halo and console APUs to further increase performance. Zen 6: Medusa Ridge with N3E technology The next [...]
NVIDIA has officially announced an updated version of the DLSS frame generation for the GeForce RTX 40 series. This optimization results in lower VRAM consumption and improves overall performance. Originally introduced with DLSS 3.0, the frame generation has been continuously developed and reaches a new level of efficiency with the latest version [...]
The new generation of AMD Ryzen AI processors marks a significant step in the development of processors specifically designed for use in AI-based applications. With a combination of advanced architecture, optimized performance cores and integrated AI accelerators, AMD is setting new standards in the fusion of conventional processor features and [...]
NVIDIA, as a leading company in the field of semiconductor technology, is increasingly focusing on robotics. In the face of growing competition in the core business of AI chip production, the company is planning to develop this area as a new growth market. The launch of the next generation of compact computers for humanoid robots, under the name [...]
The Intel Arc B580 12GB is not a workstation graphics card, I know, but I’ve decided not to chase after the 1001st supplement to the 1000 gaming benchmarks and YouTube videos that have already been published, because the added value is limited. For me, the readers and also Intel’s driver team, who should add a few things to their [...]
NVIDIA is apparently planning to present several new technologies at CES 2025 that are specifically geared towards the use of artificial intelligence (AI). This information was revealed in an announcement by Inno3D, a partner company of NVIDIA. The presentation will presumably focus on developments in the field of neural rendering and advanced [...]
NVIDIA has announced that it will bring forward the release of its next generation of chips, the Rubin architecture, by around six months, reports Taiwan Economic Daily. Originally planned for 2026, the new technology will now be launched on the market in the second half of 2025. With this decision, the company wants to respond to the increasing [...]
TensorWave, a cloud service provider focused on AMD’s AI technology, has announced that it is developing GPU clusters based on AMD’s Instinct MI300X, MI325X and MI350X accelerators. The aim is to provide a scalable computing infrastructure with an energy requirement of up to one gigawatt. The planned clusters are based on AMD’s [...]
