A prevailing — but incorrect — notion is that you need at least a $650 GeForce RTX 2080 card to get good performance with ray tracing enabled. That’s only true perhaps, if you’re the type that needs to game at ultra-high resolutions with absolute max image quality settings. However, at a FHD 1080p resolution, which is what the vast majority of mainstream gamers use, you may be surprised to learn that the lowest cost GeForce RTX card – currently the GeForce RTX 2060 — can get the job done quite well…

So what gives Adobe, Slack, Dropbox? And heck, while we’re at it, let’s throw some very popular mainstream, graphically-efficient game titles like Fortnite into the mix as well. The writing is on the wall with respect to Windows on Snapdragon and its always-connected, highly power-efficient advantages.

A few months back, I wrote about the MLPerf consortium and the release of its Inference v0.5 benchmark. MLPerf had previously disclosed some performance results from its Training v0.6 benchmark, but training is only part of the machine learning equation. It is when the training process is complete and weightings have been assigned to the dataset that a neural network can intelligently infer things from that data — this process is what is referred to as inference.

For the past couple of years, Intel has been uncharacteristically vocal about its plans to enter the discrete GPU market, to take on graphics stalwarts AMD and NVIDIA in both consumer PCs and the data center. Historically, entities like Intel have kept details of unreleased, forward-looking products hush-hush until they are much closer to being introduced.

To date, LG U+ has introduced a GeForce NOW trial in Korea, in association with a mobile subscription plan, and Softbank recently kicked-off pre-registrations in Japan for free beta that is slated to launch this winter. And today at the IgroMir Expo, a large video gaming event currently underway in Moscow, SAFMAR Group introduced the GeForce NOW service in Russia.

As part of its evolving strategy, Silicon Valley bellwether Xilinx has been integrating this adaptive technology into platform accelerator solutions for machine learning, as well as domain specific architecture solutions that incorporate various compute resources like ARM cores, high speed IO and even RF functions.

For some time now, Qualcomm has been evangelizing the idea that a next generation 5G wireless rollout will arrive significantly faster than previous generation 4G networks did back in the day. Today, putting its money where its mouth is, the company has made a number of announcements at the IFA show in Berlin that not only underscore the depth and breadth of its commitment to 5G, but also help shape the mainstream 5G landscape…

With the rapid proliferation of wirelessly connected smart devices, the need for more advanced Wi-Fi networks that offer greater capacity, reliability, range, and performance has never been greater. To that end, a number of players have announced bleeding edge 802.11ax, or Wi-Fi 6, chipsets for various networking applications, including Qualcomm, which just unveiled an extensive array of Pro Series platforms that target a number of different performance segments.

If there was any doubt in your mind that real-time ray tracing was the wave of the future for cutting-edge gaming and 3D graphics, one of the most popular game titles of all time, Minecraft, just got a full scene RT makeover and it looks gloriously good. Mojang, Microsoft and NVIDIA just announced in a Gamescom 2019 unveil that full-scene, path-based ray tracing will be made available in Minecraft

Google open-sourced BERT so that others could train their own conversational question answering systems. And today, NVIDIA announced that its AI compute platform was the first to train BERT in less than an hour and complete AI inference in just over 2 milliseconds.

At the Flash Memory Summit (FMS) currently underway at the Santa Clara Convention Center in Santa Clara, California, Xilinx is showing off a brand new member of its adaptable data center accelerator family, the Alveo U50. The Xilinx Alveo U50 is built on the company’s UtlraScale+ FPGA architecture, and targets scale-out and domain-specific acceleration of a multitude of data center workloads.

Intel just announced the second FPGA Programmable Acceleration Card (PAC) in its line-up, the Intel FPGA PAC D5005. The Intel FPGA PAC D5005 is designed for computing intensive workloads, like streaming analytics, media transcoding, financial analysis, and network security and will initially be offered in HPE ProLiant DL380 Gen10 servers.

It's not often you hear of anyone putting a stake in the sand claiming a 1000X improvement in anything, but that's exactly what Intel Chief Architect and SVP of the Architecture, Software and Graphics group, Raja Koduri did at the company's CREATE event at SIGGRAPH 2019 yesterday.

Today, NVIDIA announced multiple new NVIDIA RTX Studio laptop and mobile workstation design wins, ten in total, from manufacturers like Dell, Lenovo, HP, and BOXX, that target an array of market segments.

NVIDIA isn't taking its collective foot off the gas when it comes to its gaming GPU line-up, and this morning the company has officially launched the new GeForce RTX 2080 Super graphics card. Fleshing out the upper-end of its new Turing-based GPU lineup for gamers…

Qualcomm's Snapdragon 855 Mobile Platform powers the vast majority of flagship Android devices currently on the market, and today the company has announced a performance-optimized version of its top-end mobile SoC, dubbed Snapdragon 855 Plus.

On board the Snapdragon 855 platform lies key, previously untapped technology in silicon that will enable a new level of security compliance and features, along with more flexible and convenient multi-carrier connectivity in the future…

Another Silicon Valley bellwether just made the jump to 7nm semiconductor process technology and this time the cutting-edge chip tech is being employed on new breed of SoC (System On Chip) that can adapt to all kinds of compute and machine learning applications.

NVIDIA used the massive Computex Information and Communication Technology event currently taking place in Taipei to make a number of key disclosures. 

With Computex rapidly approaching, we had the chance to sit down and chat with one of the keynote speakers, Gregory M. Bryant, Senior Vice President and General Manager of the Client Computing Group at Intel Corporation, to discuss some of the company’s plans for the show…