Visiting Bitspower at CES or Computex is a real adventure of sorts in terms of not knowing what to expect. Last year, for example, we were greeted to LN2 pots of all things, in addition to new blocks and fans that made their way to the retail market over the subsequent months. It was also interesting to take a look back at the concept of the Premium Summit M CPU block from last year, knowing what it turned out now. This year, the company hosted us again at CES and had a large hotel suite full of products- refreshes, and new alike. We take a look at their offerings here, and be sure to read past the break for more.

The very first thing we saw was a distribution plate/case hybrid, which Bitspower wants to bring out as a test bench case later this year. There have been other such executions come out recently, including some at CES from other companies, and this is a natural extension of the distro plate development that has caught the mind of several custom watercooling companies for one reason or another. This particular version uses two Xylem DDC pumps with Bitspower heatsinks, and has four large feet to prop up the finished build horizontally for testing/display. Bitspower expects to introduce the finalized version at Computex later this year, and are seeking feedback from our readers on what features they would like to see here.

Bitspower, a leading supplier of water-cooling equipment for performance computers, has announced the release of their latest Intel CPU block, the Bitspower Touchaqua CPU Block Summit MS OLED for Intel Platform. This product combines the best of both performance and aesthetics to give you an ideal cooling solution for LGA 115x and LGA 2066 socket CPUs. In addition to a CNC-machined block top made of hardwearing acrylic and base made of high-quality copper, the block features a digital thermal sensor and a bright OLED display so you can monitor your water-cooling loop temperature in real time, as well as dazzling DRGB lighting.

The block incorporates a single piece of wiring that runs power to both the DRGB lighting, as well as the temperature gauge - ensuring that you can keep the design of your build as clean as possible. Furthermore, the water block's RGB LED Strip has been certified by ASUS AURA Sync, GIGABYTE RGB Fusion, MSI Mystic Light Sync, ASRock Polychrome and Razer Chroma-so it's highly compatible with a range of setups. Bitspower Touchaqua range offers the same quality as Bitspower's regular product lines, but features a range of unique functions and designs.

Having the world's most pervasive operating system (or office suite) is sure to leave a big mark on any company when it comes to exploitation attempts from hackers. It's a simple equation: aim your efforts at a software that runs in millions (if not billions) of machines and even a light chink in the armor could be enough to cause a cascading effect through that many users.

This principle applies to almost everything: a small effect across a billion users usually provides greater returns than a large effect on one or two players. Kaspersky labs on its security report, presented at the Security Analyst Summit, reported that the favorite target for cyber attacks was Microsoft's Office suite - a 70% figure suggests an incredible attention given to Office, really. These Office-related cyber attacks don't directly relate to the suite itself; there are other, OS-integrated components that can be targeted, or simply that Office file extensions are used as clever, headache-inducing ways of disguising malware as the second greatest evil in the world - spreadsheets.

With its 9th generation Core processors, Intel is re-introducing soldered IHS (integrated heatspreaders), at least in its top two premium models, the Core i9-9900K, and the Core i7-9700K. Intel refers to this feature as STIM (soldered thermal interface material). AMD implements soldered IHS across its Ryzen "Summit Ridge," "Pinnacle Ridge," and Threadripper families. XFastest took apart an i9-9900K to confirm that Intel is indeed using solder. Soldered IHS is generally preferred for better heat-transfer characteristics, compared to fluid TIMs. The use of fluid TIMs prompts some serious enthusiasts to even "de-lid" (run their processors without the IHS).

The 8-core "Whiskey Lake-S" die could be around 178 mm² in area, with the addition of two more cores, and 4.5 MB more cache (L2 + L3), over its predecessor. You'll recall that the 6-core "Coffee Lake" die measures 150 mm², a 25 mm² gain over the 4-core "Kaby Lake" die. We aren't expecting Intel to change the iGPU or uncore components. Intel is building these dies on the same 14 nm++ silicon fabrication node as "Coffee Lake," with the only architectural difference being silicon-level hardening against certain security vulnerabilities.

Yangtze Memory Technologies Co., Ltd (YMTC), a new player in the NAND industry, will be joining Flash Memory Summit this year for the first time, delivering a much-anticipated keynote address to reveal its ground-breaking technology - Xtacking. YMTC is the first Chinese company to take part in the high-entry-barrier NAND flash memory industry with its new architecture for unprecedented performance, higher bit density, and faster time-to-market.

Simon Yang, YMTC CEO, will deliver a keynote address, Unleashing 3D NAND's Potential with an Innovative Architecture, on August 7th, from 3:00 p.m. at the Mission Ballroom in the Santa Clara Convention Center, where he will illustrate how the company's new technology can increase NAND I/O speed up to DRAM DDR4 while delivering industry-leading bit density, marking a quantum leap for the NAND market.

We previously covered in more depth the fact that the US was gearing up to overtake China's Sunway TaihuLight, then the world's fastest supercomputer, with its Summit machine, built in collaboration between IBM (with its water-cooled Power Systems AC922 nodes with 24-core processors and 96 processing threads) and NVIDIA (GV100 GPUs).

Now, this US dream has finally come to pass, and in a big way - the Summit delivers more than double the performance of China's posterchild, coming in at 200 PetaFLOPs of computing power. Summit boasts of 27,648 Volta Tensor Core GPUs and 9,216 CPUs within its 5,600 square feet. The Summit supercomputer consumes 15 MW of power (the site where it's deployed is able to deliver up to 20 MW), which is on-par with China's Sunway - but remember, it more than doubles the peak PetaFlops from 93 to 200. A good step in the battle for supercomputer supremacy, but China still has an increasing foothold in the number of systems it has employed and registered with the TOP500.

At Adobe Summit, Adobe and NVIDIA today announced a strategic partnership to rapidly enhance their industry-leading artificial intelligence (AI) and deep learning technologies. Building on years of collaboration, the companies will work to optimize the Adobe Sensei AI and machine learning (ML) framework for NVIDIA GPUs. The collaboration will speed time to market and improve performance of new Sensei-powered services for Adobe Creative Cloud and Experience Cloud customers and developers.

The partnership advances Adobe's strategy to extend the availability of Sensei APIs and to broaden the Sensei ecosystem to a new audience of developers, data scientists and partners. "Combining NVIDIA's best-in-class AI capabilities with Adobe's leading creative and digital experience solutions, all powered by Sensei, will allow us to deliver higher-performing AI services to customers and developers more quickly," said Shantanu Narayen, president and CEO, Adobe. "We're excited to partner with NVIDIA to push the boundaries of what's possible in creativity, marketing and exciting new areas like immersive media."

China has been increasingly - and steadily - gaining relevance in the supercomputing world, with most of the top-500 entries being controlled by that country. In fact, China can boast of having the number one supercomputer in the world, the Sunway TaihuLight, which can deliver 93 PetaFLOPS of computing power - just 3x more computational power than the second most powerful machine, China's own Tianhe-2). However, supercomputing, and the amount of money that's earned by selling processing slices of these supercomputers for private or state contractors, i a very attractive pull - especially considering the increasingly more expensive computational needs of the modern world.

The Summit is to be the United State's call to fame in that regard, bringing the country back to number one in raw, top-of-the-line single-machine supercomputing power. Summit is promising to more than double the PetaFLOPS of China's TaihuLight, to over 200 PetaFLOPs. That amounts to around 11x more processing grunt than its predecessor, the Titan, in a much smaller footprint - the Titan's 18,688 processing nodes will be condensed to just ~4,600 nodes on the Summit, with each node achieving around 40 TeraFLOPS of computing power. The hardware? IBM and NVIDIA, married in water-cooled nodes with the powerful GV100 accelerator that's still eluding us enthusiasts - but that's a question for another day.