Intel Skylake-X and Kaby Lake-X CPUs, Motherboards, and Coolers: Your Complete Guide

intel corex badges intel intel core x

For nearly ten years, Intel’s Core i3, i5, and i7 brands have held strong in the PC and laptop space. Everyone immediately gets the idea of a hierarchy, with Core i3 at the entry-level and Core i7 at the top end. You don’t necessarily need to know that you get two cores and HyperThreading with an i3, and four cores without HyperThreading if you choose an i5. You don’t need to know clock speeds, cache sizes, or manufacturing processes – what matters is that when all capabilities are taken into consideration, moving from a lower number to a higher one gives you an incremental boost in performance. Within each generation, all chips work on the same motherboards, use the same kind of RAM, and have the same architectural advantages – with one key exception.

At the very high end, there are Intel’s High-End Desktop (HEDT) models, awkwardly clubbed into the Core i7 family. These are often called Extreme Edition chips, though not all of their model numbers actually have that suffix. These CPUs need a different platform altogether, lack integrated graphics, and are based on a souped-up version of the previous-generation architecture. For example, the Core i7 6950X was codenamed Broadwell-E, which is a derivative of the 5th-gen Broadwell architecture, whereas the rest of the 6th generation was codenamed Skylake.

While you can still only get up to four CPU cores with the top mainstream Core i7 after all these years, Intel has pushed out six-, eight-, and ten-core models under the HEDT Core i7 umbrella over the years, and has charged enormous amounts of money for them. In the beginning, it was rumoured that Intel would use the name Core i9 for these products since they were so obviously different from the rest of the lineup, but that did not come to pass. They were called Core i7 even though they didn’t really fit that name.

Adding to the confusion, HEDT generations have been launching later and later each year. In our review of the 10-core Intel Core i9-6950X, we pointed out how the much cheaper quad-core Core i7-6700K outclassed it in some tests. Broadwell-E, technically based on the 5th-gen architecture, launched at around the same time as the 7th-gen Kaby Lake lineup.

Now, in 2017, Intel also needs to fend off new competition from a resurgent AMD, which is now selling octa-core Ryzen 7 CPUs and gleefully pointing out how they outperform Intel’s latest and greatest while costing a fraction of the price. With Ryzen Threadripper shipping soon, clearly, Intel has a lot of cleaning up to do.

Introducing the Intel Core X-series
Finally, we have a Core i9 brand to distinguish HEDT CPUs from the mainstream. Unfortunately, there are a few new layers of complication to replace the old ones. For starters, all new Core i9 models fall under what Intel is calling its new Core X-series, which also includes a few Core i7s and one Core i5 model. Core X clearly isn’t about superior performance. Core X also mixes two Intel architectures, Skylake-X and Kaby Lake-X (Intel swapped out the -E suffix, but the idea is the same). The two CPU lines are referred to collectively by the code name Basin Falls.

The new Core i5-7640X and Core i7-7740X are based on Kaby Lake-X, while the other two Core i7s and all Core-i9 models are based on Skylake-X. All Core-X CPUs use the new LGA2066 socket and X299 platform controller – which means that not only is the Core i7 family just as fragmented as before, the Core i5 family is too. Core i9 encompasses a few new parts and a few renamed ones, and Intel hasn’t improved its segmentation at all in the bargain. The only advantage of this is that the company gets to leapfrog the generation-lag issue and use a shiny new name against AMD’s Threadripper when it comes out.

As for the differences between Skylake-X and Kaby Lake-X, things can get quite confusing, but we’ll try breaking it all down. Skylake-X is what we’d usually consider the “extreme” CPUs for any generation There are seven models in all, starting with the six-core i7-7800X and going up to the top-of-the-line 18-core Core i9-7980XE. Models with up to 10 cores are already available, and the rest are set to begin shipping between August and October this year.

Last year’s top-end CPU, the Core i7-6950X had only 10 cores. It was initially rumoured that we’d get a 12-core model this year, but now we’re going all the way up to 18 cores. The higher-end Core i9 chips are likely derived from Intel’s Xeon designs, and should they serve to counter AMD’s 16-core Threadripper quite nicely. Intel hasn’t released complete specifications for these unreleased models, but here’s what we do know about the Core i7-7800X and 7820X, and the Core i9-7900X .

The Core i7-7800X and i7-7820X have six and eight cores respectively and can work with 28 PCIe lanes, just like last year’s i7-6800K. One of the biggest criticisms of that CPU was that its PCIe lanes crippled compared to the rest of the lineup which had 40 lanes. You can’t use multiple high-end graphics cards, SSDs, and other peripherals with only 28 lanes of connectivity, but it seems that Intel still sees a reason to segment its chips this way. The Core i9-7900X has 44 lanes, and it’s possible that the more expensive chips will have more. The Core i9-7900X is the least expensive model to support Turbo Boost Max 3.0, which is Intel’s scheme for raising the speed of one single core beyond the rated boost level to benefit single-threaded workloads. All Core i7 and i9 Skylake-X CPUs support quad-channel RAM, though the maximum speed varies.

On the other hand, Kaby Lake-X doesn’t seem high-end at all. The Core i7 model tops out at four cores, and the i5 model doesn’t even do HyperThreading. They are for all practical purposes part of their respective mainstream Core families except that they have higher TDP ratings and can be overclocked further. You’re limited to the mainstream 16 PCIe lanes and dual-channel DDR4 RAM, and it’s important to note that the integrated graphics silicon has been bumped off. You do get all the architectural improvements that went into Kaby Lake, but its biggest improvements over Skylake were improved integrated graphics and better battery life, which aren’t applicable here.

Name Architecture Cores/
PCIe 3.0 lanes Base clock Turbo clock Cache TDP RAM
Core i5-7640X Kaby Lake-X 4/4 16 4.0GHz 4.2GHz 6MB 112W 64GB, Dual-channel DDR4
Core i7-7740X 4/8 4.3GHz 4.5GHz 8MB
Core i7-7800X Skylake-X 6/12 28 3.5GHz 4.0GHz 8.25MB 140W 128GB, Quad-channel DDR4
Core i7-7820X 8/16 3.6GHz 4.3GHz 11MB
Core i9-7900X 10/20 44 3.3GHz 4.3GHz 13.75MB
Core i9-7920X 12/24 TBC 2.9GHz TBC 16.5MB TBC TBC
Core i9-7940X 14/28 TBC TBC
Core i9-7960X 16/32
Core i9-7980XE 18/36

The motherboard story
It gets even more complicated when you get to factoring in the motherboards that pair with these CPUs. There is only one platform controller (ie chipset), called the X299, but since so many of the platform’s capabilities depend on the CPU, anyone building a PC without understanding exactly what is what will have problems. Motherboards have to have quad-channel RAM slots to support Skylake-X, but half of them will be useless if you have a Kaby Lake-X CPU. Similarly, a motherboard might have 44 lanes’ worth of PCIe and M.2 slots, which will just not be usable if your CPU that can’t provide that level of connectivity.

Users with less capable CPUs at their disposal will have to carefully select a board that has their preferred distribution of PCIe and M.2 slots, or at least allows for lanes to be allocated differently. You could choose to spend less on a lower-end Kaby Lake-X CPU first and then upgrade to Skylake-X down the line, but then you will have to get new RAM and components that take advantage of the differences, pushing costs way higher.

Motherboard manufacturers have to choose between supporting all Basin Falls CPUs and fragmenting the market. They will also have to carefully document which combinations of features users can expect with 16, 28, or 44 PCIe lanes, and support the inevitable flood of angry customers who didn’t get exactly what they wanted. Some companies are experimenting with LEDs that tell users exactly which slots will work based on the CPU currently inserted, and we expect to see more creativity in this area. On the other hand, Intel does allow for X299 boards to ship with only dual-channel RAM slots for people who want Kaby Lake-X without the costs of supporting Skylake-X. Such boards will launch later this year, and anyone who buys one is foregoing the Skylake-X upgrade path.

The X299 platform controller itself has 24 PCIe lanes of its own, which are distributed between USB, SATA, Ethernet, and other IO devices at the motherboard maker’s discretion. You can have up to eight SATA 3.0 ports, 10 USB 3.0 ports, multiple Gigabit Ethernet ports, and other onboard goodies such as Wi-Fi, all without touching the bandwidth used by the CPU for graphics and storage. These 24 lanes will exist regardless of the CPU in use, but it is up to motherboard vendors to publish details of the topology they design for their products.

One final little complication has to do with a feature that’s rather niche: VROC, or Virtual Raid on CPU, which lets X299 users create ridiculously fast bootable RAID 0 out of a large number of ultra-fast NVMe SSDs. Asus has demonstrated an array of eight SSDs achieving 12GBps sequential reads. NVMe uses the PCIe bus, which means these SSDs are connected directly to the CPU, either using M.2 slots, U.2 ports, PCIe lanes or adapters for any of these interfaces. With enough PCIe lanes originating from the CPU, a system can bypass the platform controller entirely and there’s no need for PCIe splitters which share bandwidth and introduce minor latency. This, of course, means that you need Skylake-X for its PCIe lanes.

You need to use specific SSDs and, for now, it appears that only some of Intel’s own SSDs are compatible – but you’ll also need to buy an extra hardware dongle from Intel that snaps onto the motherboard. There’s a standard mode dongle which enables RAID 0/1/10, but if you want RAID 5 there’s a premium mode dongle – the costs of each are unknown. This feature isn’t likely to have mass appeal, but it’s frustrating to have to pay extra for it on top of the cost of a top-end Skylake-X CPU and motherboard.

Basin Falls introduces a new CPU socket called LGA-2066, upping the number of contact points from 2011 for the first time since 2011’s Sandy Bridge-E generation. This isn’t a surprise, given the increase in PCIe lanes and a higher-speed interconnect between the CPU socket and chipset. Thankfully though, the socket is similar in size to LGA-2011-v3 so you can reuse any cooler that is compatible with the previous HEDT generation.

High-end air and liquid coolers can be really expensive, and they cannot become obsolete unless mounting mechanisms change or Intel changes specifications like the amount of clearance that motherboards need to have around the CPU socket. This is the reason that CPUs of this class don’t come with their own coolers. Some companies offer aftermarket adapter kits for older models, and Noctua famously has a policy of giving adapters away for free to anyone who asks for them, so you might be in luck with even a very old model.

Current Kaby Lake-X and Skylake-X CPUs have TDP ratings of 112W and 140W respectively. You will want a beefy air cooler with copper heat-pipes and lots of surface area at the very least, and liquid cooling would be a very good idea if you plan to overclock these chips – which is one of the most appealing reasons to buy them.

Final words
If you feel as though a regular Core i7 CPU isn’t going to cut it and you need the power of a Core-X model, you’re going to have to choose your CPU and motherboard very carefully. Do you want to run two or more graphics cards in PCIe x16 slots? Do you want multiple fast NVMe SSDs? Do you need as much memory bandwidth as possible? All these factors need to be considered.

Costs can quickly become prohibitive – the Kaby Lake-X Core i5-7640X and i7-7740X cost around Rs. 20,000 and Rs. 27,000 respectively, and the Core i9-7900X is currently selling at just over Rs. 80,000 (that’s a far sight better than the Rs. 1,69,000 that last year’s top-end Core i7-6950X launched at, and we have AMD to thank). Then add on the cost of a decent X299 motherboard (Rs. 25,000 and up), a quad-channel RAM kit (at least Rs. 30,000 for 32GB), and an AIO cooler (Rs. 8,000 onwards) plus a good enough power supply, case, SSD and graphics card.

So should you actually spend so much money, and should you choose Skylake-X or Kaby Lake-X? That depends entirely on what kind of performance you need out of your PC. Kaby Lake-X CPUs are just slightly faster versions of their Kaby Lake counterparts with higher thermal headroom and better power delivery for overclocking. If you don’t need multiple cores, these are the fastest clocks you can get and they aren’t terribly expensive. The Skylake-X models that we know of right now are aimed at serious multi-taskers (or “mega-taskers” as Intel likes to call them) and creative professionals for whom time more valuable than money.

With Ryzen Threadripper right around the corner and very little yet known about the 14-, 16- and 18-core Skylake-X models, the high-end PC space is just about to get exciting. Stay tuned for our reviews of the new quad-core Kaby Lake-X Core i7-7740X and 10-core Skylake-X Core i9-7900X, coming up soon.

Toshiba to Build Chip Plant Without Partner Western Digital

Toshiba to Build Chip Plant Without Partner Western Digital

Toshiba is moving forward with plans to build a new memory chip plant without partner Western Digital, another escalation of the fight over the future of their joint venture.

The electronics maker will spend JPY 195 billion ($1.8 billion or roughly Rs. 11,275 crores) on construction of Fab 6 of its Yokkaichi semiconductor facility in western Japan, the Tokyo-based company said in a statement Thursday. Toshiba, which owns the land, buildings and the production know-how at the factory, has split investments in production equipment with SanDisk since the joint supply venture started in 2004. Western Digital acquired SanDisk last year.

Toshiba and Western Digital are locked in a legal fight over Toshiba’s plan to sell its share of the business to make up for multibillion-dollar losses in its nuclear power operations. Western Digital argues that it has a say in the sale, as well as right of first refusal. Further legal wrangling could delay the sale to a group of preferred bidders, putting Toshiba at risk of being delisted.

Western Digital, based in San Jose, California, needs to retain access to output from new Toshiba factories as improvements in manufacturing technology are one of the key determinants of success in the memory chip industry. Newer plants and equipment typically produce better semiconductors more cheaply.

“If you keep going down the path and you play to its finality it may not be a good path for either company,” said Amit Daryanani, an analyst at RBC Capital. “The hope would be cooler heads prevail.”

Toshiba said the two sides are far apart on the new project.

“Toshiba has held discussions with SanDisk over several weeks, but could not arrive at an agreement because of the vast difference in opinions over capital spending,” said Kaori Hiraki, a spokeswoman for Toshiba. “We need to boost our production capability to meet increasing demand for Nand flash.”

Toshiba said it is spending JPY 15 billion more than originally planned, but going it alone won’t impact production or development. Installation of fabrication equipment to produce so-called 3D Nand flash will begin in December 2017 and output will ramp up to 90 percent of capacity in the fiscal year ending March 2019, Toshiba said.

“While we are disappointed by Toshiba’s announcement, the agreements governing the JVs give us the right to participate in investments,” Western Digital said in a statement. “That is exactly what we intend to do.”

Shares of Toshiba closed 0.8 percent lower in Tokyo Thursday, paring losses prior to the announcement. The stock is down 8.5 percent this year. Western Digital’s stock was little changed at $84.53 in New York trading.

Toshiba clinched a preliminary agreement in June to sell its memory chip unit to a group led by the Innovation Network of Japan, Bain Capital and other investors. The consortium of preferred bidders is offering JPY 2.1 trillion, people with knowledge of the matter have said.

Western Digital in May invoked an arbitration clause in the business agreement, seeking to block Toshiba’s transfer of ownership of the unit to a separate legal entity in preparation for a sale. Toshiba, which has since reversed that transfer, had its lawyers send a letter demanding that the US company stop its “harassment” as it seeks to sell the business.

The legal spat is threatening the very existence of the venture, according to people familiar with the matter. The breakup would increase the financial burden on Toshiba and reduce cost advantages that come with scale. It would also deprive Western Digital of access to advanced chips necessary to compete in the storage business.

If Toshiba is forced through arbitration to sell its stake in their venture to Western Digital, the move could trigger the dissolution of their legal partnership, according to regulatory filings. That would cancel the supply agreement under which the US company gets chips and make it the owner of only some equipment, according to people close to Toshiba and the terms of their relationship. That equipment would be useless without other machinery and the plant itself, which would remain Toshiba’s property. Production of memory chips for Western Digital would stop, the people said.

“Any slowdown of the joint development projects – even if temporary – could result in severe and lasting consequences,” Mark Long, chief financial officer of Western Digital, told a California court last month.

The largest maker of flash memory chips, Samsung Electronics, is currently completing what it says is the world’s largest chip plant. Underlining the speed at which Toshiba needs to move, Samsung’s plant is already in production, even before construction is complete.

GoPro Bets on Upcoming Hero6, Fusion Cameras in Bid for Profitability

GoPro Bets on Upcoming Hero6, Fusion Cameras in Bid for Profitability

GoPro forecast a smaller-than-expected loss, and said it was on track to launch the latest version of its flagship camera by the holiday season, as the company seeks to achieve its goal of returning to adjusted profitability in 2017.

The company’s shares were up 12 percent at $9.28 in after-market trading, and were set to wipe out their year-to-date decline if they open at that level on Friday.

Once a Wall Street favorite, GoPro’s body-mounted point-of-view cameras won a huge following among surfers, skydivers and other action junkies.

But the company came under pressure following a series of missteps including a delay in the launch of its Karma drone and production issues with its Hero5 camera.

GoPro also lost ground to feature-rich smartphones and rival products from companies such as Garmin and Sony.

The company is now betting on the Hero6 and the Fusion 360 action-cameras to help bring back its followers.

GoPro said during a post-earnings call on Thursday that it would realize revenue from the new products in the current quarter by shipping to customers ahead of the launch.

“Hero6 will probably perform in line with expectations, and at a higher gross margin than Hero5,” said Alicia Reese, an analyst at Wedbush Securities, adding that the management seemed to have a better handle on inventory levels this year.

GoPro forecast third-quarter adjusted loss of 1 cent to 11 cents per share and revenue of $290 million to $310 million (roughly Rs. 1,973 crores), compared with analysts’ estimate of a loss of 12 cents per share and revenue of $278.5 million.

“We expect margins to continue to improve throughout 2017, with the introduction of new products,” Chief Executive Nicholas Woodman said on the post-earnings call.

GoPro’s net loss narrowed to $30.5 million, or 22 cents per share, in the second quarter ended June 30 from $91.8 million, or 66 cents per share, a year earlier.

Excluding items, GoPro reported a loss of 9 cents per share, smaller than analysts’ estimate of a loss of 25 cents, according to Thomson Reuters I/B/E/S.

The company’s operating expenses fell 35.5 percent to $130.6 million.

Revenue rose to $296.5 million from $220.8 million.

Analysts on average had expected revenue of $269.6 million.

Google, MIT Researchers Create New AI-Based Real-Time Photo Editing

Google, MIT Researchers Create New AI-Based Real-Time Photo Editing

Scientists from MIT and Google have developed a new artificial intelligence system that can automatically retouch images like a professional photographer in real time, eliminating the need to edit images after they are clicked with smartphones.

The data captured by today’s digital cameras is often treated as the raw material of a final image. Before uploading pictures to social networking sites, even casual cellphone photographers might spend a minute or two balancing colour and tuning contrast, with one of the many popular image-processing programs now available.

The system developed by researchers from Massachusetts Institute of Technology and Google in the US is so energy-efficient and fast that it can display retouched images in real-time on phones, so that the photographer can see the final version of the image while still framing the shot.

The same system can also speed up existing image-processing algorithms.

The system employs machine-learning. The researchers trained their system on a dataset created by Adobe Systems, the creators of Photoshop.

The data set included 5,000 images, each retouched by five different photographers. They also trained their system on thousands of pairs of images produced by the application of particular image-processing algorithms, such as the one for creating high-dynamic-range (HDR) images.

The software for performing each modification takes up about as much space in memory as a single digital photo, so in principle, a cellphone could be equipped to process images in a range of styles.

Researchers compared their system’s performance to that of a machine-learning system that processed images at full resolution rather than low resolution.

During processing, the full-resolution version needed about 12 gigabytes of memory to execute its operations.

The researchers’ version needed about 100 megabytes, or one-hundredth as much.

The full-resolution version of the HDR system took about 10 times as long to produce an image as the original algorithm, or 100 times as long as the researchers’ system.

“This technology has the potential to be very useful for real-time image enhancement on mobile platforms,” said Jon Barron from Google.

“Using machine learning for computational photography is an exciting prospect but is limited by the severe computational and power constraints of mobile phones,” said Barron.

This paper may provide us with a way to sidestep these issues and produce new, compelling, real-time photographic experiences without draining your battery or giving you a laggy viewfinder experience, he said.

Nintendo SNES Classic Mini Pre-Orders From ‘Late This Month’

Nintendo SNES Classic Mini Pre-Orders From 'Late This Month'

Nintendo has announced that the SNES Classic Mini console will be available for pre-order later this August. The news comes via the company’s Facebook page.

“We appreciate the incredible anticipation that exists for the Super Nintendo Entertainment System: Super NES Classic Edition system, and can confirm that it will be made available for pre-order by various retailers late this month,” the company said in a post.

Furthermore, it claims to have increased the number of units that will be made present at stores in order to avoid a scenario akin to the NES Classic, that was hard to find at launch and even tougher to find soon after.

“A significant amount of additional systems will be shipped to stores for launch day, and throughout the balance of the calendar year,” the post reads.

Earlier, Nintendo confirmed that the SNES Classic will be available for this year only. The complete quote is as follows:

“We aren’t providing specific numbers, but we will produce significantly more units of Super NES Classic Edition than we did of NES Classic Edition. Super Nintendo Entertainment System: Super NES Classic Edition is currently planned to ship from Sept. 29 until the end of calendar year 2017. At this time, we have nothing to announce regarding any possible shipments beyond this year.

Our long-term efforts are focused on delivering great games for the Nintendo Switch system and continuing to build momentum for that platform, as well as serving the more than 63 million owners of Nintendo 3DS family systems. We are offering Super Nintendo Entertainment System: Super NES Classic Edition in special recognition of the fans who show tremendous interest our classic content.”

Call of Duty: World War 2 Private Beta Bonuses Announced

Call of Duty: World War 2 Private Beta Bonuses Announced

The Call of Duty: World War 2 private beta for PS4 starts from August 25 and ends on August 28.

In the run up to this, Activision has announced what bonus items will be given to those who partake in it. Dubbed as the Call of Duty: World War 2 Private Beta Combat Pack, this is what it has.

Call of Duty: World War 2 Private Beta Combat Pack contents

  • Custom helmet
  • Beta calling card
  • Call of Duty: World War 2 emblem

These will be available to those who play the Call of Duty: World War 2 private beta in the full game when it hits in November. Keep in mind that access to the private beta is only for those who pre-order the game. Xbox One players can try it out from September 1 to September 4 along with PS4 users again. No beta date has been given for PC gamers just yet.

Take-Two Won’t Commit to Red Dead Redemption 2 on PC; Says PC Gamers Are ‘Very Important’

Take-Two Won't Commit to Red Dead Redemption 2 on PC; Says PC Gamers Are 'Very Important'

During GTA publisher Take-Two‘s latest earnings call, the company maintained that the PC market is lucrative for its titles.

When asked if the PC is on the radar for its major releases, Take-Two President Karl Slatoff stated it depends on game to game. “Some titles are actually heavily weighted to PC, for example, Civilizationand XCOM,” he said, further stressing that the computer side of the business is of great importance.

“The great news is that the PC market is vibrant for us. It’s a great market for us. It’s a big market. It’s a core market in consumers that are highly engaged. It’s a predominantly digital market, which also removes friction in terms of ongoing engagement with a consumer. So, for us, the PC market as a company is very important and very exciting and something we focus on.”

During the same call, an investor asked if Take-Two CEO Strauss Zelnick is able to share updates on the PC version of Red Dead Redemption 2. Sadly, Zelnick gave the typical “any updates about any of our titles will come from our labels” answer.

Though we speculate a PC version of this Wild West open-world adventure is the works, Take-Two has only announced it for the PS4 and Xbox One and is due next year. Previously Take-Two implied that Red Dead Redemption 2 could have a bigger online component than expected. “Obviously we know what Rockstar tends to do. And Rockstar’s activities have been transformed by Grand Theft Auto Online,” Strauss Zelnick said.

Overwatch Summer Games 2017 Start and End Time Revealed

Overwatch Summer Games 2017 Start and End Time Revealed

Earlier Gadgets 360 reported that popular hero shooter Overwatch would see the return of the Summer Games event. This was its first event last year and it’s back again.

Blizzard has announced that the Overwatch Summer Games 2017 event will take place from August 9 to August 29. No exact time has been mentioned, but if it’s anything like the Overwatch Anniversary event, expect it to begin from 8:30pm IST on August 9 and end on 12:30pm IST on August 29.

Along with it comes the obligatory game modes, loot, and of course a new ap. Lucioball – the football inspired mode that made its debut with last year’s Summer Games makes a come back. In addition to the Rio map of 2016 is Sydney.

Although Overwatch Director Jeff Kaplan did not explain why the team had selected the locale, it perhaps could pertain to lore regarding Junkrat – an Australian character in the game.

Moreover, Lucioball sees some interesting modifications. For one, there’s Capa Lucioball – a competitive playlist with a placement and ranking. Also, Lucio’s ultimate ability doesn’t pull the ball to the player, instead it increases his movement speed and reduces the cool down time on his alternate ability. Lucio’s ultimate won’t let you ‘boop’ opponents either. What this means is, you can’t use Lucio’s ultimate to push goalies away and pull the ball inside.

In a welcome move, the Overwatch Summer Games 2017 brings a stack of event loot crates featuring all new cosmetics as well as returning favourites from the 2016 event. What’s more is, last year’s gear will be cheaper than the new ones.

New Overwatch Summer Games 2017 cosmetic items price:

  • Legendary Skins: 3,000 credits
  • Epic Skins: 750 credits
  • Rare tier: 225 credits
  • Common tier: 75 credits

Overwatch Summer Games 2016 cosmetics price:

  • Legendary Skins: 1,000 credits
  • Epic Skins: 250 credits
  • Rare tier: 75 credits
  • Common tier: 25 credits

With the Summer Games making a return, it’s safe to say Blizzard will have Halloween, Christmas, and Anniversary events as regular, annual fixtures.

Mario + Rabbids Kingdom Battle for Nintendo Switch Gameplay and Modes Revealed

Mario + Rabbids Kingdom Battle for Nintendo Switch Gameplay and Modes Revealed

Ubisoft let slip some details on what to expect from Mario + Rabbids Kingdom Battle for the Nintendo Switch when it hits on August 29. From playable characters, strategies, and story mode, this is what you need to know.

Mario + Rabbids Kingdom Battle gameplay is similar to Final Fantasy Tactics 
Yes, you read that right. Here there is a team to balance, techniques to master, and a Skill Tree to develop for each hero. The enemies you face at the very beginning are tougher than you’d expect. However it seems to be a whole lot more cerebral. Encounters in Mario + Rabbids Kingdom Battle are puzzles with the visual reward being the opportunity to see your favourite Nintendo characters like Mario, Luigi, Peach, and Yoshi pull off their trademark moves in a new genre. This is compounded by the Rabbids companions that enliven the proceedings with dollops of humour.

Mario + Rabbids Kingdom Battle combat, skills, and weapons 
The gamut of available weapons and tools in Mario + Rabbids is wide, and each hero can bring Primary and Secondary techniques to the battlefield that can be combined, such as using Luigi’s Magnetic Dance to set up a Grenade Duck ambush for Princess Peach.

In one turn, a hero can slide into and trip over an enemy, before Team Jumping off an ally’s shoulders to land behind cover from which to launch another attack. This can set up a chain of attacks that could be anything from a rocket launched Spike, to ink damage from a Bwahntasaurus Wrecks sentry.

Mario + Rabbids Kingdom Battle story mode, co-op mode, environments and enemies
The game’s story mode takes you from a host of varied locales such as the sun-drenched Ancient Garden close to Prince Peach’s castle to spooky moonlit graveyards. Enemies are equally aplenty ranging from shield-bearing Rabbids that require sneaking up from behind, and Boos that teleport team members to dangerous parts of the map. Plus, there’s a co-op mode that supports every Nintendo Switch controller ensuring that the Joy-Con included with the console get a fair work out.

IBM Develops Tape Cartridge That Stores 330TB of Uncompressed Data in the Palm of Your Hand

IBM Develops Tape Cartridge That Stores 330TB of Uncompressed Data in the Palm of Your Hand

In an interesting development, scientists at IBM have managed to develop a prototype magnetic tape drive that fits in the palm of your hand and can store a whopping 330TB of uncompressed data. IBM Research worked alongside Sony Storage Solutions for years to achieve increased areal recording densities and develop this magnetic tape.

If you are wondering about the recording density, the prototype comes with an unprecedented 201Gb per inch density. “Tape has traditionally been used for video archives, back-up files, replicas for disaster recovery and retention of information on premise, but the industry is also expanding to off-premise applications in the cloud,” said IBM Research fellow Evangelos Eleftheriou in a statement.

“While sputtered tape is expected to cost a little more to manufacture than current commercial tape, the potential for very high capacity will make the cost per terabyte very attractive, making this technology practical for cold storage in the cloud,” he said.

In its release about the magnetic tape cartridge, Sony said that this achievement was made possible by bringing together it’s “new magnetic tape technology employing lubricant with IBM Research – Zurich’s newly developed write/read heads, advanced servo control technologies and innovative signal-processing algorithm.” It further pointed out that closing the gap between the magnetic tape and magnetic head is critical to achieving high-density recording capabilities for tape storage media.

Interestingly, IBM has also said that this achievement is also reflective of the viability of scaling up storage on tapes continuously for another decade.

Our storage needs have been growing for years with advancement in technology sector and if you didn’t realise this till now, just imaging how iPhones with 8GB base storage would feel in this day and age. This sort of achievement gives hope that our data requirements will be met with advancement in technology and we won’t be ill-equipped to tackle the situation going forward.