Innodisk's New Fire Shield SSD Can Withstand 800C for 30 Minutes

Innodisk has announced the company’s latest Fire Shield SSD, which is capable of withstanding temperatures of up to 800 degrees Celsius for more than 30 straight minutes.

Credit: InnodiskCredit: Innodisk

The Fire Shield SSD features a 3.5-inch design and has up to three layers of protection. The first enclosure consists of a flame-resistant copper alloy material, while the second is made from a heat-isolating lining material. The SSD itself is housed inside a third protective enclosure. With over 20 heat-resistant materials in its design, the Fire Shield SSD is not only fireproof but also resistant to vibrations and impacts.

Credit: InnodiskCredit: Innodisk

Innodisk created the Fire Shield SSD for the sole purpose of protecting valuable data in Black Box applications. The drive can survive for more than 30 minutes at temperatures the scale up to 800 degrees Celsius. Obviously, you can’t just go connecting the Fire Shield SSD to your system to recover the data after it has been exposed to extreme heat during a prolonged period. Instead, an Innodisk engineer will have to extract the integrated chips and perform data recovery on them with a specialized machine.

Inside of the robust enclosure, the Fire Shield SSD is everything you would expect from a normal SATA SSD. It still speaks with your system through a SATA III port. However, the drive does utilize SLC (single-level cell) or iSLC (Inno single-level cell) NAND chips for maximum endurance and reliability. According to Innodisk, the SLC chips have a program-erase (PE) cycle of 60,000 cycles while the iSLC ones are rated for 20,000 cycles. The Fire Shield SSD delivers sequential read and write speeds up to to 520 MBps and 360 MBps, respectively. Innodisk offers the drive in capacities of 8GB, 16GB, 32GB, 64GB and 128GB.

Innodisk didn’t mention the pricing or availability of its Fire Shield SSDs.

Apple Recalls 15-Inch MacBook Pro Laptops Over Battery Concerns (Update: FAA Ban)

Credit: AppleCredit: Apple

Updated, 8/14/19, 6:05am PT: The Federal Aviation Administration (FAA) told airlines not to allow MacBook Pro models affected by this recall onto their flights, Bloomberg reported, in accordance with a 2016 order prohibiting devices with recalled batteries from taking to the skies. That ban doesn’t allow the affected MacBook Pro models to be included with carry-on luggage or checked as cargo. It’s not clear how well the airlines would enforce this ban, though, considering the identification of recalled batteries is a multi-step process.

Original article, 6/20/19, 1:24pm PT:

Apple today voluntarily recalled “a limited number of older generation 15-inch MacBook Pro units which contain a battery that may overheat and pose a safety risk.” The company is now offering free battery replacements for MacBook Pro (Retina, 15-inch, mid-2015) laptops. 

Batteries are wonderful components that enable many of the devices we use every day. But until companies announce recalls like this, it can be easy to forget that batteries can also be dangerous, given their occasional acts of apparently spontaneous combustion. Samsung’s exploding Galaxy Note 7 from 2016 helped remind us of that fact with smartphone batteries, but laptop batteries can also have real safety problems.

See: Lenovo’s recall of certain Thinkpad X1 Carbon models in February 2018, HP’s expanded recall of various laptop models in March and now Apple’s recall of 15-inch MacBook Pro units “sold primarily between September 2015 and February 2017.” All three recalls were prompted by concerns about the batteries overheating. The best-case scenario is for the laptop itself to be ruined; the worst-case would be for its owner to be injured in the process.

How to Get a Replacement Battery

Apple said in a support article that only MacBook Pro (Retina, 15-inch, mid-2015) models are subject to this recall. Owners can determine what model of MacBook Pro they have by checking the “About This Mac” page found via the Apple menu in the top-left corner of the screen. If you confirm you’re using the affected model, you can submit you serial number via the support article to see if your particular unit needs a new battery. 

There are three ways to participate in the battery replacement program:

  • Find an Apple Authorized Service Provider.
  • Make an appointment at an Apple Retail Store.
  • Contact Apple Support to arrange mail-in service via the Apple Repair Center.

Apple said that units “will be examined prior to any service to verify that it is eligible for this program” and that the service can take 1-2 weeks. While a free battery replacement is a decent gesture (paying for a replacement part that’s less likely to blow up isn’t fun), we wonder how many people will continue using the defective units simply because they can’t go that long without their laptop. Hopefully the answer turns out to be ‘not that many.’

AMD Announces BIOS Fix for Ryzen 3000 Boost Clocks, Update Comes September 10th

Credit: AMDCredit: AMDAfter an extended period of silence, AMD has finally publicly acknowledged that “some” of its customers aren’t receiving the expected boost frequencies with the Ryzen 3000 series processors.

As you can read in the statement below, the company simultaneously announced that it would issue a fix for the BIOS issues. The company will update the community on September 10, 2019 about the availability of the fix.

“AMD is pleased with the strong momentum of 3rd Gen AMD Ryzen™ processors in the PC enthusiast and gaming communities. We closely monitor community feedback on our products and understand that some 3rd Gen AMD Ryzen users are reporting boost clock speeds below the expected processor boost frequency. While processor boost frequency is dependent on many variables including workload, system design, and cooling solution, we have closely reviewed the feedback from our customers and have identified an issue in our firmware that reduces boost frequency in some situations. We are in the process of preparing a BIOS update for our motherboard partners that addresses that issue and includes additional boost performance optimizations. We will provide an update on September 10 to the community regarding the availability of the BIOS.”

The statement comes amidst a growing chorus of complaints from Ryzen 3000 owners on forums, reddit, and social media that their chips aren’t reaching the advertised boost clocks. In response to the growing number of complaints, YouTuber Der8aeur recently conducted a survey that garnered 2,700 respondents, of which only 5.6% were able to reach the advertised boost clocks for AMD’s flagship Ryzen 9 3900X processor.

Our own investigation of AMD’s new boost clock behavior also found that only one core on any given Ryzen 3000 CPU is guaranteed to hit the rated boost clock, which AMD confirmed. That means the Ryzen 3000-series processors contain a mix of faster and slower cores. Unfortunately, users must have the latest version of Windows 10 to use the Ryzen-aware scheduler, which targets the fastest cores with lightly-threaded applications, further complicating matters for frustrated customers trying to attain the advertised boost frequencies.

In either case, even with the presence of the necessary BIOS, driver, and Windows 10 scheduler, most customers have been unable to attain Ryzen 3000’s advertised speeds with any of the models.

We don’t expect the new BIOS fix to change the requirement for Windows 10 or AMD’s driver, or the new binning tactic of using a mix of faster and slower cores. Hopefully it will expose the best performance possible from the fastest core, but we’ll have to wait until September 10 to find out.

Breaking News, more to come….

Trump Delays Tech Tariffs to August 2020

Credit: ShutterstockCredit: ShutterstockThe U.S. Trade Representative (USTR) announced Friday that it lifted tariffs on $200 billion worth of goods imported from China. Impacted products included graphics cardsmotherboards and PC cases, as well as some gaming peripherals. Now those goods will be exempt from the increased tariffs levied as part of the trade war between the U.S. and China–at least until the temporary exemption comes to an end in August 2020.

U.S. tariffs on Chinese-manufactured goods have expended to cover hundreds of billions of dollars worth of goods over the last few months. The most recent expansion affected $110 billion worth of products, including the Apple Watch and wireless audio devices, on September 1. A tariff increase from 25% to 20% was also supposed to affect $250 billion worth of goods starting October 1, but the Trump administration delayed the bump to October 15.

Now much of the PC industry–as well as the gaming console market–has been granted a brief reprieve. USTR said in a public docket published to the Federal Register on September 20 that many electronics devices will be excluded from additional tariffs until August 7, 2020. The executive office also published a one-paragraph statement about deputy-level trade discussions between the U.S. and China that took place on September 19 and 20.

“These discussions were productive,” USTR said in the statement, “and the United States looks forward to welcoming a delegation from China for principal-level meetings in October.” The delayed tariffs along with the exemptions granted on Friday were likely meant to help with those discussions. (President Donald Trump had previously framed the delay of the October 1 tariffs as a “gesture of good will [sic]” to China on its 70th anniversary.)

This delay is the latest way in which the U.S. and China trade war has been unpredictable for U.S. tech companies. Tariffs expand, increase and are then delayed based on how talks between the countries have gone. Huawei goes from persona non grata to a fine company on a whim. Unless the October talks go well, there’s no resolution to this conflict in sight. Just more delays of the seemingly inevitable as the countries go head to head.

Tiny TPM Promises to Secure IoT Devices

TCG MARS Subgroup showcasing Radicle TPM prototype. Credit: TCGTCG MARS Subgroup showcasing Radicle TPM prototype. Credit: TCGThe Trusted Computing Group, founded by companies such as AMD, HP, IBM, Intel and Microsoft in 2003 to protect cryptographic keys on computers against tampering, recently announced its work to develop the specification for the “world’s tiniest Trusted Platform Module (TPM).”

What’s the Point of a Tiny TPM?

The TCG said that many Internet of Things (IoT) device makers can’t afford to add a full TPM on an IoT board either because of space, cost or power constraints. At the same time, TCG said, many manufacturers still want or need the TPM features in their devices, such as Roots of Trust for Measurement (RTM), Storage (RTS) and Reporting (RTR), so that the devices can work securely within the TCG Measurement and Attestation framework. Through these features, the TPM ensures platform authentication and integrity, making it much more difficult for malicious software to take over a system.

TCG’s new Measurement and Attestation RootS (MARS) Subgroup has been formed to develop specifications for compliant TPM chips with very little overhead for IoT devices. 

“In a nutshell, we want to specify what the tiniest TPM needs to be so it can be integrated directly within the host chip,” Tom Brostrom, Chair of the MARS Subgroup, said in a statement. “This will ensure that devices that aren’t big enough to integrate a separate TPM will still be able to retain the required RTS/RTR capabilities. In turn, this will allow greater reach of trusted computing technologies over a wider set of devices and use cases.”

In a meeting held in Warsaw, Poland during the inaugural session of the MARS Subgroup, the TCG members announced the first prototype for this type of TPM, called “Radicle.” The team also agreed on the scope of its work, which will focus on the hardware requirements needed to operate RTS and RTR, as well as on the software API needed to access those features.

The Internet of Threats

Four years ago, Eugene Kaspersky, the founder of the Russian security firm Kaspersky Labs, called IoT the Internet of Threats. As we’ve seen over the past few years, the vast majority of IoT devices don’t even seem to take security into account, and many manufacturers don’t issue patches for them on time or at all to fix security issues that may have been found on them after being shipped to customers.

If the devices don’t ship with strong security by default from the factory, it becomes that much less likely that the manufacturers will want to deal with the dozens and dozens of security issues they could have in their first few years of life. Ensuring strong security out of the box should significantly reduce the number of serious issues that will need to be fixed down the line.

The TCG is playing a small part in trying to secure IoT devices. Besides creating a TPM specification for IoT devices (that’s presumably as secure or at least almost as secure as the regular TPM), they have other projects involving secure firmware and software updates for embedded systems. Arm, the largest provider of microcontroller and CPU IP for IoT devices, is also playing a role in trying to secure IoT devices, through its Platform Security Architecture that involves using a reference open source firmware and other built-in hardware and firmware security features.

When we talk about IoT we’re not just talking about surveillance cameras that can be easily taken over by bad actors to spy on you in your own home, “smart toys” that can be hacked into, or zombie routers that can be used in DDoS attacks. IoT in the coming years will also mean industrial robots that attackers could take over to cripple a large portion of a company’s production, or self-driving cars that can be driven off the highway.

Security will no longer be a feature that’s nice to have, but something that will be critical for the product’s existence and survivability in the market. 

Corsair’s Stylish Black Mirrored Tempered-Glass RGB Case Is $60 Off

Your case says a lot about your PC building personality. Are you more serious, sleek and efficient or all about the flash? If you’re part of the latter category, you’ll be pleased to hear the Corsair Crystal Series 570X RGB case is currently on sale at Newegg ahead of Amazon Prime Day.

The black, red and white versions are all at a discount, but the most showstopping design “Mirror Black” comes at the biggest discount: $139.99 after $20 rebate card.

– Get the Corsair CORSAIR Crystal Series 570X RGB ATX PC Case for $139.99 (reg. $199.99).

As mentioned, the 570X comes in four different colors, but Mirror Black has always been the most expensive with an MSRP of $199.99. With Newegg’s sale you can finally feel less guilty about paying for appearances. Without any lights on, the 570X RGB in Mirror Black looks like a stylish PC case made of, well, mirrors. With RGB lights activated, it goes from looking cool to being mesmerizing.

With the mirrored tempered glass, all your components will be on display for you and passersby to appreciate. And if those components come with some RGB of their own, the show’s even more lively.

The 570X RGB comes with three SP120 RGB fans, plus an LED controller for selecting dozens of preset effects. The RGB fans are also compatible with Corsair Link software if you buy the Corsair Lighting Node Pro.

The case can hold up to six fans and 360mm, 280mm or 120mm radiators. The fan trays are removable from the front and top for making PC building a breeze. A vertical GPU mount (you’ll need you own PCI-e riser cable) can also come in handy when it’s time to install your graphics card.

Should You Buy This Case?

Before committing to the 570X RGB in any color, be sure to read our review of the Corsair Crystal 570X (the non-RGB version) for in-depth looks at building and performance. Our testing found that this case gets good ventilation and dust filtration. However you might have trouble getting to the power supply filter.

For more recommendations and tips see our Best PC Cases page.

Photo Credits: Newegg

12-Core for the Masses: AMD Threadripper 1920X Drops to $260

Credit: NeweggCredit: Newegg

If you need a CPU with an abundant number of cores, the AMD Ryzen Threadripper 1920X may be the perfect fit. The 12-core chip, which usually sells for over $350, can be yours today for just $259.99.

Part of AMD’s first-generation of Ryzen Threadripper lineup, the Threadripper 1920X boasts 12 cores, 24 threads and up to 32MB of L3 cache. The processor has a 3.5 GHz base clock and a boost clock that hits the 4 GHz. It comes with a completely unlocked multiplier, so you can overclock the core-heavy monster to gain even more performance.

With a discounted price of $269.85, the Threadripper 1920X is currently the cheapest 12-core chip in AMD’s arsenal and on the market.

AMD Threadripper 1920X vs. Threadripper 2920X vs. Ryzen 9 3900X

Model Cores /
Base / Boost Clock (GHz) L3 Cache 
PCIe  Unlocked Multiplier DRAM TDP Price (at time of writing) Price 
Per Core
AMD Ryzen 9 3900X 12 / 24 3.8 / 4.6 64 PCIe 4.0 x 24 Yes Dual DDR4-3200 105W $499 $41.58
AMD Threadripper 2920X 12 / 24 3.5 / 4.3 32 PCIe 3.0 x 64 Yes Quad DDR4-2933 180W $399.99 $33.33
AMD Threadripper 1920X 12 / 24 3.5 / 4.0 32 PCIe 3.0 x 64 Yes Quad DDR4-2667 180W $259.99 $21.67

As you would expect from a HEDT (high-end desktop) processor, the Threadripper 1920X supports quad-channel memory and memory speeds up to 2,667 MHz. When paired with the right X399-based motherboard, the Threadripper 1920X can support up to 256GB of memory, so you can open tabs to your heart’s content and still have plenty of memory leftover.

In regards to PCIe connectivity, the Threadripper 1920X supports to 64 PCIe 3.0 lanes. You can run multiple graphics cards configurations, PCIe storage arrays and a good number of PCIe-based expansion cards without worry.

Should You Buy This CPU?

We highly recommend you check out our in-depth AMD Ryzen Threadripper 1920X review before opening your wallet.

You can also review our CPU buying guide for help. To see where this processor ranks among others currently available, including from rival Intel, check out our CPU hierarchy page. And for other CPUs we love, see our favorite gaming CPUs and favorite CPUs for productivity performance

U.S. Federal Agencies Issue Ambitious Roadmap For AI Development

Credit: ShutterstockCredit: Shutterstock

On Monday, the White House and the National Institute of Standards and Technology (NIST) issued a federal plan to develop technical standards for artificial intelligence (AI). The plan follows up on a mandate contained in President Trump’s executive order on Maintaining American Leadership in Artificial Intelligence, issued last February.

NIST’s AI Standardization Plan

The executive order directed NIST to develop a plan that would: “ensure that technical standards minimize vulnerability to attacks from malicious actors and reflect Federal priorities for innovation, public trust, and public confidence in systems that use AI technologies; and develop international standards to promote and protect those priorities.”

The plan is meant to bolster AI standards-related knowledge, leadership and coordination among agencies that develop or use AI. It also states that the government should prioritize efforts that are “inclusive and accessible, open and transparent, consensus-based, globally relevant and nondiscriminatory, and that use multiple approaches.”

The NIST plan also promotes the focused research on the trustworthiness of AI; the support and expansion of public-private relationships; and an increase in collaborations on this matter with other countries. NIST’s AI plan also comes with guidelines for federal agencies for how to embrace AI technologies to further their mission. 

The federal government will also help with the standardization of some AI-development tools “to advance the development and adoption of effective, reliable, robust, and trustworthy AI technologies.”

These tools include, but are not limited to: data sets in standardized formats, including metadata for training, validation and testing of AI systems; tools for capturing and representing knowledge and reasoning in AI systems; fully documented use cases and best practices for AI technologies so that others can learn from them.

CCC 20-Year AI Research Roadmap

Last week, the Computing Community Consortium, which is associated with the National Science Foundation, also released a 20-year Community Roadmap for Artificial Intelligence Research in the United States. 

This roadmap is based on three main themes: creating an “AI infrastructure” to serve academia, industry and government; training a capable AI workforce, and investing in basic AI research (the type of research that needs to be done for decades before significant results are seen). 

The CCC believes that all of these efforts will require substantial investments from the government, but it noted that the results will be transformative.

Report: China-Based Yangtze Memory Starts 64-Layer NAND Production

Credit: YMTCCredit: YMTCYangtze Memory Technologies Co. (YMTC) has reportedly started volume production of 64-layer 3D NAND. The triple-level cell chips with 256GB capacity use the company’s in-house Xtacking architecture for bonding two dies together.

It is the company’s first process to go into high volume production. Digitimes’ sources indicate that the Chinese firm has gradually increased its 3D NAND yield and will be ramping its Wuhan factory to 100,000 wafers per month in 2020; although, that could further grow to 150,000 wafers per month. The sources also stated that its state-owned parent company, Tsinghua Unigroup, is constructing a 3D NAND memory fab in Chengdu with a similar wafer capacity, to come online in 2021-2022.

YMTC’s 256GB chips contain 64-layer TLC 3D NAND (which other companies started manufacturing in 2017) and are manufactured with its proprietary Xtacking architecture. The company unveiled the technology last year at the Flash Memory Summit. The chips are manufactured on two wafers. One wafer contains the ‘periphery’ CMOS logic, and the other one has the actual 3D NAND, based on common charge trap technology. The wafers are then bonded together with a process step, with billions of what it calls metal Vertical Interconnect Accesses (VIAs).

The benefit of this technology is a much more efficient use of die area. YMTC claims that die size is reduced by 25% because the area efficiency is over 90% for Xtacking, compared to less than 65% for conventional 3D NAND. It even estimates that at 128 layers, the periphery circuits would take up over 50% of the die area. Xtacking makes YMTC’s 64-layer 3D NAND within 80-90% the density of other’s 96-layer 3D NAND, it claimed. Other companies are also using or working on CMOS under the array technology though, albeit not by bonding two separate wafers.

Additionally, the company says that it allows for a modular approach to technology development, shortening product development by three months and reducing the manufacturing cycling time, the time the product is in the fab, by 20%.

Both wafers are manufactured by YMTC, with the logic wafer produced on a 180nm process. The architecture is also capable of a 3 Gbps I/O speed, which is over two times faster than common today and similar to DDR4. YMTC is working on multiple generations of Xtacking simultaneously; however, it intends to skip the 96-layer generation and move directly to 128 layers.

YMTC is a new, ambitious player in the NAND industry and plays a key role in China’s semiconductor ambitions. The company was founded in 2016 and has over 1,500 R&D engineers. Its CEO had previously worked at Intel and the manufacturing company XMC, now YMTC’s wholly-owned subsidiary.

EU: Right to Be Forgotten Doesn't Apply to US

Photo Source: GooglePhoto Source: Google

The Court of Justice for the European Union today ruled that a privacy law commonly known as the “right to be forgotten” doesn’t apply beyond the EU’s borders. Internet activists hailed the decision as a victory against online censorship, because it means that EU laws can’t force search companies to de-list results made in other countries, which would have given the EU outsized control over what data is available online around the world.

The premise behind the right to be forgotten is simple: people shouldn’t be haunted by their pasts forever. But in the modern era, companies like Google have made it near impossible for people to leave their pasts behind, with many of their most embarrassing moments just a few clicks away. That wasn’t a problem just a few decades ago; the right to be forgotten is supposed to give people the same chance of moving on.

Implementing the right to be forgotten is another matter. Some people, like celebrities and politicians, live in the public eye. How should their requests to have search results de-referenced (as the CJEU put it) be handled? Other people might want to bury search results, too, but at what point do they become a matter of public interest? And who’s going to make sure the rule isn’t abused to cover up information that should remain available?

This wasn’t a small problem. Google said in 2018 that it received requests to de-list 2.4 million URLs in the first three-and-a-half years since the right to be forgotten’s introduction. We doubt those requests have stopped in the interim, which means it’s important for the company to know exactly how the EU wanted de-listed results to be handled. Was it truly attempting to remove the results from the global internet, or just within its borders?

CJEU published an initial opinion in January saying it believed that applying the right to be forgotten outside the EU would set a dangerous precedent when it came to online censorship. The EU might have good intentions regarding the removal of certain URLs from search results. Would more oppressive governments like China, Russia and the like use the precedent set by the right to be forgotten to step up their own censorship efforts?

Here’s what CJEU said in today’s ruling:

The Court emphasises that, in a globalised world, internet users’ access — including those outside the EU — to the referencing of a link referring to information regarding a person whose centre of interests is situated in the EU is likely to have immediate and substantial effects on that person within the EU itself, so that a global de-referencing would meet the objective of protection referred to in EU law in full. However, it states that numerous third States do not recognise the right to dereferencing or have a different approach to that right. The Court adds that the right to the protection of personal data is not an absolute right, but must be considered in relation to its function in society and be balanced against other fundamental rights, in accordance with the principle of proportionality. In addition, the balance between the right to privacy and the protection of personal data, on the one hand, and the freedom of information of internet users, on the other, is likely to vary significantly around the world.