NPU: What is it and why is it gaining importance in semiconductor industry

NPU: What is it and why is it gaining importance in semiconductor industry


Neural Engine on the Apple M4 chip

Apple debuted the M4 chip with the iPad Pro on May 7. While Apple stated significant improvements in chip’s overall performance, the most prominent change in the M4 is the 16-core Neural Engine, which is Apple’s term for Neural Processing Unit (NPU). Apple called the M4 an “outrageously powerful chip for AI.” It is the improvements in the NPU that is backing Apple’s claim of the M4 being significantly more powerful than any other chip powering an AI PC. So, what exactly is NPU and why is it gaining prominence in the chip industry? Let us find out:


What is NPU (Neural Processing Unit)


An NPU, or a Neural Processing Unit, is a dedicated processor designed specifically for accelerating neural network processes. A neural network is essentially a type of machine learning algorithm that mimics the human brain for processing data. Therefore, the NPU is highly capable for handling machine learning operations that form the basis for AI-related tasks, such as speech recognition, natural language processing, photo or video editing processes like object detection, and more.


In most consumer-facing gadgets such as smartphones, laptops and tablets, the NPU is integrated within the main processor, adopting a System-on-Chip (SoC) configuration. However, for data centres, the NPU might be an entirely discrete processor, separate from any other processing unit such as the central processing unit (CPU) or the Graphics processing unit (GPU).


How is NPU different from CPU and GPU


CPUs employ a sequential computing method, issuing one instruction at a time, with subsequent instructions awaiting the completion of their predecessors. In contrast, NPU harnesses parallel computing to simultaneously execute numerous calculations. This parallel computing approach results in swifter and more efficient processing. That said, CPUs are good at sequential computing, executing one process at a time, but running AI tasks requires the processor to execute multiple calculations and processes simultaneously.


This is where the Graphic Processing Units, or the GPUs, come in. These have parallel computing capabilities and have integrated circuits embedded to carry out AI workloads, but they are generally meant for handling other processes, such as graphic rendering and resolution upscaling. It essentially makes a case for NPUs, which simply replicate those circuits and make it dedicatedly work on carrying out machine learning operations. This allows the AI workload processing to be more efficient and less power consuming.


GPUs are still used in the initial development and refinement of AI algorithms, while NPUs later on takes the mantle by running those refined language models on the consumer’s device.


NPU and on-device AI


Large language models (LLMs) are too big to run on-device and service providers generally take up the processing to the cloud to offer AI features based on their language models. However, recently big technology companies have released small language models such as Google’s Gemma, Microsoft’s Phi-3 and Apple’s OpenELM, indicating a trend towards small scaled AI models, capable of running entirely on-device. As on-device AI models gain more prominence, the role of NPUs get even more crucial as they are the one deploying AI-powered applications on the hardware.

First Published: May 08 2024 | 5:21 PM IST



Source link

Google launches Wallet app in India to help users store digital documents

Google launches Wallet app in India to help users store digital documents



Tech giant Google on Wednesday launched the Google Wallet app in India, enabling users to store and retrieve digital documents like boarding passes, loyalty cards, movie tickets, and more.


However, unlike its US offering, the Google Wallet app won’t allow users to store payment cards and access the tap-and-pay feature in stores.


The wallet app, which has gone live in the country, will cater to non-payment use cases and complement Google Pay – the payment app of the tech giant, said the company.


The company has partnered with 20 brands including PVR & INOX, Air India, Indigo, Flipkart, Pine Labs, Kochi Metro, Abhibus, among others.


Ram Papatla, general manager and India engineering lead, Android at Google, said, “The arrival of Google Wallet in India marks an important milestone in Android’s India journey, bringing innovative and convenient experiences to simplify people’s daily lives.”

“We are delighted to partner with many of India’s top brands to offer a comprehensive solution that helps you securely access and manage your everyday essentials. From boarding passes to loyalty cards, and event tickets to public transport passes – they’re there when you need them,” Ram said.


Apart from accessing movie tickets and boarding passes, the app will also allow users to redeem loyalty points and gift cards. Google has partnered with platforms like Flipkart, Dominos, Shoppers Stop, and loyalty programme enablers like Pinelabs, EasyRewardz and Twid to enable this feature.


Google has also collaborated with integrators Wavelynx and Alert Enterprise to provide users with the option to store and access a corporate badge inside the wallet app.


If the users turn on the smart personalisation settings in Gmail, the event or ticket confirmations sent to a user’s Gmail account will automatically appear in their Google Wallet. 


The company has collaborated with airlines like Air India, Air India Express, and online travel companies including MakeMyTrip, EaseMyTrip, and ixigo to allow travellers to save and access their mobile boarding passes on the wallet app. 


Pixel users will have an additional feature of adding boarding passes to their Google Wallet by simply taking a screenshot and tapping “Add to Google Wallet,” said the company.

First Published: May 08 2024 | 4:53 PM IST



Source link

US eyes restrictions on China's access to AI behind apps like ChatGPT

US eyes restrictions on China's access to AI behind apps like ChatGPT


That computing power threshold could become the basis for determining what AI models would be subject to export restrictions, according to two U.S. officials (Photo: Bloomberg)


The Biden administration is poised to open up a new front in its effort to safeguard U.S. AI from China with preliminary plans to place guardrails around the most advanced AI Models, the core software of artificial intelligence systems like ChatGPT, sources said.


The Commerce Department is considering a new regulatory push to restrict the export of proprietary or closed source AI models, whose software and the data it is trained on are kept under wraps, three people familiar with the matter said.

 


Any action would complement a series of measures put in place over the last two years to block the export of sophisticated AI chips to China in an effort to slow Beijing’s development of the cutting edge technology for military purposes. Even so, it will be hard for regulators to keep pace with the industry’s fast-moving developments.

 


The Commerce Department declined to comment. The Chinese Embassy in Washington did not immediately respond to a request for comment.

 


Currently, nothing is stopping U.S. AI giants like Microsoft -backed OpenAI, Alphabet’s Google DeepMind and rival Anthropic, which have developed some of the most powerful closed source AI models, from selling them to almost anyone in the world without government oversight.

 


Government and private sector researchers worry U.S. adversaries could use the models, which mine vast amounts of text and images to summarize information and generate content, to wage aggressive cyber attacks or even create potent biological weapons.


To develop an export control on AI models, the sources said the U.S. may turn to a threshold contained in an AI executive order issued last October that is based on the amount of computing power it takes to train a model. When that level is reached, a developer must report its AI model development plans and provide test results to the Commerce Department.

 


That computing power threshold could become the basis for determining what AI models would be subject to export restrictions, according to two U.S. officials and another source briefed on the discussions. They declined to be named because details have not been made public.

 


If used, it would likely only restrict the export of models that have yet to be released, since none are thought to have reached the threshold yet, though Google’s Gemini Ultra is seen as being close, according to EpochAI, a research institute tracking AI trends.

 


The agency is far from finalizing a rule proposal, the sources stressed. But the fact that such a move is under consideration shows the U.S. government is seeking to close gaps in its effort to thwart Beijing’s AI ambitions, despite serious challenges to imposing a muscular regulatory regime on the fast-evolving technology.

 


As the Biden administration looks at competition with China and the dangers of sophisticated AI, AI models “are obviously one of the tools, one of the potential choke points that you need to think about here,” said Peter Harrell, a former National Security Council official. “Whether you can, in fact, practically speaking, turn it into an export-controllable chokepoint remains to be seen,” he added.

 


BIOWEAPONS AND CYBER ATTACKS?

 


The American intelligence community, think tanks and academics are increasingly concerned about risks posed by foreign bad actors gaining access to advanced AI capabilities.

 


Researchers at Gryphon Scientific and Rand Corporation noted that advanced AI models can provide information that could help create biological weapons.

 


The Department of Homeland Security said cyber actors would likely use AI to “develop new tools” to “enable larger-scale, faster, efficient, and more evasive cyber attacks” in its 2024 homeland threat assessment.

 


Any new export rules could also target other countries, one of the sources said.

 


“The potential explosion for [AI’s] use and exploitation is radical and we’re having actually a very hard time kind of following that,” Brian Holmes, an official at the Office of the Director of National Intelligence, said an export control gathering in March, flagging China’s advancement as a particular concern.

 


AI CRACKDOWN

 


To address these concerns, the U.S. has taken measures to stem the flow of American AI chips and the tools to make them to China.

 


It also proposed a rule to require U.S. cloud companies to tell the government when foreign customers use their services to train powerful AI models that could be used for cyber attacks.

 


But so far it hasn’t addressed the AI models themselves.

 


Alan Estevez, who oversees U.S. export policy at the Department of Commerce, said in December that the agency was looking at options for regulating open source large language model (LLM) exports before seeking industry feedback.

 


Tim Fist, an AI policy expert at Washington DC based think tank CNAS, says the threshold “is a good temporary measure until we develop better methods of measuring the capabilities and risks of new models.”

 


The threshold is not set in stone. One of the sources said Commerce might end up with a lower floor, coupled with other factors, like the type of data or potential uses for the AI model, such as the ability to design proteins that could be used to make a biological weapon.

 


Regardless of the threshold, AI model exports will be hard to control. Many models are open source, meaning they would remain beyond the purview of export controls under consideration.

 


Even imposing controls on the more advanced proprietary models will prove challenging, as regulators will likely struggle to define the right criteria to determine which models should be controlled at all, Fist said, noting that China is likely only around two years behind the United States in developing its own AI software.

First Published: May 08 2024 | 4:41 PM IST



Source link

Lenovo Yoga 7i 2-in-1 review: Windows laptop brimming with useful features

Lenovo Yoga 7i 2-in-1 review: Windows laptop brimming with useful features


The Lenovo Yoga 7i 2-in-1 is a convertible Windows laptop in the thin-and-lightweight segment. Priced at Rs 101,990, the laptop boasts all-aluminium design, OLED touchscreen of 120Hz refresh rate, fullHD camera with shutter button for privacy, backlit keyboard, and Dolby Atmos audio system. On the software side, the laptop is packed with useful features from Lenovo and Microsoft, but most importantly a dedicated button on the keyboard to wake Microsoft Copilot. Since the laptop has a convertible form factor and a touchscreen, it supports stylus and doubles up as a tablet. Seems like Lenovo has ticked all the right boxes for a convertible laptop with the Yoga 7i 2-in-1. Has it? Let us find out:


Design


The Lenovo Yoga 7i straight up ticks the first box of being an ultra-portable 2-in-1 laptop by being lightweight, tipping the scales at 1.49 Kg. Its sleek all-aluminium design looks premium and the matte finish on the surface balances style with substance. Moreover, the 360-degree hinge, a hallmark of 2-in-1 design, is well executed by Lenovo. It is flexible for easy display lid opening with a single finger, yet maintains sufficient stiffness to securely hold the display in place without any wobbling. However, the lack of adequate grip at the bottom may cause slight sliding on smooth surfaces when opening or closing the laptop.


With the lid open, the 14-inch OLED panel flanked by thin bezels come to notice. Lenovo has extended the lid enclosure from the top-centre to accommodate the front camera, slider for physical shutter, and infrared sensor. Despite its size, the extended portion does not compromise device stability when folded as a table-top. Beneath the display is a full-sized keyboard flanked by speakers on left and right sides, accompanied by a generously sized trackpad sporting the same matte finish as the laptop’s body.


The Lenovo Yoga 7i 2-in-1 excels in versatility, offering multiple usage modes such as tent mode for tabletop use or a complete 360-degree fold for a tablet-like experience. Lenovo paid attention to detail in enhancing the user experience in each mode, such as incorporating rounded edges in the lower section for improved ergonomics when using it as a tablet. However, the placement of the power button on the side, while convenient for accessibility in any posture, may require a moment’s pause to locate when in standard laptop orientation.


Display and audio


The Lenovo Yoga 7i 2-in-1 laptop sports a 14-inch 2.8K resolution OLED touchscreen of 120Hz refresh rate. Since the display is stretched in 16:10 aspect ratio, it shows more vertical content besides improving the device display-to-body ratio.


OLED displays are known for rich colours and excellent contrast, and the touchscreen panel on the Lenovo Yoga 7i 2-in-1 is no different. In fact, the display is enhanced by value-added features for better performance. These features include support for Dolby Vision and Super Resolution. Both these entertainment-focused features elevate the content viewing experience.


Perhaps the most important feature that makes a noticeable impact on everyday use is 120Hz refresh rate. Because of it, the display delivers a smooth experience, especially evident during touch-based scrolling.


The only downside of the OLED touchscreen on the Lenovo Yoga 7i 2-in-1 is the glossy profile, which hampers sunlight legibility despite the panel being adequately bright.


In the audio department, the laptop boasts a dual-speaker system with support for Dolby Atmos. These are facing upwards, thus, sound clear, loud, and balanced. Speaking of loudness, the speakers are good for everyday use but do not expect an immersive experience out of them since they are lacking in depth and clarity. Not that the speakers sound shallow, but they are not rich sounding as well.


Camera and microphone


The Lenovo 7i 2-in-1 laptop has a FHD camera with a physical shutter for added privacy. While delivering decent video output during well-lit conditions, the camera’s performance may suffer in low-light conditions, resulting in slightly hazy images. While suitable for video conferences, it may not meet the requirements for streaming or recording purposes. Regarding facial unlock, the laptop’s detection speed may occasionally lag, albeit it maintains overall accuracy.


Likewise, the microphone provides satisfactory audio quality during calls, although recordings may sound somewhat muffled. However, the microphone excels in background noise reduction, ensuring clear audio transmission during calls.


Keyboard, trackpad and digital pen


The Lenovo Yoga 7i 2-in-1 sports a full-sized backlit keyboard without a dedicated numeric pad, a common omission in compact laptops. Typing experience on the keyboard is a delight, requiring minimal adjustment time. Despite its slim profile, the keys offer sufficient downward travel, providing tactile feedback with each press. Functionality-wise, apart from the new Copilot button adjacent to the left arrow key, there are few dedicated buttons. This Copilot button swiftly launches Microsoft’s AI chatbot.


Coming to the touchpad, it is adequately sized and is accessible with both thumbs while typing. In default settings, the touchpad sensitivity is set quite high, leading to occasional accidental clicks. While sensitivity adjustments can be made in settings, this may slow down scrolling speed.


Lenovo includes a pen in the box, powered by a AAAA battery, made from plastic and equipped with a pressure-sensitive tip, ideal for creators for sketching. Testing it on Paint revealed no noticeable lag, with smooth performance on the display. While the pen is a valuable inclusion, most touch controls will likely still be managed with fingers. Unfortunately, the pen lacks magnetic attachment, instead relying on a plastic holder that slides into the USB-A port for storage.


Software


Lenovo provides various customisation options and settings through its Vantage software. Users can access features like “Super Resolution” mode to enhance display resolution and preset Dolby Atmos sound profiles tailored for various activities such as Movie or Gaming.


The Lenovo Vantage app also offers a range of modes suited to different tasks, including Media, Meeting, Learning, and Gaming modes. Additionally, it features an “Automatic” mode powered by Lenovo’s AI engine, which it said adjusts settings based on current activity and user behaviour. Three power modes—Performance, Battery Saver, and Adaptive—are available, with the Adaptive mode dynamically adjusting based on usage patterns. The Vantage software also offers multiple widget options for easy access to these modes and personalisation settings.


Beyond Lenovo Vantage, the Yoga 7i 2-in-1 laptop features “Flip to Boot” function, which instantly powers up the laptop when flipped open. Additional features include Windows Studio Effects, providing background effects like background blur during video calls for enhanced privacy and auto framing to keep the camera focused on the user’s face regardless of position. During testing, the background blur effect functioned effectively, though the impact of auto framing was less effective.


Performance


When it comes to performance, the Lenovo Yoga 7i 2-in-1 lives up to expectations as a portable workstation. Powered by Intel Core Ultra 7 155H processor and 16GB LPDDR5X RAM, the laptop effortlessly handles multitasking while maintaining efficient thermal management. Even with numerous tabs, windows, and background apps open, the laptop remains responsive and cool to the touch, operating silently throughout.


Surprisingly, the laptop delivers a satisfactory gaming experience even with casual gaming titles like eFootball 2024. What’s particularly impressive is its battery life. During testing, the laptop sustained continuous multitasking for 9 hours, still retaining 12 per cent battery life. This longevity is further enhanced when using the Automatic mode, which optimises power consumption based on usage patterns. The included 65W charger utilises the USB-C port, replenishing the battery to 80 per cent capacity in just over half an hour, ensuring sufficient power for a full day’s office use in optimised mode.


Verdict


Priced at Rs 1,01,990, the Lenovo 7i 2-in-1 offers compelling value for those seeking a Windows-powered work machine that doubles as an entertainment hub. It boasts a long-lasting battery, and supports fast charging, making it a versatile choice for people on the go. However, for consumers seeking a portable laptop for content creation, alternative options with discrete graphics may better suit their needs.



Source link

Google Wallet now available in India: How is it different from Google Pay?

Google Wallet now available in India: How is it different from Google Pay?



Google Wallet is now officially available in India on Android smartphones. With it, you can now save essential digital documents such as boarding passes, loyalty cards, event tickets, public transport passes, gift cards, and more. Google Wallet app is now available on the Google Play Store in India for Android smartphones.


Currently, Google Wallet supports access to digital documents from 20 Indian brands including e-commerce platform Flipkart, Airline companies such as Air India and Indigo, cinema multiplex chain PVR and INOX, and more. Below are the details:


Google Wallet: What you can do with it


Movie and event tickets: Android smartphone users will be able to save their tickets from PVR and INOX movie theatres to their Google Wallet.


Boarding pass: In collaboration with airlines like Air India, Air India Express, Indigo, and online ticketing platforms such as MakeMyTrip, Easymytrip, Ixigo, Google allows Wallet users to save their mobile boarding pass into their Google Wallet for quick access. Additionally, Pixel smartphone users can add their boarding pass to Google Wallet simply by taking a screenshot.


Loyalty and Gift cards: Users can save and access their loyalty points and gift cards from Flipkart (Supercoins), Dominos, Shoppers Stop, and brands supported by loyalty program enablers such as Pinelabs, EasyRewardz and Twid, directly from Google Wallet.


Public transport: Google has partnered with Kochi Metro, Hyderabad Metro, VRL Travels, and Abhibus to provide access to transit tickets of travellers using these services.


Documents: Google Wallet users can create new passes in wallets by simply clicking a picture of any document containing a barcode or QR code. This includes luggage tags and parking receipts.


Apart from these, Google automatically adds tickets and reservations to Google Wallet by scanning details from linked Gmail accounts. However, this functionality is optional and users will have to enable “smart personalization” from their Gmail settings.


How is it different from Google Pay


Google Pay is a payments platform with some features that overlap with Google Wallet. However, it is a more comprehensive digital service for payments. It encompasses peer-to-peer payments, in-store payments, online purchases, and even transit ticketing. Google Pay is designed to be a versatile digital wallet that users can use for a wide range of transactions, including contactless payments at physical stores, in-app purchases, and online shopping.


Google Wallet, India version, is a platform to store digital assets from multiple sources at one place. With it, you can save essential digital documents such as boarding passes, loyalty cards, event tickets, public transport passes, gift cards, and more.


“Google Pay is not going anywhere. It will remain our primary payments app. Google Wallet is specifically tailored for non-payment use cases,” said Ram Papatla, GM & India Engineering Lead, Android at Google.

First Published: May 08 2024 | 1:45 PM IST



Source link

Google introduces private digital wallet for Android users in India

Google introduces private digital wallet for Android users in India



Google has introduced a private digital wallet for Android users in India, allowing users to securely store cards, tickets, passes, keys and IDs, an official said on Wednesday.


Google Wallet can be downloaded from Play Store and would allow users to store their debit cards, credit cards, loyalty cards, and gift cards among other things.


It is different from Google Pay app which helps manage money and finances.


“Google Pay is not going anywhere. It will remain our primary payments app. Google Wallet is specifically tailored for non-payment use cases,” said Ram Papatla, GM & India Engineering Lead, Android at Google.

(Only the headline and picture of this report may have been reworked by the Business Standard staff; the rest of the content is auto-generated from a syndicated feed.)

First Published: May 08 2024 | 12:45 PM IST



Source link

YouTube
Instagram
WhatsApp