Connect with us

Technology

UC San Diego, Tsinghua University researchers just made AI way better at knowing when to ask for help

Published

on

UC San Diego, Tsinghua University researchers just made AI way better at knowing when to ask for help

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More


A team of computer scientists has developed a method that helps artificial intelligence understand when to use tools versus relying on built-in knowledge, mimicking how human experts solve complex problems.

The research from the University of California San Diego and Tsinghua University demonstrates a 28% improvement in accuracy when AI systems learn to balance internal knowledge with external tools — a critical capability for deploying AI in scientific work.

How scientists taught AI to make better decisions

“While integrating LLMs with tools can increase reliability, this approach typically results in over-reliance on tools, diminishing the model’s ability to solve simple problems through basic reasoning,” the researchers write in their paper. “In contrast, human experts first assess problem complexity using domain knowledge before choosing an appropriate solution approach.”

Advertisement

The new method, called “Adapting While Learning,” uses a two-step process to train AI systems. First, the model learns directly from solutions generated using external tools, helping it internalize domain knowledge. Then, it learns to categorize problems as either “easy” or “hard” and decides whether to use tools accordingly.

The two-step process researchers developed to teach AI systems when to use tools versus rely on internal knowledge, mirroring how human experts approach problem-solving. (Credit: UC San Diego / Tsinghua University)

Small AI model outperforms larger systems on complex tasks

What makes this development significant is its efficiency-first approach. Using a language model with just 8 billion parameters — far smaller than industry giants like GPT-4 — the researchers achieved a 28.18% improvement in answer accuracy and a 13.89% increase in tool usage precision across their test datasets. The model demonstrated particular strength in specialized scientific tasks, outperforming larger models in specific domains.

This success challenges a fundamental assumption in AI development: that bigger models necessarily yield better results. Instead, the research suggests that teaching AI when to use tools versus rely on internal knowledge — much like training a junior scientist to know when to trust their calculations versus consult specialized equipment — may be more important than raw computational power.

Examples of how the AI system handles different types of climate science problems: a simple temperature calculation (top) and a complex maritime routing challenge (bottom). (Credit: UC San Diego / Tsinghua University)

The rise of smaller, smarter AI models

This research aligns with a broader industry shift toward more efficient AI models in 2024. Major players including Hugging Face, Nvidia, OpenAI, Meta, Anthropic, and H2O.ai have all released smaller but highly capable models this year.

Hugging Face’s SmolLM2, with versions as small as 135 million parameters, can run directly on smartphones. H2O.ai’s compact document analysis models have outperformed tech giants’ larger systems on specialized tasks. Even OpenAI entered the small model arena with GPT-4o Mini, offering similar capabilities at a fraction of the cost.

This trend toward “AI downsizing” reflects growing recognition that bigger isn’t always better — specialized, efficient models can often match or exceed the performance of their larger counterparts while using far fewer computational resources.

Advertisement

The technical approach involves two distinct learning phases. During training, the model first undergoes what the researchers call “World Knowledge Distillation” (WKD), where it learns from solutions generated using external tools. This helps it build up internal expertise.

The second phase, “Tool Usage Adaptation” (TUA), teaches the system to classify problems based on its own confidence and accuracy in solving them directly. For simpler problems, it maintains the same approach as in WKD. But for more challenging problems, it learns to switch to using external tools.

Business impact: More efficient AI systems for complex scientific work

For enterprises deploying AI systems, this research addresses a fundamental challenge that has long plagued the industry. Current AI systems represent two extremes: they either constantly reach for external tools — driving up computational costs and slowing down simple operations — or dangerously attempt to solve everything internally, leading to potential errors on complex problems that require specialized tools.

This inefficiency isn’t just a technical issue — it’s a significant business problem. Companies implementing AI solutions often find themselves paying premium prices for cloud computing resources to run external tools, even for basic tasks their AI should handle internally. On the flip side, organizations that opt for standalone AI systems risk costly mistakes when these systems attempt complex calculations without proper verification tools.

Advertisement

The researchers’ approach offers a promising middle ground. By teaching AI to make human-like decisions about when to use tools, organizations could potentially reduce their computational costs while maintaining or even improving accuracy. This is particularly valuable in fields like scientific research, financial modeling, or medical diagnosis, where both efficiency and precision are crucial.

Moreover, this development suggests a future where AI systems could be more cost-effective and reliable partners in scientific work, capable of making nuanced decisions about when to leverage external resources — much like a seasoned professional who knows exactly when to consult specialized tools versus rely on their expertise.

The power of knowing when to ask for help

Beyond the immediate technical achievements, this research challenges the bigger-is-better paradigm that has dominated AI development. In demonstrating that a relatively small model can outperform its larger cousins by making smarter decisions about tool use, the team points toward a more sustainable and practical future for AI.

The implications extend far beyond academic research. As AI increasingly enters domains where mistakes carry real consequences – from medical diagnosis to climate modeling – the ability to know when to seek help becomes crucial. This work suggests a future where AI systems won’t just be powerful, but prudent – knowing their limitations just as skilled professionals do.

Advertisement

In essence, the researchers have taught AI something fundamentally human: sometimes the smartest decision is knowing when to ask for help.


Source link
Continue Reading
Advertisement
Click to comment

You must be logged in to post a comment Login

Leave a Reply

Technology

Vodafone-Three merger could get green light, watchdog says

Published

on

Vodafone-Three merger could get green light, watchdog says

The regulator says a merger between Vodafone and Three could go ahead – if both companies make price promises for consumers and commit to boosting the UK’s 5G rollout.

The Competition and Markets Authority (CMA) had previously said that creating what would be the country’s biggest mobile network could drive up prices and harm competition.

But it has now provisionally concluded those concerns could be addressed – and the merger could proceed – if the firms agree to its proposed remedies.

A Vodafone spokesperson said both companies would need to study the CMA’s proposal more closely but believed on first impressions it “provides a path to final clearance”.

Advertisement

They insisted, as they have throughout, that the deal was in everyone’s interests.

“It will bring significant benefits to businesses and consumers throughout the UK, and it will bring advanced 5G to every school and hospital across the country,” they said.

The CMA’s findings are the latest step in its probe into the merger, which began in January.

Vodafone and Three announced their plans to merge their UK-based operations in June last year.

Advertisement

Their combined network would have around 27 million customers.

“We believe this deal has the potential to be pro-competitive for the UK mobile sector if our concerns are addressed,” said Stuart McIntosh, who is leading the CMA panel investigating the merger.

“We anticipate in the longer term that the significant commitment to upgrade the merged companies network over the next 10 years or so will ultimately create a competitive environment that will maintain the competition we’ve seen in mobile in recent years,” he told the Today programme, on BBC Radio Four.

But he also made clear that short term commitments not to increase the price of certain existing mobile tariffs and data plans for at least three years were also key to making sure consumers did not lose out.

Advertisement

The regulator also said upholding pre-agreed deals or prices with Mobile Virtual Network Operators such as Sky Mobile, Lyca and Lebara could protect consumers and and wholesale customers alike.

Industry analyst Paolo Pescatore told the BBC it marked “another key step towards approval” and showed all concerned were trying to find a way to make the deal happen.

The two largest players in the market are currently EE and 02 – Mr Pescatore said a merged Vodafone and Three would be in a better place to take them on.

“To date, both parties are demonstrating that this is genuinely in the interest of UK plc, the economy, and users which paves the way for a far stronger three-player market than the current imbalance,” he said.

Advertisement

The CMA is seeking responses to its proposed remedies by 12 November, with a deadline of 7 December for a final decision on the merger.

Source link

Continue Reading

Technology

NYT Mini Crossword today: puzzle answers for Tuesday, November 5

Published

on

NYT Mini Crossword today: puzzle answers for Saturday, September 21

The New York Times has introduced the next title coming to its Games catalog following Wordle’s continued success — and it’s all about math. Digits has players adding, subtracting, multiplying, and dividing numbers. You can play its beta for free online right now. 
In Digits, players are presented with a target number that they need to match. Players are given six numbers and have the ability to add, subtract, multiply, or divide them to get as close to the target as they can. Not every number needs to be used, though, so this game should put your math skills to the test as you combine numbers and try to make the right equations to get as close to the target number as possible.

Players will get a five-star rating if they match the target number exactly, a three-star rating if they get within 10 of the target, and a one-star rating if they can get within 25 of the target number. Currently, players are also able to access five different puzzles with increasingly larger numbers as well.  I solved today’s puzzle and found it to be an enjoyable number-based game that should appeal to inquisitive minds that like puzzle games such as Threes or other The New York Times titles like Wordle and Spelling Bee.
In an article unveiling Digits and detailing The New York Time Games team’s process to game development, The Times says the team will use this free beta to fix bugs and assess if it’s worth moving into a more active development phase “where the game is coded and the designs are finalized.” So play Digits while you can, as The New York Times may move on from the project if it doesn’t get the response it is hoping for. 
Digits’ beta is available to play for free now on The New York Times Games’ website

Source link

Continue Reading

Technology

The pros and cons of using a render farm

Published

on

The pros and cons of using a render farm

As a 3D artist, you’re likely aware of how time-consuming and resource-demanding the rendering process can be. This is where a render farm proves invaluable. A render farm is essentially a network of computers dedicated solely to rendering visual effects and 3D animations.

With render farms, artists can transform simple sketches into realistic images that incorporate shadows, textures, lighting, and intricate details. By distributing rendering tasks across hundreds of machines, rendering times are drastically reduced, making adjustments, edits, and the entire production process faster and more efficient.

In this article, we’ll explore the advantages and drawbacks of using a render farm for animation projects. Read on to learn more!

What are the advantages of using a render farm

Image 83949848349883

As you can probably guess, a render farm has some obvious advantages.

Advertisement

Faster rendering

The primary advantage is the boost in rendering speed. Rendering alone can take up hours or even days, and during that time, your computer is often tied up, leaving you unable to use it for other tasks. Using a render farm accelerates the rendering process significantly, allowing you to devote more time to animation or design work.

Access to the latest technology

Render farms often use the latest hardware and software updates to stay competitive, which means you get access to cutting-edge technology without the need to constantly upgrade your own equipment. This can improve rendering quality and speed, as high-performance machines can handle more complex scenes and higher resolutions with ease, providing results that might be difficult to achieve on an average workstation.

Ease of Use

You can submit your projects for rendering through an application or online interface at the majority of render farms, if not all of them. This eliminates some of the trouble involved in rendering files, particularly when working locally and transferring data between computers for rendering.

Technical assistance for clients

Render farms often provide 24/7 customer service to assist with any problems or specific questions you might have about rendering times, costs, or other needs.

Advertisement

Enhanced quality control

With faster rendering, you gain more time to focus on design and testing. Rather than working in the dark and hoping everything aligns in the final render, you can conduct multiple test renders, allowing for quality control and experimentation.

Scalability

Render farms are built to be scalable, giving you access to more nodes to match your project’s demands.

Remote rendering

As long as you have an internet connection, you can work from anywhere and access a vast render farm remotely, offering greater flexibility.

What are the disadvantages of using a render farm?

While there are many benefits to render farms, there are also some downsides to consider.

Advertisement

Security

Submitting your project to a render farm means sharing intellectual property. While most render farms will sign an NDA to protect your work, bringing a third party into the process still involves some level of risk.

Software compatibility

Different production teams use various applications and plugins to create renders. This means that a render farm must support the software and plugins you’re using; otherwise, its services won’t be compatible with your project.

Limited customization

Using a render farm means working within the parameters of their setup, which can sometimes limit your customization options. If your project requires specific software configurations, unique plug-ins, or particular hardware requirements, a render farm may not be able to accommodate these fully. This can restrict the creative control and flexibility you might have when rendering locally.

Internet dependence

Render farms rely heavily on stable internet connections for uploading and downloading files. If you have a slower or inconsistent internet connection, large uploads or downloads can take significant time and may even disrupt the workflow. This dependence on connectivity can create delays, particularly for large projects or in areas with limited internet infrastructure.

Advertisement

Cost

Running a render farm is costly, as it requires high-end hardware on a large scale, constant power, and potentially different plugins and software. Software licenses alone can be expensive, even in bulk.

Depending on your rendering needs, setting up one or two dedicated workstations for occasional rendering might save time and money over time. However, as projects grow more complex, a render farm can still be cost-effective. Hardware and maintenance expenses can skyrocket as workstation numbers increase. In most cases, working with a professional render farm remains easier and, in some cases, less costly—unless you’re a large production company with a generous budget.

To conclude

Outsourcing rendering to a render farm can be a fantastic way for animators and designers to save both time and money. Compared to local rendering, render farms often allow you to create higher-quality work faster.

If you’re looking to streamline your rendering, Fox Renderfarm offers dependable, fast cloud rendering solutions that can help you save time and resources. Take advantage of their $25 free render coupon and see how much easier your projects can become!

Advertisement

Source link

Continue Reading

Science & Environment

Silicon anodes are ahead of solid-state batteries in race to power EVs

Published

on

Silicon anodes are ahead of solid-state batteries in race to power EVs


A Wallbox EV charger for electric car is displayed during the “Mondial de l’Auto” at Parc des Expositions on October 15, 2024 in Paris, France.

Chesnot | Getty Images News | Getty Images

Silicon anodes appear to be leading the way in the race to commercialize next-generation battery technologies for electric vehicles.

The buzz around silicon-based anodes, which promise improved power and faster charging capabilities for EVs, has been growing in recent months — just as the hype around solid-state batteries seems to have fizzled.

Advertisement

It comes as increasing EV sales continue to drive up global battery demand, prompting auto giants to team up with major cell manufacturers on the road to full electrification.

While some OEMs (original equipment manufacturers) have inked deals with solid-state battery developers, carmakers such as Mercedes, Porsche and GM have all bet big on silicon anodes to deliver transformative change in the science behind EVs.

A recent report from consultancy IDTechEx described the promise of advanced silicon anode materials as “immense” for improving critical areas of battery performance, noting that this potential hadn’t gone unnoticed by carmakers and key players in the battery industry.

It warned, however, that challenges such as cycle life, shelf life and — perhaps most importantly — cost, need to be addressed for widespread adoption.

Advertisement

Venkat Srinivasan, director of the Collaborative Center for Energy Storage Science at the U.S. government’s Argonne National Laboratory in Chicago, said silicon anodes appear to have the edge over solid-state batteries.

“If there’s a horse race, silicon does seem to be ahead at least at this moment, but we haven’t commercialized either one of them,” Srinivasan told CNBC via videoconference.

How silicon could enable cheaper EVs, electric flight and more powerful batteries

Srinivasan said five years ago silicon-anode batteries had a calendar life of roughly one year, but recent data appears to show a dramatic improvement in the durability of these materials, with some tests now projecting a three to four-year calendar life.

Unlike the cycle life of a battery, which counts the number of times it can be charged and discharged, the calendar life measures degradation over time. Typically, the calendar life of a battery refers to the period in which it can function at over 80% of its initial capacity, regardless of its usage.

Srinivasan said solid-state batteries, long billed as the “holy grail” of sustainable driving, still have a long way to go before they can match the recent progress made by silicon anodes.

Advertisement

“That transition still has to be made in solid-state with their metal batteries and that’s why I think you’re hearing from people that, hey, it looks like that promise hasn’t panned out,” Srinivasan said.

“That doesn’t mean we won’t get there. It may happen in a few years. It just means that it feels like today silicon is in a different part of the technology readiness level.”

Silicon anodes vs. solid-state batteries

Analysts say silicon anodes theoretically offer 10 times the energy density as graphite, which are commonly used in battery anodes today. Yet, these same materials typically suffer from rapid degradation when lots of silicon is used.

“Silicon anodes and solid-state batteries are two emerging technology trends in the EV battery market aimed at pushing the boundaries of high-performance battery cells,” Rory McNulty, senior research analyst at Benchmark Mineral Intelligence, told CNBC via email.

Advertisement

A researcher checks the electromagnet de-ironing machine at the Daejoo Electronic Materials Co. R&D center in Siheung, South Korea, on Thursday, June 22, 2023.

Bloomberg | Bloomberg | Getty Images

It has typically been the case that better battery performance comes at the cost of longevity or safety, McNulty said. Silicon anodes, for example, are known to swell significantly during charging, which reduces the battery’s longevity.

By comparison, McNulty said solid-state batteries were claimed to greatly improve the stability of the electrolyte to high performance electrode materials, combating the challenges of using high energy density materials such as silicon and lithium.

Advertisement

As the name suggests, solid-state batteries contain a solid electrolyte, made from materials such as ceramics. That makes them different from conventional lithium-ion batteries, which contain liquid electrolyte.

Especially in the West, advances in the area of silicon anodes [are] seen as strategic opportunity to catch up with China.

Georgi Georgiev

Battery raw materials analyst at Fastmarkets

Japan’s Toyota and Nissan have both said they are aiming to bring solid-state batteries into mass production over the coming years, while China’s SAIC Motor Corp reportedly said in early September that its MG brand would equip cars with solid-state batteries within the next 12 months.

Advertisement

Nonetheless, analysts remain skeptical about when solid-state batteries will actually make it to market.

A strategic opportunity?

“Silicon based anodes promise to be the next-generation technology in the anode field, providing a solution for faster charging,” Georgi Georgiev, battery raw materials analyst at consultancy Fastmarkets, told CNBC via email.

Georgiev said several industry players have been looking into the potential of silicon anodes, from well-established anode suppliers in China and South Korea to new players like Taiwan’s ProLogium and U.S. manufacturers Group14 and Sila Nanotechnologies.

“Especially in the West, advances in the area of silicon anodes [are] seen as strategic opportunity to catch up with China, which dominates the graphite-based anode supply chains with Chinese anode producers holding 98% of the global anode market for batteries,” Georgiev said.

Advertisement

“However, there are significant technical challenges going to 100% silicon anode such as silicon expansion affecting the longevity of the batteries and currently there are several routes to produce silicon anodes,” he added.

A FEV x ProLogium Technology Co. 100% silicon composite anode next-generation battery at the Paris Motor Show in Paris, France, on Tuesday, Oct. 15, 2024.

Bloomberg | Bloomberg | Getty Images

Taiwanese battery maker ProLogium debuted the world’s first fully silicon anode battery at the Paris Motor Show last month, saying it’s new fast-charging battery system not only surpassed traditional lithium-ion batteries in performance and charging efficiency but also “critical industry challenges.”

Advertisement

ProLogium, citing test data, said it’s 100% silicon anode battery could charge from 5% to 60% in just 5 minutes, and reach 80% in 8.5 minutes. It described the advancement as an “unmatched achievement in the competitive EV market,” which will help to reduce charging times and extend the range of EVs.

Fastmarkets’ Georgiev said a big question mark over the commercialization of silicon anodes is the cost of production and whether any of the major silicon-anode producers “could produce material at scale with a consistent quality and at a competitive price — [a] major requirements of OEMs.”

“At this stage silicon anodes are used more as an additive to graphite-based anodes and in the years to come we expect to see increase of silicon share in anode, but in combination with graphite, while 100% silicon anodes will take longer time to enter the mass market,” he added.



Source link

Advertisement

Continue Reading

Technology

The missing link in Amazon’s ereader lineup

Published

on

The missing link in Amazon’s ereader lineup

Well, it finally happened. After years of waiting and requests, Amazon debuted the $280 Kindle Colorsoft, its first ereader with a color display. The company’s ereaders have dominated this space since the original Kindle came out 17 years ago, but in this case, it feels like Amazon is playing catch-up. Color E Ink displays aren’t novel: we’ve tested and reviewed a number of color ereaders and E Ink tablets from Kobo, Boox and reMarkable in recent years. But Amazon is essentially trying to pull an Apple with the Colorsoft: with claims that color E Ink technology just wasn’t good enough to put into a Kindle until now, Amazon’s promising the Colorsoft gets this implementation right thanks in part to the custom tweaks it made to the display. And, unsurprisingly, Amazon’s ready to charge you a premium for it. So is it all it’s cracked up to be? As you might suspect, the answer isn’t as simple as yes or no.

Update 11/4/24, 11:54am ET: There have been multiple reports of Kindle Colorsoft owners seeing a yellow band at the bottom of their ereader’s display. My review unit was not affected during the time I spent testing the Colorsoft, but upon checking this morning, it has appeared for me, too. I reached out to Amazon for comment and a spokesperson told Engadget: “We’re aware of a small number of reports from customers who see a yellow band along the bottom of the display. We take the quality of our products seriously and are looking into it. If customers notice this on their device, they can reach out to our customer service team.”

Let’s get the tech details squared away first. The Kindle Colorsoft’s seven-inch screen is based on E Ink Kaleido 3 technology, but a representative from the Kindle team explained to me that they developed a custom display stack for this device. That means they made quite a few changes to the tech in order to achieve things like higher-contrast pigments and improved speeds overall. The Colorsoft’s custom oxide backplane uses 24 driving volts to move pigments around more quickly and it helps those pigments appear with better contrast. Nitride LEDs enhance colors and brightness, and a custom coating in between the display’s layers helps focus light through each pigment so there’s less color mixing. Some of the same tech helps make page-turns quicker and supposedly reduces excessively noticeable screen refreshing when you go from one color page to another, or pinch-and-zoom on an image.

Engadget / Amazon

The Kindle Colorsoft (finally) brings color to Amazon’s ereader lineup. It’s a solid premium ereader that will be ideal for those who primarily read things like comics, graphic novels and other material best experienced in full-color glory.

Advertisement

Pros

  • Color on a Kindle, finally!
  • Quick page-turns and load times
  • Pinch-to-zoom feature lets you get closer to details
  • Auto-adjusting front light
  • No lock screen ads by default
Cons

  • Expensive
  • Screen has a noticeable blue skew to it when the warm light is off
  • Slight reduction in sharpness and contrast when reading black-and-white text

$280 at Amazon

That’s all to say that Amazon would like you to believe that this E Ink Kaleido 3 screen is not like the other girls in this space, and while I do not doubt the Kindle team’s efforts, the differences are not as dramatic as the story would suggest. Until Kobo updates the Sage with color, the closest competitor to the $280 Kindle Colorsoft is the $220 Kobo Libra Colour (in size, platform and overall experience), so I did a lot of side-by-side comparisons of the two.

The biggest difference I saw was that the Kobo’s screen skews warmer than the Kindle’s; I kicked the brightness up to the maximum and turned all warm/natural light settings down to zero on both devices and the difference was noticeable, regardless of if the displays were showing color images, black-and-white text or a mix of the two. This would suggest that the Kindle will show more accurate colors more often since there’s less of a warm lean to its display.

But on the flip side, the Kindle screen’s blue tint was just as noticeable, particularly in low-light situations (like a dark office or a dimly lit living room). At max brightness with warmth down to zero, the Kindle’s screen was borderline uncomfortable to read in those environments — but all it took was a slight adjustment to warmth level four (out of 24) to get it to match the Kobo’s display in warmth almost exactly (at least to my eyes). That made it more comfortable to stare at in dark spaces. I also compared the Colorsoft’s screen to my personal Kindle Paperwhite (previous generation) and the blue skew was noticeable there too.

Ultimately, how much warm or cool light you prefer while reading is up to personal taste. My preferences would lead me to adjust the warmth on the Kindle to be a bit higher than zero, mimicking that of the Kobo. In an unscientific poll of the Engadget staff, both in person and with device photos, everybody preferred the screen on the Kobo. I think the slight added warmth in Kobo’s screen makes colors appear a bit more saturated and more inviting overall. It also gets close to mimicking the look of actual physical pages (as much as one of these devices could, at least), and that’s the experience I’m going for when I read pretty much anything. But I could understand why some hardcore comic fans would want to start off with the most color-accurate baseline as possible, and then adjust from there to fit their preferences.

In addition to adjusting the warm light, the Kindle Colorsoft has vivid mode, which “enhances color in less saturated images.” For the sake of efficacy, most of my time was spent in standard mode when testing the Colorsoft. But in trying out vivid mode, I noticed that its enhanced saturation was most noticeable in warm-toned images: reds appeared ever-so-slightly more striking, while oranges and yellows had a more bronze effect and the like. But I had to turn vivid mode on and off a few times to clock the effect because it’s quite subtle.

Then there’s the question of actually reading words on the Colorsoft. Even the Kindle team representative I spoke with acknowledged that, due to the extra physical layer in the screen that enables color, one might notice a bit less sharpness and contrast in black-and-white text on the Colorsoft’s screen. That’s not unique to this particular Kindle — every color ereader will have this issue to some degree.

When comparing the text-only experience of the Kindle Colorsoft to that of the Kobo Libra Colour, I found both to be quite good and comparable to one another. Where there’s a bigger possibility for discrepancies is in a comparison of the Colorsoft to, say, the new Kindle Paperwhite. I did not have the latter device to compare to, but I did have my personal, previous-generation Paperwhite, and the difference was ever so slight, with the standard Paperwhite having the (small) upper hand in the contrast and sharpness departments.

Advertisement

Amazon Kindle Colorsoft

Photo by Valentina Palladino / Engadget

The color screen is the most consequential thing about the Kindle Colorsoft, and chances are if you’ve had a Kindle in the past, the reading experience on this new device will feel quite familiar. The Kindle UI hasn’t changed much, still dividing the main screen into Home and Library options. The former is basically a space for Amazon to serve you personalized book recommendations and promote new Kindle releases, while the latter shows your entire digital library including books, documents, Audible audiobooks and library loans. The Library page is striking in color and there is something undeniably satisfying about seeing all of your title covers in full-color glory.

While reading a book, you can still customize and save different themes with fonts, font sizes and page layouts that best suit your preferences. You still have the option to quickly navigate within a title by page, location, chapter and even popular highlights. Page turns are speedy and will likely be an improvement for anyone coming from an older Kindle or other ereader.

Annotations and your own highlights are collected in the same place for easy reference, and with the latter, you can filter by highlight color as well. You have four colors to choose from on the Colorsoft — orange, yellow, blue and pink — so if you use the yellow highlighter to mark favorite quotes, you can then filter by just that color. Note that all highlighting and note-taking must be done with your fingers because, unlike the Kindle Scribe, the Colorsoft has no stylus support.

The pinch-to-zoom feature on the Colorsoft is good for those who read a lot of graphic novels and comics. Amazon developed a custom algorithm to make this motion as smooth as possible, and it’s a pretty good experience, albeit not a unique one. You can also pinch to zoom on the Kobo Libra Colour, which mostly helps get in closer to comic panels to read small text or better see minute details.

Advertisement

On both devices, there are full-screen refreshes when you pinch to zoom on color images and the speed of completion is roughly the same. I also found image quality to be quite similar as well, and it’s worth noting that art style can skew your impression of an image’s quality. A comic that employs clean, distinct lines compared to one that’s more grungy and watercolor-like will always come off more crisp.

The Colorsoft, like the regular Kindle Paperwhite Signature Edition, also has an auto-adjusting front light that promises to illuminate the display just right depending on if you’re reading outside on a sunny day, in the dark cabin of an airplane or anywhere else. It’s a nice hardware perk to have and, with the feature turned on, removes much of manual fiddling that some might find annoying to do when they take their Kindle into different environments.

Amazon Kindle Colorsoft

Photo by Valentina Palladino / Engadget

While we already went through the screen comparisons for the $280 Kindle Colorsoft and the $220 Kobo Libra Colour, there are plenty of other differences between the two that you should consider when picking your next ereader. I’ve added a spec list below to break down the basics, and the most consequential to me are the fact that the Kobo has page-turn buttons and stylus support.

The former is really a matter of preference — you either love physical buttons or you find no use for them — but the latter is pure added value even if you do have to purchase the $70 Kobo stylus separately. It essentially allows you to turn the Libra Colour into a makeshift Kobo Sage or Kindle Scribe, which could be useful for anyone in academia (students and educators alike) or anyone who just loves the feeling of putting pen to “paper.” Meanwhile, the Kindle has the upper hand in its wireless charging capabilities and its slightly cleaner flush-front design.

Advertisement

When it comes to actual content available on Kindle and Kobo devices, the libraries you can purchase from on both are vast: both have ebooks and audiobooks available, and both the Colorsoft and Libra Colour support Bluetooth, so you can listen to audiobooks directly from the device with your wireless headphones. At the time of writing this review, all of the top five New York Times bestseller titles were available on both platforms at the same prices, with the only discrepancies being one that was on sale as a Kindle ebook and one that was available to read for free for Kindle Unlimited and Kobo Plus subscribers. Both of those monthly subscriptions give you unlimited reading access to thousands of titles, but I would give the advantage to Amazon on this one since Kindle Unlimited has been around for much longer. Amazon also has Prime Reading and Kids+ subscriptions that work with Kindle devices and provide even more content to paying subscribers.

As far as borrowing ebooks from your local library goes, it might be a draw. Kobo integrates elegantly with Overdrive, making it nearly seamless to get borrowed books on your ereader. Simply connect your Overdrive account and library card in the settings menu and you can then either browse your library’s offerings directly on device, or use the Libby mobile app to borrow titles and those will appear automatically on your Kobo. My only gripe with this system is that it works best if you have just one library card, since you can only connect one at a time. I’m a unique case where I have at least three library cards and I switch among them in Libby depending on which has the title I’m looking for.

Those like me might prefer the “send to Kindle” option in Libby, which just takes a couple more clicks to get any book from any library network to your Kindle device. The biggest downside here is that my Colorsoft review unit did not show all of my library book covers in full-color glory on the lock screen. After troubleshooting with a representative from Amazon, it was determined to be a title-specific issue. According to Amazon: “For library books, the lock screen is taken from the designated marketing cover for the book used by the library, which may not be the same as the actual book cover.” So just know that if you get most of your reading material from your local library, there’s a chance some of the covers may not display properly on the Colorsoft.

I’d be remiss if I didn’t mention the $250 Boox Go Color 7, which earned a spot in our best ereaders guide after my colleague Amy Skorheim tested it. First, it’s imperative to know that Boox devices require a bit more tech-savvy than a Kindle or a Kobo, as well as a willingness to experiment. They are full-blown Android tablets after all, and that might be exactly what you’re looking for if you get your reading material from many different sources, since it offers access to the Google Play Store and all of its apps. As for specs, the Go Color 7 has a seven-inch Kaleido 3 display with 300 ppi in black and white and 150 ppi in color (similar to the Colorsoft), along with a warm light, page-turn buttons, a splash-resistant design and 64GB of storage (and a microSD card slot for more space!). It’s even more similar to the Kobo Libra Colour in feature set and price, so it’s another option for anyone willing to think a bit outside the standard ereader box.

Advertisement

The decision between Kindle, Kobo and Boox is infinitely more complicated than choosing between the Colorsoft or the standard Kindle Paperwhite — and that’s precisely because of the stark price difference. The Colorsoft is $120 more than the regular Kindle Paperwhite ($100 more if you pay to remove the lock screen ads from the Paperwhite, and $80 more than the Signature Edition) and aside from the color panel, it only adds wireless charging, the auto-adjusting front light sensor and doubles the storage. Remove color from the equation, and arguably the most useful of all of those is the extra storage but, make no mistake, 16GB of space on the Kindle Paperwhite is nothing to scoff at and will be just fine for most people. Unless you’re 100-percent certain that color will make a huge difference in your day-to-day reading experience, the standard Paperwhite is the better value.

Amazon Kindle Colorsoft

Photo by Valentina Palladino / Engadget

While it’s very late to the color E Ink party, the Kindle Colorsoft is a solid premium ereader that provides an excellent experience both in color and black and white. I focused a lot on comparisons in this review because most people will not have the opportunity to have the Colorsoft and any of its contemporaries side by side (unless you’re my dad, who buys almost every ereader and small tablet under the sun). But to be clear, the Colorsoft is a good Kindle, and in many ways, Amazon did pull an Apple here. If you’re already heavily entrenched in the Kindle ecosystem and have been holding out for a color ereader, this is the device to get — just be prepared to pay a premium for it.

Also, like Apple and its various operating systems, there’s something to be said about the convenience and ubiquity of the Kindle ecosystem. The library is seemingly endless (with discounts galore), supplemented by Prime Reading and Kindle Unlimited, so it’s easy to get sucked in when you’re already shopping on Amazon for household goods and holiday gifts. But when it comes to value for your money (and a pure spec breakdown), you can get more from Kobo and Boox devices. Kindle isn’t the only name in town anymore for ereaders, and it hasn’t been for a long time, and that seems to be a more pronounced fact now with the introduction of the Colorsoft. If nothing else, Amazon has finally filled a glaring hole in its ereader lineup with this device.

Amazon Kindle Colorsoft

Advertisement

Kobo Libra Colour

Boox Go Color 7

Price

$280

Advertisement

$220

$250

Display size

7-inch color E Ink Kalaido 3 (with custom display stack)

Advertisement

7-inch color E Ink Kalaido 3

7-inch color E Ink Kalaido 3

Pixel density

300 ppi (black-and-white), 150 ppi (color content)

Advertisement

300 ppi (black-and-white), 150 ppi (color content)

300 ppi (black-and-white), 150 ppi (color content)

Storage

32GB

Advertisement

32GB

64GB

Battery life

Up to 8 weeks

Advertisement

Up to 6 weeks

2,300 mAh capacity; “long-lasting” battery life

Page-turn buttons

No

Advertisement

Yes

Yes

Adjustable warm light

Yes

Advertisement

Yes

Yes

Auto brightness adjustments

Yes

Advertisement

No

No

Auto warm light adjustments

No

Advertisement

Yes

No

Waterproof rating

IPX8

Advertisement

IPX8

Not provided

Pinch-to-zoom support

Yes

Advertisement

Yes

Not provided

Stylus support

No

Advertisement

Yes

No

Audiobook support

Yes, Audible audiobooks

Advertisement

Yes, Kobo audiobooks

Yes, via Android apps

Library support

Yes, via Overdrive’s “Send to Kindle” option

Advertisement

Yes, via built-in Overdrive integration

Yes, via Android apps

Wireless charging

Yes

Advertisement

No

No

USB-C charging

Yes

Advertisement

Yes

Yes

Wi-Fi

Yes

Advertisement

Yes

Yes

Bluetooth

Yes

Advertisement

Yes

Yes

Source link

Continue Reading

Technology

Nintendo’s 1st half sales drop 34% as Switch shows its age

Published

on

Nintendo's 1st half sales drop 34% as Switch shows its age

Nintendo reported that sales for the first half of 2024 fell 34.1% to 523.2 billion yen ($3.43 billion) as it saw a slowdown in sales for Nintendo Switch hardware and games.

And in the mobile and movie-related intellectual property business, sales decreased by 43.3% year-on-year to 31.2 billion yen ($204.8 million), mainly due to the decrease in income from visual content related to The Super Mario Bros. Movie.

Meanwhile, R&D expenses went up 15% in the half year — perhaps a sign that the company’s next-generation game console is coming soon. (It’s expected in 2025). Foreign currency expenses were also higher. Overall, net profit was 108.6 billion yen ($713 million, down 59.9%).

As a result of the weaker quarterly results, Nintendo downgraded its forecast from 1,350 billion yen in the fiscal year ending March 31, 2025 to 1,280 billion yen, down 5.2%. It also expects net profit to be flat.

Advertisement

Instead of selling 13.5 million Nintendo Switch units, Nintendo now expects to sell 12.5 million, down 7.4%, in the fiscal year. And it expects to sell 160 million units of software, down 3% from the earlier expected 165 million units.

So far in the half year, Nintendo has sold 4.72 million Switch devices (down 31% from a year ago) and 70.28 million software copies (down 27.6% from a year ago).

Nine titles have sold more than a million units in the first half, including 2.58 million for The Legends of Zelda: Echoes of Wisdom, 2.31 million for Mario Kart 8, and 1.94 million for Paper Mario: The Thousand-Year Door.

Digital sales for the first half of the fiscal year decreased by 26.5% year-on-year to 159.9 billion yen, but as a proportion of total software sales for the dedicated video game platform, digital sales increased 6.1 percentage points to 56.3%.

Advertisement

Digital sales declined year-on-year mainly due to a decrease in sales of downloadable versions of package software and add-on content for Nintendo Switch.


Source link
Continue Reading

Trending

Copyright © 2024 WordupNews.com