Just another news site
The Lenovo Legion Tab (Gen 3) is a tablet made for gamers, with an 8.8 inch, 2560 x 1600 pixel, 165 Hz LTPS display, a Qualcomm Snapdragon 8 Gen 3 processor, 12GB of LPDDR5x memory and 256GB of UFS 4.0 storage. It supports WiFi 7 and Bluetooth 5.4 and …
The Lenovo Legion Tab (Gen 3) is a tablet made for gamers, with an 8.8 inch, 2560 x 1600 pixel, 165 Hz LTPS display, a Qualcomm Snapdragon 8 Gen 3 processor, 12GB of LPDDR5x memory and 256GB of UFS 4.0 storage. It supports WiFi 7 and Bluetooth 5.4 and has two USB Type-C ports (one […]
The post Daily Deals (4-07-2025) appeared first on Liliputing.
$450 system isn’t being sold at a loss, wasn’t priced with tariffs in mind.
Among the many surprises during last week's wider unveiling of the Nintendo Switch 2 was the pricing: $450 for the console itself and $70 to $80 for many first-party games. Now, in a set of interviews posted today (but conducted during last week's unveiling event), Nintendo executives are explaining and defending those prices, even as Trump's tariffs are apparently forcing the company to pause and reassess its whole launch strategy.
Nintendo of America President Doug Bowser was speaking to CNBC just as Trump's tariffs were being announced, and said in the moment that "we're still all trying to really understand [the tariffs] better and understand what possible impacts may rise from that." At the same time, he said that the company "didn't consider tariffs into that equation" when choosing the Switch 2's $450 price and instead went with what "we felt that was going to be the right price point for our consumers and the right value proposition if you will for the device that we're creating."
Elsewhere in that CNBC interview, Bowser suggested that Nintendo isn't following the Wii U example of selling hardware at a loss in order to gain more potential software customers. Instead, Bowser said the company is "trying to find a way to maintain... margins on the hardware even though they may be more slim than they are on software," and then "to make sure that they're seeing the value in their investment in one of our devices" through software.
Independent developers are keeping the command prompt alive on PCs new and old.
We're used to updating Windows, macOS, and Linux systems at least once a month (and usually more), but people with ancient DOS-based PCs still get to join in the fun every once in a while. Over the weekend, the team that maintains FreeDOS officially released version 1.4 of the operating system, containing a list of fixes and updates that have been in the works since the last time a stable update was released in 2022.
FreeDOS creator and maintainer Jim Hall goes into more detail about the FreeDOS 1.4 changes here, and full release notes for the changes are here. The release has "a focus on stability" and includes an updated installer, new versions of common tools like fdisk, and format and the edlin text editor. The release also includes updated HTML Help files.
Hall talked with Ars about several of these changes when we interviewed him about FreeDOS in 2024. The team issued the first release candidate for FreeDOS 1.4 back in January.
The AOOSTAR GT37 is a mini PC with an AMD Ryzen AI 9 HX 370 processor, OcuLink and USB4 ports, two 2.5 GbE LAN ports, support for up to three 4K displays, and support for WiFi 7. First introduced last year as the AOOSTAR GEM10 370, the computer is desi…
The AOOSTAR GT37 is a mini PC with an AMD Ryzen AI 9 HX 370 processor, OcuLink and USB4 ports, two 2.5 GbE LAN ports, support for up to three 4K displays, and support for WiFi 7. First introduced last year as the AOOSTAR GEM10 370, the computer is designed to pack a lot of performance into […]
The post AOOSTAR GT37 mini PC with Ryzen AI 9 HX 370 now available appeared first on Liliputing.
The AOOSTAR GT37 is a mini PC with an AMD Ryzen AI 9 HX 370 processor, OcuLink and USB4 ports, two 2.5 GbE LAN ports, support for up to three 4K displays, and support for WiFi 7. First introduced last year as the AOOSTAR GEM10 370, the computer is desi…
The AOOSTAR GT37 is a mini PC with an AMD Ryzen AI 9 HX 370 processor, OcuLink and USB4 ports, two 2.5 GbE LAN ports, support for up to three 4K displays, and support for WiFi 7. First introduced last year as the AOOSTAR GEM10 370, the computer is designed to pack a lot of performance into […]
The post AOOSTAR GT37 mini PC with Ryzen AI 9 HX 370 now available appeared first on Liliputing.
The AOOSTAR GT37 is a mini PC with an AMD Ryzen AI 9 HX 370 processor, OcuLink and USB4 ports, two 2.5 GbE LAN ports, support for up to three 4K displays, and support for WiFi 7. First introduced last year as the AOOSTAR GEM10 370, the computer is desi…
The AOOSTAR GT37 is a mini PC with an AMD Ryzen AI 9 HX 370 processor, OcuLink and USB4 ports, two 2.5 GbE LAN ports, support for up to three 4K displays, and support for WiFi 7. First introduced last year as the AOOSTAR GEM10 370, the computer is designed to pack a lot of performance into […]
The post AOOSTAR GT37 mini PC with Ryzen AI 9 HX 370 now available appeared first on Liliputing.
Touted 10M token context proves elusive, while early performance tests disappoint experts.
On Saturday, Meta released its newest Llama 4 multimodal AI models in a surprise weekend move that caught some AI experts off guard. The announcement touted Llama 4 Scout and Llama 4 Maverick as major advancements, with Meta claiming top performance in their categories and an enormous 10 million token context window for Scout. But so far the open-weights models have received an initial mixed-to-negative reception from the AI community, highlighting a familiar tension between AI marketing and user experience.
"The vibes around llama 4 so far are decidedly mid," said independent AI researcher Simon Willison in a short interview with Ars Technica. Willison often checks the community pulse around open source and open weights AI releases in particular.
While Meta positions Llama 4 in competition with closed-model giants like OpenAI and Google, the company continues to use the term "open source" despite licensing restrictions that prevent truly open use. As we have noted in the past with previous Llama releases, "open weights" more accurately describes Meta's approach. Those who sign in and accept the license terms can download the two smaller Llama 4 models from Hugging Face or llama.com.
What they delivered is a grey wolf genome carrying a handful of genetic edits.
On Monday, biotech company Colossal announced what it views as its first successful de-extinction: the dire wolf. These large predators were lost during the Late Pleistocene extinctions that eliminated many large land mammals from the Americas near the end of the most recent glaciation. Now, in a coordinated PR blitz, the company is claiming that clones of grey wolves with lightly edited genomes have essentially brought the dire wolf back. (Both Time and The New Yorker were given exclusive access to the animals ahead of the announcement.)
The dire wolf is a relative of the now-common grey wolf, with clear differences apparent between the two species' skeletons. Based on the sequence of two new dire wolf genomes, the researchers at Colossal conclude that dire wolves formed a distinct branch within the canids over 2.5 million years ago. For context, that's over twice as long as brown and polar bears are estimated to have been distinct species. Dire wolves are also large, typically the size of the largest grey wolf populations. Comparisons between the new genomes and those of other canids show that the dire wolf also had a light-colored coat.
That large of an evolutionary separation means there are likely a lot of genetic differences between the grey and dire wolves. Colossal's internal and unpublished analysis suggested that key differences could be made by editing 14 different areas of the genome, with 20 total edits required. The new animals are reported to have had 15 variants engineered in. It's not clear what accounts for the difference, and a Colossal spokesperson told Ars: "We are not revealing all of the edits that we made at this point."
YouTube restricts some, but not all, videos depicting the poker-ish indie game.
Balatro is certainly habit-forming, but there's nothing to be won or lost, other than time, by playing it. While the game has you using standard playing cards and poker hands as part of its base mechanics, it does not have in-app purchases, loot boxes, or any kind of online play or enticement to gambling, beyond the basics of risk and reward.
Yet many YouTube creators have had their Balatro videos set to the traffic-dropping "Age-restricted" status, allegedly due to "depictions or promotions of casino websites or apps," with little recourse for appeal.
The Balatro University channel detailed YouTube's recent concerns about "online gambling" in a video posted last weekend. Under policies that took effect March 19, YouTube no longer allows any reference to gambling sites or applications "not certified by Google." Additionally, content with "online gambling content"—"excluding online sports betting and depictions of in-person gambling"—cannot be seen by anyone signed out of YouTube or registered as under 18 years old.
Google’s AI Mode can now understand images as part of your searches.
Google started cramming AI features into search in 2024, but last month marked an escalation. With the release of AI Mode, Google previewed a future in which searching the web does not return a list of 10 blue links. Google says it's getting positive feedback on AI Mode from users, so it's forging ahead by adding multimodal functionality to its robotic results.
AI Mode relies on a custom version of the Gemini large language model (LLM) to produce results. Google confirms that this model now supports multimodal input, which means you can now show images to AI Mode when conducting a search.
As this change rolls out, the search bar in AI Mode will gain a new button that lets you snap a photo or upload an image. The updated Gemini model can interpret the content of images, but it gets a little help from Google Lens. Google notes that Lens can identify specific objects in the images you upload, passing that context along so AI Mode can make multiple sub-queries, known as a "fan-out technique."