Mobile Tech & Gadgets

Sweden has just found another damaged underwater cable in the Baltic. The fourth case already has a suspect

by

Last week there were two news that gave an idea that everything that surrounds the cuts and damage of underwater cables in the Baltic is far from ending. After an investigation carried out in the United States, it was slid that the cases in recent months They were not due to a Russian sabotage, but to inexperience and the deterioration. Hours later, United Kingdom put the focus on Russia After detecting a ship in its waters for the second time. Sweden too. A Sweden-Leonia cable. Apparently, A new underwater data cable, now between Sweden and Latvia, was damaged yesterday Sundaymarking the fourth recent incident in the Baltic Sea, one that generates concern in NATO about the vulnerability of critical infrastructure. Latvia’s prime minister, Evika Siliņa, has described the damage as significant and probably “caused by an external force”which has led to the opening of a criminal investigation. As we have explained, previous incidents They have been linked to Russian and Chinese ships. Ongoing research and measures. The Financial Times had a few hours ago That the systemal authorities already have a suspect. In fact, they sent a patrol boat to inspect a vessel near the cable at the time of damage, while two other ships are also being monitored. Swedish prime minister, Ulf Kristersson, confirmed the damage to at least one cable and said that Sweden is collaborating closely with Latvia and NATO to clarify the incident. In response to these attacks on critical infrastructure, NATO has deployed drones, submarines, ships and airplanes in the region Under the “Baltic Sentry” operationan action that aims to detect and prevent possible sabotage attempts. As we also count, United Kingdom has deployed its p8 posidon fighters. Background and suspicions. The situation, by repetitive, leads to the affected nations that there is something else. The Recent incidents in the Baltic They have fueled suspicions about possible acts of sabotage by Russian and Chinese vessels. Finland recently seized An oil tanker of the so -called “Shadow Flot” by Russiacomposed of old ships registered in remote jurisdictions such as the Cook Islands, which evade international sanctions. Finland Foreign Minister Elina Valtonen declared that It is unlikely that the damage will be accidentalalthough he avoided pointing directly to Russia. Not just that. In previous incidents, a Chinese ship, the Newnew Polar Bear, damaged a gas pipeline between Finland and Estonia by dragging your anchor, while Chinese load Yi Peng 3 crossed on two data cables In the same region in November. Subsequently, he remained a month in international waters between Denmark and Sweden, where he was approached by Chinese researchers, although access to Swedish inspectors was denied, generating criticism from Stockholm. Why are they so important. It We have underlined On more than one occasion, but now something else joins. This last incident occurs in a high voltage context, since Estonia, Latvia and Lithuania are preparing to disconnect in February from the Soviet Electricity Inherited and join the European continental system, a key step for its total integration in the West. The three countries, NATO members and the EU, consider this transition as His last separation from Russian influencewhich has generated fears of possible sabotage attempts to hinder the process. In this regard, Lithuania Defense Minister Kęstutis Budry A review of navigation standards in the Balticespecially when it comes to the use of anchors, underlining the most obvious: that the high number of recent incidents reduces the possibility of accidents. Repairs and contingency measures. Unlike gas pipelines and power lines, whose repair can take months, Fiber optic cables damaged in the Baltic are usually restored in weeks. In fact, the State Radio and Television Center in Latvia has assured that it has already found alternative routes to maintain communications in operation. Be that as it may, the repetition of these incidents, The room in the Baltic in such a short timeposes a significant strategic challenge for the region, one that has intensified its cooperation with NATO to strengthen the protection of its critical infrastructure. In fact, the organization has made it clear that any new sabotage action will be faced with firmer measures to guarantee the protection of communications and energy supply in the region. Image | BORN In Xataka | United Kingdom will display the “Frankenstein” of the electronic war to protect its submarine cables: the P8 Poseidon In Xataka | A British nuclear submarine has discovered a Russian ship in front of its submarine cables. The second time in three months

It is the greatest existential threat that Silicon Valley has faced

by

They have barely spent five days since its launch, but Deepseek R1 has caused An earthquake in markets (-6.5% NVIDIA, -3.5% Microsoft, -8% ASML) that goes beyond its technical efficiency. The true threat to the American industry is in its open nature: Depseek is democratizing technology that Silicon Valley has kept jealously stored after its own walls. Why is it important. The western industry of AI has been built on two pillars that Depseek has just dynamit: The need for multi -million dollar investments in hardware. The extreme secretism about the architecture of its models. The money trail. Wall Street is reacting hard because Depseek directly threatens the dominant business model that has taken us here. Investors ask questions: how to justify projects such as Stargate (and its half billion dollars) If there is an open and efficient alternative? What value intellectual property when your competitors openly publish their advances? Between the lines. The Chinese strategy seems clear: to use the open source as a Trojan horse to destabilize Western domain in AI. It is no coincidence that Depseek has chosen this moment, when: American models show signs of stagnation (where is it GPT-5?). Inversiones in infrastructure are triggered. Business models remain without clarifying. The big question. Several experts have pointed out that we are facing Sputnik moment For AI: when the United States discovered that it did not have the monopoly of advance in the space race. R1 has that magnitude, or do you have to relax expectations about its implications? Perhaps “only” (add all the quotes they want) we are rather at a Linux time: their arrival democratized the operating system starting from open source as a disruptive tool, but the industry continued to thrive. Windows did not disappear, he simply adapted to a new scenario in which open and closed source coexist. The difference is in the rhythm. Linux took years to seriously impact the market. R1 has caused immediate shaking even in stock assets of giants. Perhaps the market understands that AI is too valuable and promising to afford as closed and expensive development as the current one. Whether it is a Sputnik as if it is a Linux, R1 marks a before and after: it shows that the toe can be created more openly and efficiently. Yes, but. There are certain persistent doubts about the real costs of Deepseek. THE CEO OF SCALE AI He has suggested They have more hardware access than they say, but hide it for violating commercial restrictions. Even so, even if their costs were greater, R1 would continue to have a strong impact on the industry to choose the open model for such a powerful model. The time of truth. The American’s industry now has a dilemma: or maintain its current model, based on very high investments and a strong secrecy, risking being displaced by Chinese efficiency … or pivoting towards a greater opening to compete on the board raised by Deepseek . Deepen. The analogy with Sputnik beyond technological surprise. As then, this moment can catalyze a transformation in how technology develops and markets. The difference is that this time disruption does not come from the state apparatus but from the Chinese private sector, and its weapon is not the space race but the open source for AI. China, frequently criticized for its control over information, is using the openness and transparency to stand up to Western technological domain. In Xataka | Two years and 60 million users later: how Luzia has become the greatest success produced by Spain Outstanding image | Deepseek, Xataka with Mockuuuups Studio

What is it, how does this artificial intelligence have

by

Let’s explain What is Deepseekthe artificial intelligence chat created by the company of the same name. It is an AI that at the beginning of 2025 is calling a lot of attention, and that even in mobile stores of applications already managed to overcome Chatgpt. And not only that, but also in performance and reasoning tests with its model R1. In recent months we have talked a lot about this company and its models, especially because They are open sourceallowing it to be more accessible to everyone, and because it shows how much China is progressing in artificial intelligence technologies. What is Deepseek Deepseek is the name of a Chinese company specialized in artificial intelligence. It is also The name of your ia chatan alternative of Chatgpt, Gemini, Copilot and company. Therefore, you can listen and read Deepseek mentions referring to both. In this article we are going to focus on the artificial intelligence chatbot, which is a Great linguistic model o Large Language Model (LLM), created to help with software development, natural language processing and business automation. Deepseek stands out for open source. This means that anyone can see how it works internally, It is totally transparentand anyone can be installed locally or use it with total freedom. A developer or researcher can Download the github and modify it for various scenarios, including some commercials .. This totally contrasts with the secrecy of private models. There is usually the false perception that one of the things in favor of the private and opaque code of the majority of developers is that the quality of their products is superior. However, Depseek is proof that open source can match and even exceed these companies in some aspects. Another of Deepseek’s main characteristics is that its native chatbot, that of its official website, It is totally free and does not require any subscription to use its most advanced model. If we compare it with Chatgpt asking the same questions, Depseek can be a bit more concise when answering, it goes more to the grain. However, it has the same flexibility as other models, and you can ask you to explain things more widely or adapt them to you. Two models of AI as spearhead Deepseek has given much to talk at the end of 2024 and early 2025 for two specific models. First it is Deepseek v3a LLM long scale model that exceeds most of AI, and also surpasses some owners. In evidence such as programming, this model managed to overcome Llama 3.1 405B, A GPT-4O and QWEN 2.5 72B, although they all have much less parameters and that can influence performance and comparisons. But what has really surprised this model is that “only” needed 2,788 million hours of training, and this process cost only 5.5 million dollars. To give an example, with the OpenAI GPT-4 about 80 million dollars were invested. Come on, what cost much less to trainwhich makes it a very efficient model, and 60% larger that calls 3.1 405b, of finish line. The other great model has been Deepseek R1which specializes in reasoning, and has been able to achieve or overcome the performance of OpenAI’s most advanced models in key mathematics and programming tests. With him, it is the first time that an open and free Chinese model Equal Western leadersbreaking the monopoly of Silicon Valley. To put some figures, this R1 model has cost to develop between 90% and 95% less than its competitors, and has 671,000 million parameters. In addition, in mathematical tests Math-500 has achieved 97.3% hits, compared to 96.4% OpenAi. However, not everything is good. As with Gemini or Chatgpt, Deepseek It is also subject to the values ​​of the country where it has been developed. In this case it has been developed in China, and adapts to its regulation, and must “embody fundamental socialist values.” To give an example, this does not answer sensitive questions about politics and geopolitics, such as those about Tiananmen or Taiwan’s autonomy. In this regard, in China they are more ferrous when controlling the answers, there is greater state control. Therefore, while Western models can have more liebrtad to discuss controversial issues for the values ​​with which it has been trained, in Chinese models They must always align with ideology and narrative of the party that governs the country. How Deepseek works Deepseek has been developed using pure reinforcement learningwithout previously labeled data. This is a technical feat that until now was considered impossible, and that opens new doors to training this type of systems. This referez learning allows the model to be able to Learn for yourself by proof and errorjust like you can learn to ride by bike or perform some tasks. This makes the initial result more erratic and inaccurate, but the model itself discovers and develops unique reasoning strategies to continue improving. Once this training was done, Depseek refined the model combining it with other supervised training to polish it and create the final version of R1, which maintains this component but adding consistency and polished. Facing the user its operation is the same than that of other models. You have a home screen with a writing field and several operating options, and then when you see the results you have a left column with a history of all your chats. To use it you have to Write a question in natural languagejust as you would do it with a question. Then, Depseek will analyze the words of your question to find the result, and then look in the database with which he was trained or on the Internet to obtain data with which to compose an answer also in natural language. Thanks to the way it has been created, this model can understand complex contexts of very long and elaborate questions. In addition, you can also learn and improve. Its application can be used for free online or lowering your mobile app, and has no fee. What can you do with Depseek … Read more

up to 4,000 euros in aid (and it doesn’t matter if it is electric or not)

by

Without him MOVES III Planone of the main incentives to change from a combustion car to an electric one is fading. Without the approval of the omnibus decree last Wednesday in the Congress of Deputies, aid to change electric car has disappeared. Shortly before the year ended, The Government approved the extension of these subsidies until June 30, 2025. The aids They have continued to be delivered in the first three weeks of January but the green light in the Congress of Deputies was still necessary. The approval of this aid was framed within the omnibus decree which, among other measures, also included transport aid which, consequently, They have also fallen into the part that corresponds to the State. Despite this, some autonomous communities keep them active in its entirety, partially or as a temporary measure. Although it had been approved previously, with the fall of the MOVES III Plan, regional aid for mobility or, as in this case, the replacement of an old vehicle with a newer one becomes especially interesting. In a project to modernize the vehicle fleet, Galicia has its own aid. Galician aid for the purchase of a car The main difference that exists between Galician aid to replace a vehicle and those collected by the MOVES III Plan is that in Galicia a financial amount is provided regardless of the type of technology chosen. It is the volume of CO2 emissions that really defines whether or not a vehicle receives a greater discount. These discounts are available for passenger cars (M1) and vans (N1) whose total costs (including taxes) are 42,000 euros. If the car is electric or a plug-in hybrid with more than 30 kilometers of electric range, aid is available for cars worth a maximum of 47,000 euros. Furthermore, if the car is adapted to a person with reduced mobility or has eight or nine seats, the cost may exceed the aforementioned figures by another 6,000 euros. The discounts will be the following: Car or van Help from the Xunta Dealer Help (Before Taxes) Help for large families Total amount of aid Cars that approve less than 100 gr/km of CO2 3,000 euros 1,000 euros 600 euros 4,000/4,600 euros CARS THAT APPROVED between 100 and 120 gr/km of CO2 2,000 euros 1,000 euros 400 euros 3,000/3,400 euros To the above we must add that the buyer will be obliged to scrap a vehicle that is ten or more years old. The Xunta de Galicia activates these aids for those who meet these requirements: a) Natural persons with habitual residence in the territory of the Autonomous Community of Galicia b) Legally constituted companies and self-employed entrepreneurs, who have a registered office or a work center in Galicia. Finally, it must be taken into account that the deadline to request aid is open until September 30, 2025. However, the aid is delivered in the order in which the application is received until the funds are exhausted, which will be 2,925. 180 euros for individuals and 250,000 euros for self-employed workers and companies. Regarding the application for aid, in this link You can check all the requirements that must be met and documents to fill out, but you should know that the concessionaires themselves will be in charge of making the aid request. Photo | BYD In Xataka | Buying a car: in-depth guide with types, aids, models and everything you need to know

This is what the affected rates will cost as of today

by

The year 2025 has begun as 2024 for Premium operators: with price increases. They suffered it In Vodafone and In Yoigothen it was his turn To Movistarand now it is Orange who has applied An increase in monthly installments of several rates. Although the operator began warning the affected customers a month and a half ago, the day marked by Orange to apply it was today, January 27, 2025. And that has not happened a year since it entered into force The last price increase. The good news is that it is not a general increase throughout Orange’s commercial offer, but It affects only certain services and rates Fiber, mobile and television. Let’s see what those plans are and how much they will cost from now on. You will pay between 2 and 6 euros more per month Unlike Movistar and Vodafone, which every year apply their price increase in January, Orange had accustomed us to make it effective in spring, specifically, in April. But That has changed this year: In 2025, following the example of the competition, the Masorange group operator will also increase its quotas in the first month of the year. Image: Orange That increase, which was already communicated to its clients in advance so that they could unsubscribe if they were not sane, it enters into force today, January 27. The climb affects both the current customers of the affected rates and those that hire them from now on. And what are those rates? Because mainly the most complete fiber, mobile and television love, whose monthly fee It will rise between 2 and 6 euros per month. Of course, in some cases, they will bring an associated improvement (the famous strategy of “more for more”): Initial TV love: duck CIBER Protection and rises 59 to 61 euros per month (2 euros more per month). Initial TV love 2: It rises from 79 to 81 euros per month (2 euros more per month). Love TV Cinema and Series: Add cyber protection and rise 70 to 73 euros per month (3 euros more a month). Love TV Cinema and Series 2: It rises from 90 to 93 euros per month (3 euros more a month). Love TV Cinema and Series Total 2: It rises from 100 to 103 euros per month (3 euros more a month). Love TV Cinema and Series Total 4: It rises from 130 to 136 euros per month (6 euros more per month). Love TV Soccer: Add cyber protection and rises from 108 to 110 euros per month (2 euros more per month). Love TV Soccer 2: It rises from 123 to 126 euros per month (3 euros more a month). Love TV Total Soccer 2: It rises from 138 to 141 euros per month (3 euros more a month). Love TV Total Soccer 4: rises from 168 to 174 euros per month (6 euros more per month). Some tariffs discontinued such as the initial, medium, unlimited, total, lite, original and base also raise their prices. Of course, in the case of the base plans, they will double the fiber speed, which goes from 300 to 600 Mbps. In addition, the additional mobile lines They round their price, so that the Smartphone 0 rate goes to cost 9 euros, while the additional unlimited stays at 16 euros per month. My fixed rises its quota of 12.95 to 14 euros per month. And finally, the price of International callswhose establishment becomes 90 cents. International messages will rise to 90 cents/SMS from February 24, while Roaming services outside the EU will also increase the price from March 16. Cover image | Xataka In Xataka | Three of the great Chinese telecos in the US are in the eye of the hurricane. The reason: the government believes that they spy In Xataka | The largest European fiber giant has just been born. And it is Spanish

Xiaomi sweeps the high range with this mobile phone with a Leica camera. A real bargain during these sales

by

If you have been thinking about renewing your old mobile for a while, now is a good time to get good deals, since many stores have discounts available on numerous terminals. Such is the case of PcComponentes, which has available the Xiaomi 14Tone of the best mobile phones from the Chinese manufacturer, on sale. You can get it even cheaper than on Black Friday, for 449.99 euros. *Some prices may have changed since the last review A mobile with an excellent quality-price ratio This Xiaomi 14T It is a terminal that was launched in the second half of 2024. It stands out for having a screen 6.67 inch AMOLED and that offers 1.5K resolution. In addition, its refresh rate is 144 Hz and it reaches a maximum brightness of 4,000 nits. Its photographic system is another of its great assets. It is signed by Leica and has a 50+50+13 MP triple rear camera and the front one is 32 MP. The processor it has is the MediaTek Dimensity 8300 Ultra, accompanied by a RAM 12 GB and 256 GB of internal storage. In the connectivity section, this Xiaomi mobile is also very complete. Integrates WiFi 6E, NFCUSB-C charging port and Bluetooth 5.4. In addition, it comes packed with artificial intelligence and the operating system it comes with is Xiaomi HyperOS. The best accessories to protect this mobile XTCASE Magnetic Case for Xiaomi 14T *Some prices may have changed since the last review Wayneer Screen Protector compatible with Xiaomi 14T, (3 + 3 Pieces) *Some prices may have changed since the last review Some of the links in this article are affiliated and may provide a benefit to Xataka. In case of non-availability, offers may vary. Images | Xiaomi In Xataka | I am an amateur photographer and this is the accessory that made my photos jump in quality. In Xataka | Best large phones: which one to buy and 12 recommended large screen phones

This is how this silent (and advanced) system called XProtect works

by

Malware is not an exclusive threat to PCs. Macs can also become infected. For this reason, there are different third-party security solutions, such as Bitdefender, Intego, Malwarebytes or Avast, among others. What not many people know is that Apple computers have, for more than a decade, had a integrated antivirus called XProtect. XProtect It has been running silently in the background since it landed on Mac OS Unlike Windows Defender, there are no icons in the menu bar, nor do we have an application to manually perform system scans. How does XProtect work on macOS? The Apple ecosystem has a variety of protection measures, and XProtect is one of them. If we focus on macOS, the Mac operating system, Apple tries to avoid the malware from the App Store. The company reviews the applications so that no malicious software lives in its store, but let’s be honest, we don’t always use the App Store. Computers have always given us the freedom to install any compatible external software. To address the issue of external threats, Apple has a certification mechanism, where developers who do not use the App Store to distribute their applications can obtain a trusted certificate for macOS. Apple does not take this task lightly. Scans the applications and grants the certificate only when it cannot find known malware. Once this instance is passed, developers receive a certificate that they can attach to their application so that the target system can verify it, even when it does not have an Internet connection. At this point we find another of the protection measures. This is Gatekeeper, which looks for that certificate in the application that the user is trying to open or install. If the program does not have the corresponding certificate, the system warns users that they are trying to use software that cannot be verified. Since not all developers go through the certification processwe can ignore that message to use the application in question. Sometimes we will have to allow the installation of applications from the App Store and well-known developers from System settings > Privacy and security > Security > Allow applications from… XProtect is responsible for preventing the execution of malicious software. If for any reason the protection measures mentioned above have been violated, XProtect is responsible for preventing the execution of malicious software for the first time or detecting malware that has been executed on the computer. This integrated antivirus uses YARA signatures that are automatically updated periodically. XProtect comes into action at three key moments: when opening an application for the first time, when modifying an application in the file system, and when updating security signatures. If you detect known malware, blocks it immediatelynotifies the user and offers the option to send the software in question to the recycle bin. XProtect’s scope does not end there. Apple has improved the system to address problems that certain infections can cause. Once malicious software is removed, the built-in antivirus continues to scan for threats using a behavioral analysis engine. Of course, it does not have the ability to restart the computer. How do you update macOS built-in antivirus? XProtect updates automatically. However, you can buy if the system is installing the updates automatically. To do this, click the apple icon in the toolbar and then go to System settings > Software Update. Then tap the information icon under Security Answers. In the Automatic Updates category make sure the Install security responses and system files switch is turned on. So am I completely sure now? Although macOS has a solid security system, it is important to clarify that no system is 100% secure. In the world of cybersecurity it is often said that what is sought is to increase the level of difficulty for cybercriminals. And we achieve this in many ways. Integrated system tools are only part of this. The security features of the applications themselves, our security practices and, if necessary, third-party security tools also come into play. The level of difficulty varies according to the needs of each user. For example, for a long time iPhones and Macs can activate Isolation Mode. This is an extreme solution, but it is a good example of what it means to put higher barriers. This mode limits the functionality of applications, web pages, and certain system features. As a result, we obtain a more protected system, but much less useful than in its standard configuration. Images | Apple | Jakub Zerdzicki | Screenshot In Xataka | The Bank of Spain warns of malware capable of “capturing banking credentials.” His name: TrickMo

This is what the law says about the controversy over million-dollar fines

by

Installing a security camera at home or in a business is an increasingly attractive option. The price of security systems makes them affordable, but it is also easy to control them from your mobile phone and they have live alert systems. It is a good way to gain security and are a deterrent against possible crimes. The problem is that it is not as simple as placing a camera wherever we want and starting to monitor: the Data Protection Law comes into play. It is something that happens in Spainbut also in Mexico, and in recent days controversy has erupted: if you install surveillance cameras, you will face a million-dollar fine. The commotion that has arisen is such that the Government itself has come out to clarify the issue. The controversy. A few days ago, a popular TikToker posted a video in which he talked about the situation. “Did you know that you can get into legal trouble because of the cameras you have in your business or home?” This is how the video begins, which, with a pejorative tone, accuses the Government of not acting to guarantee the safety of the citizen. Subsequently, and throughout the video, it clarifies the scenario in which these fines can be applied: if the cameras record public spaces or you as the owner disseminate the images. The seed, however, had already been planted. Media and users used this video to affirm that Mexico could denounce the owners of the surveillance cameras, causing the payment of million-dollar fines: up to 18 million pesos. Some of these media clarified the fact of the dissemination of the images or the orientation of the cameras, but others did not. The answer. The commotion has been tremendous, so much so that different institutional bodies have rushed to address the issue. They have done it in two ways: one, through the INAI, or National Institute of Transparency, Access to Information and Protection of Personal Data. In a videocomment that they have received several calls asking for information in this regard, commenting on the following: “The regulated subjects are those responsible who use personal data for purposes of disclosure or commercial prospecting. “If a person has cameras in their home and they do not have that purpose, but are used for personal treatment, the INAI will not impose any sanction.” The second, during an information section from the Treasury Room of the National Palace in which the Government denied the information. “The Federal Law on Protection of Personal Data held by individuals applies only to private individuals or legal entities that process personal data,” they comment. Basically, the same as the INAI. Cover of the PROFECO guide The buying guide. In fact, a few months ago it was the Government itself that published, through a document from the Federal Consumer Prosecutor’s Office, a purchasing guide with cameras for both internal and external use. In this studythey analyze various models, pointing out their pros and cons in a complete buying guide that would not make much sense if they were prohibited. The keys. But at the end of the document is the key. In a section called “Is the use of security cameras legal in Mexico?”, they confirm that it is. There are nuances: if it is inside a home or pointing towards private property, there is no problem, but if they record from the house to the outside, there are certain rules and restrictions to protect the privacy of others. What must be taken into account, according to the Federal Law on Protection of Personal Data Held by Private Parties, is: It must aim to protect assets and facilities, as well as the prevention and detection of crimes. Images of people should not be captured without their consent, especially in public areas, unless they are police forces. When you can face a fine. In the document, the LFPDPPP clarifies that “it is legal to install surveillance cameras inside a property, as long as people’s rights to privacy and personal data are not violated.” Regarding outdoor cameras, it states that “if you plan to install security cameras, we recommend you notify your neighbors, either by putting up a sign notifying you of the existence of the camera or through a meeting. This ensures that they will be aware that they are being monitored and can take steps to protect their privacy.” However, it is true that there are establishments in which it is illegal to place cameras: Dressing rooms, bathrooms and private business rooms. Schools. Hospitals. Churches. Furthermore, in these last three, use is restricted if you do not have authorization from the authorities. Therefore, it is legal to have a security camera in Mexico if you do not broadcast the recordings and if, pointing to common areas, you warn in advance with a sign or a meeting with people likely to be recorded. Image | Eufy In Xataka | Best surveillance cameras: which one to buy and 11 recommended models for indoors, outdoors, babies and pets

Chips that recover energy when working

by

Michael Frank has been working in a technological discipline that seems like science fiction for more than three decades. One that apparently pursues something impossible. And yet, it may not. The goal? Than using a chip manages to recover energy. Chips have always done the opposite. They consume energy. A lot. More and more, especially now that we are immersed In the era of glottone artificial intelligence. And yet Frank is immersed in the development of the call “reversible computing“As they explain In IEEE SpectrumLast summer Frank left his work as a scientific engineer in Sandia National Laboratories to join a very special startup called Vaire Computing. It is there where he intends to become what he has already estimated in his studies: according to his investigations, reversible computing could Allow energy efficiency 4,000 times higher to that of alternative options. In Vaire Frank and his team are working in a chip prototype that will be manufactured in the first quarter of 2025 and that is very ambitious. Even so, for the first time a chip is expected to recover energy when used in an arithmetic circuit. Then, in 2027, they hope to have a supply processor designed for artificial intelligence inference. The advanced version of said chip, which will theoretically be 4,000 times more efficient than the current ones, It will take 10 or 15 years to arrivetheir creators warn. What is reversible computing In 1961 Rolf Landauer, an IBM engineer, discovered that deleting a bit of information on a computer had an energy cost and produced heat. For him there was a solution: try to compute tasks without erasing information. A traditional xor door is not reversible: you cannot recover entries just knowing the exit. If an additional output is added, a copy of one of the inputs is reversible. Thus, the two outputs can be used to “disassemble” the door Xor and recover the entrances and, with them, the energy used in the calculation. Source: IEEE Spectrum. Landauer made clear that it seemed unchanged, because store all the information You would fill the memory With unnecessary data. More than a decade later Charles H. Bennett, physicist and information theoretical in IBM, discovered a solution. Instead of storing intermediate results in memory, you could reverse the computing process, “disbelieve it” to store only the final inputs and outputs. “Almost 100% of the energy used by a chip ends up being dissipated as heat, so we are basically evil. But in a reversible chip you never dissipate energy. Do not allow energy to become heat, and recycles internally.” The idea was taking shape in academic environments, and in the 90 Frank ended up being one of those who immersed themselves in their promises. The problem is that when developing that concept they saw that although energy was recovered in the circuit, it was lost in the external power supply. Vaire Computing was founded by researcher Hannah Earley and entrepreneur Rodfo Rosini. Earley explained In TechCrunch How “almost 100% of the energy used by a chip ends up being dissipated as heat, so we are basically missing it. But In a reversible chip you never dissipate energy. Do not allow energy to become heat, and recycles internally. “ In Varie they have apparently propose na solution with traditional CMOS transistors. To do this, they use the so -called “resonator”, a system that manages to recover energy by changing a bit from 0 to 1 so as not to turn it into heat and store most of it into the transistor itself. That idea has already been explored in the past, but in Vaire they seem to have managed to integrate that resonator into the chip. They will start with a little ambitious chip and then work on those chips for inference to which they will be theoretically what will be what will be theoretically really demonstrate How far can you get with reversible computing, also well explained in this small essay published in the Newsletter Exponential View. Of course, It is one thing to promise those efficiencies and another very different to achieve them. The proposal is very interesting and of course could raise a revolution that accelerated the development of much more powerful chips and that were not so voracious in terms of energy consumption. It remains to be seen if they will achieve it. Hopefully yes. Image | Wu Yi In Xataka | The 2 Nm chips race will start in 2025. And it will be the most fierce of all

China is advancing at breakneck speed in nuclear fusion. It already has something ready that until now only the Netherlands had

by

The path to a destination as challenging as it is nuclear fusion commercial must necessarily be full of small conquests. Of achievements that may seem modest, but that, in reality, are milestones that put us a little closer of an ambitious objective that seeks nothing more than to help us solve our energy needs without continuing to emit greenhouse gases. In this context ITER attracts much of the attention. And it is understandable that this is so. After all, it is a project of enormous magnitude, which is also led by the European Union. In fact, this organization is jointly assuming approximately 50% of the total cost of a plan in which the United States, Russia, China, Japan, India and South Korea also participate. However, the public commitment to nuclear fusion is not condensed solely into ITER. And it is not limited only to the European Union either. Not at all. Europe is signing up very important scientific milestonesbut there are other countries that are also bidding very high, and that, precisely, do not move in the orbit of the West. In fact, two of them, probably the most advantaged, are China and South Korea. China has a very sophisticated linear plasma generator to advance fusion In the field of nuclear fusion, plasma is the extremely hot gas that contains the nuclei of deuterium and tritium, the two isotopes of hydrogen, which are involved in the reaction. For these nuclei to overcome their natural electrical repulsion and the strong nuclear interaction to fuse them, they must acquire a very high kinetic energy. And this is only possible if the plasma reaches a temperature equal to or greater than 150 million degrees Celsius. As we can guess, very few known materials are capable of withstanding such a high temperature. However, this is not all. When a deuterium nucleus fuses with a tritium nucleus, they produce a helium nucleus and a neutron that is ejected with an energy of about 14 MeV (megaelectronvolts). The problem is that the neutron lacks a net electrical charge, so it cannot be confined inside the magnetic field which, however, does manage to retain the deuterium and tritium nuclei, which have a positive electrical charge. The components that will be most affected by the direct impact of high-energy neutrons and the most intense heat flow are the inner wall of the vacuum chamber and the mantle. This is the reason why when it originates as a result of the nuclear fusion reaction, this neutron is ejected towards the walls of the vacuum chamber with enormous energy. This particle is very important because in practice it will be closely linked to the production of electrical energy in nuclear fusion reactors, but, at the same time, it represents a very aggressive form of radiation that can significantly degrade the materials used in the reactor. . The components that will be most affected by the direct impact of high-energy neutrons and the most intense heat flow are the inner wall of the vacuum chamber and the blanketwhich is a mantle that covers it and whose purpose is regenerate tritium which is necessary to use as fuel in the nuclear fusion reaction. This is why it is crucial to develop new materials that are able to withstand the neutron flux and therefore ensure that the reactor will have a long operational life. Until now, only the Netherlands had a device capable of generating a high-flow plasma similar to what occurs in the vacuum chamber of a nuclear fusion reactor. But now China has it too. The Hefei Institute of Physical Sciences has successfully built a highly advanced linear plasma generator capable of accurately recreating the extreme conditions found inside fusion reactors. Its purpose is to use it to test candidate materials to be used in vacuum chamber constructionfor which it is essential to subject them to the interaction of plasma. Fortunately, China has confirmed that this machine will be available for international collaboration. Image | Hefei Institutes of Physical Science More information | Hefei Institutes of Physical Science In Xataka | Spain’s milestone in nuclear fusion: the first plasma produced by the SMART reactor invites us to optimism

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.