face

He has found a way to shoot China’s competitiveness in the face of the US

by

Two days ago we told you something very interesting: the Chinese state medium Securities Times had revealed that Huawei was about to present a technological advance that pursued Reduce China dependence of HBM memory chips (High Bandwidth Memory) from abroad. According to this source Huawei was going to officially announce his technological milestone a few hours later, during the celebration in Shanghai (China) of the Applications Forum and Development of Financial Reasoning 2025. Huawei has fulfilled what was promised, although not as we had planned. In any case, before getting into flour it is important that we remember that Chinese memory chips manufacturers They are not producing solutions capable of competing with the most advanced memories manufactured by South Korean companies Samsung and SK Hynix, or the American Micron Technology. GPUs for Ia work side by side with HBM memory chips. In fact, its performance is largely conditioned by these memories. As the editors of SEMIANALYSISthe total bandwidth of the HBM3 memory chips that live with some of THE GPU FOR THE MOST ADVANCED Nvidia or AMD exceeds 819 GB/s, while DDR5 and GDDR6X memories reach much more modest 70.4 GB/Sy 96 GB/s. HBM3E memories and future HBM4 are even better. Chinese manufacturers of this type of chips do not yet produce this kind of memories, but it seems that Huawei will deeply alter this scenario. An algorithm expressly designed to accelerate inference in AI The filtration that occurred scarcely 48 hours suggested that probably what Huawei was going to present was a avant -garde packaging technology that, perhaps, would rival those used by SK Hynix, Samsung or Micron to produce their HBM3 and 3E memories. And it is that manufacturing these integrated circuits is complex because they require stacking several DRAM chips and implementing an interface between the XPU (Extended Processing Unit) or extended processing unit and extraordinarily dense HBM chips. As a button shows: in a HBM3E stack the XPU and the HBM memory are linked through more than 1,000 drivers. According to Huawei, the UCM algorithm is capable of drastically accelerating inference in the great AI models However, finally Huawei has presented a different technology: an advanced algorithm called UCM (Unified Cache Manager) that, according to this company, it is capable of drastically accelerate inference In the great models of artificial intelligence (AI). A relevant note: inference is broadly the computational process carried out by language models with the purpose of generating the responses that correspond to the requests they receive. To achieve its purpose, the UCM algorithm displays a very ingenious strategy: decide in what type of memory it is necessary to store each data taking as a fundamental indicator the latency requirements. In practice, this algorithm behaves as a gigantic cache that guarantees that each data will go to the right memory, including HBM3, with the purpose of minimizing latency during inference. If it is a very often used data, it will be stored in a very fast memory, such as HBM3. According to Huaweithis technology is able to reduce the latency of inference by 90%. Interestingly, this company plans to do the UCM Source Open Algorithm in September. More information | SCMP In Xataka | Nvidia has to deal with the absolute distrust of several US legislators. His plan in China is in danger In Xataka | The US wants to end the chips for the Chinese that are sold abroad. And China knows how to defend oneself

The AI has converted work interviews into a cheat circus. So companies are returning to the face -to -face

by

The virtual work interviews They are already normal in job search processes. The problem is that in certain areas, such as software and programmers engineers, candidates usually take advantage of AI tools to cheat. The trend is so worrying that some companies are already returning What always worked: Interviews in person. Google wants to meet you in person. SUCTAR PICHAI, CEO of Google, He explained in June In the Lex Friedman podcast that renewal of the company’s policies at the time of Hire certain profiles. “We are making sure that we introduce at least one round of the candidate for the candidates, just to ensure that the fundamental aspects are fulfilled.” Return to the face -to -face. Pichai’s message is the same as They are adopting other companies such as Apple, goalCisco or the consultant McKinsey, who are among a growing number of companies that are recovering the interviews face to face with the candidates in various stages of the selection process. Mike Kyle, from the Coda Search/Staffing Employment Agency, explained how the quota of companies that have the requirement to make face -to -face interviews has become 30%, when in 2024 it was only 5%. If you want what salary I offer you, I need to meet you. That return to face -to -face interviews usually also focus on the last part of the selection process, but in reality it can occur at any time. If you do not present yourself physically at some point in that process, You won’t know what conditions (Salary included) offers you the company. Cheat programmers. As we said, that is especially true for profiles such as programmers. During some phases of the interviews, real -time sessions usually consider in which candidates must solve a programming problem. In virtual interviews, what has happened is that the interviewees cheat And they use AI to solve the problem, which is not clear whether or not those candidates meet the requirements. AI has become a problem. The resurgence of interviews in person tries to mitigate the problem that AI has raised. Companies that offer jobs have ended up using AI systems for Filter candidateswhich usually flood those offers because they also use AI to generate their CVS adapted to each position and automate the request for various positions quickly. THE ART OF THE TRAPS A few months ago we counted how a student managed to overcome a Amazon technical interview Thanks to the use of AI. That gave him a disturbing idea, because he created a startup called Cluelly to help others do the same or to make exams. One that in fact advertises with an unusual message: “Cheat everything.” The idea seems to have liked, because the investment firm A16z has already injected 15 million dollars In the project. Deepfakes that are interviewed for you. That is a good example of a dangerous trend. It is not that candidates try to deceive companies about their real capacities thanks to AI: there are people who are using these methods much more worrying. The FBI already warned in 2023 of a fraud that involved thousands of North Koreans that simulated being North Americans seeking to work remotely in US companies. His way of doing so was really striking and complex. Be careful with murmuring. Companies that use virtual interviews in their selection processes are also paying special attention to these traps. In fact, some try to detect them by monitoring indications that can aim at fraud: muttering out of the screen or typing and then pauses before answering (waiting to see what the chatgpt tells them on duty to read that answer) It is usually a clue to hunt checkery. Image | RAD MINE In Xataka | Hide the holes in the curriculum of an intermittent work career: the art of not deceiving without telling the whole truth

Openai’s plan to get more money with GPT-5 has exploded in the face: total back

by

Shitting not only affects streaming. We begin to see indications that the chatbots of AI are no longer so beneficial for users, and they are not for a simple reason: they must be monetizing them. It is what OpenAi has just done when launching GPT-5, a model that promised to be easier to use and powerful than ever, but has ended up returning to what worked with GPT-4. The controversial router. When Openai launched GPT-5, he did it with a great novelty: to raise it as a unique model that adapted only to the needs of each user according to the question we asked him. The router detected if that question was more or less complicated and theoretically activated the most appropriate mode of operation in each case, but there was a problem: Always using the cheapest operating mode. Reverse. People-especially those who used chatgpt-quickly criticized both that decision and the option to use ancient models such as GPT-4O. The mutiny had effectbecause: He has recovered the old models he had killed (such as the aforementioned GPT-4O), although only for payment users Has enabled the function of choosing GPT-5 variant to free users It is a spectacular revenue just when Openai had sold us that unique and off -road model (and its router) as something differential. The options are good. Altman himself explained that from now on Chatgpt users can choose between “Auto”, “Fast” and “Thinking” when using GPT-5. “Most users will want to use a car,” he said, “but additional control will be useful for some.” He also remembered that GPT-4O is again available for payment users and clarified that “if at any time we deactivate it, we will warn with a long time.” Less root, more customizable. The OpenAi CEO also talked about another of the problems that were highly criticized in GPT-5: it was too neutral. Too cold and robotic. That could change very soon because as I said, “we are working on an update of the GPT-5 personality that should be perceived warmer than the current but not so annoying (for many users) as GPT-4O.” In Openai they have understood something important: people love to be able to customize everything they use … although many do not. A GPT-5 hypothesis. In Semiianalysis They have a theory Curious that they would explain the way Openai has launched GPT-5. According to them, the model of the model is not the model, but the router. This component was intended to monetize the service much more and convince users to pass the free service to one of the payment subscriptions. Altman already pointed to that approach. In fact, Sam Altman shared first revealing data on Sunday. The percentage of free users who were using the “reasoning” variant of GPT-5 had gone from less than 1% to 7%, while for Chatgpt Plus users that percentage went from 7% to 24%. That can imply that the basic model was not as good as users expected and preferred to make him think, but there is another striking fact. More subscriptions. According to Semiianalysis, the router and that best behavior when the “thinks” model seems to have convinced many more users, and subscriptions, they point out, have multiplied by 3.5. The router may have caused criticism of intensive chatgpt users, but it also seems to be a key element to try to achieve something OpenAi needs like water: convert free users – it has about 700 million – into paid users. Shit. Openai’s tactics, if really this, is not new. To degrade the free service with respect to the payment usually makes more users go to the payment (in addition to the initial criticisms). We have seen it for example in Netflix: when it began to close shared accounts and put internet ads, it was thrown over and the service seemed to have a complicated future. Today Netflix is more reference than ever and the fucking its service has worked perfectly. You may want to copy the idea in Openai. In Xataka | Sam Altman and Elon Musk hate each other publicly, so Altman has attacked where it hurts most: Neuralink

Its sales continue to fall in Europe in the face of the advance of the Asian electric car

by

What Tesla It is not going through its best moment In Europe it is no secret. Elon Musk’s company still resents of the fleeting, although media, controversial and Extremely changing support From his CEO to the president of the United States, Donald Trump, and the unstoppable advance of the Chinese electric car. The latest data reflects: sales in the main markets of the old continent remain in free fall. Although not everything is bad news. The data. The most important European markets for Tesla (and for the car market in general) are the United Kingdom, France and Germany. In these countries, the fall in Tesla’s registration, a strong indicator of sales, has been pronounced. Why does an electric car have less autonomy than the announcing Image | Charlie Deets According to information shared by Reuters and Bloombergonly in July the enrollments fell 60% in the United Kingdom, 55% in Germany and 27% in France. The rest of the countries They do not fall short: 86% in Sweden, 52% in Denmark, 62% in Holland, 58% in Belgium, 5% in Italy and 49% in Portugal. On the whole. If we look at the full photo, the figure is even more striking: Tesla sales have fallen 45% In the ten European markets that, together, accumulated 80% of Tesla sales in Europe during the first half of the year. In other words, Tesla’s sales continue to go downhill in the most important European countries. So far this year, the sales of Elon Musk’s firm have fallen a third in the old continent. And there are no countries in which it grows? Yes, Spain and Norway, to be exact. In Spain, Tesla car registrations increased by 27% (702 cars) in July, which represents a 1.1% growth so far this year. It is a percentage that does not accompany the growth of the electric car market in our country: 89.6%. In Norway, a country in which 97.2% Of the cars sold in July they were electric, increased 83% or what is the same, 838 cars. Image | Xataka Byd advances Tesla on the right. As it could not be otherwise, Asian manufacturers have taken advantage of the situation to continue eating the toast to Tesla. According to the data of the Jato consultancy of those echoing from The countryonly in the month of June Byd sold 15,565 cars in Europe of which 9,153 were electric (132% more). The semiannual figure is much more interesting: 41,270 electric cars, 143% more. The panorama has changed a lot. Best selling cars. The good news for Tesla is that, despite the fall in sales, the best -selling electric car continues to sign during the first semester of the year: the Tesla Model and. He Volkswagen ID.4 He is closely followed with a dangerous and threatening second place. The photo, however, is very different if we look at our country. Best -selling electric cars in July | Image: ANFAC The Spanish case. If we take a look at the most recent data from the ANFACwe will see that the best -selling electric cars in July have been the TOYOTA BZ4X (604 units), the Byd Dolphin Surf (531 units) and the KIA EV3 (493 units). It is not until the fourth position that we find a Tesla, Model 3, with 456 units sold. The Tesla Model and, meanwhile, falls to the tenth position. Plug -in hybrids best selling in July | Image: ANFAC In the accumulated January-Julio, Tesla goes quite well stopped. The Tesla Model 3 is the best selling electric with 4,545 units. They are followed by Model and (3,282 units) and KIA EV3 (3,206 units). In plug -in hybrids, the undisputed leader is the Byd Seal U.whose sale in July added 738 units (1,616% more). Cover image | Martin Katler In Xataka | Elon Musk has suggested his next goal: “Countries that do not manufacture their own drones will be vassal states”

DGT recommendations to face the output operation on August 1

by

They say that the flow of travelers has changed over our country. The holidays are usually chosen for weeks and not for fortnights, as before. Therefore, it is Fridays and on Sundays the days that are usually registered on roads in summer. But this year we have a particularity. This Friday is August 1, the day that was traditionally used to go on vacation. One more reason for The DGT theme that the roads become hell this weekend. The usual summer traffic between Friday and Sunday must be added that it begins (or ends) the fortnight, so a huge flow of workers who go out to enjoy their vacations or begin to return to them is expected. These movements, the DGT estimates, will add some 6.88 million journeys Between yesterday, Thursday, July 31, and Sunday, August 3. During all this time the Operation of August 1 will be activated. A campaign that gives the starting gun to a month in which they are expected to count about 52.87 million displacements. These are the recommendations of the DGT. Better with full tank What can we expect from an operation holiday? Traffic. A lot of traffic. That is why the DGT already advances that the most complicated hours will be condensed in the August 1 (starting at 3:00 p.m.), throughout the morning of Saturday, August 2 and, of course, on the afternoon of August 3 (between 4:00 p.m. and 23:00 hours) when they are expected to meet both those who go out to spend the day and those who have lengthened the entire weekend and, of course, those who return on vacation. What remains in our hands? The DGT gives a series of Recommendations. The most basic is that we move outside these schedules, whenever possible. Thus we will reduce the risk of seeing ourselves wrapped in a jam. But, in addition, they ask for a certain forecast: Prepare the trip: check the car components well, such as the pressure of the tiresbrakes or lightsthinking that you will demand much more than usual. In addition, have an eye on the browser to avoid possible jams and be aware of the possible accidents or breakdowns of other vehicles. During the trip the DGT emphasizes a “circulate with the Full fuel tank. It will be of great help in case of being faced with any anomalous situation as retentions, accidents or unforeseen weather. “This recommendation is usually given in winter where the fuel tank It is key to keeping heating alive of the car but it is always a good idea because, in case it is complicated and the night is over, we can have problems to find an open gas station or face an alternative route if, for example, for example, for example, for example, for example, A road has been cut by a fire. Food and drink: always carries water inside the car and avoids heavy meals since they can sleep. Of course, try to sleep as well as possible before and drive rested. And the DGT ends with a sentence: “Do not take the mobile and avoid any other distraction. Your only task is to drive“ Photo | Tsai Derek In Xataka | The problem is not that the air conditioning of the car breaks more in summer. The problem is that the rest of the year we do not look at it

The US has realized how risky it is to continue pressing China. His reverse looks for a “face to face”

by

The US is softening its export restrictions to China. In early July, the administration led by Donald Trump allowed Nvidia to sell Your H20 GPU To its Chinese clients. At the time This movement surprised us because it clearly indicated that an obvious change was consolidating in the US strategy in front of the nation governed by Xi Jinping. According to the newspaper Financial Times This relaxation of export restrictions to China seeks to “avoid damaging trade negotiations with Beijing.” And it is presumably this week US and China will resume their negotiation to establish the rules that will resolve their future commercial interests. As expected, this rudder of the US administration favors the interests of Nvidia, AMD, Broadcom and other companies, but not all members of the Trump administration support this initiative. According to Financial Timessome White House advisors are pressing the government to maintain their blockade to China. Nvidia needs TSMC to manufacture 300,000 H20 chips for China Donald Trump wants to meet “face to face” with Xi Jinping before he ends this year, again according to the information he has Financial Times. The intention of the US president is to file roughness with his Chinese counterpart with the purpose of promoting negotiation in commercial matters that both countries have maintained for many months. Just a few hours ago they have gathered in Stockholm (Sweden) Scott Besent, the US Treasury Secretary, and He Lifeng, the Vice Prime Minister of China, so it is evident that the negotiation follows its course. After many months of uncertainty the current scenario clearly benefits Nvidia. The company directed by Jensen Huang saw its survival in China threatened by The sanctions approved by the Department of Commerce from the US, but This volantazo changes everything. And its protagonist is the chip for artificial intelligence (AI) H20. The reception that Chinese clients of Nvidia initially gave to this GPU was very good despite the fact that the capabilities of this chip are clearly inferior to those of the other proposals for this company. After many months of uncertainty the current scenario clearly benefits Nvidia In fact, initially the Department of Commerce allowed its sale in China because this integrated circuit met the restrictions it had imposed. And despite its limitations its sales in China grew by 50% quarter to quarter since it reached this market in mid -2024. Everything was complicated for Nvidia in the middle of last April. And is that the US Department of Commerce imposed new restrictions To the export to China of the H20 GPU, which in practice caused this chip to stop reaching the Chinese clients of this company. This news sank Nvidia’s actions by 6% because it could no longer address the commitments linked to the H20 GPU it had acquired. Among the Chinese customers who had bought it Large amounts of this GPUand presumably planned to continue doing it, were Tencent, Alibaba or Bytedance. Finally Nvidia has achieved That the Department of Commerce review its regulation and allow you to sell the H20 GPU in China again. Surprisingly this scenario has put Nvidia in trouble. And, once again, the reception of the H20 GPU in China is being so good that this American company has run out of units. According to ReutersNvidia has commissioned TSMC for the manufacture of no less than 300,000 units of this chip to be able to respond to the current demand from China. Blessed problem. However, the company’s engineers have been working on A new GPU with Blackwell microarchitecture The latest generation that is intended to occupy in China the place of the H20 chip. In fact, presumably TSMC began in June the production of The B30 GPUwhich is how it seems to be called. What we know at the moment is that this chip for AI will be less capable than the H20 GPU, as is logical, and also that its price will move Between 6,500 and $ 8,000. For domestic users it is a lot of money, but in the field of professional GPUs for AI it is a moderate price. In fact, the H20 GPU costs between $ 10,000 and $ 12,000. Image | Gage Skidmore | Wikipedia More information | Financial Times | Reuters In Xataka | The US gives Huawei a great opportunity: to get its new chip for AI with the Nvidia market in China

Renfe trusted the Avril trains to face the Low Cost of Ouigo and Iro. They do not stop giving problems

by

When the competition has pressed, Renfe has seen how his commitment to a low -cost service to the AVE had been as necessary. The problem is that this service, called Avlo, is operated by trains that in just over a year have not stopped giving problems. Retired avlos. It has returned to normal with the five Avlo trains frequencies between Madrid and Barcelona for the month of August but for a few days the cheap offer of the company in the corridor faded completely. First the five trains retired. Then the frequency was in three options in the day. Finally, they can already be bought normally. It is what Víctor de Elena accountjournalist electionomista.eswho had written an article last Friday explaining that Renfe had stopped offering his services of the Avlo line between Madrid and Barcelona. Since last Friday, when the information was published, the sale of tickets has been returning to normal. What happened? Everything indicates that Renfe has been putting patches to a complicated situation. He explained about Elena that Renfe had found a fissure in one of the components of the Bogies of an Avril train. This would have caused that, temporarily, it would be suspended The sale of all Avlo tickets in Madrid-Barcelona. Hours later, Renfe sources would confirm to the medium that during one of the maintenance reviews last Monday, July 21, one of the five 106 series presented “a fissure in an element of the tractor bogie called a bast.” Subsequently, a review of the five trains that serve and the affected systems were replaced. Now, according to Renfe “trains are operating without any anomaly.” More slowly. The trains operate, yes, but below the expected speed. They point out in electionomista.es That according to the State Railway and Adif Security Agency, Renfe operates the trains at 250 km/h of maximum speed, instead of the 300 km/ha those who were operating in a stretch of 190 kilometers that extends between Puerta de Atocha and Ariza (Zaragoza). In your communicationRenfe points out that the company “through its Renfe Travelers and Renfe engineering and maintenance, as a railway company and entity in charge of maintenance, respectively, as well as Talgo in its condition as manufacturer and the State Railway Security Agency, guarantee that the circulation and operation of trains is carried out with total security for travelers.” In the statement they point out that the offer is being reorganized with the S106 trains but do not specify anything about the speed at which trains circulate and emphasize that the sale of tickets is being made “normally.” In Xataka We have contacted the company but so far we have not obtained an answer. The Avril. The S106 series in which the fault has been detected corresponds to the trains known as Avril. Trains that since they were launched in 2024 They have caused various incidents. The S106 series It is designed by Renfe to fight the lowcost but also to be able to offer a service in Galicia that You will not have competition in the short term. For this last line, Renfe has 15 trains in his hands that can change the wide track automatically. Then, it has another 15 trains to operate between Madrid and Barcelona. Of the latter, they carry the specifications to operate as AVE but in five other trains the places have increased from 507 to 581 seats and thus be able to compete as a low cost option. A controversial arrival. When the Avril trains made an appearance they have already arrived surrounded by controversy. They launched in May 2024 but The material had been delivered with three years delay. Then, of the 30 trains, 12 of them had been delivered in a first lot. The first steps of the Avril were not the best. They promised to improve the connection between Galicia and Madrid but in the first days they registered delays of more than two hours and in later days of more than 30 minutes. Neither In the Asturian line The promised time sales were being fulfilled. But, in addition, the trains did not like among the experts either. Accounts in social networks that follow the current world of the train and its novelties portrayed the excess noise and vibrationsas well as They questioned the quality of some materials used in the rolling material. More incidents. The incidents did not stay here. Above all, Avril trains are remembered by the fault in the Galician line that on the first day of this year he left thousands of passengers on the ground or in the middle of nowhere. In a kind of “2000 effect”, trains were retained by a software problem With the year jump. And Barcelona-Madrid has not been the only one. Just a few days ago, a S106 series train collapsed at low speed at the Gijón station without consequences for travelers. At the moment, Renfe has not given explanations about what happened. Photo | André Marques 432 In Xataka | “In 1961 it took Bilbao for three hours and five minutes. Now it takes three and ten”: Cantabria and the drama of Spain with the train

The United States has lost almost all its advantage in the face of China. And in just over a year

by

American leadership in AI wobbles. The data of the AI Index 2025 Stanford reveals that China has trimmed the technical advantage of the United States to levels that seemed difficult to reach. Above all, because it has taken a little over a year to close most of the gap. Why is it important. The United States maintains the volume –40 notable models against 15 Chinese at the end of the year in the Epoch list-but China has shown that quality is no longer American monopoly, and that almost-convergence rewrites the rules of geopolitics in the most promising and strategic technology of the 21st century. The figures. In January 2024, the best American model exceeded Chinese in 9.26% in Benchmarks of chatbots. Just over a year later, that advantage has collapsed to 1.7%. Convergence is repeated in reasoning, mathematics and programming. The case Deepseek It is a perfect example. While Google spent 192 million dollars on the specific training of Gemini 1.0 Ultrathe Chinese startup said they have achieved competitive results with just 6 million. That caused A media and stock market earthquake At the beginning of the year. The contrast. They are two different approaches: The United States dominates for quantity and resources: it leads private investment (150,000 million dollars in 2024) and concentrates the most valued companies in the sector. China is responding efficiently: less models, much lower costs, equivalent results. The narrative of “Chinese technological backwardness” falls apart. Yolanda Gil, co -director of AI Index, He explained Thus his surprise: “I hoped that a more efficient version of the LLMS appeared at some point. We simply did not know who would build it or how.” Yes, but. American hegemony persists on other fronts. American companies continue to create the most influential models and monopolize global investment. And Europe is relegated: only three notable models in 2024. All French. The shadow of Mistral. The background. This race reflects a broader battle for the future of the digital economy. China demonstrates that efficiency can compensate for huge spending, challenging the “brute force” model that has characterized US development of AI. In Xataka | Openai is going better than expected thanks to payment users. And that marks an evident direction Outstanding image | Aerps.com in Unspash

Apple has not washed the iPhone’s face. Has subjected it to cosmetic surgery

by

Forget iOS 19 and stay with this name: iOS 26. This is the beginning of a new stage. The protagonist operating system of the WWDC 2025 that Apple is celebrating in Cupertino is that of the iPhone, and drinks directly from an interface we already knew: The Vision Pro. Apple’s most niche device is the one that will mark the way for the rest of the products. Some that, in this new stage, will have a name and A unified design language. Why iOS 26. The logical jump after iOS 18 It was iOS 19, that’s over. Apple will begin to rename its software platforms by adding a +1 a year in which they are launched. We are in the year (20) 25, adding that one we already have the 26. It is the name that the rest of their brothers will receive: Visionos 26, Watchos 26, Tvos 26 and Ipados 26. Next year, we will have versions 27 above the table. A name change that meets a concrete need of Apple: to try to make its operating systems to the fullest. Liquid Glass. The first novelty of iOS 26 has to do with its change at aesthetic level. Apple has been inspired by visionos to share its design language with the rest of the software platforms. And he has decided to call him Liquid Glass. The translation is simple: a system starring transparencies, rounded icons and even softer animations. The changes range from the application pitcher to the apps themselves, a work that the developers will have to begin to adapt to maintain coherence with the renewed system. News at Apple Intelligence. Apple wanted to start (in a very short way) its Keynote with news in its artificial intelligence functions. A good part of them, not yet available in Europe. Among them, The new Siri. One of the big changes is how applications will interact with Apple Intelligence. Developers can configure their apps to directly access Apple Intelligence, privately and even without connection. Image | Apple

Three watches for less than 200 euros to run. We face them on the ground and we already have a winner

by

Going to run with a smartwatch sounds tempting. But when prices range from less than 200 euros up to about 900 euros, it is normal to ask if it is worth paying more. What do we lose when we opt for a cheaper model? And what do we continue to win? In our new video we face three affordable watches that seek to offer us several answers. That is why in Xataka We have prepared a new video For our YouTube channel With one of our favorite formats: the Versuswhere we compare face -to -face devices to discover which one deserves more. In this new comparison we face them Garmin Forerunner 55he Amazfit Active 2 and the newcomer Huawei Watch Fit 4. For several weeks, our partner Mario Arroyo has run about five times a week with each of them. “I have been testing what are the three best watchesTo start running and I’m going to tell you why this is the best ”, starts in the video. The goal is clear: check which best meets in key aspects such as comfort, drums, sports functions, experience of use and mobile applications. As for comfort, the three models fulfill well, although one of them feels “slightly thicker and heavier.” The biggest difference is in the way of interacting: Huawei and Amazfit bet on Amoled touch screens, while Garmin resorts to buttons. At first it may seem uninitive. “Uncle, it is very little intuitive,” said Mario, but soon found an unexpected advantage. After running, the analysis comes into play. The apps of the three offer data and statistics, But with clear differences. Garmin Connect is the most technical and deep, while Huawei Health and Zepp (Amazfit) are more visual and intuitive. In this Versus We will try to discover which one that can help us continue to progress. Mario also explores which proposals fall short as smartwatch, as well as related to autonomy. “I have trained a lot with them for a lot of time to be able to make the most precise possible comparison and having so many an hour training, I think they are very good data,” he summarizes. Will anyone stand out? You can discover all the details and the final conclusions In the new Versus On the Xataka YouTube channel. Images | Xataka In Xataka | iPhone 16E vs Samsung A56: the mid -range duel that many asked (and finally we have)

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.