You are reading the article How Google Is Powering The World’S Ai updated in March 2024 on the website Cattuongwedding.com. We hope that the information we have shared is helpful to you. If you find the content interesting and meaningful, please share it with your friends and continue to follow and support us for the latest updates. Suggested April 2024 How Google Is Powering The World’S Ai
Google unveiled its second-generation TPU at Google I/O earlier this year, offering increased performance and better scaling for larger clusters. The TPU is an application specific integrated circuit. It’s custom silicon designed very specifically for a particular use case, rather than a general processing unit like a CPU. The unit is designed to handle common machine learning and neural networking calculations for training and inference; specifically matrix multiply, dot product, and quantization transforms, which are usually just 8 bits in accuracy.
While these kinds of calculations can be done on a CPU and sometimes even more efficiently on a GPU, these architectures are limited in terms of performance and energy efficiency when scaling across operation types. For example, IEEE 754 8-bit integer multiplication optimized designs can be up to 5.5X more energy and 6X more area efficient than 16-bit floating-point optimized designs. They’re also 18.5X more efficient in terms of energy and 27X smaller in terms of area than 32-bit FP multiply. IEEE 754 being the technical standard for floating point computations used in all modern CPUs.
What being an “AI first” company means for Google
Furthermore, many neural networking use cases require low latency and almost instantaneous processing times from a user perspective. This favors dedicated hardware for certain tasks, as opposed to trying to fit typically higher latency graphics architectures to new use cases. Memory latency accessing external RAM can be hugely costly too.
Earlier in the year, Google released a comprehensive comparison of its TPU’s performance and efficiencies compared with Haswell CPUs and NVIDIA Tesla K80 GPUs, giving us a closer look at the processor’s design.
In terms of numbers, Google’s TPU can process 65,536 multiply-and-adds for 8-bit integers every cycle. Given that the TPU runs at 700MHz, it can compute 65,536 × 700,000,000 = 46 × 1012 multiply-and-add operations or 92 TeraOps (trillions of operations) per second in the matrix unit. Google says that its second generation TPU can deliver up to 180 teraflops of floating point performance. That’s significantly more parallel throughput than your typical scalar RISC processor, which usually only passes a single operation with each instruction over a clock cycle or more.
The 16-bit products of the Matrix Multiply Unit are collected in the 4 MiB of 32-bit Accumulators below the matrix unit. There’s also a unified buffer of 24MB of SRAM, which work as registers. Instructions to control the processor are sent from a CPU to the TPU via the PCIe bus. These are complex CISC type instructions in order to run complex tasks which each instruction, such as numerous multiply-add calculations. These instructions are passed down a 4-stage pipeline. There are only twelve instructions for the TPU in total, the five most important of which are simply to read and write results and weights in memory, and to begin a matrix multiply/convolution of the data and weights.Working with Intel for edge compute
Google’s hardware efforts have given it a major head start in the cloud space, but not all AI applications are well suited to transferring data such great distances. Some applications, such as self driving cars, require almost instantaneous compute, and so can’t rely on higher latency data transfers over the internet, even if the compute power in the cloud is very fast. Instead, these type of applications need to be done on device, and the same applies for a number of smartphone applications, such as image processing on RAW camera data for a picture.
Google’s Pixel Visual Core is primarily designed for HDR image enhancement, but the company has touted its potential for other future machine learning and neural networking applications.
With the Pixel 2, Google quietly launched its first attempt at bringing neural networking capabilities to dedicated hardware suitable for a lower power mobile form factor – the Pixel Visual Core. Interestingly, Google teamed up with Intel for the chip, suggesting that it wasn’t entirely an in-house design. We don’t know exactly what the partnership entails; it could just be architectural or more to do with manufacturing connections.
Intel has been buying up AI hardware companies, nabbing Nervana Systems in 2024, Movidius (which made chips for DJI drones) last September, and Mobileye in March 2023. We also know that Intel has its own neural networking processor in the works, codenamed Lake Crest, which falls under its Nervana line. This product was the result of Intel’s purchase of the company of the same name. We don’t know a lot about processor, but it’s designed for servers, uses a low-precision number format called Flexpoint, and boasts a blazing fast memory access speed of 8 Terabits per second. It’s going to compete with Google’s TPU, rather than it’s mobile products.
What is machine learning?
Even so, there appear to be some design similarities between Intel and Google hardware based on images floating around online. Specifically, the multi-core configuration, use of PCIe and accompanying controller, a management CPU, and close integration to fast memory.
At a glance, the Pixel’s hardware looks quite different to Google’s cloud design, which isn’t surprising given the different power budgets. Although we don’t know as much about the Visual Core architecture as we do about Google’s Cloud TPUs, we can spot some similar capabilities. Each of the Image Processing Units (IPUs) inside the design offers 512 arithmetic logic units, for a total of 4,096.
Again, this means a highly parallelized design capable of crunching lots of numbers at once, and even this trimmed down design can perform 3 trillion operations per second. Clearly the chip features a far smaller number of math units than Google’s TPU, and there are no doubt other differences as this is primarily designed for imaging enhancements, rather than the variety of neural networks Google is running in the cloud. However, it’s a similar, highly parallel design with a specific set of operations in mind.
Whether Google sticks with this design and continues to work with Intel for future edge compute capabilities, or returns to relying on hardware developed by other companies remains to be seen. However, I would be surprised if we don’t see Google’s experience in neural networking hardware continue to evolve silicon products both in the server and small form factor spaces.
The future according to Google: AI + hardware + software = ?
Google may be best known for its software, but when it comes to powering this new generation of AI computing, Google is equally embedded in the hardware development and deployment side.
The company’s custom TPU silicon provides the necessary energy efficiency savings needed to deploy machine learning on a large cloud scale. It also offers up notably higher performance for these specific tasks than more generalized CPU and GPU hardware. We’re seeing a similar trend in the mobile space, with SoC manufacturing increasingly turning to dedicated DSP hardware to efficiently run these mathematically intensive algorithms. Google could become a major hardware player in this market too.
We’re still waiting to see what Google has in store for its first generation smartphone AI hardware, the Pixel Visual Core. The chip will soon be switched on for faster HDR processing and will no doubt play a role in some further AI tests and products that the company rolls out to its Pixel 2 smartphones. At the moment, Google is leading the way forward with its Cloud TPU AI hardware and software support with TensorFlow. It’s worth remembering that Intel, Microsoft, Facebook, Amazon, and others are all vying for a piece of this quickly emerging market too.
With machine learning and neural networks powering an increasing number of applications both in the cloud and on edge devices like smartphones, Google’s early hardware efforts have positioned the company to be a leader in this next generation field of computing.
You're reading How Google Is Powering The World’S Ai
By 2012, Hank Green had already been creating content on the VlogBrothers YouTube channel for five years, but he wanted to do more. Among the most popular videos he and his brother John have made are educational explainers, but creating those takes significantly more time than a normal video, due to the need for research and fact-checking. “You need to try harder to make sure stuff is right in an educational video than when you’re making a funny video,” Hank tells Popular Science. But despite being more resource-intensive to produce, they continued to make these videos because “We felt like it was good for the world.”
So that year, the Greens applied for and received a grant to create several new channels as part of YouTube’s Original Channel Initiative. One of the ideas they came up with was SciShow, a YouTube channel with the express purpose of educating the public about the science of the world around them. “For SciShow, we’re just excited about understanding people, understanding the world, understanding the universe, and that is the mission,” according to Green. And that mission is working—SciShow has nearly 3.7 million subscribers and its more than 1,000 videos have been viewed over 505 million times to date.
As video ideas are pitched, no matter the source, the staff considers how many principles must be explained to fully express the idea. The most interesting pitches, and therefore the most interesting videos, come about when the answer is counter-intuitive. De Pastino believes that’s especially true in science, because, “there are a lot of things out there that people either don’t know or think they know and is commonly misunderstood.” Rarely, he says, does an idea come up that the staff thinks is too difficult to convert into a video. Rather, if there are too many principles that need to be explained within a video (the more concepts that need to be explained, the longer a video will be), the staff will break the idea up into chunks and explain each principle individually.
SciShow and its spin-offs, SciShow Space and SciShow Kids, produce several different types of videos, and production for each of these takes different amounts of time. De Pastino estimates that it takes about a month on average between the time a script is assigned and when its filmed. However, scripts for the weekly segment “SciShow News” and “Space News” are written on Monday, filmed on Tuesday, and uploaded on Friday.
SciShow also does longer videos that dive deeper into subjects. Green has a particular affinity for one deep dive episode, “The Science of Anti-Vaccination,” which has been viewed over one million times. The episode is not actually about the effects of vaccination, but rather the psychology of the anti-vaxxer mindset and trying to understand the phenomenon. The reason he enjoys that episode so much is, “because it was an opportunity to frame that conversation in a scientific way,” He continued, “Instead of in a culture war way, where we’re like, ‘Here we are being angry at each other,’ we can be like, ‘Well how do we use science to help understand each other?’ and maybe use those sets of tools to heal that wound a little bit.’”
In this age of viral internet trends though, not every story can be worked on for weeks or comes with an embargo that gives science communicators time to prepare—sometimes a story or trend will take the world by storm, like “The Dress” did in early 2024. “We were all obsessing over it and fighting over what colors we saw, so we were like ‘Well, we have to do it,’” says Caitlin Hofmeister, SciShow’s producer. “Luckily that caught on first thing in the morning and we were able to write a script and shoot and edit it.”
While one editor will normally “carry” a video through the end of post-production, as Hofmeister puts it, multiple staff members may work on individual components simultaneously in such a case in order to get a time-sensitive episode up quickly. “We’re not worried about being the first to report on the gold and blue dress,” she continued, “but we know our audiences wants to know what we think about it so we’ll do a video.”
As with any creator on the internet, SciShow deals with its own set of “trolls,” internet users who express (or feign) outrage against content. Science can be particularly prone to this, between particularly vocal anti-vaxxers, the flat Earth society, those who don’t believe in climate change, and members of several other movements. According to Green, the best way for creators to protect their audiences from trolls, “is to try and create content for people who are a little more thoughtful and to try to not make content for people who aren’t.” While he acknowledges it isn’t the best solution, and can even allow trolls to find each other and create their own communities, he simply states, “you gotta do what you gotta do to protect your community.”
Green has seen many changes in the online video landscape during his tenure, the latest of which is the rise of Facebook as a video platform. He says both the viewer and creator experiences are very different on the social network than on YouTube. “Facebook content tends to be much more like a lighter touch,” he explains. “It’s the kind of content you can pop in and out of without any trouble. Whereas, YouTube, it seems like people are there to spend time watching video and they’re happy to spend five, 10, 15 [minutes], even an hour-long session on SciShow videos, which is much better for us.” He also says that there are SciShow viewers who only watch on Facebook, that hasn’t resulted in a dip in viewer numbers on YouTube.
While some episodes of SciShow are natively uploaded to Facebook, don’t expect to see much original SciShow content made for the social network. In October 2024, Facebook announced it was testing a “video matching technology,” similar to YouTube’s ContentID, which would allow creators to report a duplicate of their video uploaded to the social network. That feature, which Facebook calls “Rights Manager,” launched in April. Unlike ContentID though, Rights Manager doesn’t ensure that the creators of videos are paid for the views of their videos. In fact, Facebook doesn’t monetize the majority of the videos being uploaded to the social network (It was reported in June that Facebook does pay some media companies, including Buzzfeed and The New York Times, to produce videos). Until a revenue model is in place, SciShow won’t prioritize Facebook as a platform. “At the moment, it’s not a high-strategic priority for us to be growing on Facebook,” Green says, “because it seems like it’s really good for Facebook and not for creators.”
Last week during Google’s quarterly earnings call CEO Larry Page laid out a new vision for the company that seemed to signal that the era of Google Labs was at an end. After years of throwing a lot of projects against the wall to see what sticks, Page said that major priority for the company was focus. Page called the approach “more wood behind fewer arrows.”
This new level of focus will mean more support for major Google projects and ideas but it also meant the search giant trim away projects that didn’t work. Page tied this new strategy to Google’s recent decision to shut down Google Health and Google PowerMeter Services Google introduced to give users easier access to their health care and energy usage information respectively. Now Google Labs is the latest casualty of the “more wood fewer arrows” approach.
In Video: Say Goodbye to Google Labs
It’s hard not to feel like we’ve lost something great as the search giant shuts down Google Labs. Labs launched way back in 2002 and in its almost 10 years of existence it’s been responsible for some of Google’s biggest services. Even if a lot of the projects from Google Labs never went beyond the test tube, it was still great for Google to establish a breeding ground for strange and wonderful ideas that didn’t fit in anywhere else in the organization.
But a second look at Google Labs shows that many of these projects didn’t fit anywhere for a reason. It’s hard for even the most diehard Labs supporter to stay enthusiastic about lab experiments like DataWiki with the exciting description of “a wiki for structured data.” I don’t mean to imply that it’s all been failures or boring experiments, just that it’s important to remember when mourning Google Labs that it’s not what it once was.
Still, the highs got pretty high, the project has also been responsible for some enormous successes over the years. Both Gmail and Google Maps originally started as humble lab projects. Recently Google announced Swiffy, a tool that converts Flash files automatically to HTML5. It’s also a godsend for stranded iPad users everywhere. In recent years the experiments coming out of Google Labs have been less revolutionary and it’s wise of the company to focus on the products that really could make as much of a splash as Gmail.
I agree that Google should give more time and support to some of its big name projects, and I can already see how that strategy has served them well on projects like Google+. With all the ideas that Google throws at the populace it’s not surprising they’ve had trouble finding the resources for all of them.
If Google had been able to put more of its programmers on a project like Google Wave, for instance, then we might still be using it today. Instead, after the company hyped Wave as the next generation of e-mail, Google unceremoniously dumped Wave after less than a year. The technology had promise (it gained quite the following among PCWorld staffers) but with numerous bugs and almost no updates to fix them, the service never really found the audience it needed. Who knows, a more focused Google might even have created something like Google+ right off the bat instead of surprising and annoying Gmail users with Google Buzz.
If you still have a favorite project in Google’s Labs all hope isn’t lost. Google has said that while it will be quietly shutting down some of the projects currently in the Labs others will find a new life integrated into the search giant’s other projects and services.
We’ve got our own slideshow of our favorite Google Labs experiments but we want to hear from you. What were your favorite Google Labs projects over the years? What do you hope makes the cut as Google transitions into the new era?
Google has dominated the search engine market for most of its 20-year existence. Today, most SEO efforts mainly revolve around the popular search engine.
Google holds a massive 92.74 percent search engine market share worldwide, according to StatCounter, as of October.
While Google is truly a force to be reckoned with, some view its dominance in the internet search space as problematic.
The company, with its large network of Internet-related services and products, owns a vast wealth of information on its users and we don’t exactly know all the ways they are using it.
Privacy concerns are among the top reasons why some people prefer using other search engines instead of Google.
We wanted to know which Google search alternative is favored by marketers, so we asked our Twitter community.What Is Your Favorite Google Search Alternative?
Here are the results from this #SEJSurveySays poll question.
According to SEJ’s Twitter audience:
36 percent chose DuckDuckGo as their favorite Google search alternative.
32 percent said their top pick is Twitter.
30 percent their favorite alternative search engine is Bing.
2 percent favor Yandex as a Google search alternative.Here Are a Few Comments from Our Twitter Followers
A few followers explained the reason behind their vote:
DDG hands down, it respects your privacy which is why I use it.
— Denpafighter978VGCP (@DAXISAWINNER) October 29, 2023
But in number of search queries @YouTube is on 2nd position. 🙂
— Digital Prem (@DigitalPrem1) November 1, 2023
For me, Bing is as good as Google. I have started using Bing a lot from last 4 months.
However, I am looking forward to install DuckDuckGo (after seeing the poll result). It’s not prominent in India, so it will be interesting to see what results it gives for Indian search terms.
— Mihir Vedpathak🚀 (@VedpathakMihir) October 29, 2023
I actually don’t use anything other than google
— Imtanan Tech Tips (@ImtananTech) October 30, 2023
Other followers also shared a few other Google search alternatives such as:
Mojeek.Which Search Engine Is Right for You?
Whatever your reason is for deciding not to use Google, you have plenty of other search engine options.
Check out the post that inspired our poll, by Chuck Price: 14 Great Search Engines You Can Use Instead of Google.
Learn more about the most popular search engines worldwide with these posts from our SEJ contributors:Have Your Say
What is your favorite Google search alternative? Tag us on social media to let us know.
Be sure to have your say in the next survey – check out the #SEJSurveySays hashtag on Twitter for future polls and data.
Chart created by Shayne Zalameda
Pokémon Go took the world by storm with the Artificial Reality powered game back in July 2024. Although the game’s popularity saw a small downfall, it is still widely popular among players. Many events, raids, and shiny Pokémon have still kept the flames alive for the game, and many players like me go back to the game for the nostalgia it brings.
Here, we shall discuss the best Pokémon Go locations in the world, which may come in handy when you plan your next trip.
Locally, the natural scenery relates to the type of Pokémon available in the area. Thus, you will find water type Pokémon near water parks, grass and bug type Pokémon near parks, etc. You need to keep this in mind before looking for Pokémon locally.
Some of the best Pokémon Go locations also have gyms close to them. Gyms host the raids, winning which grants you special items and a chance of catching the gym boss Pokémon as well. Keep this in mind as well while looking for the good Pokémon Go locations.
Here is a list of some of the best Pokémon Go locations for rare spawns around the world, where you are more likely to find rare spawns.
There are many Asia specific Pokémon that you will find in japan, which include the following:
Farfetch’d, Corsola, Torkoal, Pachirisu, Chatot, Pansage, and Basculin.
Starting with the electronic shopping hub, Akihabara is one of the best places to play Pokémon Go in Japan. The area is filled with Pokéstops, and you can also add lures and incense sticks to boost the spawn rate of rare Pokémon. The popularity of Pokémon Go has led to many establishments in the area offering promo codes, you will also find Pokémon themed cafes, restaurants, and shops in the area.
One of the best things about Akihabara is the popularity of the game itself. You will find many fellow players, with whom you can team up and go on legendary Pokégym battles. Make sure to check out the collectibles as well while you are in the vicinity.
Japan’s Ueno Park is known for the Pokéstop clusters that you will find. At any time of the day, you will find these bundled Pokéstops filled with lures, and you can also use the incense sticks to further increase the spawn rate of rare Pokémon.
Ueno Park is one of the many hotspots in the capital of Japan, and you will find your time used up by the frequent spawns of many Pokémon from different generations. The place is also a popular tourist attraction due to the spectacular Sakura Tree blooms, and is usually filled with tourists in spring. Do not miss Ueno Park if you come to Tokyo.
The Meiji Jingu forest shrine is another popular place in Tokyo, Japan when it comes to Pokémon Go. The Shinto shrine is dedicated to the sacred spirits of Emperor Meiji and Empress Shoken, and the place is a popular tourist spot all year round. The place is a popular rare Pokémon hotspot, so you can catch many monsters during your visit here.
Make sure to check out the spots surrounding the Shrine, such as Yoyogi Park, Tokyo Tower, and Shinjuku. Here, you will find more grass and flying type Pokémon among others, such as Jigglypuff, Electrabuzz, Tangela, etc. Shinjuku is popular in Japan for being the Pikachu hotspot, and you will be able to catch a bunch of Pikachu and collect candies for the same.
The most popular Pokémon Go locations in USA are situated in NY, California, and Chicago. There are some exclusive Pokémon in North America, which include the following:
Tauros, Heracross, Seviper, Illumise, Pachirisu, Carnivine, Basculine, Maractus, Sigilyph, Throh, Solrock and Shellos.
Here are some places in North America where you can find rare Pokémon and common Pokémon with high CP.
The Central Park in Manhattan, NY is one of the most iconic places to play Pokémon Go. The place is filled with Pokestops in close vicinity, and you can place lures to attract many Pokémon to this place as well. You can catch a variety of grass and water type Pokémon due to the water areas around the park.
Make sure to read the safety instructions on the Central Park website before going to play the game there, as the game is approved by them and even hosts events from time to time, given the players follow the required safety measures. While in NY, also check the Metropolitan Museum, Brooklyn bridge, and Times Square.
The pier became a Pokémon hotspot in the summer of 2023, when many rare Gen 1 Pokémon suddenly spawned in abundance. Since then, Santa Monica has regularly been one of the best places to catch rare Pokémon in USA. Although recently there had been complaints about network reception in the Pier, which had led to many players not going to the pier as often nowadays.
You can still find Tyranitar and Unown among other rare Pokémon from different generations, so make sure you do not forget to visit this location.
In a sweet combination of magic and Pokémon, Disneyland resorts are some of the best places to play Pokémon Go and catch powerful Pokémon. The areas are filled with Pokéstops, and you can place lures to attract powerful Pokémon to the vicinity to catch them.
There also gyms present in which you can battle for supremacy. You also get rare item drops, which should help you level up quicker in the game. Make sure you follow the safety measures and proper etiquettes, so as to not disturb the decorum of the place while playing games.
Tip: You will usually find good Pokémon Go locations having many Pokéstops. There are some other places in North America like Chicago lake Front (60 PS), Long Beach, Washington DC, Toronto (16), Vancouver (7), and Miami that you can also check out for the same.
The European continent has many beautiful locations, and good hotspots for Pokémon Go players as well. Europe also has some regional exclusives, which include the following:
Mr. Mime, Mime Jr. (hatched from eggs), Volbeat, Lunatone, Zangoose, Tropius, Relicanth, and Sawk.
If you are planning a Europe trip post-pandemic, here are the locations you may want to check out to catch that rare Pokémon.
As many Pokémon Go lovers will tell you, Paris city is one of best places to find rare Pokémon and grind to level up quicker. Almost any scenic location is Paris (spoiler: there are many) is filled with Pokéstops. One of the most common places is of course the Eiffel Tower, where you will find many Pokéstops.
On your way to the Eiffel Tower, wait around the water areas to catch some powerful water type Pokémon. You will most likely find Dratini, Goldeen, Golduck, and Polywag among many others. While at Paris, head over to Trocadéro Gardens, Louvre Museum, and the Arc de Triomphe for some scenic pleasure along with Pokéstop clusters.
The capital city London hosts many landmarks and sceneries, which accompany Pokémon hotspots. Around the Big Ben, you are more likely to find many rare and powerful Pokémon, which can be increased by placing lures around the Pokéstops. Similar to most other landmarks, you will find Pokégyms here, which give you a shot at legendary raid attacks.
Combining scenic locations with Pokémon battles is a fun experience, and you will surely find this in London. Apart from Big Ben, also check out Hyde Park, the University district, Leicester Square, and the Arsenal Stadium.
The capital city of Denmark has seen some Safari events in Copenhagen, which saw many Pokéstops and gyms being added in clusters. In and around the city, you will find many common Generations 3 and 4 Pokémon spawns, so you can catch Pokémon like Torchic, Makuhita, Gulpin, and Skitty among others while taking a stroll.
Two of the most popular areas in Copenhagen for Pokémon Go include the Amager City Quarter and Fisketorvet. Both these areas have a ton of Pokéstops, and you can place lures to increase the spawn rate of the powerful Pokémon as well. Stroll around the city using the metro, or rent a bicycle there and you will surely find good Pokémon along with some beautiful sceneries as well.
While in Copenhagen, make sure to check out Tivoli Gardens, The National Museum, Amalienborg Palace and the Strøget Shopping Mile for tourist scenic spots while also playing Pokémon Go.
Another beautiful city that is full of beautiful sceneries and powerful Pokémon is the capital city Rome. There are certain areas of the city, especially around natural monuments, which have clusters of Pokéstops. You will find many starters Pokémons from Gen 1 and 2, and some evolved Pokémon like Espion and Umbreon spawning as well.
You will find these clusters around the monuments of national importance, such as the Colosseum, Pantheon, Trevi Fountain and St. Peter’s Basilica. Placing lures should get you powerful rare spawns, and you also get a chance at the legendary Pokégym raid battles.
There are certain Pokémon which you will find in high concentration in the African continent. These Pokémon include:
Corsola. Tropius, Shellos, Pansear, Throh, Basculin, Sigilyph, and Heatmor.
The African continent is filled with beautiful landmarks and scenic routes, so you will have fun while travelling around and looking for Pokémon as well. The beautiful city of Cape Town is one of the best places to play Pokémon Go. There are many areas around the city where you will find rare and powerful Pokémon, so make sure to look for Pokémon near the hotspots.
The city is filled with Pokéstops, so you will be able to restock your supplies while looking for rare Pokémon. If you place lures around these clusters, you will also find rare and powerful Pokémon spawning more often. These clusters are usually around the national monuments, so you can catch Pokémon while sightseeing as well.
Th continental group Oceania houses many important tourist spots from Australia and other countries within the area. Although not immensely popular, there are many regional exclusives that you will find in this area. These exclusives include:
Kangaskhan, Corsola, Zangoose, Volbeat, Tropius, Lunatone, Durant, and Relicanth (rare).
If you are planning a trip to the southern pacific islands post pandemic, here are some of the places you should add to your travel list to catch the regional exclusives and powerful Pokémon.
The Opera house is probably one of the first things that comes to mind when you think of Australia. The Australian capital city Sydney is one of the best Pokémon Go locations, and you are sure to find a tonne of rare and powerful Pokémon. The popularity of the game in the region has led to some local establishments selling promo codes to the trainers as well.
The most abundant Pokémon here is the Pikachu, and Kangaskhan is exclusive to this country. You will find many Pokéstops in the popular Australian locations, such as the Opera House, Sydney Harbour, Circular Quay, and the Bondi Beach among other locations.
After its launch, Melbourne quickly became obsessed with the Pokémon Go game. It is evident from the hotspots you will find in the city of Melbourne, so you can add it to the list of locations you want to catch Pokémon while in Australia. You will find many water type Pokémon along the beautiful banks of Yarra or Port Phillip, and grass and bug type near the Botanic Gardens.
Similar to Sydney, you will have a higher chance of catching Kangaskhan which is a regional exclusive. Check out the various beautiful scenic locations around the State Library, Federation Square, Queen Victoria Market, and the vineyards along the Flagstaff Gardens and Yarra River.
Home to the exclusive Relicanth, Auckland is one of the best areas for Pokémon Go players around the world. Filled with water bodies and scenic locations, you will enjoy your time catching rare Pokémon around the hotspots located all over the city. Near the coastal areas, there are higher chances of finding Relicanth and Durant, although these Pokémon are really rare to find.
The Asian subcontinent is primarily known for the variety of food and tourist spots, and you can add Pokémon to this list as well. The regional exclusives in the Asian continent include the Pokémon mentioned above, and here are some of the places that you should add to your travel plan if you wish to catch Pokémon along with sightseeing as well.
Indonesia has seen influence from many diverse cultures around the world, and so is the diversity in the Pokémon that you will find in this city. There are many places in Jakarta where you can find clusters of Pokéstops, along with cafes and restaurants to help you catch a break when you catch them all.
The most popular areas in Jakarta include the National Museum, The Central Park Mall, and the Plaza Festival. You will also find Pokégyms in around Lotte Shopping Avenue and CityWalk.
While in Indonesia, check out some other popular spots such as Sumatra and Jayapura, where people have reported higher spawn rate of legendary Pokémon such as Moltres and Ho-oh.
The beautiful city of Seoul is mainly known for the tourist spots, and now most of them are filled with clusters of Pokéstops. Farfetch’d is one of the Pokémons exclusive to this region along with Japan, while Torkoal is limited to South Asia, and you will find many others around the popular locations.
While in Seoul, make sure to check the Pokéstop clusters around Myeong-dong, Seoul tower, and Gyeongbokgung Palace among other areas. Here, you can meet other Pokémon Go players as well, who will participate in the Gym Raid battles for a chance at legendary Pokémon.
The small country of Singapore is filled with scenic locations, most of which host huge clusters of Pokéstops. Many events have been arranged in this area as well, with the Sentosa Safari event bringing closer more than 95000 trainers from around the world. Players caught a wide range of Pokémon including Tropius, Unown, and Shuckle among others.
While in Singapore, make sure to check the Pokémon clusters in and around Sentosa, Tao Payoh Dragon Playground, ION Orchard, and the Hougang area for those rare spawns. You will also be able to refill your materials, as you go through the Pokéstop clusters around the country.
So, there you have it. Now you know the best Pokémon Go locations around the world, that you can use to plan your next travel. Comment below if you found this useful, and tell us where you caught your most powerful Pokémon in the game.
Spring that year had come early, and growers predicted that the warmth would produce a memorable vintage—but that heat dried the earth, and set the stage for a particularly bad wildfire year. On the night of August 14, a lightning storm started a blaze at the southern edge of the valley. By morning, high winds were driving the flames north, through the Golden Mile’s dry sage and grass.
On September 10, the valley was under an evacuation order, and Fairview’s Bill Eggert was only partly done with harvesting for the year. The vineyard owner, who retired earlier this year, stuck around. “There’s a golf course between myself and the fire,” he explains, and that highly watered turf wasn’t likely to feed the flames in a hurry. In the end, they stopped within a half mile. “It was just a forest fire,” he says when asked if he remembers anything from the day. “Lots of helicopters, lots of planes bombing, lots of bureaucrats running around justifying their existence. You just carry on, right?”
When the fire was over, Eggert harvested the grapes, pressed them, and set them up in barrels. That winter, when he cracked open the containers for a first taste, there was an odor he didn’t recognize.
“I never really smelled it in my wines before,” he says. He thought a wayward colony of wild yeast might be to blame—a genus called Brettanomyces is known to create savory funks that range from smoke to rot. But he opened another and tasted the same thing. Same with the next. “I asked quite a few different winemakers, and none of them could identify what it was,” he says. Then he realized what it was: He was tasting the fire.
Grapes thrive in dry summers. So do fires. Over the last 30 years, persistent drought caused by climate change has turned the West Coast of North America into a tinderbox. Between 2000 and 2023, fires were twice as common and burned four times as much land in the Western US than in the previous two decades. Sometimes, the damage to grapes is direct: In 2023, fires around California’s Napa and Sonoma destroyed at least 25 wineries; 2023’s Glass Fire wrecked dozens more.
Plenty of winemakers are figuring out how to avoid smoke in their vintages, but Eggert is among a faction of vintners who think wine drinkers can learn to accept, and maybe even love, these tasting notes the same way whiskey lovers embrace a peaty scotch. “I consider it to be part of the terroir now,” he explains.
Adapting to a changing planet in this way presents a two-part problem. First, chemists and horticulturalists are analyzing grapes to understand how smoke in the air becomes a taste on the palate. Armed with that information, winemakers then must puzzle out if they can tame the flavor—or embrace it and hope the consumers catch on.
On a scale from zero to 100-percent-ashtray, Eggert places his 2024 harvest at about a 30—think Dewar’s. So he bottled and sold it as Fumé Franc. It was a hit. “It’s been sold out now for four years, but there’s still people that show up and hope I have a bottle or two kicking around somewhere,” he says.
A LITTLE BIT of smoke in a wine isn’t really a novelty. Many chardonnays are aged in lightly charred barrels, and winemakers talk about bacon, hoisin, and mesquite as tasting notes. Plenty of spirits go even further: Peaty scotches carry the boggy essence of the flames used to dry barley, and mezcal is born out of slow-roasted agave. Tom Collins, a professor of winemaking at Washington State University and the president of the American Society for Enology and Viticulture, has been digging into smoke’s effects on wine since 2008, when fire tainted grapes at the Napa Valley vineyard where he worked. He says two crucial things distinguish wildfire smoke: Drinkers don’t expect it, and winemakers can’t control it.
When a wildfire burns, it doesn’t fully vaporize its fuel. Instead, the heat energy shakes up raw plant material (sap, grass, wood, leaves) and turns it into thousands of chemicals called volatile phenols. These have a tendency to evaporate, making them easy to smell. They’re what give smoke—a cocktail of ash, carbon gasses, and assorted other organic detritus—its “smokiness.”
When the phenols settle over vines, some of them wind up soaking into the grapes. Phenols are toxic to plants, so the fruit begins to seal the excitable chemicals up with sugars, which disguises their smell. “Unless it’s so severely smoke-tainted that the grape is actually reeking of smoke when it’s fresh, you’re not going to taste smoke at harvest,” says Wesley Zandberg, a wine chemist at the University of British Columbia Okanagan.
It’s the fermentation that eventually releases the smoke back into the world. The main engine of that process is cultured yeast. Over the course of months, yeast eats sugar and dumps out ethanol; in fact, the microorganisms are so hungry that they’ll munch the sugar coating off of a phenol, releasing it naked into the wine. That’s the smell that Eggert noticed when he first opened his barrels in 2024. The longer a vintage sits, either in a barrel or bottle, the more phenols come unwrapped.
That time-release reaction helps explain why winemakers hesitate to work with smoked grapes. “This is a really, really big risk if you make a $60 bottle of wine that you expect people are going to buy and put in their cellar and pull out when they really want to impress their guests,” Zandberg says. “If it gets pulled out after five years and it’s just rotten, you’ve completely destroyed your brand permanently. And that is worst-case scenario.”
Some of those challenges would be solved by just being able to predict the effects of smoke. Horticulturalists in Australia have been working on this since 2003, when a drought-fueled bushfire season destroyed tens of millions of dollars’ worth of grapes. And agricultural schools up and down the dry interior valleys of California, Oregon, Washington, and British Columbia have been dialing in on the variables that might impact a vintage’s smokiness for more than a decade.
Variation, they’ve found, can start with the fire itself. The Australian Wine Research Institute has gathered up published findings into a series of fact sheets that many winemakers turn to when first trying to understand what’s happened to their grapes. That research has sketched the outlines of how smoke works: Late-season grapes that have already reddened appear more likely to pick up off flavors. Smoke from far away loses its phenol bouquet, even if it’s thick enough to turn the sky orange—and so soot that blows out over the ocean before settling in interior valleys is going to have less impact than smoke from the next valley over. “The age of the smoke, the distance it travels, whether it’s attached to particulate or water droplets—all of this stuff has a lot of value,” says Zandberg.
“It’s actually been really rare for people to smell the wine and be like, ‘Oh, that is clearly tainted.’”
—Aron Weinkauf, Spottswoode Winery
The Aussie fact sheets have created a palate of specific chemicals (cresols, syringols, guaiacols) that commercial labs can test for, but they’re also full of gaps. They don’t pinpoint, for instance, how close a fire needs to be to impact grapes, what kinds of fruits are at risk, and whether anything can be done to treat specific wines—at least not with any degree of certainty. More questions became apparent as researchers studied North American fires: Differing forests, like North America’s conifer-filled ones, produce different flavors than Australian eucalyptus. “If you smoke fish or smoke meat, there are all kinds of ideas about what’s the best kind of wood to use,” says Collins. “Lots of people spend a lot of time online debating that.”
Collins started running experiments to connect discrete plants to distinct smoke effects in 2024. He set up huge tents across rows of grape vines and bathed them with the exhaust from a wood-fired smoker. So far, he can say that, yes, a grape hit by smoke from western red cedar ends up with a different chemical bouquet than sagebrush.
He’s also trying to hone in on the way phenols blossom as they sit inside bottles. In one experiment, Collins separated the tiny chemicals with a microscopic filter—“stripped them out” with carbon—until he could no longer measure the free-floating smoke compounds in the wine. Then he and his lab rebottled 10 gallons of each. “We could watch month over month, and see the [phenol] levels coming back,” he says. In some cases, there was so much smoke stored in the wine’s sugar that, after a few months, the wine was back where it had started before filtering.
Yet, even if people like Zandberg and Collins can nail down the particular chemistry of a smoky vintage, there’s still no guarantee that everyone will taste the same thing. “We really, really, still have an early understanding of how exactly we can connect an objective chemical measurement with a subjective sensory one,” Zandberg says. Humans—even highly trained ones—can reliably recognize only about 10 flavor notes at a time, while there could be 25,000 chemicals in a glass of red wine. And those a person does detect can accentuate or mask one another. In its pure state, guaiacol might be smoky, or even medicinal, but in smaller concentrations it’s part of what makes a tomato skin spicy; in combination with pungent ethanol, bitter tannins, and fruity yeast esters, it could be transformed even more.
As wildfire becomes a part of the winemaking reality, knowing the molecular makeup of a particular bottle is only half the puzzle. What even the best flavor chemistry can’t can’t tell you is who—if anyone—will enjoy it.
ON A SUNDAY NIGHT this summer, I sat down with my partner and two friends to taste a few bottles of smoke-affected wine: a riesling grown in 2023 in the Columbia River Gorge, on the Washington-Oregon border, and a 2023 pinot noir and cabernet franc from Oregon’s Willamette Valley. The white, which had been affected by the massive Eagle Creek Fire, wore its smoky origin on its bottle. It was branded as a “rauchwein,” a nod to toasted German beers called rauchbiers. The two reds were made from less-smoky grapes that had still been grown during one of Willamette’s first brushes with megafires, including the 200,000-acre blaze known as Lionshead.
None of us—a former cheese-seller, a food journalist, and two enthusiasts—can claim any expertise in wine, but naming every complexity of the bottles wasn’t the point. We wanted to know if we liked what we were drinking.
My three tasting partners agreed that the pinot was overly metallic, with lingering, sharp tannins. ”Rusty pipe,” one said. I found a mushroomy umami note enjoyable, and I didn’t taste even a hint of metal. Two tasters liked the way the riesling balanced green apple and a bitter finish, while I found it a little cloying. We agreed that the cabernet franc was the most surprising of the three, with a musty-sweet smell that my partner read as smoky and others thought was more similar to the funk of kombucha or apple cider vinegar.
What we didn’t taste was smoke, at least not in any way we would recognize in a drink like mezcal or scotch. I got a hint of rubber in the pinot. Was that it? I wondered? The clearest shared quality was an aftertaste that, depending on the sip, leaned bitter or acidic.
It turns out that people who work with wine for a living have had the same experience. “It’s actually been really rare for people to smell the wine and be like, ‘Oh, that is clearly tainted,’” explains Aron Weinkauf, manager and winemaker at Napa’s Spottswoode Winery, which was blanketed in smoke for nearly two months in 2023. “Oftentimes they’ll taste the wine and find that maybe the tannins are a little more dry, or there’s a little bit more of an acrid finish.” That could be a symptom of smoke—but it doesn’t have to be. Picking grapes a little early or late can produce the same effect.
And while the verdict on the wines was overall negative (I was the outlier), our tasting group agreed that with an enthusiastic guide telling us the story of each bottle, we probably could have been swayed. If anything, I want to try a smokier bottle.
Some winemakers, like Doug Tunnell, of Brickhouse Wines in Oregon, have decided that those notes don’t have a place in his lineup. “With our customers, it’s not something that they’re going to embrace.” After a fire in late September 2023, he discarded some batches of heavily affected grapes and blended the unaffected ones with leftover barrels from 2023. “We take a hit, six bucks a bottle, or so,” he says. “But our feeling was, ‘We need to make something out of this vintage.’”
Not everyone can afford to ditch some, or all, of a harvest, though. Fairview Cellars’ Eggert went ahead with Fumé Franc—as a pragmatic farmer rather than a romantic vintner. He makes just 5,000 cases a year, which puts Fairview at the small end of the “small winery” category—bigger than around 80 percent of US outfits, but tiny compared to the 71 biggies that produce more than 500,000 cases a year. Larger wineries value predictability, so they might treat wine with filters or carbon to remove phenols. “A lot of places would rather strip it out and not talk about it,” says Rhys Pender, a master of wine and winemaker in British Columbia. But, he says, that often removes other sparks from the wine. “These small-production wineries don’t necessarily want to, and they don’t have to.”
There’s a parallel to the growing popularity of natural winemaking, a philosophy that avoids treating wine and highlights idiosyncrasies. “Natural wine is all about an open-mindedness among consumers,” Pender says. “You’ve got this new market that’s more interested in story, focused on where their food comes from, and what does that taste like?” In 2009, for instance, Eggert made a vintage from grapes that had been split open by a hailstorm and allowed to dry on the vine. He marketed it as The Wrath.
The Mount Kobau fire which raged for weeks in 2024 immediately behind the vineyards on the Golden Mile Bench. The British Columbia Wildfire Service initiated a countermeasure called a controlled back-burn, seen here, that greatly reduced the blaze. Luke Whittall
Lots of things can go wrong in winemaking, and the resulting flavors are called faults. Some are climatological, others technical, and others bacterial. And plenty of flavors that could be faults are fine in moderation. “Oak itself isn’t technically defined as a flaw,” says Chad Stock, a winemaker at Oregon’s Limited Addition Wines. Too much, though, can be sickly and overpowering. “There are an infinite number of what I would call artistic examples of delicious wine that can defy and bend the rules,” he says.
Stock and his winemaking partner and wife, Bree, tasted Eggert’s Fumé Franc on a 2024 visit to British Columbia and encouraged him to work with the vintage. “The smoke profile actually felt a lot more like a barrel toast than it did acrid or ashy.”
According to Bree, a master of wine, many of the rules about what good wine tastes like were revised recently, by modern palates. Twenty years ago, red wines from Burgundy and Bordeaux were renowned for tasting like the floor of a stable—like the barnyard funk of a ripe goat cheese with a creamy core and a fluffy rind—because of Brettanomyces, the same yeast that Eggert first suspected of tainting his barrels. “I remember when I first had Burgundy in a restaurant,” she says. “My wine director gave me one, and I came back to her and I said, ’This tastes like horseshit.’ And she said, ‘Yeah, that’s burgundy.’”
In the past few decades, fermentation chemists developed tools to test for and eliminate the yeast—in part because American palates didn’t favor it. “Now it’s very difficult to find a wine from Burgundy with Brettanomyces,” Bree says. “It’s been that quick of a shift. Modern winemakers started showing sellers in these old traditional regions that, ‘Oh, yeah, this flavor is actually a fault.’”
Winemakers who see smoke as a fault have found different ways to adapt. Some release vintages that are specifically marketed to be drunk within a few months of bottling to head off gradually intensifying smoke. Some blend vintages to keep production running when they lose some berries. Others actively treat wine via reverse osmosis and carbon to remove volatile phenols.
But some have come back to smoke after experimenting. JR Fletcher, one of the winemakers at the Hatch, near Fairview, has explored filtration in the years since the 2024 fires. “Honestly, sometimes it works, sometimes it doesn’t,” he says. They’d had success in just releasing their 2024 cabernet franc, which they called Smokeshow. “It was kind of funny, a little cheeky,” Jones says. When the tasting room heard about it, they wondered if anyone would like it. But the sellers dug in. “It was really a logical progression for us to point out the smoky and be like, ‘Hey, this is technically a wine fault. But why?’”
If a winemaker is working in pursuit of some ideal, Fletcher says, they’re going to miss surprises in front of them. And maybe they’re missing a way that regional flavors will change with the climate. The Okanagan Valley and the surrounding region have been covered in smoke every August for the past several years.
In fact, Fletcher and Jones want to keep making wine with smoke—on purpose. “We did set aside some cab franc this year for another Smokeshow,” he says. “It’s our biggest cult-following wine we’ve ever put out,” Jones adds. And, she jokes, they’re going to sell it for a lot more than they did last time.
Read more PopSci+ stories.
Update the detailed information about How Google Is Powering The World’S Ai on the Cattuongwedding.com website. We hope the article's content will meet your needs, and we will regularly update the information to provide you with the fastest and most accurate information. Have a great day!