The United States only borders two countries. The country that borders the United States to the north is Canada. The United States’ 5525 -mile (8891 kilometers) long border with Canada is the longest land border in the world.

The country that borders the United States to the south is Mexico. This 1,954-mile (3145 kilometers) long border separates two of the most socioeconomically linked countries in the world.

The US and Mexico do a lot of business together. The United States buys more from Mexico than any other country, Mexico buys more from the United States than any other country, and Mexico sells to more to the United States than any other country. Also, there are 37 million US residents with Mexican ancestry. That means there are more Mexicans living in the US than there are Canadians living in Canada.

The border between Mexico and the United States is crossed more than any other border in the world. There are 50 spots where you can legally cross the border by land. The smallest of these legal border crossings is the Boquillas Port of Entry. This location processes 11 thousand trips per year remotely using video inspection kiosks. The largest, the San Ysidro Port of Entry, processes 11 million pedestrians and 26 million cars per year. This makes the San Ysidro Port of Entry the busiest land-border checkpoint in the western hemisphere. The San Ysidro Port of Entry is part of the primary route between San Diego and Tijuana, 2 cities in which many people commute between on a daily basis.

Hundreds of thousands to millions of people illegally enter the united states each year. The vast majority of illegal border crossings are unsuccessful. An estimated 55% to 85% of illegal crossers get caught.

Why come to the United States?

There is an enormous economic disparity between the United States and Mexico. In the United States, the average income is $65,000 per year. In Mexico, the average income is $9,500 per year. Also, The United States has greater social mobility than Mexico. This means that it is easier to move up the social latter in the United States than it is in Mexico. In the United States, its easier to be born poor and end up rich. By contrast, in Mexico, if you are born poor, you typically die poor. The United States has a social mobility rating of 70.4 compared to Mexico’s rating of 52.6. The pursuit of money and the escape from poverty are the strongest reasons to immigrate. Some other reasons why someone might want to come to the United States is to reunify with family or to escape violence.

Why not do it legally?

For some people, there is no legitimate route to legal status in the United States, and for others, there is no practical way. Many forms of American immigrant Visas have annual caps that are not proportional to immigration demand for a given country. For example, a married Mexican citizen who is the child of an American citizen would have to wait 24 years to have their visa processed. The current backlog means that in 20202, immigration offices are just now looking at applications submitted in 1996.

Should I cross illegally?

Probably not. But your choices are few. You can either wait 25% of a lifetime or take the faster, riskier option, which is to cross illegally. 95% of first-time-crossers higher a smuggler.

How much does a typical smuggler charge?

A smuggler costs about $3,500 per person, which is about 4 months’ worth of wages for the average Mexican.

What is included in the smuggling fee?

For this fee, a smuggler will typically

• Put you in a group
• Take your group to the border
• Instruct your group on how to cross
• Arrange transportation on the other side

This article is currently being written. Check back for updates!

The post How to illegally enter the United States appeared first on TechReportArticles.

This argument is often made as ‘Apple vs Android’, ‘iPhone vs Android’, or even ‘Apple vs Samsung’, so I wanted to clear a few things up. Apple manufactures consumer computer hardware. They have a line of desktop and laptop PCs, and they have several tablets and phones to choose from. To maintain tight control over their products, Apple also provides each platform with an Operating System. For desktops and laptops, Apple provides Mac OS. For phones and tablets, Apple provides iOS.

That’s the end of the line for iOS. There is no version of iOS that a consumer or a business can purchase to install on their own device. No amount of money in the world is going to see Apple license its operating system to a competitor.

On the other end of the spectrum you have Google. Google is predominately a software company and is well known for, ya know, the internet. They develop and maintain a totally free to use operating system. Android can be freely downloaded by anyone to tinker with, or by any company wishing to create a phone. As it turns out, its pretty easy to build a phone if you have the right equipment. Making a compelling operating system that people are willing to switch to is a completely separate challenge.

Because of this fact, you are able to go to your closest grocery store and get a $10 to $50 dollar android phone. Sure, your user experience on that device will range from frustrating to abysmal, but think about it:

You can walk into a grocery store and spend less than $50 on a device that has at least 4 processors, 2GB of RAM, Wi-Fi, Cellular Data, 2 cameras, a compass, speakers, and more. Not to mention it includes a display that provides a clearer image than a $3000 TV did 20 years ago. All powered by an included, rechargeable battery. Its an internet browser, a TV camera, a music player, and well, a phone, among many more things. Hell, I think I paid more for an actual guitar tuner a few years ago than what it costs today to buy a phone and use an app.

Android and only android provides this level of accessibility. This fact, of course, is the root of the assumption that Android is for ‘poor people’. No, Android is for everyone, even poor people. There are expensive Android phones. In fact, some of them cost more than iPhones do.

There is something fishy about this chart

Notice that it doesn’t show how much RAM each device has? You may think, ‘Well, all you have to do is look up each device.’ And yes, that works for all items on this list except the top 2. The 2 modes of the Snapdragon 865 that are referenced at the top of the chart shows only the name of the chip, rather than the device that is housing it. This is more than likely because the chip was benchmarked on a Snapdragon 865 development board. Development boards aren’t subjected to the constraints that all the below devices operate in.

Development boards have large heatsinks and generally come with more RAM than you will find available on actual products. Because of this, clock speeds on a development board won’t have to throttle in the same way that ones that are installed in actual phones, do.

So, what does the chart show, exactly?

All that chart shows me is that it takes Qualcomm 8 cores to do a little better than what Apple can do with 6. Pound for pound, core for core, watt for watt, Apple’s platform is superior.

This is understandable, considering the fact that Apple licensed the ARM instruction set and built a custom ARM CPU, where Qualcomm and other ARM CPU vendors use an off the shelf mix of ARM Cortex reference CPUs.

I have to admit, I do agree with the statement somewhat. It is definitely hard to get humans to change their habits and what they are comfortable with. This meme, however, implies that the iPhone is superior to Android, and the only reason people don’t switch is because they are stubborn.

The truth is much more complicated than that. There are factors to consider such as cost, capability, and availability. Some other things to consider are performance, battery life, and ease of use.

Build Quality – Winner: iPhone

As someone who has taken apart many types of phones over the last decade, I can tell you that bar-none, Apple has the best build quality. It’s much more difficult to get into an iPhone than it is your run-of-the-mill Android device.

Do you ever notice that it’s hard to find screws on an Apple device? If they are obvious, like in a Macbook Pro, they are really nice looking and there are only a few of them that are visible. This is because Apple takes great care in how a device looks. They want their devices to look a certain way, but they also want them to have a high build quality. That means using screws in place of clips, so, sometimes they have to put those screws in some out-of-the-way places.

Hardware Performance – Winner: iPhone

Qualcomm’s Snapdragon 865

Qualcomm has, in the past, licensed the ARM instruction set and made a killer CPU, but lately, they have been using off-the-shelf designs provided by ARM. In the case of the Snapdragon 865 which, make no mistake is an excellent chip, Qualcomm chose to use 4 ‘Kryo 585 Silver‘ cores operating at 1.8 GHz, 3 ‘Kryo 585 Gold‘ cores at 2.42 GHz, and a single ‘Kryo 585 Gold‘ core clocked at 2.84 GHz.

Even though real-word tasks don’t scale this way, benchmarks surely do. So, that means the Snapdragon 865 has an accumulative 17.3 billion cycles of CPU power at its disposal every second.

Apple’s A13 Bionic

Apple, who designs their chips from scratch, didn’t need to pile on that many cores to achieve an acceptable level of performance. So, Apple chose a 6 core design for the A13 Bionic.

It has 2 ‘Lightning’ cores running at 2.66 GHz, and 4 ‘Thunder cores’ clocked at 1.82 GHz. This core and clock speed layout provides a total of 14.28 billion cycles of CPU power per second.

Benchmark points per core

The Snapdragon 865 takes 8 cores to get a score of 557,378. That comes out to 69,672 points per core. The A13 Bionic has 6 cores and is able to get a score of 498,296. That provides a score of 83,049 points per core. This puts Apple’s chip in a core-for-core lead of 19.2%.

Benchmark points per clock cycle

The Snapdragon 865 takes 17.3 billion cycles of CPU power to get a score of 557,378. That comes out to 31,038 clock cycles per point. The A13 Bionic has 14.28 billion cycles of CPU power per second and is able to get a score of 498,296. That’s 28,658 clock cycles per point. This means it takes Apple’s chip 7.69% fewer clock cycles for each benchmark point.

How apps run

iOS apps are written in Objective-C and run directly on the CPU. Android apps, on the other hand, are written in Java. Android uses a JIT (Just in Time) compiler to translate Java into machine language. This translation process impacts performance and it’s simply something that iOS apps don’t have to do to run.

This article is currently being written. Check back later for updates!

The post iPhone vs Android: A definitive breakdown of the age-old argument appeared first on TechReportArticles.

The patent was filed as ‘Device Having integrated Interface System.’ This is a continuation patent, so its based on an earlier approved or pending patent application. Apple detailed a laptop that makes use of a touch-based surface where the keyboard would usually go.

With this next-gen MacBook having a glass input surface also means that the MacBook could take on completely new interfaces beyond a keyboard and trackpad… It could theoretically present users with a game controlling interface when playing a video game. It could provide DJ’s with a touch-based turntable and audio controls. It may offer video editors with an interface that could take advantage of touch instead of a keyboard. It could be a clean canvas to allow artists to use their Apple Pencil.

Jack Purcher

The patent shows how a thin physical keyboard can be set on top of the capacitive touch surface. This ‘thinboard’ will then translate physically typing movements to the touchscreen.

This all seems great. Apple is excellent when it comes to user interfaces. Sure, if anyone else did this, it would be 3 years ago and awful. Apple, however, has a habit of waiting for a technology to become mature enough for them to put their name on it. Once that happens, they innovate like crazy and make a great, overpriced product. What I am more excited about, are Android and Windows devices that use this setup.

In addition to the touch-based key entry, the patent also shows that this laptop will have wireless charging coils built into the palm rest. For some reason, people are starting to refer to this as ‘reverse wireless charging,’ but I don’t think that name is appropriate. Horizontal wireless charging, maybe? I’m not sure. I have not put much time into thinking about it, but we have to come up with something better than ‘reverse wireless charging.’

Of course, this is just a patent application. This technology may never see the light of day, but honestly, that’s unlikely. This is the direction that computing has been moving in for quite some time. Others have done similar things before, and when you see Apple file a patent, they usually act on it. It will, however, probably take a while for this thing to show up on the market.

The post Apple’s Next Macbook Might Not Have a Keyboard appeared first on TechReportArticles.

TAIWAN — A provider of high-performance GPGPU (General Purpose Graphics Processing Unit) and AI computing solutions, Aetina Corporation, released its AN810-XNX edge AI computer that leverages the capabilities of the nVidia Jetson Xavier NX SoC (System on Chip). This AI processing board is available in several I/O configurations that are designed for things like drone, UAV, and industrial inspection, deep learning, and medical imaging.

The AN810-XNX has a Nano-ITX form factor and is composed of an NVIDIA Jetson Xavier NX board on Aetina’s new AN810 carrier board. It has a full M.2 slot which is ‘M’-keyed, another than is ‘B’-keyed. It includes support for PCI Express, SATA, USB 3.2 Gen2, and of course, USB 2.0. The AN810-XNX SBC includes support for 4G and 5G modules. The board has a built-in SIM card slot.

The AN810-XNX includes a 120-pin connector that fully supports the MIPI CSI-2 (Mobile Industry Processor Interface Camera Serial Interface) interface. The board can manage some rather intensive AI workloads like using ultra-high-resolution cameras for image analysis.

The Jetson Xavier NX’s CPU is capable of 21 TOPS (Trillion Operations per Second) which is great for running modern AI workloads. It is able to do so while consuming just 10 watts of power. The NX has the processing power to run several neural networks in parallel while processing data coming in from multiple high-resolution sensors.

“AIoT environment is diverse and complex. Throughout the edge AI systems, GPU-based platform has the most flexibility and Azure cloud services enhance the overall performance. Benefits the customer with shorter development period,” said Joe Lo, general manager of Aetina. AN810-XNX, high compatibility platform with well-backup service from edge to cloud, could deploy high-quality, software-defined features on embedded and edge devices.

The post AN810-XNX: Aetina’s new nVidia Jetson-based platform appeared first on TechReportArticles.

Researchers ‘Prevost’ and ‘Ghose’ have designed a deep-learning algorithm that can generate an engaging soundtrack that is synchronized to an otherwise silent video.

Conventionally, sound effects in movies are added in post-production (after the recording) in a process called ‘Foley’. Some Texan researchers, however, decided to use deep learning to automate this process. The team was able to train a neural network to understand 12 popular movie events where Foley effects are usually added. Their neural network detects the type of sound to generate, then it uses a sequential network to generate that sound. To sync things up just right, the team then used neural networks to sync up the generated effect with its respective frames of video.

This was accomplished by creating a dataset that contained short video clips that featured the 12 distinct movie events. The team recorded 1000 videos that were about 5 seconds long. For things like footsteps, cutting, and clock sounds, they physically generated and recorded in a studio. Other sounds, like a horse running, gunshots, and those involving fire were downloaded from YouTube. 

The next thing that needs to be done is to predict the correct class of sound. To do this, they compared two methods. A TRN (Frame Relation Network) and an FSLSTM (Frame Sequence Network).

Frame Sequence Network

In the first approach, they take each video frame and interpolate them between the other frames in the video to increase granularity. They used a ResNet-50 CNN (Convolutional Neural Network) to extract the image’s features. Then, the sound class was predicted using a Fast-Slow LSTM (Recurrent Neural Network) and then computed with the image features. In the FRN, they attempted to capture the object’s actions and the transformations of detail using less computational time.

Frame Relation Network

The MSTRN (Multi-Scale Temporal Relation Network) then compares features from frames that are at a controlled, but changing number of frames apart. Finally, MP (Multilayer Perceptron) is used to combine all the features.

Then, a sound has to be generated for this class. To do so, the Texan researchers used a technique known as Inverse SFFT (Short Time Fourier Transform). Using SFFT, they determined the average of all captured spectrograms related to each sound class. Doing so gave them a decent starting point to generate a sound. The only job left for the neural network at this point is to predict the change to this average sound bias for each sample step of the sound. Moden video uses over 44,000 sound samples per second.

Four methods were used to review the performance of the project. One of which was a human-based qualitative evaluation. Local college students were asked to choose the most appropriate sound, the most lifelike sound, the most accurately synchronized sound, and the sound with the least noise.

As it turns out, the synthesized sounds were preferred over the actual sound 73.71% of the time for one of the models. In another, 65.96% of students prefered the generated sounds.

The post Artificial Intelligence systems can now create soundtracks for totally silent videos appeared first on TechReportArticles.

Poco, a Xiaomi spin-off, is releasing another budget smartphone, the Poco X3 NFC. The new device will be the first phone in the world to use Qualcomm’s Snapdragon 732G SoC (System on Chip).

Processors

The Snapdragon 732G is an 8-core CPU that uses Qualcomm’s semi-custom Kryo architecture. It has two 2.3Ghz Kryo 470 Gold (Corex A57-based) cores along with six 1.8Ghz Kryo 470 Silver (Corex A76-based) cores. As far as graphics processing, the Snapdragon 732G features an Adreno 618 CPU that is capable of running a 120Hz display.

The GPU found in the Poco X3 is actually pretty decent. The Adreno 618 is usually reserved for upper mid-range smartphones and tablets. It’s the same graphics processor that you will find in more expensive phones that use the Qualcomm Snapdragon 730 and 730G SoCs.

The Adreno 618 GPU fully supports Vulkan 1.1, OpenGL ES 3.2, OpenCL 2.0 and DirectX 11.

Display

Speaking of which, the Poco X3 has a 1080 x 2400 (20:9) 6.67″ 120Hz display that comes in at 395 PPI (Pixels Per Inch). Its touchscreen interface updates quickly, taking 240 touch samples per second. That amazing display and touch system is protected by Corning Gorilla Glass 5.

Cameras

The Poco X3 has an impressive 4K-video-capable 64MP quad-camera module in the usual place on the back. You won’t be doing 60FPS 4K video with this phone, though, as it is limited to 30FPS when recording video in 4K mode. It does, however, support slow-motion capture at 960 frames per second. The Poco X3 employs artificial intelligence for scene detection, filters, and studio lighting. On the front, you will find a 20MP selfie camera.

Memory

This budget phone has a whopping 6GB of LPDDR4 RAM and can be configured with up to 128GB of storage. It, of course, includes an SD card slot that will accept SD cards up to 256GB in size. Also, it’s got a huge 5160mAh battery that supports 33W fast charging.

The Poco X3 in an Android 10 device that has twin self-cleaning speakers, a standard 3.5mm headphone jack (the port that refuses to die), and an obligatory USB-C port. It’s also got a side-mounted fingerprint scanner, which I happen to favor over the rear-mounted ones and full support for Google Pay.

What is a self-cleaning speaker?

A pass-down from Xiaomi, the Poco 3 uses low frequency (800Hz) vibrations to clear the speaker of dust and other debris that may have settled on the dual mesh layers of speaker.

Price

Poco is providing a discount for early adopters, selling the 64GB version of the device on AliExpress for $238.80 (₱11,601.86). You can pick up the 128GB version for $298.80 (₱14,516.90).

The post The Poco X3 is the first Qualcomm Snapdragon 732G phone and has self-cleaning speakers appeared first on TechReportArticles.

When nVidia introduced the GeForce 10-Series, it began the process of limiting SLI support to certain GPUs. Since then, Team Green has been doing this at an ever-increasing pace. With the 2000-Series, nVidia made it so that only the GeForce RTX 2070 ‘Super’ and higher GPUs supported the feature. The GeForce RTX 2070 was never really considered an ‘enthusiast-class’ GPU to begin with, so setting the bar there was reasonable. Also, the overwhelming majority of modern games didn’t include SLI support, anyway.

What graphics cards are SLI compatible?

Now, though, with the GeForce RTX 3000-series, if you want SLI support you are going to have to spend at least $3000 (₱145,890) dollars. That’s right. The only nVidia GeForce 3-Series graphics card that supports SLI is the $1,499 ₱72,945 dollar GeForce RTX 3090.

To add to the cost, most power supplies would struggle to support a single RTX 3090, let alone two of them.

SLI is still a thing, but only barely. Now only tech enthusiasts with fat wallets can afford the minimum GPU configuration needed to have a multi-GPU setup. This change, of course, will discourage game developers from including support for SLI in their games due to SLI’s market percentage shrinking to almost non-existent levels.

Nvidia is making this move after recently releasing its RTX 3000 series video cards which, in their defense, offers an amazing price to performance ratio over previous-generation hardware. The removal of SLI, such an adorned feature that has been around since 2004, is going to upset quite a few people.

SLI, which stands for Scalable Link Interface, allows a user to combine the performance of 2 or more graphics cards to increase gaming performance by splitting the workload across multiple cards. This would, in turn, increase frame rates. Way ahead of its time, and using a restrictive PCI express interface, SLI never really hit the mainstream. Even though it’s a great idea that works well both on paper and in practice, it always represented a small part of the gamer market due to the requirement to purchase 2 or more graphics cards. It was hard to find an SLI compatible game, as it was usually only reserved for AAA titles in the real of Battlefield and Call of duty.

The post nVidia drops SLI (multi-GPU) support for its GeForce RTX 3000-Series graphics cards appeared first on TechReportArticles.

Its largest investor is Creo Capital, which is the finance wing of Hong Kong conglomerate, ‘New World Development.’ Jumping on the bandwagon, ‘Linear Venture’ and ‘Redpoint Ventures China,’ also invested in the Chinese AI startup.

Supremind states that its goal is to ‘enhance industrial solutions, product lineups, and technical development with NWS Holdings.’ NWS is the infrastructure department of Supremind’s main partner, New World Development. Supremind will use this money to expand its business and to further increase its share of the market.

Supremind started in 2019, and its core members have taken part in the research and development of still images, video deep learning, and large-scale application systems for over a decade. The company tackles problems that face major urban areas. Supremind does so by leveraging AI technology and video analysis.

The company focuses on 5 areas which are urban governance, transportation management, traffic control, public service administration systems, and internet content censoring. Supremind’s products and services have been implemented by over 200 Chinese cities, already. Among them, its most popular products are those related to smart transportation and have been implemented in over 100 Chinese cities.

Local governments in Zhejiang Province and Shanghai City are utilizing Supremind’s Urban governance solutions. Also, Supremind’s public service administration software is in use by several public security system departments in Shanghai. The company’s internet content censoring platform is, of course, widely used across China by the Offices of the Central Leading Group for Cyberspace Affairs.

Human video monitoring is very ineffective and inefficient. It’s hard for a human to keep up with so many factors. Due to this fact, traffic management has become difficult and has led to a lot of accidents occurring.

Using data resources regarding transportation and road systems provided by NWS Holdings, Supremind has been able to develop its ‘VisionMind’ platform. VisionMind facilitates efficient highway management by enabling the quick analyzation and identification of quite a bit of data in a short period of time. Visionmind is able to identify accidents, illegal acts, traffic volume, and specialized vehicles.

Supremind is expecting sales from its smart transportation management system will make up most of its sales in 2020. Supremind is placing its crosshairs on road-operators, government offices, and institutions. Supremind predicts that it will take only about a month to recoup its investments. Supremind aims to, at the very least, increase its market’s total size by 100% while at the same time boosting sales to more than 100 million yuan ($14,640,216 USD) this year.

Supremind formed an alliance with NWS Holdings in April to combine their technological and industrial expertise. This move helped Supremind gain an edge in the market. Now, the company can more easily develop smart highway and architecture management systems.

Supremind will first apply AI video analysis to smart transportation management. Then, it will focus toward smart city construction.

The post Communist China Builds Artificial Intelligence Monitoring System For Smart Cities appeared first on TechReportArticles.

US trade-bans on Chinese companies like ZTE and Huawei appear to be only the beginning. The US DOD (Department of Defense) confirmed that the agency is in talks as to whether or not they will add China’s Semiconductor Manufacturing International Corporation (SMIC) to the growing list of banned Chinese companies. SMIC is China’s biggest semi-conductor manufacturer, and it would suffer a critical blow if it lost all of its US-based business.

The DOD is concerned that SMIC might be helping China’s defense infrastructure. Defense contractor ‘SOS International’ recently issued a report that claimed that SMIC worked with one of the largest defense companies in china. It also claims that researchers at university associated with the Chinese military among the design teams.

SMIC, of course, rejected the allegations in a statement. The company instead insisted that its products and services are ‘solely’ for civilian use. It also claims to have ‘no relationship with the Chinese military,’ but we all know that its impossible to own a company in China without having to report directly to the Chinese communist party. SOS, however, claimed that SMIC is indeed ‘deeply embedded’ in Chinese military projects.

There are levels to this S$%!.

If and when the United States adds SMIC to the ban-list, it will escalate an already pretty tense trade war between the two countries. In a lot of cases, there are no alternatives to using American parts, so SMIC could face major difficulties in growing or even maintaining its business. That would then inhibit SMIC customers like Huawei as they run out of manufacturing options. China might, then, retaliate by hurting US companies that depend on Chinese manufacturing and parts.

The post The United States is considering banning China’s largest chip maker appeared first on TechReportArticles.

Electronic skins already have the ability to react to touch, but they can’t so very well react to things like burns. This is because computers have a very limited understanding of pain. This is, of course, an issue for prosthetics and robots that need to have human-like responses. Thanks to RMIT, however, these devices will be more sensitive in the future. University researchers developed an artificial skin that can react to pain in the same way that humans do. The system provides ‘near-instant’ when it detects that pressure and temperatures hit critical levels.

The prototype is composed of very thin, stretchable, extremely thin electronics that employ a bio-compatible form of silicon. The device includes the ability to sense pressure and has several temperature-reactive coatings and brain-inspired memory cells. Ataur Rahman, a researcher, said that its accurate enough to be able to distinguish the difference between a gentle poke a painful stab. The design is modeled after neural pathways, receptors, and neurons themselves, that are responsible for enabling human senses.

Don’t get excited too soon yet, though. This research has a long way to go before it gets out of the lab and into the wild. The potential uses are obvious and vast, though. The technology will enable a prosthetic arm to better understand things while keeping its users out of danger. Industrial robots could be more attractive to the public by exhibiting a more human-like, gentle approach.

The post New developments in electronic skin allow it to react to pain like a human appeared first on TechReportArticles.