Connect with us

Featured

GPU is the new CPU

Published

on

In the past, the CPU was king. But as games required more processing power, developers looked at ways to improve the GPU, to the point where it is now a mini server within the computer, writes BRIAN PEREIRA of Digital Creed.

The CPU (central processing unit) was king in the PC era. I remember the hotly contested race between two chipmakers (Intel and AMD) to make the fastest CPU. Megahertz was the metric for clock speed then and today it’s gigahertz. Using clever tricks like ‘Turbo mode’ the CPU clock speed could be further bumped up.  The math co-processor and graphics processing unit or GPU played second fiddle or sidekick roles. Recently, I attended the NVIDIA GTCx India 2016 conference in Mumbai, and was surprised to learn how the GPU has matured almost into a graphics server within the computer. In fact, there were demos showing how GPUs are much more efficient than CPUs in processing data. The GPU has cut down processing time significantly, and many real-life applications across industries have benefitted. I will talk about some examples later in this post. But first, let’s get into a time machine and go back to the 90s to see how GPUs evolved.

As gaming software got more defined (realism) and monitor resolutions improved, more power and more memory were needed to process all those mathematical equations that generate those stunning graphics. The CPU (and its IBM PC architecture) could not keep up. Hence we witnessed the emergence of graphics cards. My earliest memory of one is the Hercules Graphics Card. Later we had CGA, VGA, and XGA versions. These graphics cards were mounted in expansion slots on the motherboard.

With better graphics came better sound. There was demand for surround sound on PCs, and multi-speaker systems became available (my favourite ones were manufactured by a company called Creative Technologies). Again, the CPU could not process the complex sound algorithms for surround sound, and hence we got sound cards with onboard sound processors and Digital to Analog (DAC) converters. Back to graphics.

Today’s graphics cards have so much power and memory, they would give a 1990s PC a complex. Sample this: The NVIDIA GTX 1080 graphics card has 2560 CUDA cores, a base clock speed of 1607 MHz, 8GB memory and it consumes 180W of power! If your geekiness makes you curious about CUDA you can read more about it here. Essentially, CUDA is a parallel computing platform and programming model that adapts a GPU for general computing.

And because of the sheer power that today’s GPUs have, they are transforming everything from deep learning and AI to visual computing, high performance computing (HPC) and VR.

Demos & Examples

Walking along the exhibit area at GTCx India 2016 I was thrilled to discover a number of interesting examples. Apart from the various startups showing some amazing demos, there was also a special zone called VR Village.

I was lucky to experience an interactive model of the Bank of England building that was destroyed in 1920. The lobby of the bank was recreated using an interactive model, photo-realistically rendered using NVIDIA’s Iray VR software.

When the HTC Vive VR headset was placed on my head, I was stunned with what I saw. It looked like I was in the deserted lobby of the bank in the early morning hours. The sun’s rays filtered through the glass dome in the atrium.  I looked up and saw the gold gilded edges on the ceiling reflecting the sunlight at just the right angles, relative to my viewing angle. Shadows were created around the grooves and niches on the elegant statues placed around the lobby. And as I walked around or rotated my head, the shadows changed in relation to  changing viewing angles.

Later, as I walked around the exhibit area I visited some of the stalls and witnessed some demos. EDGE Networks has been using the power of GPU processing for a business application. Interestingly, they use data science and AI for workforce transformation and an HR application. I spoke to Rahul Kulhari, their Lead Data Scientist. He told me that their talent acquisition app scans through thousands of resumes and filters candidates based on skills and experience that match a job description. This kind of processing used to take days to complete with a CPU. But now, with the power of a GPU it takes only a few hours.

I sat through an interesting presentation by Mandar Kulkarni, Head Automotive CAE, TCS. He spoke about some product development challenges and how GPUs and HPC have helped cut down production time and cost in automotive manufacturing.

“Earlier we had to make 100 prototypes of an engine and test each prototype. So development used to take years. But now we create virtual prototypes and do various simulations on the models. There are 500 types of simulations for every car — all the components have to be tested. We even do simulations on the coffee cup holder on the dash,” said Kulkarni.

The various simulation include vehicle crash, external aerodynamics, and aero acoustic analysis (to name a few). And billions of equations must be solved to create these simulations. This is only possible through parallel processing and HPC — which a modern GPU can perform.

There are numerous other examples and demos I can describe here to show the power of GPU, but that’s going to make this post a lot longer than I expected! So I will write about these separately in future posts.

I met and interviewed Greg Estes, Vice President, Enterprise Marketing, NVIDIA. Lookout for the interview that I’ll post on this site very soon.

* Brian Pereira has been tracking technology since 1989. He is the former editor of CHIP and InfomationWeek magazines in India. He now blogs at www.digitalcreed.in

Featured

Prepare your cam to capture the Blood Moon

On 27 July 2018, South Africans can witness a total lunar eclipse, as the earth’s shadow completely covers the moon.

Published

on

Also known as a blood or red moon, a total lunar eclipse is the most dramatic of all lunar eclipses and presents an exciting photographic opportunity for any aspiring photographer or would-be astronomers.

“A lunar eclipse is a rare cosmic sight. For centuries these events have inspired wonder, interest and sometimes fear amongst observers. Of course, if you are lucky to be around when one occurs, you would want to capture it all on camera,” says Dana Eitzen, Corporate and Marketing Communications Executive at Canon South Africa.

Canon ambassador and acclaimed landscape photographer David Noton has provided his top tips to keep in mind when photographing this occasion.   In South Africa, the eclipse will be visible from about 19h14 on Friday, 27 July until 01h28 on the Saturday morning. The lunar eclipse will see the light from the sun blocked by the earth as it passes in front of the moon. The moon will turn red because of an effect known as Rayleigh Scattering, where bands of green and violet light become filtered through the atmosphere.

A partial eclipse will begin at 20h24 when the moon will start to turn red. The total eclipse begins at about 21h30 when the moon is completely red. The eclipse reaches its maximum at 22h21 when the moon is closest to the centre of the shadow.

David Noton advises:

  1. Download the right apps to be in-the-know

The sun’s position in the sky at any given time of day varies massively with latitude and season. That is not the case with the moon as its passage through the heavens is governed by its complex elliptical orbit of the earth. That orbit results in monthly, rather than seasonal variations, as the moon moves through its lunar cycle. The result is big differences in the timing of its appearance and its trajectory through the sky. Luckily, we no longer need to rely on weight tables to consult the behaviour of the moon, we can simply download an app on to our phone. The Photographer’s Ephemeris is useful for giving moonrise and moonset times, bearings and phases; while the Photopills app gives comprehensive information on the position of the moon in our sky.  Armed with these two apps, I’m planning to shoot the Blood Moon rising in Dorset, England. I’m aiming to capture the moon within the first fifteen minutes of moonrise so I can catch it low in the sky and juxtapose it against an object on the horizon line for scale – this could be as simple as a tree on a hill.

 

  1. Invest in a lens with optimal zoom  

On the 27th July, one of the key challenges we’ll face is shooting the moon large in the frame so we can see every crater on the asteroid pockmarked surface. It’s a task normally reserved for astronomers with super powerful telescopes, but if you’ve got a long telephoto lens on a full frame DSLR with around 600 mm of focal length, it can be done, depending on the composition. I will be using the Canon EOS 5D Mark IV with an EF 200-400mm f/4L IS USM Ext. 1.4 x lens.

  1. Use a tripod to capture the intimate details

As you frame up your shot, one thing will become immediately apparent; lunar tracking is incredibly challenging as the moon moves through the sky surprisingly quickly. As you’ll be using a long lens for this shoot, it’s important to invest in a sturdy tripod to help capture the best possible image. Although it will be tempting to take the shot by hand, it’s important to remember that your subject is over 384,000km away from you and even with a high shutter speed, the slightest of movements will become exaggerated.

  1. Integrate the moon into your landscape

Whilst images of the moon large in the frame can be beautifully detailed, they are essentially astronomical in their appeal. Personally, I’m far more drawn to using the lunar allure as an element in my landscapes, or using the moonlight as a light source. The latter is difficult, as the amount of light the moon reflects is tiny, whilst the lunar surface is so bright by comparison. Up to now, night photography meant long, long exposures but with cameras such as the Canon EOS-1D X Mark II and the Canon EOS 5D Mark IV now capable of astonishing low light performance, a whole new nocturnal world of opportunities has been opened to photographers.

  1. Master the shutter speed for your subject 

The most evocative and genuine use of the moon in landscape portraits results from situations when the light on the moon balances with the twilight in the surrounding sky. Such images have a subtle appeal, mood and believability.  By definition, any scene incorporating a medium or wide-angle view is going to render the moon as a tiny pin prick of light, but its presence will still be felt. Our eyes naturally gravitate to it, however insignificant it may seem. Of course, the issue of shutter speed is always there; too slow an exposure and all we’ll see is an unsightly lunar streak, even with a wide-angle lens.

 

On a clear night, mastering the shutter speed of your camera is integral to capturing the moon – exposing at 1/250 sec @ f8 ISO 100 (depending on focal length) is what you’ll need to stop the motion from blurring and if you are to get the technique right, with the high quality of cameras such as the Canon EOS 5DS R, you might even be able to see the twelve cameras that were left up there by NASA in the 60’s!

Continue Reading

Featured

How Africa can embrace AI

Currently, no African country is among the top 10 countries expected to benefit most from AI and automation. But, the continent has the potential to catch up with the rest of world if we act fast, says ZOAIB HOOSEN, Microsoft Managing Director.

Published

on

To play catch up, we must take advantage of our best and most powerful resource – our human capital. According to a report by the World Economic Forum (WEF), more than 60 percent of the population in sub-Saharan Africa is under the age of 25.

These are the people who are poised to create a future where humans and AI can work together for the good of society. In fact, the most recent WEF Global Shapers survey found that almost 80 percent of youth believe technology like AI is creating jobs rather than destroying them.

Staying ahead of the trends to stay employed

AI developments are expected to impact existing jobs, as AI can replicate certain activities at greater speed and scale. In some areas, AI could learn faster than humans, if not yet as deeply.

According to Gartner, while AI will improve the productivity of many jobs and create millions more new positions, it could impact many others. The simpler and less creative the job, the earlier, a bot for example, could replace it.

It’s important to stay ahead of the trends and find opportunities to expand our knowledge and skills while learning how to work more closely and symbiotically with technology.

Another global study by Accenture, found that the adoption of AI will create several new job categories requiring important and yet surprising skills. These include trainers, who are tasked with teaching AI systems how to perform; explainers, who bridge the gap between technologist and business leader; and sustainers, who ensure that AI systems are operating as designed.

It’s clear that successfully integrating human intelligence with AI, so they co-exist in a two-way learning relationship, will become more critical than ever.

Combining STEM with the arts

Young people have a leg up on those already in the working world because they can easily develop the necessary skills for these new roles. It’s therefore essential that our education system constantly evolves to equip youth with the right skills and way of thinking to be successful in jobs that may not even exist yet.

As the division of tasks between man and machine changes, we must re-evaluate the type of knowledge and skills imparted to future generations.

For example, technical skills will be required to design and implement AI systems, but interpersonal skills, creativity and emotional intelligence will also become crucial in giving humans an advantage over machines.

“At one level, AI will require that even more people specialise in digital skills and data science. But skilling-up for an AI-powered world involves more than science, technology, engineering and math. As computers behave more like humans, the social sciences and humanities will become even more important. Languages, art, history, economics, ethics, philosophy, psychology and human development courses can teach critical, philosophical and ethics-based skills that will be instrumental in the development and management of AI solutions.” This is according to Microsoft president, Brad Smith, and EVP of AI and research, Harry Shum, who recently authored the book “The Future Computed”, which primarily deals with AI and its role in society.

Interestingly, institutions like Stanford University are already implementing this forward-thinking approach. The university offers a programme called CS+X, which integrates its computer science degree with humanities degrees, resulting in a Bachelor of Arts and Science qualification.

Revisiting laws and regulation

For this type of evolution to happen, the onus is on policy makers to revisit current laws and even bring in new regulations. Policy makers need to identify the groups most at risk of losing their jobs and create strategies to reintegrate them into the economy.

Simultaneously, though AI could be hugely beneficial in areas such as curbing poor access to healthcare and improving diagnoses for example, physicians may avoid using this technology for fear of malpractice. To avoid this, we need regulation that closes the gap between the pace of technological change and that of regulatory response. It will also become essential to develop a code of ethics for this new ecosystem.

Preparing for the future

With the recent convergence of a transformative set of technologies, economies are entering a period in which AI has the potential overcome physical limitations and open up new sources of value and growth.

To avoid missing out on this opportunity, policy makers and business leaders must prepare for, and work toward, a future with AI. We must do so not with the idea that AI is simply another productivity enhancer. Rather, we must see AI as the tool that can transform our thinking about how growth is created.

It comes down to a choice of our people and economies being part of the technological disruption, or being left behind.

Continue Reading

Trending

Copyright © 2018 World Wide Worx