Browse content similar to Megabits. Check below for episodes and series from the same categories and more!
Line | From | To | |
---|---|---|---|
Where would any of us be without a computer? | 0:00:09 | 0:00:14 | |
From playing games, to developing medicine, | 0:00:14 | 0:00:17 | |
to watching movies, or exploring Mars, | 0:00:17 | 0:00:19 | |
computing's influence is continually breaking new ground. | 0:00:19 | 0:00:23 | |
It may interest you to know | 0:00:23 | 0:00:24 | |
that computers have also been ever present | 0:00:24 | 0:00:27 | |
in our parents' and even our grandparents' lives. | 0:00:27 | 0:00:31 | |
And it wasn't the Americans or the Japanese that first built them, | 0:00:31 | 0:00:35 | |
it was a British Post Office engineer called Tommy Flowers, | 0:00:35 | 0:00:37 | |
way back in 1943. | 0:00:37 | 0:00:41 | |
Now, I bet you didn't think it would look like | 0:00:41 | 0:00:43 | |
it had been knocked up in a back garden shed, | 0:00:43 | 0:00:45 | |
but you should always respect your elders, | 0:00:45 | 0:00:48 | |
and Colossus is as old and as big as they come. | 0:00:48 | 0:00:52 | |
Colossus was one of the first computers, yes. | 0:00:52 | 0:00:55 | |
And we can talk about it in academic or engineering terms. | 0:00:55 | 0:00:59 | |
But it was probably the most significant computer | 0:00:59 | 0:01:01 | |
because it had a worldwide impact. | 0:01:01 | 0:01:05 | |
It shortened the war, together with other activities at Bletchley Park, | 0:01:05 | 0:01:09 | |
by at least two years. | 0:01:09 | 0:01:11 | |
It saved hundreds of thousands of, not only Allied lives, | 0:01:11 | 0:01:15 | |
but German lives too. | 0:01:15 | 0:01:17 | |
So, in terms of the impact of a computer, | 0:01:17 | 0:01:21 | |
this machine is certainly at the very, very top. | 0:01:21 | 0:01:24 | |
Colossus was built to solve a colossal problem. | 0:01:24 | 0:01:27 | |
In World War II, that problem was the Nazis. | 0:01:27 | 0:01:30 | |
Hitler had overrun Europe and was communicating to his generals | 0:01:30 | 0:01:33 | |
using highly encrypted messages. | 0:01:33 | 0:01:36 | |
What we desperately needed to do was find out what he was saying. | 0:01:36 | 0:01:40 | |
You have to imagine a situation in the middle of a world war, | 0:01:40 | 0:01:44 | |
where you can hear your enemies sending critical messages, | 0:01:44 | 0:01:47 | |
but packaged in a cipher. | 0:01:47 | 0:01:50 | |
A cipher is a mathematical process, or a logical process, | 0:01:50 | 0:01:54 | |
that is used to disguise the characters in your message. | 0:01:54 | 0:01:58 | |
A bit like a padlock. | 0:01:58 | 0:02:00 | |
You can lock it up, and unlock it at the other end, | 0:02:00 | 0:02:03 | |
and the message is sent securely. | 0:02:03 | 0:02:06 | |
Even though all this was taking place over 60 years ago, | 0:02:06 | 0:02:10 | |
a lot of information is still exchanged in the same way now. | 0:02:10 | 0:02:14 | |
We live in an age of the transfer of data. | 0:02:14 | 0:02:17 | |
We buy things online. | 0:02:17 | 0:02:19 | |
We use mobile phones. | 0:02:19 | 0:02:21 | |
Data is being moved at all sorts of times. | 0:02:21 | 0:02:24 | |
Every time it's moved, it has to be moved securely. | 0:02:24 | 0:02:28 | |
So, today, we use ciphers to package our data | 0:02:28 | 0:02:32 | |
and to give it to somebody else securely. | 0:02:32 | 0:02:34 | |
And, I'm afraid, there are always people trying to break those ciphers. | 0:02:34 | 0:02:39 | |
So, the people who were able to break the German codes | 0:02:41 | 0:02:45 | |
were essentially 1940s-style hackers? | 0:02:45 | 0:02:48 | |
We could call the codebreakers one of the early hackers, yes. | 0:02:48 | 0:02:53 | |
Their job was to take apart somebody else's cipher, | 0:02:53 | 0:02:58 | |
and look inside, and gain an insight | 0:02:58 | 0:03:00 | |
to the information that they were transferring. | 0:03:00 | 0:03:04 | |
Unbelievably, at one stage, | 0:03:04 | 0:03:05 | |
all this was being done without the aid of a computer. | 0:03:05 | 0:03:08 | |
An incredible achievement, but also incredibly slow. | 0:03:08 | 0:03:13 | |
To spend weeks cracking a code with pen and paper | 0:03:13 | 0:03:16 | |
might mean you knew what your enemies were up to. | 0:03:16 | 0:03:18 | |
But by the time you found out | 0:03:18 | 0:03:20 | |
they'd already have carried out their plans. | 0:03:20 | 0:03:23 | |
However, by programming Colossus to do this task in hours, | 0:03:23 | 0:03:26 | |
the Allies managed to stay one step ahead of the game. | 0:03:26 | 0:03:30 | |
What they got so right about Colossus | 0:03:30 | 0:03:32 | |
was keeping their eye on the problem that they had to solve. | 0:03:32 | 0:03:37 | |
It's a key skill that anybody who has to program a computer today | 0:03:37 | 0:03:41 | |
still has to have. | 0:03:41 | 0:03:43 | |
And, by telling a computer to process a complex task, | 0:03:43 | 0:03:45 | |
we can increase the speed and accuracy a problem can be solved in. | 0:03:45 | 0:03:50 | |
So, even today, | 0:03:50 | 0:03:51 | |
we have to think about what the computer's being used for. | 0:03:51 | 0:03:56 | |
Do you need speed? Do you need accuracy? Do you need reliability? | 0:03:56 | 0:03:59 | |
Those factors all come together, to make the perfect machine. | 0:03:59 | 0:04:04 | |
Colossus needed all those things to efficiently break the Nazi codes, | 0:04:04 | 0:04:08 | |
and the people who built it, programmed it and used it, | 0:04:08 | 0:04:12 | |
would have been acutely aware of its precise purpose and abilities. | 0:04:12 | 0:04:15 | |
If we look at Colossus, we can see the data flow. | 0:04:15 | 0:04:20 | |
We can hear the mechanism. | 0:04:20 | 0:04:23 | |
We can understand, perhaps, using all our senses, | 0:04:23 | 0:04:26 | |
what this machine is doing. | 0:04:26 | 0:04:28 | |
But we have to remember that, as clever as they are, | 0:04:28 | 0:04:31 | |
they only reflect what we tell them to do. | 0:04:31 | 0:04:34 | |
We need someone who has an insight into solving the problem. | 0:04:34 | 0:04:38 | |
We need someone who can express that as a series of instructions. | 0:04:38 | 0:04:42 | |
And, ultimately, get the machine to reflect us as human beings. | 0:04:42 | 0:04:46 | |
And the programmers who can rise to this sort of challenge | 0:04:46 | 0:04:50 | |
in today's day and age | 0:04:50 | 0:04:51 | |
will be able to embrace infinite computing possibilities, | 0:04:51 | 0:04:55 | |
compared to the pioneers who built and first used Colossus. | 0:04:55 | 0:04:59 | |
The principle of the computer is actually flexibility, | 0:04:59 | 0:05:03 | |
and being general purpose. | 0:05:03 | 0:05:05 | |
Colossus isn't general purpose. | 0:05:05 | 0:05:07 | |
And maybe what we're doing is moving now to a world where | 0:05:07 | 0:05:10 | |
computers can be turned to do lots of different things, | 0:05:10 | 0:05:14 | |
simply by changing the instructions, the program, inside them. | 0:05:14 | 0:05:19 | |
That's what makes them unique in the world of machines. | 0:05:19 | 0:05:25 | |
And, with this kind of potential at our fingertips, | 0:05:25 | 0:05:28 | |
it's perhaps not so surprising just how many jobs | 0:05:28 | 0:05:30 | |
we're now programming computers to do on our behalf. | 0:05:30 | 0:05:34 | |
However, before we can really start to get to grips | 0:05:35 | 0:05:38 | |
with the amazing capabilities of these machines, | 0:05:38 | 0:05:41 | |
we first need to understand the machinery that we are working with. | 0:05:41 | 0:05:46 | |
Computers are not the most animate machines. | 0:05:46 | 0:05:49 | |
Looking at them, you'd barely believe their hardware was capable | 0:05:49 | 0:05:53 | |
of such wondrously diverse and clever tasks. | 0:05:53 | 0:05:55 | |
Their interiors, this hardware, is as unknown quantity to us | 0:05:55 | 0:05:59 | |
as, say, the inside of a car. | 0:05:59 | 0:06:03 | |
We rarely delve inside, but, delve inside, we must. | 0:06:03 | 0:06:05 | |
So let's lift the bonnet of your average games console, | 0:06:05 | 0:06:08 | |
to give us a sneak peek | 0:06:08 | 0:06:10 | |
at how this internal architecture works so hard in our favour. | 0:06:10 | 0:06:15 | |
The UK has the largest videogame industry in Europe. | 0:06:15 | 0:06:19 | |
Zoe Mode, based in Brighton, | 0:06:19 | 0:06:21 | |
specialise in leading dance titles, like Zumba. | 0:06:21 | 0:06:24 | |
And they are the first to admit the important role | 0:06:24 | 0:06:27 | |
that hardware plays in their day-to-day work. | 0:06:27 | 0:06:30 | |
The hardware we're making the game for | 0:06:30 | 0:06:32 | |
defines what the game can be, | 0:06:32 | 0:06:34 | |
not just in terms of the way it looks, | 0:06:34 | 0:06:37 | |
but also in terms of the way that you interact with the games. | 0:06:37 | 0:06:41 | |
All computational devices, from a laptop to a smartphone, | 0:06:41 | 0:06:45 | |
essentially rely on the following key hardware components to work: | 0:06:45 | 0:06:49 | |
So, if we are starting with a game, it has to be uploaded first, | 0:06:56 | 0:06:59 | |
and for us to do that, we need to utilise the memory. | 0:06:59 | 0:07:03 | |
On all computational devices, memory falls into two distinct areas. | 0:07:03 | 0:07:09 | |
ROM, or read-only memory, is permanently stored memory. | 0:07:09 | 0:07:13 | |
It's like a bookshelf with instructions that's always there. | 0:07:13 | 0:07:16 | |
When you turn on your console, | 0:07:16 | 0:07:19 | |
it's this internally stored ROM that helps start up the machine. | 0:07:19 | 0:07:23 | |
But it's also ROM you insert in order to play the game. | 0:07:23 | 0:07:27 | |
And this data is transferred from your DVD-ROM, into your RAM. | 0:07:27 | 0:07:32 | |
Things are stored in RAM, | 0:07:32 | 0:07:34 | |
in order to make it quicker for the CPU to take those things out. | 0:07:34 | 0:07:38 | |
So, for games consoles, | 0:07:38 | 0:07:40 | |
random access memory's job is to temporarily store programs and data | 0:07:40 | 0:07:44 | |
that the console uses to let you play the game. | 0:07:44 | 0:07:48 | |
In all computational devices, | 0:07:48 | 0:07:49 | |
random access memory is described as "volatile". | 0:07:49 | 0:07:53 | |
If you were in the middle of playing, and there was a power cut, | 0:07:53 | 0:07:56 | |
you'd have to start all over again. | 0:07:56 | 0:07:58 | |
Basically, the more memory you have, the better looking the game is, | 0:07:58 | 0:08:02 | |
the more sounds there are going to be, | 0:08:02 | 0:08:04 | |
the more inputs it can understand. | 0:08:04 | 0:08:06 | |
It's an important part of the computer's architecture. | 0:08:06 | 0:08:08 | |
And, with our game loaded up and ready to go, | 0:08:08 | 0:08:11 | |
we'll need to give it some instructions, or "input". | 0:08:11 | 0:08:14 | |
Inputs relay operating instructions back into the computer, | 0:08:14 | 0:08:18 | |
where they can be processed. | 0:08:18 | 0:08:20 | |
They can be numerous things. | 0:08:20 | 0:08:22 | |
A mouse, a keyboard, joysticks, joypads, touchscreens, | 0:08:22 | 0:08:25 | |
button presses on a teller machine. | 0:08:25 | 0:08:28 | |
Even your own voice. | 0:08:28 | 0:08:30 | |
With regards to this game, | 0:08:30 | 0:08:31 | |
the input is actually a sensor | 0:08:31 | 0:08:34 | |
which combines a web camera and an infrared camera, | 0:08:34 | 0:08:38 | |
which can do lots of clever things, like pick out human beings. | 0:08:38 | 0:08:43 | |
Your body is the input. | 0:08:43 | 0:08:46 | |
Then it makes a skeleton of that and tracks your movement, | 0:08:46 | 0:08:49 | |
which is a lot of information. | 0:08:49 | 0:08:51 | |
So, the CPU for this console needs to be very powerful | 0:08:51 | 0:08:54 | |
to deal with all that information, | 0:08:54 | 0:08:56 | |
to make gameplay interesting and fun. | 0:08:56 | 0:08:58 | |
The CPU, or central processing unit, | 0:08:58 | 0:09:00 | |
could be described as the brains of the operation. | 0:09:00 | 0:09:03 | |
It's the device within all computers that takes the input information, | 0:09:03 | 0:09:07 | |
decides what's gone on, | 0:09:07 | 0:09:09 | |
and processes a command using the memory, | 0:09:09 | 0:09:11 | |
to give you that all-important feedback. | 0:09:11 | 0:09:14 | |
You can only do one thing at a time. | 0:09:14 | 0:09:16 | |
It fetches an instruction, | 0:09:16 | 0:09:17 | |
it will execute that instruction, | 0:09:17 | 0:09:20 | |
and then it will move onto the next instruction and do that over again. | 0:09:20 | 0:09:24 | |
So, the faster your CPU, | 0:09:24 | 0:09:26 | |
the more instructions you can do within a period of time. | 0:09:26 | 0:09:29 | |
# Right hand in the air, left hand in the air... # | 0:09:29 | 0:09:32 | |
And the more instructions your CPU processes, for this game at least, | 0:09:32 | 0:09:36 | |
the more reactive and immersive your game experience will be. | 0:09:36 | 0:09:41 | |
Predominantly with Zumba, the output device that we are using is the TV. | 0:09:41 | 0:09:46 | |
The TV provides the visuals, | 0:09:46 | 0:09:49 | |
it provides feedback to you to let you know how you're doing. | 0:09:49 | 0:09:52 | |
It also provides the music for you to dance along to. | 0:09:52 | 0:09:56 | |
So the output device is really key | 0:09:56 | 0:09:58 | |
for your understanding of what's going on, | 0:09:58 | 0:10:00 | |
and your interaction with the game. | 0:10:00 | 0:10:02 | |
Output devices are perhaps the ones we notice the most. | 0:10:02 | 0:10:05 | |
They project the results of the computer processing to the user. | 0:10:05 | 0:10:09 | |
They can be speakers for sound output, | 0:10:09 | 0:10:11 | |
a monitor for visuals, | 0:10:11 | 0:10:13 | |
or, less important for a dancing game, a printer. | 0:10:13 | 0:10:15 | |
We're seeing technology developed now | 0:10:15 | 0:10:17 | |
where output is displayed on glasses, | 0:10:17 | 0:10:20 | |
or even displayed on contact lenses, so you can put things in your eyes | 0:10:20 | 0:10:25 | |
and see things in front | 0:10:25 | 0:10:27 | |
And that's a really exciting thing for games in the future. | 0:10:27 | 0:10:32 | |
The capabilities of digital electronic devices | 0:10:32 | 0:10:35 | |
are constantly increasing over time. | 0:10:35 | 0:10:38 | |
Massive increases in processing speed and memory capacity | 0:10:38 | 0:10:41 | |
now represent incredible potential in all areas of computing. | 0:10:41 | 0:10:46 | |
It's very important for us | 0:10:49 | 0:10:50 | |
to understand the capabilities of the machines, | 0:10:50 | 0:10:54 | |
of the console that the game is going to be played on, | 0:10:54 | 0:10:56 | |
because we wanted to know where our limits were, | 0:10:56 | 0:10:59 | |
but also how far we could push things. | 0:10:59 | 0:11:02 | |
But also, we need to know how the player's going to interact. | 0:11:02 | 0:11:06 | |
That's a really important thing, because that's where the game is, | 0:11:06 | 0:11:09 | |
that's where the fun is, the player experience is all about fun. | 0:11:09 | 0:11:12 | |
And that is driven by the hardware, | 0:11:12 | 0:11:14 | |
and the limitations of the hardware decides on what we can do | 0:11:14 | 0:11:18 | |
to entertain the player. | 0:11:18 | 0:11:21 | |
With the global games industry expected to keep on growing, | 0:11:21 | 0:11:24 | |
it's not hard to understand | 0:11:24 | 0:11:26 | |
just how important it is for the next generation of games designers | 0:11:26 | 0:11:30 | |
to appreciate how these incredible machines work. | 0:11:30 | 0:11:33 | |
Three stars?! | 0:11:38 | 0:11:40 | |
It's not just the games industry that's manipulating | 0:11:44 | 0:11:46 | |
the ever-increasing power of computers in our favour. | 0:11:46 | 0:11:50 | |
They are playing an ingenious role in space exploration too. | 0:11:50 | 0:11:54 | |
Mars, the final frontier. | 0:11:55 | 0:11:59 | |
Well, certainly our nearest frontier, | 0:11:59 | 0:12:01 | |
and even then we're talking of, at best, | 0:12:01 | 0:12:03 | |
35 million miles away of nearest frontier. | 0:12:03 | 0:12:06 | |
That said, it's still the best and most likely place | 0:12:06 | 0:12:09 | |
we'll be living on after Earth. | 0:12:09 | 0:12:11 | |
And huge efforts have, and are, being made | 0:12:11 | 0:12:14 | |
to land robots there to research that very possibility. | 0:12:14 | 0:12:19 | |
But, with no maps to work off, | 0:12:19 | 0:12:20 | |
and GPS a technology that's only available back on planet Earth, | 0:12:20 | 0:12:24 | |
it's proving very hard to tell these robots where they should be going. | 0:12:24 | 0:12:29 | |
It's a problem that's been engaging space engineers for many years. | 0:12:30 | 0:12:35 | |
Currently, NASA has at least two rovers on Mars. | 0:12:36 | 0:12:40 | |
But the real problem that those have is | 0:12:40 | 0:12:42 | |
that they don't think for themselves. | 0:12:42 | 0:12:43 | |
All the thinking is done by the scientists on Earth. | 0:12:43 | 0:12:46 | |
And it takes a very long time to send a command, | 0:12:46 | 0:12:49 | |
and then get a response back. | 0:12:49 | 0:12:52 | |
Opportunity has only gone 21 miles in eight years. | 0:12:52 | 0:12:57 | |
And Spirit has only gone five miles in eight years. | 0:12:57 | 0:13:02 | |
To travel around on Mars just is a very, very slow, painstaking process. | 0:13:02 | 0:13:09 | |
Meanwhile, back on Earth, at this lab just outside Oxford, | 0:13:09 | 0:13:12 | |
the UK is pioneering the idea that, one day, | 0:13:12 | 0:13:15 | |
robot explorers on Mars will be able to navigate themselves. | 0:13:15 | 0:13:20 | |
It's a natural progress of robotics and artificial intelligence. | 0:13:20 | 0:13:26 | |
Nowadays, we have better sensors, faster computers. | 0:13:26 | 0:13:29 | |
We have all the experience previous programmers have done. | 0:13:29 | 0:13:32 | |
We have all the experience from previous NASA missions. | 0:13:32 | 0:13:36 | |
I think we just needed all this background knowledge | 0:13:36 | 0:13:38 | |
and the availability of new cutting-edge hardware, | 0:13:38 | 0:13:41 | |
to be able to accomplish this. | 0:13:41 | 0:13:43 | |
Having this prior knowledge of where their robot was travelling to | 0:13:43 | 0:13:48 | |
has given programmers a valuable head start | 0:13:48 | 0:13:50 | |
in being able to design and execute a program that works. | 0:13:50 | 0:13:55 | |
The rover needs to find its way around, using local reference points. | 0:13:55 | 0:13:59 | |
And it does this, typically, | 0:13:59 | 0:14:02 | |
from looking at local features which are fairly close to it. | 0:14:02 | 0:14:06 | |
So, looking for rocks that can be recognised, | 0:14:06 | 0:14:08 | |
it goes from one rock to another, as it travels around. | 0:14:08 | 0:14:13 | |
Every time we encounter an obstacle, | 0:14:13 | 0:14:16 | |
or some zone which we haven't seen before, | 0:14:16 | 0:14:21 | |
which the computer is uncertain about, | 0:14:21 | 0:14:23 | |
it has to make a decision based on all the input it gets. | 0:14:23 | 0:14:26 | |
So it applies logic to the inputs, | 0:14:26 | 0:14:28 | |
and converts them into the appropriate outputs. | 0:14:28 | 0:14:31 | |
And the appropriate "if then, or else" logic commands | 0:14:31 | 0:14:35 | |
that a navigation system might execute, would simply be, | 0:14:35 | 0:14:37 | |
"stop, reverse, go right, go left, or continue forwards." | 0:14:37 | 0:14:42 | |
Depending, obviously, on what it finds in front of it. | 0:14:42 | 0:14:46 | |
By using a camera, | 0:14:46 | 0:14:48 | |
the machine is able to assess the hazards in front of it. | 0:14:48 | 0:14:52 | |
Choose a route around it, | 0:14:52 | 0:14:54 | |
and then get to where it knows it's got to go to, | 0:14:54 | 0:14:57 | |
just by using that simple information. | 0:14:57 | 0:15:00 | |
Its tasks are all logical task. | 0:15:00 | 0:15:03 | |
And, although this basic decision-making rings true, | 0:15:03 | 0:15:06 | |
new state-of-the-art visual sensors | 0:15:06 | 0:15:08 | |
are feeding back incredible detail to the computer. | 0:15:08 | 0:15:12 | |
Programmers need to be able to apply logic to all of this information. | 0:15:12 | 0:15:16 | |
I think the hardest part of this project | 0:15:16 | 0:15:20 | |
is just the huge number of inputs you get from a real environment. | 0:15:20 | 0:15:26 | |
Computer programs are really good at making logical decisions | 0:15:26 | 0:15:30 | |
when the number of inputs is limited. | 0:15:30 | 0:15:32 | |
But, when you're dealing with the real world, | 0:15:32 | 0:15:34 | |
you can encounter so many different obstacles, | 0:15:34 | 0:15:37 | |
you just never know what will be next. | 0:15:37 | 0:15:40 | |
So it's vitally important that the team of scientists here | 0:15:40 | 0:15:42 | |
work out programs that can deal with all that Mars might throw at them. | 0:15:42 | 0:15:47 | |
It's absolutely crucial the programming is right for this mission | 0:15:47 | 0:15:52 | |
because there's no chance of repairing it. | 0:15:52 | 0:15:54 | |
So if we allow the machine to make all the decisions on its own, | 0:15:54 | 0:15:58 | |
which is what we are trying to do, | 0:15:58 | 0:15:59 | |
then all those decisions have to be right | 0:15:59 | 0:16:02 | |
because, if it makes the wrong decision, | 0:16:02 | 0:16:04 | |
then it could be catastrophic, and we could lose the rover. | 0:16:04 | 0:16:07 | |
And, with NASA's latest Mars rover | 0:16:10 | 0:16:12 | |
planning to touch down in August 2012, | 0:16:12 | 0:16:16 | |
costing a staggering 2 billion, | 0:16:16 | 0:16:19 | |
it's not something you want to get wrong. | 0:16:19 | 0:16:22 | |
This is a massive leap forward, compared to previous technology. | 0:16:22 | 0:16:26 | |
This would enable Mars rovers to go much further, | 0:16:26 | 0:16:30 | |
and get to targets which are in parts of Mars | 0:16:30 | 0:16:34 | |
where landing would be impossible. | 0:16:34 | 0:16:36 | |
And, although navigating around Mars is, without doubt, | 0:16:36 | 0:16:40 | |
the prime objective, the team designing this navigation system | 0:16:40 | 0:16:43 | |
are more than aware of potential applications back here on Earth. | 0:16:43 | 0:16:47 | |
This work has never been done by anybody looking for | 0:16:47 | 0:16:51 | |
sufficient brains in a machine, that it can travel itself | 0:16:51 | 0:16:55 | |
for six kilometres without any intervention from the ground. | 0:16:55 | 0:17:00 | |
The technology which we are developing here can be | 0:17:00 | 0:17:02 | |
directly used on Earth, | 0:17:02 | 0:17:04 | |
in places, for example, where GPS is not available. | 0:17:04 | 0:17:08 | |
Computing devices all operate | 0:17:08 | 0:17:11 | |
entirely on instructions prepared by someone who has done the thinking, | 0:17:11 | 0:17:15 | |
and reduced the problem to a point | 0:17:15 | 0:17:17 | |
where logical decisions can deliver the correct answer. | 0:17:17 | 0:17:20 | |
How you can use this programming know-how | 0:17:20 | 0:17:23 | |
depends on what problem you want to solve. | 0:17:23 | 0:17:27 | |
The link between computing and robotics | 0:17:27 | 0:17:30 | |
is one many would expect to encounter. | 0:17:30 | 0:17:32 | |
However, computers increasingly play a vital role | 0:17:32 | 0:17:35 | |
in our very own health and well-being. | 0:17:35 | 0:17:37 | |
In recent years, biologists have made | 0:17:40 | 0:17:42 | |
a massive scientific breakthrough in discovering, for the first time, | 0:17:42 | 0:17:45 | |
how to sequence species' entire genetic code. | 0:17:45 | 0:17:49 | |
We can see DNA, you can see it by eye. | 0:17:49 | 0:17:52 | |
You can't decode it, you can't understand what it's saying. | 0:17:52 | 0:17:56 | |
It's only by sequencing that DNA, you can decode it, | 0:17:56 | 0:17:59 | |
you can understand the messages, the sequences, | 0:17:59 | 0:18:01 | |
the genes that are encoded by that DNA. | 0:18:01 | 0:18:05 | |
What sequencing a species' entire DNA into a genome means | 0:18:05 | 0:18:09 | |
is that scientists are effectively now capable | 0:18:09 | 0:18:12 | |
of examining the building blocks of life. | 0:18:12 | 0:18:15 | |
Sequencing gives you a glimpse, | 0:18:15 | 0:18:17 | |
it's actually a very privileged view | 0:18:17 | 0:18:20 | |
of the instruction manual for an organism. | 0:18:20 | 0:18:25 | |
To give you some idea of the challenge computer programmers face, | 0:18:25 | 0:18:29 | |
a single genome can be composed of over three billion base pairs of data. | 0:18:29 | 0:18:34 | |
In computing terms, | 0:18:34 | 0:18:35 | |
that would equate to over four gigabytes of storage space. | 0:18:35 | 0:18:38 | |
Or, if you were to read it out loud, | 0:18:38 | 0:18:41 | |
it would take you over nine years of continuous reading. | 0:18:41 | 0:18:45 | |
Using the current sequencing technologies, | 0:18:47 | 0:18:49 | |
we generated, in six months, | 0:18:49 | 0:18:51 | |
more data than we produced in the previous 10 years. | 0:18:51 | 0:18:55 | |
This tells you the pace, the development of both sequencing, | 0:18:55 | 0:18:58 | |
and it also gives you an idea of how computing has had to keep up | 0:18:58 | 0:19:02 | |
with that pace of sequencing. | 0:19:02 | 0:19:04 | |
But to explore all the possibilities that reading this data gives them, | 0:19:04 | 0:19:08 | |
scientists have needed to be able to compare thousands and thousands | 0:19:08 | 0:19:12 | |
of separate individual codes. | 0:19:12 | 0:19:14 | |
That's so much data, it's making my head hurt just thinking about it! | 0:19:14 | 0:19:20 | |
Both the processing power, | 0:19:20 | 0:19:21 | |
and the memory for storing all the data we are generating, | 0:19:21 | 0:19:24 | |
is absolutely huge. | 0:19:24 | 0:19:26 | |
And, without that, | 0:19:26 | 0:19:27 | |
we wouldn't even be able to entertain many of the projects | 0:19:27 | 0:19:31 | |
that we are now initiating. | 0:19:31 | 0:19:33 | |
We wouldn't even be to start those projects. | 0:19:33 | 0:19:35 | |
So, whilst it's biologists | 0:19:35 | 0:19:37 | |
who initially have been able to create the data, | 0:19:37 | 0:19:40 | |
it's people with the knowledge and understanding | 0:19:40 | 0:19:43 | |
of how to manipulate some serious computer hardware | 0:19:43 | 0:19:45 | |
that has allowed them to analyse it efficiently, and effectively. | 0:19:45 | 0:19:51 | |
Here, you are looking at over 16 petabytes of storage. | 0:19:52 | 0:19:56 | |
That's 16 million gigabytes. | 0:19:56 | 0:19:59 | |
Or the equivalent of 250,000 top-of-range iPads. | 0:19:59 | 0:20:04 | |
Or, if you like, over a million of your average smartphones. | 0:20:04 | 0:20:08 | |
Through programming rigorous data architecture within these systems, | 0:20:11 | 0:20:15 | |
computer technicians are able to store data, move data, | 0:20:15 | 0:20:20 | |
test data, analyse and display data, | 0:20:20 | 0:20:22 | |
in ever more ingenious ways. | 0:20:22 | 0:20:25 | |
Effectively handing biologists specialist computing tools | 0:20:25 | 0:20:28 | |
to extract maximum value from this most precious resource. | 0:20:28 | 0:20:33 | |
(COMPUTERISED VOICE): Now, some impressive genetic detective work | 0:20:34 | 0:20:37 | |
has analysed the most recent seven cholera pandemics, | 0:20:37 | 0:20:40 | |
and traced all of them back to the same source. | 0:20:40 | 0:20:44 | |
We looked into cholera, | 0:20:44 | 0:20:46 | |
because we wanted to understand where it originated from. | 0:20:46 | 0:20:50 | |
It is important to know the source population for cholera because, | 0:20:50 | 0:20:53 | |
once you know the source, you have the possibility | 0:20:53 | 0:20:55 | |
of controlling and preventing its spread from that source. | 0:20:55 | 0:20:59 | |
Even bacteria like cholera have DNA that is traceable | 0:20:59 | 0:21:03 | |
from one generation to the next. | 0:21:03 | 0:21:05 | |
In being able to compare cholera bacteria from all over the world, | 0:21:05 | 0:21:08 | |
Nick had valuable insight in tracing where it was originating from. | 0:21:08 | 0:21:13 | |
Easier than it sounds. | 0:21:13 | 0:21:17 | |
Just the sequence of one genome is five million base pairs for cholera. | 0:21:17 | 0:21:20 | |
Within those five million base pairs of DNA sequence, | 0:21:20 | 0:21:24 | |
we're looking, maybe even for just one base pair change | 0:21:24 | 0:21:27 | |
between two different bacteria. | 0:21:27 | 0:21:29 | |
When you have 100, 150 genomes, | 0:21:29 | 0:21:32 | |
it becomes an unimaginable amount of sequence data. | 0:21:32 | 0:21:36 | |
It's quite literally like looking for a needle in a haystack. | 0:21:36 | 0:21:39 | |
But, because of programmers' clever manipulation of these immense datasets, | 0:21:39 | 0:21:43 | |
it's been possible to find that needle. | 0:21:43 | 0:21:47 | |
By looking at huge amounts of sequence data, | 0:21:47 | 0:21:49 | |
we could understand how they related to each other. | 0:21:49 | 0:21:52 | |
And, by understanding, just like a family tree, | 0:21:52 | 0:21:55 | |
understanding how they relate to each other, | 0:21:55 | 0:21:58 | |
we can actually find out where they were originating from. | 0:21:58 | 0:22:01 | |
And, whilst cholera and other diseases | 0:22:01 | 0:22:03 | |
are not going to vanish off the face of the earth overnight, | 0:22:03 | 0:22:06 | |
computing has allowed us to plan ways of starting to make it happen. | 0:22:06 | 0:22:11 | |
Computing is effectively central to everything we do, | 0:22:11 | 0:22:14 | |
when you are trying to answer large questions relating to human health, | 0:22:14 | 0:22:18 | |
rely on large computing facilities and processing power. | 0:22:18 | 0:22:23 | |
And, with the kind of technologies that biologists will use | 0:22:23 | 0:22:27 | |
likely to improve over time, | 0:22:27 | 0:22:29 | |
the opportunities available for the next generation of computer scientists are enormous. | 0:22:29 | 0:22:34 | |
In biological research, | 0:22:34 | 0:22:36 | |
I think, mostly, computers are taken from granted. | 0:22:36 | 0:22:39 | |
We assume there will be processing power, | 0:22:39 | 0:22:42 | |
we assume there will be memory. | 0:22:42 | 0:22:44 | |
We assume there will be good people | 0:22:44 | 0:22:45 | |
that can actually write programs to analyse the data. | 0:22:45 | 0:22:49 | |
But that's an unsafe assumption. | 0:22:49 | 0:22:51 | |
From a biological point of view, you need those people. | 0:22:51 | 0:22:54 | |
Because, as I say, you are asking really key questions | 0:22:54 | 0:22:58 | |
that have fundamental importance to human health. | 0:22:58 | 0:23:00 | |
So, programming a computer could allow you to play a role | 0:23:00 | 0:23:04 | |
in fighting killer diseases as diverse as cystic fibrosis, cancer, | 0:23:04 | 0:23:08 | |
and even alcoholism. | 0:23:08 | 0:23:11 | |
Computers, without doubt, have a huge impact on the real world. | 0:23:11 | 0:23:15 | |
But they are proving just as key | 0:23:15 | 0:23:17 | |
in merging the worlds of reality and make-believe. | 0:23:17 | 0:23:20 | |
Once you understand the fundamentals of how a computer works, | 0:23:23 | 0:23:26 | |
and the language that it speaks, | 0:23:26 | 0:23:28 | |
then you can challenge it to do a wider range of tasks. | 0:23:28 | 0:23:32 | |
However, programming a computer to carry out what you want it to do | 0:23:32 | 0:23:34 | |
is as much about being able to make sense of the world around us, | 0:23:34 | 0:23:38 | |
as it is about understanding the computers themselves. | 0:23:38 | 0:23:41 | |
To be able to impress a world-class director is an a difficult thing. | 0:23:41 | 0:23:44 | |
And that's what we have to do on a regular basis. | 0:23:44 | 0:23:47 | |
We are there, from the word go, during film production, | 0:23:47 | 0:23:51 | |
and all through post-production, throughout the edit. | 0:23:51 | 0:23:54 | |
Visual effects is entirely integrated into film production now. | 0:23:54 | 0:23:58 | |
Visual effects is a really central part of the film-making process now. | 0:23:58 | 0:24:03 | |
Every artist will be designing something, | 0:24:03 | 0:24:05 | |
or deciding how a shot is composed, how it breaks down. | 0:24:05 | 0:24:10 | |
Those decisions make it up on to the big screen. | 0:24:10 | 0:24:12 | |
We are as central a part of the film-making process as the actors, | 0:24:12 | 0:24:16 | |
the set designers, the costume designers. | 0:24:16 | 0:24:20 | |
And, with the ever advancing power of computers, | 0:24:20 | 0:24:23 | |
visual effects companies, like DNeg, | 0:24:23 | 0:24:25 | |
are capable of creating ever more complex animations. | 0:24:25 | 0:24:28 | |
Imagination simply knows no bounds. | 0:24:28 | 0:24:31 | |
My job, as effects supervisor, | 0:24:31 | 0:24:33 | |
is to take any part of a film that needs to be computer-generated, | 0:24:33 | 0:24:36 | |
and that moves under the forces of physics. | 0:24:36 | 0:24:39 | |
So, things like natural phenomena, water, fire. | 0:24:39 | 0:24:43 | |
Anything that follows the rules of nature. | 0:24:43 | 0:24:46 | |
Newton's laws of motion. | 0:24:46 | 0:24:48 | |
It's something that we can definitely represent inside the computer. | 0:24:48 | 0:24:52 | |
We take things in the real world, and you make a computer program. | 0:24:52 | 0:24:55 | |
That's was a programmer will do. | 0:24:55 | 0:24:56 | |
The person is the intermediary that writes the software. | 0:24:56 | 0:25:00 | |
The software dictates what the computer does. | 0:25:00 | 0:25:03 | |
But the software is a set of rules derived from the real world. | 0:25:03 | 0:25:06 | |
Computer modelling, or simulation, | 0:25:06 | 0:25:08 | |
underpins much of their initial work. | 0:25:08 | 0:25:11 | |
We start by looking at the effect that we want to achieve. | 0:25:11 | 0:25:14 | |
So, in the case of something like fire, | 0:25:14 | 0:25:16 | |
we look at how it behaves in the real world. | 0:25:16 | 0:25:18 | |
What are the forces of physics that define how fire moves? | 0:25:18 | 0:25:23 | |
What's the chemistry of what's going on inside the flame? | 0:25:23 | 0:25:28 | |
And we would break it down to the maths that dictates that motion. | 0:25:28 | 0:25:31 | |
And then we return that maths into a list of exact instructions | 0:25:31 | 0:25:35 | |
that the computer needs to follow. | 0:25:35 | 0:25:37 | |
That process is called abstraction. | 0:25:37 | 0:25:39 | |
Abstraction essentially tries to reduce and factor out irrelevant details, | 0:25:39 | 0:25:44 | |
so that the programmer can focus on a few concepts at a time. | 0:25:44 | 0:25:47 | |
It's really important to shoot reference wherever possible. | 0:25:47 | 0:25:50 | |
It's very easy to think, I know what lightning looks like. | 0:25:50 | 0:25:53 | |
But you probably don't. | 0:25:53 | 0:25:55 | |
If I said, how long exactly in seconds does it exist? You wouldn't know. | 0:25:55 | 0:26:01 | |
So, to program a computer how to replicate lightning, | 0:26:01 | 0:26:04 | |
we need to understand lightning ourselves. | 0:26:04 | 0:26:07 | |
We'd look at that reference footage, and work out what we want to inherit | 0:26:07 | 0:26:10 | |
and what we'd don't want to inherit from real life. | 0:26:10 | 0:26:12 | |
We would then program those rules into the computer, | 0:26:12 | 0:26:16 | |
and encourage it to follow them over and over again. | 0:26:16 | 0:26:19 | |
And usually we will have to adapt it, rework it. | 0:26:19 | 0:26:22 | |
And shape it to where we need it to be to do the effects work. | 0:26:22 | 0:26:26 | |
But, even then, we are by no means dealing with the finished article. | 0:26:26 | 0:26:30 | |
Directors can be a picky bunch, | 0:26:30 | 0:26:33 | |
and changes will often be required. | 0:26:33 | 0:26:36 | |
There will be notes about the movement of the flames. | 0:26:36 | 0:26:39 | |
They will say, "Can flames be more chaotic at this point? | 0:26:39 | 0:26:41 | |
"Can the flames be calm at this point?" | 0:26:41 | 0:26:43 | |
So you're always balancing a line between stylised and natural. | 0:26:43 | 0:26:48 | |
Say, if you wanted to have the flames more swirly, | 0:26:48 | 0:26:52 | |
more curled shapes rather than sharp shapes, | 0:26:52 | 0:26:55 | |
you might go into the algorithm, | 0:26:55 | 0:26:57 | |
look at where rotational shapes have been introduced, and exaggerate them. | 0:26:57 | 0:27:02 | |
It's not something that would happen in real life, | 0:27:02 | 0:27:05 | |
but it could change the character of the flames. | 0:27:05 | 0:27:07 | |
So, writing a program's one thing. | 0:27:07 | 0:27:10 | |
But being able to go back into a pre-written program, | 0:27:10 | 0:27:12 | |
and adjust it to your requirements has many advantages. | 0:27:12 | 0:27:15 | |
Reinventing the wheel is incredibly wasteful. | 0:27:15 | 0:27:18 | |
Once you have software which is able to create flames and modify flames, | 0:27:18 | 0:27:22 | |
you should use that to your best advantage. | 0:27:22 | 0:27:25 | |
I think debugging skills, reasoning skills, testing skills, | 0:27:25 | 0:27:30 | |
are even more important than writing software in the beginning. | 0:27:30 | 0:27:34 | |
It's great to have people who can pick those tools up, | 0:27:34 | 0:27:37 | |
try using them, say, "This works but this didn't. | 0:27:37 | 0:27:41 | |
"This creatively hinders me. | 0:27:41 | 0:27:43 | |
"This part could be much more useful if it worked like this." | 0:27:43 | 0:27:46 | |
And can feed that stuff back, so we end up writing the best software. | 0:27:46 | 0:27:50 | |
It's not just about writing accurate software. | 0:27:50 | 0:27:53 | |
You need to be creative enough | 0:27:53 | 0:27:55 | |
to manipulate it in the director's favour. | 0:27:55 | 0:27:58 | |
It's important we try to represent real-life in our simulations. | 0:27:58 | 0:28:02 | |
But it's also important that we do things that help the story, | 0:28:02 | 0:28:05 | |
that help the drama. | 0:28:05 | 0:28:06 | |
It's really important everyone in the visual effects industry | 0:28:06 | 0:28:09 | |
has a technical side to them and an artistic side to them. | 0:28:09 | 0:28:11 | |
So, if you have any doubts about the creative possibilities | 0:28:11 | 0:28:16 | |
opened up by a career in computer programming, | 0:28:16 | 0:28:18 | |
simply look to the stars. | 0:28:18 | 0:28:21 | |
I think it's absolutely vital | 0:28:21 | 0:28:23 | |
that kids today learn all these concepts, | 0:28:23 | 0:28:26 | |
get really into computer programming, understand the logic behind it. | 0:28:26 | 0:28:29 | |
It's really fantastic, | 0:28:29 | 0:28:30 | |
And when you get to use it | 0:28:30 | 0:28:32 | |
you can do absolutely amazing stuff, like the work that we do. | 0:28:32 | 0:28:35 | |
Subtitles by Red Bee Media Ltd | 0:28:42 | 0:28:45 |