3DTV in your face

You can’t get away from it. 2010 has been billed as being the year of 3D. This certainly seems to continue to be the case at the CES show in Las Vegas this week with lots of blog posts about 3DTV already popping up. Obviously there have been many more 3D films in the cinema recently, in particular the major blockbuster Avatar, and consumers are eager to get this technology into their living room.

But what exactly are we talking about when we say 3D? It’s not to be confused with computer generated models such as in the original Toy Story animation. These are often referred to as 3D models or CGI, but when you view them on a screen you still see a flat image, because it’s a flat screen. Both of your eyes are seeing the same image and therefore there is no difference in depth. What we are talking about now is officially known as Stereoscopic 3D.

Stereoscopics, or Stereo 3D, has been around for many many years, starting with people creating stereo 3D photographs taken with two cameras a few inches apart and then placing the photos in a special viewer box like binoculars which forces each eye to look at it’s respective photo. The brain does the rest. Things have moved on from there though and we now have about 4 main viewing styles. A quick synopsis…

Anaglyph glasses (red and blue)
This is what many people still think of when they think of 3D video. Not really very impressive and a bit cheap. Channel 4 in the UK got people excited when they announced a 3D week on their channel recently, only for people to then discover it was just going to be with those annoying red/blue images again. This technology has been around for a while and was very popular in the 50s and 60s. Cheap glasses but you lose a lot of the colour information of the image. This is only used on current monitors as it is the only method most of them can handle.

Polarised glasses (passive viewing)
This is what you get when you go to the cinema these days. Two projectors project an image each onto the screen but the light from each is polarised in a different way. The glasses are lightweight and inexpensive. Each lens blocks out one of the polarised lightwaves and therefore each eye gets the correct image. One slight drawback with this method is that the image is in essence ‘interlaced’ which means that each eye only gets half the number of lines in the image. HD will not be ‘full’ HD therefore. The quality difference, however, is negligible.

Shutter glasses (active viewing)
This method has no quality loss at all. Each eye’s full res image is played alternately at very quick succession. The glasses, however, are bulkier and more expensive as they need to synchronise with the projector/monitor so that it blocks off each eye at the same rate as the images are being displayed. This method can be used with current monitors that have a high enough refresh rate of 120hz (only a small number of expensive computer monitors). A synch unit is also required to make them work.

No glasses (autostereoscopic / lenticular)
This is the latest breakthrough technology and there are already a few models available which use tiny lenses on the monitor itself to split the image before it hits your eyes. It’s very complicated to make and in terms of viewing it’s great not having to wear glasses but you do have to be in a sweet spot to view it correctly and not move too much. This is definitely the future though and as the technology improves and the price comes down I think we’ll see these models become mainstream within 5 years.

  • I have used anaglyph since I was a young child and never really been that impressed with it
  • I made stereo 3D videos and computer game levels with the shutter glasses at University about 9 years ago and while it was good quality I still found the glasses gave me headaches. There was also often hardware issues with the glasses not synching up correctly
  • I recently went to Inition in London and received a demo of the Autostereoscopic lenticular method which I had never seen before. It is a very strange sensation to see 3D images without wearing glasses
  • The polarised glasses are great and don’t induce nausea. For me these are the winners right now

So how do I view 3D?
If you have a great computer monitor and are happy installing new software and drivers then you can use something like Nvidia’s GeForce 3D Stereo kit. Otherwise, and if you want to view it in the luxury of your living room the bottom line is that you will have to buy a new TV.

A new TV?
Yes. I’m afraid so. You probably just bought a new HDTV too didn’t you? The good news is that after CES there should be a larger number of manufacturers bringing out 3DTVs so hopefully a price war might begin. Further good news is that existing set top boxes (such as the Sky+HD box) will be able to handle the 3D signal. Playstation3s will also be able to show them after downloading a firmware update. TV manufacturers are, however, split over which technology to go with. Some are launching shutter glasses systems while others favour the polarised glasses. The good news is that content which is produced in 3D will be compatible with all the hardware.

But is there any content to watch?
Not a huge amount yet but there has been some great news recently about organisations working on new content. Sky in Europe will be launching a 3D channel this year, primarily filming live events in 3D such as sport and concerts. ESPN has followed suit by announcing their 3D plans. The big news at CES yesterday was that IMAX will be linking up with Sony and Discovery to produce 3D content later this year.

Will everything be produced in 3D?
This is debatable. My recent review of Avatar highlighted how action movies maybe aren’t the best exponents of the 3D technique. The fast cuts and changes in camera angles don’t go well with 3D as the brain can struggle to keep working out the new depths on screen. Better content is things like sport, live events and science documentaries. This also seems to be the way things are going with the recent Sky, ESPN and IMAX/Discovery announcements.

Will it become mainstream?
Everywhere we look just now we are seeing 3D. YouTube has also just recently launched a 3D element to its videos. As I said above I don’t think every form of video is suited to 3D. Others are perfectly suited. As directors and producers gain more experience in what shooting styles work, we should see a better and more focussed range of content.

What will be the killer use?
For me the killer app will be games. When you interact with a 3D image it makes it much more ‘believeable’ as your brain almost moves with it. It works. You control it as it moves and you feel a part of it. You also control the speed of the movement which is crucial. This will also link in with the recent swathe of Augmented Reality apps we are seeing and incredible games such as EyePet.

What’s next?
Viewing without glasses is the next big thing for 3DTVs but in terms of the user experience watch out for head-tracking and haptic interfaces. They let you move round an object and get a new perspective on it as well as touching it with force feedback. Industry and Medicine will lead the way with this but it’s only a matter of time until we see that in our living rooms too.

Avatar

After all the hype about the stereoscopic 3D effects in James Cameron’s Avatar movie it was surprising how that wasn’t the thing that impressed me the most about the film when I saw it this week. It was the character animation and totally seamless blending of real actors with CGI that blew me away.

PERFORMANCE CAPTURE

CGI characters have historically been shiny round things ever since Luxo Jr and Toy Story many moons ago. Then, when hair was possible, we had Monsters Inc to show off that effect. Polar Express attempted lifelike characters but produced almost expressionless mannequins. Final Fantasy looked great but it was still obvious you were watching CGI. And let’s not even talk about Jar Jar Binks in Star Wars Episode I. What James Cameron and Weta Digital have achieved in Avatar is a realism never seen before. Ok, the aliens are still blue (affectionately referred to by many as Thundersmurfs) but human skin tones are still notoriously difficult to render. It’s the motion capture which is the ground breaking element in this film. It’s even now referred to as ‘performance capture’ as it truly does capture the actors’ performances, right down to their eyes. It’s very strange watching Sigourney Weaver’s Avatar when you totally see her mannerisms and facial expressions in the animation. You really will forget, for much of the film, that you are watching CGI characters.

People have said that this means the end of actors, and that we can keep actors appearing on screen in new films after they die because we will have their facial structure and expressions saved on computer. I have to say I’m really not sure about that to be honest. What about their voices? That’s probably about 70 percent of the acting. And the reason the CGI characters work in Avatar is because of the performance capture from the real actors, not just because of any master animators.

3D
Now we come to the 3D. This was perhaps the most hyped part of the film and it really is fantastic. But to be honest, as a big fan of stereoscopic 3D and someone who has seen many 3D movies, it didn’t seem groundbreaking. Don’t get me wrong, it was incredible in places. Truly amazing and enthralling. But in other places it just almost got in the way. I absolutely love stereoscopic 3D, but there’s a time and a place. Some bits of the film like the floating mountain wideshots or the walking on the cliffside were greatly enhanced by the 3D. Other bits, however, were just obvious tricks to try to show off the 3D effect. Man sticking gun into screen and camera rotating round gun. Man putting golf ball and camera panning down to be almost hit by ball. These shots were just too obvious and only succeeded in reminding you that it was a movie that was trying to show off 3D. 3D is truly awesome for documentaries, music events and sport but I’m just not yet convinced its totally suitable for action films. Even James Cameron, with all his directorial experience and 3D research couldn’t get away from some basic 3D principles…

  • Every time you cut to another camera or scene, your brain needs to work out new depth perception. So when you keep cutting extremely quickly between camera angles it can be hard to keep up. It’s unnatural, and the whole point of 3D is to make the film more natural
  • When there are lots of things happening on screen, like in a dogfight, this gets even worse. Often it’s good to tone down the 3D effect for these sequences. I think this had to be done a few times in the film
  • If you’re protruding things out of the screen you need to be careful they don’t cross the border of the frame, otherwise the illusion is lost. Cameron for this reason has lots of sequences where the depth into the screen is highlighted (positive parallax), such as the view over the edge of the cliff, rather than objects protruding out of the screen (negative parallax). This works really effectively when done well
  • It’s best not to have subtitles if you can help it. It reminds the viewer there is a plane between them and the action
  • Focus-pulling really confuses the brain. This is one thing which let’s 3D down. If there is more than one thing happening on screen your brain assumes that it can focus your eyes on whichever one you look at as you do in real life. Obviously you can’t as it is up to the director and cameraman to control the focus and so once again you realise you’re watching a film and the 3D illusion is lost. Unfortunately James Cameron started the movie with precisely this faux pas in the first 20 seconds.

The 3D really does help to bring you into the film but I’m not convinced it makes the film a ‘game-changer’ in that regard. It doesn’t convince me that every film should be in 3D yet. There are too many things to have to be aware of which means that not every film will translate well to 3D and not every director will be willing to accommodate it. But it certainly does show that 3D is a stunning effect when used in the appropriate places. As mentioned earlier, however, this may be best for documentaries and sporting events.

VIRTUAL WORLD

One of the things that intrigued me most about this film was the remote avatar dynamic. It’s the thing after which the movie was named after all. It was also the only thing to really set the story apart from just being another remake of Dances with Wolves or Pocahontas. And yet for me it was the area that was least touched upon. The main character was paralyzed from the waist down but when his mind went into the body of the physical avatar he could leap around like an olympic athlete. He learns skills, makes friends, builds relationships in this other world. It’s not quite a virtual world as it does actually exist but his mind is merely projected remotely into the physical avatar. His ‘second life’ is completely dependant on the technoloy back in the base and if the plug is pulled on the technology link his physical avatar drops down lifeless. When his avatar and the rest of his new alien friends sleep he wakes in the real world and debriefs his superiors. But this was all just touched upon lightly as if it was a totally normal situation. So much more could have been made of the split personality and double life that this caused. How he became almost addicted to this other world, which gave him freedom in so many ways he couldn’t get in his real world. How it alienated him from his fellow humans.

This was one of the most interesting things for me from the film and I don’t feel it was fully addressed. In this day and age of people spending more and more of their lives in virtual worlds and social networks pretending to be someone they’re not I felt that this could have been a much bigger moral input to the plot. Maybe he’s saving some of that for the sequel. In stunning 4D Smell-o-vision I’m sure. This is a James Cameron film so you are just along for the ride. And what a ride it is.