Technology Overview
NVIDIA has been working on stereoscopic 3D gaming for several years but now is the official release of retail product that you can buy NOW! Come in and see how we reate the new technology offering from NVIDIA and see how it could change the way you game forever.
|
|
3D stereoscopic technologies have been around for ages – I am sure you remember going to see a movie and wearing those red and blue paper glasses at a drive-in at some point in your life. Theme parks have used 3D movies and attractions for quite some time as well though with little advancement over previous implementations. In recent years we have seen a resurgence of 3D effects in younger audience movies like Spy Kids and Hanna Montana and while the technology used in those movies was better (as we didn’t have the colored hues to everything) they were still less than fantastic to say the least.
At last fall’s Intel Developer Forum, Intel and DreamWorks announced a partnership bringing 3D to all future DreamWorks animated films and putting 3D back in the spot light once again. If you have been watching the NFL playoffs here in the US over the past week you have likely seen the promotion that instructs viewers to pick up some free 3D glasses at retailers in order to view some 3D content during football half time shows.
NVIDIA has its own take on 3D technology, but unlike Intel the GeForce technology focuses on gaming and uses a much higher quality viewing solution. Unfortunately that also means it is much more expensive – you won’t find anyone giving away NVIDIA’s 3D Vision glasses anywhere. But what I expected to be a mostly gimmicky attempt to get some more PC gamers’ money has impressed me beyond my expectations.
Passive and Active 3D Solutions
The idea of creating stereoscopic 3D image using a two dimensional display media like a TV or movie screen has been around for years, as I mentioned above, and the theory behind all 3D methods are basically the same. To produce the effect correctly, each eye must be presented with a slight variation of the same image in order to simulate depth perception. The same process is used when your brain processes each image from your eyes to help you perceive depth out of a pair of flat two dimensional images.
Some slightly less cool active 3D shutter glasses
The original “red/blue” method of producing a stereo 3D image is called “anaglyph” and is able to give each eye its own image by superimposing each frame of a video with a different color hue. The viewer then wears correctly tinted glasses that filter out every other frame so that each is presented with a unique view. Unfortunately the dramatic tinting of the image is nearly impossible for the brain to remove completely and thus seeing any kind of high quality content in this format was not acceptable.
The most recent and popular stereo 3D format uses polarization in much the same way the anaglyph works. Instead of using colored hues to filter the image for each eye, in this method the user is wearing mostly-translucent glasses that have polarization filters on each side, slightly varied in a way to allow each eye to see a unique image on the display. When a frame with the left eye polarization is displayed then the left lens on the glasses effectively blocks the image from coming through, leaving the right eye to see the image on the screen. The downside to this type of stereo 3D method is that issues can arise based on the users head position and orientation to the screen – the polarization filters are less effective at varying angles.
Both of these methods are considered passive stereo 3D effects and require that the content be precompiled for 3D. This precompile process is necessary to take the video from its original 60 Hz (or 24 Hz for standard film) to a split frame rate, half of the frames rendered for the left eye and half for the right eye. This could be done using a pair of cameras (if filming live action) that are a set distance apart or by re-rendering the video (if animated with computers). If the content was built with 3D technologies from the start, producing the stereo effect is impossible.
The most effective, and most expensive, method for producing stereo 3D image is to use active glasses that “turn off” each eye in time with a display’s refresh rate. Each lens in the glasses is effectively a 1 pixel LCD screen that is turned off (black) or on (clear) depending on which eye should see the image displayed on the screen.
In this image the right lens is essentially blacked out blocking the right eye from seeing the screen leaving the left eye open.
The key to an active glasses solution is that the lenses must be properly synchronized with the display requiring some form of communication between the two devices. When the left eye frame is shown on the screen then the left lens must be completely open and during the time the screen refreshes with the next frame for the right eye the glasses must also switch lens opacity. This communication could be done with a cable of some kind or using infrared signals, as the NVIDIA solution we are looking at today uses.
Active glasses for stereo 3D viewing are not a technology created by NVIDIA and in fact they have been around for some time as well. However, the quality of the glasses and the user experience has been low due to low frame rates (30 Hz to each eye usually) and bulky hardware. The GeForce 3D Vision product plans on breaking that mold by using 120 Hz displays (60 Hz to each eye) and much improved hardware.
Display Options for NVIDIA GeForce 3D Vision
Because NVIDIA is aiming for quality much higher than that of other stereo 3D solutions display options are going to be somewhat limited out of the gate. As of today, NVIDIA’s 3D technology will work with only two types of displays: true 120 Hz LCD monitors and 3D-Ready DLP projection televisions.
The 120 Hz LCD panel display option is the easiest to understand – if you have 120 frames per second of video then you can display 60 for each eye per second on the monitor. This produces an effect that is much smoother than other 3D technologies that used 60 Hz panels but does limit buyers to one of only a pair of certified displays: one from Samsung and the other from Viewsonic. Both of these monitors are 22” in size, reasonable even for extreme enthusiast gamers, but some may have become used to 24” and above (or at least the option of) displays. Because of the large amount of data that must be processed as the resolution increases larger panels are prohibitively expensive and I think it will be well into 2009 that we see larger true 120 Hz panels on the market. Keep in mind that while you have seen other 120 Hz LCD TVs on the market most do not accept a 120 Hz input over HDMI and are in fact only taking in a 60 Hz signal and inserting frames between the content to increase quality.
Over the past 18 months or so you might have seen DLP TVs for sale with a feature called “3D Ready” or something to that effect and wondered what that even meant. TVs that have that label (currently only Samsung and Mitsubishi are supported) have Texas Instruments DLP 3-D format support that allows the screen to accept two independent left and right image channels, filters them and then samples them using a grid pattern. The images are then combined into a single frame and displayed on the screen without any degradation in resolution.
The magic behind keeping the full resolution image lies with the TI SmoothPicture technology that uses the mirror array and optical actuator of a DLP display system to slightly modify the left and right channel frames and then optically offset the grid pattern created in the above diagram. With the high switching speed of the mirrors of the DLP technology all of the original pixels in the image can be displayed with a 8ms field time and thus allowing the display to show 60 effective frames in under a second (8ms x 120 = 960 ms). An added benefit of this offset and displacement method that the DLP technology uses is that the images on the screen are somewhat softened thus lessening edge artifacting. Oh, and also, you can get DLPs in sizes up to 73-in; who doesn’t want to take on the zombies with that kind of screen?
I should also note that there is at least one projector model (the DepthQ HD 3D Projector) that is purpose built for 3D stereoscopic video and has the necessary 120 Hz refresh rate for supporting NVIDIA’s GeForce 3D Vision. While I am sure it offers a great visual experience, I wasn’t able to test it, and you’d be bring your daddy’s check book with you for this one.
GREAT review. Super video for
GREAT review. Super video for download. MANY THANKS!!
hey there buddy o pal i saw
hey there buddy o pal i saw xpand 3d cinema it gives me an idea what nvidia would be like but when you take the glasses off ur u tend to feel dizzay but at least i dident feel naseau so affter all xpand 3d is better than dolby 3d because it dosent you thos crappy colors anaglyph and dolby have
I loathe those 3D glasses so
I loathe those 3D glasses so a 3D monitor is a refreshing change indeed, although you will have to wait and see if it lives up to what it claims to do.
I’m a fence sitter on 3D; on
I’m a fence sitter on 3D; on the one hand I am excited by the prospect of an extra sensation when a hand reaches out for me while watching ‘The Ring’ but on the other the 3D experience offered by many mainstream movies don’t impress at all. http://www.optegra.com
http://www.theinquirer.net/in
http://www.theinquirer.net/inquirer/news/1050192/nvidia-picks-wrong-3d-glasses-technology
Good article about 3d technology. Especially the sub header 😉