Some Projects that Need More Pixels
This week we saw a couple of R&D projects that point towards different technical destinations, but both of which have some relevance to our work at the 8K Association. We are aiming to ensure that the infrastructure and ecosystem exist to support very high resolution immersive display experiences, using lots of high quality pixels.
Some Folks Really Need 8K
The first item picked up on a topic that we wrote about on our Discover 8K site a couple of weeks ago, that one particular group of people really needs 8K. That group is those that make autostereo 3D (AS3D) displays which means 3D without glasses (or headsets). If you are going to have views for, at least, one pair of eyes, then you need more pixels. That’s especially the case if you are using optical methods to simultaneously send different images to each eye.
However, there is another way to create multiple images, and that is to separate them by time. Historically, time-multiplexed 3D has created the separation between the eyes by pairing the display with shutter glasses that are synchronized to the images. The view of one eye is blocked when the image for the other is being shown. This technology goes back to the 1970s and was used with CRTs in the 1980s. TV companies including Samsung and Sony used this kind of technology for their 3D TVs a decade or more back.
Our attention was caught by a demonstration from a small 3D specialist firm at CES that was interviewed on YouTube. The company with the new technology is called Realfiction and it has been around since 2008, working on a variety of ‘3D’ displays. The firm is Swedish-owned but operates in Denmark. It is listed on Nasdaq.
Pointing at the Eye
Realfiction’s speciality is in developing technology that can direct display images towards the users’ eyes, but with the direction changing fast enough that multiple left and right images can be displayed on a single imager. This means that, for example, if you track the position of several users’ eyes, you can send the correct images to each eye, allowing multiple viewers to see different 3D images without needing glasses. The company previously exhibited a system using switchable parallax barriers at SID in 2023 and at CES in 2024.
Now, the idea of sending multiple pairs of images to allow AS3D for multiple viewers is not a new one, but historically this has been done using lenses or fixed parallax barriers to create the multiple images. These optical techniques have the disadvantage that the total resolution of the display has to be shared between the images (which is why 8K displays are popular – there are simply more pixels to play with). The other disadvantage of fixed optics is that there tend to be ‘sweet spots’ where particular viewing directions work well, but ‘dead spots’ where the 3D effect is not great. In other implementations there may be darker regions or artifacts between the sweet spots.
The Demonstration
In the demonstration at CES, a low resolution display was used to show multiple different 3D images to different viewers and to move those images as the viewer moved around the display. There’s a good animation on the Linkedin page of the firm’s Co Founder. If you want to get a look at the firm’s technology, it is planning a demonstration for shareholders in early February in Denmark.
Now, to have time-multiplexed images, you need a fast switching system and Realfiction uses patented ferroelectric liquid crystal (FLC) technology that is licensed from the Hong Kong Uninversity of Science & Technology. FLCD is an interesting technology and you can get more information on what the HKUST has been doing with the material here.
In the very early days of liquid crystal displays, it wasn’t clear whether FLCDs would be a better option than TN LCDs, which were the main technology until the development of VA and IPS panels for monitors and TVs. FLCDs are very fast in operation and are bistable – that is the pattern is retained even when the power is off, like E Ink. However, they can be tricky to make and have limited viewing angles. Canon developed the first 1280 x 1024 LCD panels for monitors using FLCD, but eventually abandoned the technology.
No Free Lunches
The demo by Realfiction is, of course, very low resolution. Moving to high resolution would mean a lot of data at high refresh rates, which will be one challenge. The second challenge is that if the display is only visible for a short time during the frame, there will be less light available and brightness of the displays might be a barrier in moving to commercial products.
However, there are a number of applications where having different images available to separate viewers is a real advantage – in the consoles of autos and game play are two, but there are more.
(After our article was published, Realfiction got in touch to point us to this new patent which has more detail.)
Everything in Focus
A second interesting technology that we saw this week was a system that puts a spatial light modulator in a large lens configuration. The system adjusts the focus of the lens on a pixel-by-pixel basis based on a 3D map.
The technology is interesting partly because, if developed (and it’s just a research project at this point), it offers the possibility of capturing a scene completely in focus, and then allowing the depth of field in the image could be changed later for artistic effect. In seeing this, we were reminded of a project by the Fraunhofer HHI to capture scenes with a genuine light field to allow the same kind of manipulation. The new approach looks much more cost effective, on the face of it, although the Fraunhofer idea would allow more flexibility later in changing viewpoints etc.
Still, it’s good to see projects going on that reinforce the need for lots of high quality pixels!
