Why 8K isn’t the endpoint for resolution

OculusRift1

I’ve been greatly enjoying following the progress of visual fidelity over the past 15 years.  For such a long time TVs were just TVs, and most people never really thought about them changing.  But, of course, ever since 720P/1080i HDTVs hit store showfloors in the late 90s advancements have been happening at a staggering rate.

For a while now it seemed that 8K would be the resolution endpoint for the popular television sizes (mainly due to the fact that NHK and others set 8K as the goal as far back as 2003).  And for tablet sized devices on down it seemed that 8K would be more than sufficient to reach “retina” status.

But when we’re talking about virtual reality, even an 8K-by-8K screen isn’t enough pixel density to create true verisimilitude.

Oculus VR founder and CEO Palmer Lucky recently explained why this is so in an interview with Ars Technica. In short, the reason is thin diagonal lines:

“There is a point where you can no longer distinguish individual pixels, but that does not mean that you cannot distinguish greater detail.  You can still see aliasing on lines on a retina display. You can’t pick out the pixels, but you can still see the aliasing. Let’s say you want to have an image of a piece of hair on the screen. You can’t make it real-size… it would still look jaggy and terrible. There’s a difference between where you can’t see pixels and where you can’t make improvements.”

Since 16K+ resolution would be necessary for perfect VR, it follows that a perfect “life wall” (a screen that doubles as a wall in a room that can display any desired scenery) would need to be made out of thousands of 16K-by16K head-mounted display sized screens tiled together. That’s an obscene amount of K’s.

Luckey also explains why 8K resolution in screens of the small size required for a head-mounted display will be possible 10 years from now:

“To get to the point where you can’t see pixels, I think some of the speculation is you need about 8K per eye in our current field of view [for the Rift],” he said. “And to get to the point where you couldn’t see any more improvements, you’d need several times that. It sounds ridiculous, but HDTVs have been out there for maybe a decade in the consumer space, and now we’re having phones and tablets that are past the resolution of those TVs. So if you go 10 years from now, 8K in a [head-mounted display] does not seem ridiculous at all.”

The article’s top comment calculates that “perfect” VR for someone with 20/10 eyesight would require a display with 108,000 horizontal pixels, or 108K per eye, which is significantly higher than most estimates I’ve read (most calculations fall somewhere between 8K and 32K).

There’s also an in-depth discussion about resolution over at the Oculus VR developer forums.  And, for a slightly more pessimistic take, Valve’s Michael Abrash has a blog entry arguing that 8K screens aren’t just around the corner.

You can read more about some of the various criteria that scientists/engineers have come up with to figure out the limits of the human eye at this page over at 100fps.com.

And it’s always good to get a reminder of how amazing “mere” 8K resolution is (and keep in mind that this video shows an 85-inch screen!):

Here’s some recent HFR Movies coverage of the march towards an 8K world:

• 8K, 60 fps, 3D documentary: To Space And Back

• NHK Shows Off Compact 8K Camera

• Major Transformers 4 sequences will shoot with new 8K+ 3D IMAX cameras

• Japan plans 8K broadcasts in 2016 – 2 years ahead of schedule

• 4K dubbed ‘Ultra HD’ by the CEA, but Sony sticks with 4K

 

Major Transformers 4 sequences will shoot with new 8K+ 3D IMAX cameras

michael-bay-directing-transformers-4-as-reboot-to-franchis-01Collider has a video interview with Michael Bay (at CinemaCon) where he reveals that Transformers 4 will be the first movie to make use of IMAX’s new 3D digital cameras. Major sequences will be shot with these cameras, which Bay says have a resolution topping 8K:

Collider: What kind of resolution can the IMAX 3D cameras achieve? Is it 4K, 8K?

Bay: It’s like…I think it’s higher than that. It’s actually going to break the….you know, my finishing company said it takes, it’s some zigabyte, whatever the hell it is- it’s some name that I don’t even know what that name is. They just go, “we are screwed with the amount of computing power we’re going to need here.”

Collider: Are you going to use that camera for the whole shoot, or for specific scenes?

Bay: We’re going to do several big showcase scenes. [The screen aspect ratio] is going to open up. Because it’s too expensive to finish IMAX quality FX.

Bay also tells Collider that one of these new IMAX cameras costs more than the entire budget of his most recent film, Pain and Gain.

Here’s all that IMAX’s site currently has to say about their new digital cameras:

Even better for filmmakers, IMAX is expanding its repertoire of capturing tools by perfecting a next-generation 2D and 3D Digital camera. It will be small and nimble enough to go anywhere. It will also be hand-held, take longer shots. Directors will be able to work without interruptions, because there’s no film to change, yet still deliver quality that’s a match for the beauty of film. And as the technology develops, we expect it will ultimately go far beyond what film can do.

16K+ resolution would be amazing.  I suspect that their statement that they expect their digital tech to “ultimately go far beyond what film can do” is a hint regarding the potential of ultra-high resolutions and/or computational photography.  Hopefully we’ll get some more details from IMAX soon.

320 gigapixel image of London shows potential of ultra high resolutions

panoramalondonThis 320 gigapixel “photo” of London is billed as the world’s largest panorama:

http://btlondon2012.co.uk/pano.html

from http://btlondon2012.co.uk:

The 320 gigapixel image – taken by expert photography firm 360Cities – comprises 48,640 individual frames which have been collated into a single panorama by a supercomputer.

If printed at normal photographic resolution, it would be almost as tall and as wide as Buckingham Palace.

Playing with the zoom makes me feel like I have Superman’s telescopic vision. Although it’s not a perfect image (there are some seams in the stitching and people/objects are sometimes duplicated) it nonetheless give an idea of the power and versatility that super high resolutions offer.  Future filmmakers will be able to capture 16K+ video of an entire set or scene without worrying about framing since they would be able to view the footage in post production, zoom in, and extract whatever particular piece of the scene they want.

If you then combine this ultra high resolution with computational photography technology such as that found in the Lytro camera, all sorts of possibilities open up, including the ability to adjust focus after the fact and mimick camera moves without actually moving the camera.

Although it is far below the gigapixel resolution level and doesn’t incorporate computational photography, RED has been working on a 28K camera. Hopefully we’ll get an update on this camera soon, as RED hasn’t said much about it since announcing it in 2008.