Difference between 8K and 4K televisions

I've long maintained that the value of 8K displays does not come from their increased pixel count. There is a limit to the resolution that humans can perceive on video screens at standard seating distances, and increasing the pixel density beyond that limit provides no benefit.

However, where is that boundary precisely? More precisely, does 8K offer any perceived detail advantage over 4K displays under normal viewing conditions? Warner Bros. recently addressed this question in a well-designed, double-blind study in collaboration with Pixar, Amazon Prime Video, LG, and the American Society of Cinematographers (ASC). The study examined whether people could tell the difference between 4K and 8K with a variety of content.

"4K" refers to a resolution of 3840 x 2160 for the purposes of this article, while "8K" refers to a resolution of 7680 x 4320. As you may already know, these definitions are somewhat misleading; technically, "4K" refers to 4096 x 2160 pixels, while "8K" refers to 8192 x 4320 pixels. However, because the consumer electronics industry has appropriated these terms to refer to the corresponding television resolutions, I will use them in this context.

Choice and preparation of content

Seven clips were created in total, each in native 8K and running for approximately ten seconds without compression. Two clips from Warner Bros.' Dunkirk (8K scans of 70mm film) featured a close-up of a character and a wide shot of the beach. This study utilized 8K renderings of animated clips from Pixar's Brave and A Bug's Life. Additionally, two clips from Amazon's live-action series The Tick—one shot in a cave and another in a spaceship—were shot in 8K on a Red digital-cinema camera, as was a clip of nature footage shot by Stacey Spears.

Instrumentation and methodology of testing

The LG 88Z9 88-inch 8K OLED TV, which I reviewed here, was used to display all clips. All clips were loaded onto a Windows PC powered by an Intel i9 18-core processor, SSD RAID storage, and an Nvidia 1080Ti GPU. The video was sent from a Black Magic 8K Pro video interface via four 12G SDI links to four AJA Hi5-12G SDI-to-HDMI converters, which output HDMI 2.0. The four converters routed HDMI 2.0 signals to an Astrodesign SD-7075, which converted them to a single HDMI 2.1 bit stream for transmission to the television.

The files were delivered in their original uncompressed state in the DPX professional format at a progressive frame rate of 24 frames per second (24p). They were encoded with a precision of ten bits in BT.2100, which employs the BT.2020 color gamut and the PQ EOTF (electro-optical transfer function); this is standard for HDR10. These files require a sustained file-read rate of at least 3GBs (that's three gigabytes per second! ), which explains why such a large hardware system was required.

The study, which lasted three days, enrolled a total of 139 participants. Each session, which lasted approximately 30 minutes, had five participants. Two participants sat in the front row, approximately five feet away from the screen, and three in the back row, approximately nine feet away from the screen. To put this in perspective, seven feet from the screen equals two screen heights, and the ITU (International Telecommunication Union) has determined that seven feet is the average viewing distance in the home, regardless of screen size.

Each clip was played in its 4K and 8K versions in three sequences during each session, though the sequences for each clip were not presented consecutively. In two of the sequences, the 4K and 8K versions were randomly assigned the labels "A" and "B" and played twice in alternating fashion—that is, A-B-A-B—followed by participants indicating which one looked better on a scoring form (see Fig. 3). In the third sequence, the 4K version was played four times, but the participants continued to see the labels "A" and "B" alternate and scored them as they had previously. This established a control group for the purpose of ensuring more robust statistics.