What Would a Potential 8K Disc Format Look Like?

Introduction

I am aware of the discussion around viewing distances and whether higher resolutions make sense beyond a certain point. While I won’t go into detail on this topic in this article I want to state the following:

• 8K still very much makes sense, at least for home theatre projection dependent on viewing distance.
• 16K and even 32K also makes sense, but as a production, archival and possibly Digital Cinema format.

There are now both 8K TVs and 8K capable projectors on the market, and more are coming. 8K content – and even an 8K disc format – is also coming. The only questions are when and how. In this article, I wanted to take a look at the current market trajectory for disc-based media, as well as possible technical solutions for bringing such a format to market.

Market Trajectory

If we have a look at the total sales revenue in the US from movie disc sales, we can see that there’s a decline of 20% year over year. In fact, the market is shrinking so fast that many Blu Ray player manufacturers have exited the market such as Samsung and Oppo, after not being able to compete with Sony and Panasonic in quality and price. More recently Best Buy said they would be stopping sales of DVD and Blu Ray discs.

Year	Revenue from Disc Sales	% Drop
2019	3.29 billion	–
2020	2.45 billion	25%
2021	1.97 billion	20%
2022	1.58 billion	20%
2023	1.26 billion expected	20%
2024	1 billion projected	20%
2025	800 million projected	20%
2026	640 million projected	20%

At the same time, streaming revenue is increasing with a projected sales revenue of $95 billion for 2023, and now makes up 98% of the market, with disc sales making up less than 2%.

The reason I wanted to start here is that – regardless of our ability to invent and produce another disc format – there might not be a very large market for it by the time 8K would come out. It is entirely possible that the decline of physical media will slow over time, but at this point in time, the decline doesn’t show a slowdown.

Potential Release Date

To calculate a potential release date for 8K media – whether on disc or otherwise – it is a good idea to look at past formats. There were 10 years between DVD, Blu Ray and UHD Blu Ray disc release dates. Each of the disc formats are still in use today with DVD having the largest market share still both in terms of sales and install base.

Media	Release Date	Current Market Share
DVD	Nov 1996	50%
Blu Ray	Jun 2006	35%
UHD Blu Ray	Feb 2016	15%
8K	Jan 2026 earliest	N/A

If we were to take as many years between the next format release, the earliest release date for 8K discs would be sometime in 2026. However, there are a few things that are different this time:

1. 4K UHD growth was a lot slower than Blu Ray and especially DVD. It has still only penetrated around 12 – 15% into an already shrinking market.
2. 4K TVs were incredibly popular in the consumer market as the benefits were clear to see. In fact, the whole market shifted to 4K within a few years. 8K TVs on the other hand are facing a major uphill battle: 8K resolution isn’t as much of a jump visually at TV sizes, so they are a much harder sell.
3. 8K TVs are both more expensive to produce and more expensive to run than 4K TVs are, although these costs are slowly coming down. However, there is also increasing regulation around energy use which will make 8K TVs take somewhat longer to come down in price than 4K TVs did.
4. Streaming has taken over 98% of the market. By the time an 8K disc format would come out, disc sales would sit at less than 1%!

Even with all the above, I predict that an 8K disc format is inevitable at some point. It might just be a niche consumer product – such as Laserdisc, or DVHS were – or a professional archival product. Laserdisc had a market-penetration of around 17 million players sold worldwide. In 2026-2028, an 8K disc format would need to capture around 1-3% of the disc-based market to achieve the same penetration, which is entirely possible.

Potential Format Specifications

OPTION 1: Use What We have

The best compression method we have available today is the successor to UHD Blu Ray’s HEVC (H.265) called VVC (Versatile Video Coding) or H.266.

It was designed to halve the bitrate required for 8K and 16K content compared to H.265 at the same quality level. This was achieved mainly by allowing much more sophisticated spacial (within the same frame) and motion (across frames) prediction across 4x the number of pixels – exactly the size increase for 8K resolution. In addition, H.266 does have more sophisticated film grain removal and simulation compared to H.265, whereby film grain would not need to be encoded directly, but could be added back at decode time.

Film-grain simulation might actually be a good feature to use, as some people prefer a more pristine image, while others would prefer the original look of the movie – especially those shot on optical film. Having this be configurable on the playback side could work in the favour of an 8K format. However, it needs to be implemented in a way that the original look of the movie can be accurately reproduced – and the user knows exactly how much grain was present in the original, with the playback device able to default to that setting during playback. There are multiple ways to implement this with H.266, including metadata or using enhancement layers, which we will discuss under option 2 in more detail.

H.266 requires around 50Mbps for the 8K content to be of high quality. Since existing high-quality 4K UHD releases use this bitrate already on either 66GB or 100GB Blu Ray discs, we could use the exact same drives, lasers and disc manufacturing facilities to create an 8K media format. The only thing that would need to change would be the chipsets used to decode the content on the discs.

However, to maximise 8K quality, it would be advisable to use the currently available 100GB discs with special features going on a second disc. This is generally what happens with 4K UHD releases, but the majority of the movies go on a 50GB or 66GB disc.

Using a 100GB disc with only 2 audio tracks could get us closer to an 80Mbps bitrate for the video stream, which is much more likely to maximise the picture quality for the 8K stream above what a 4K stream would look like, because 50Mpbs is really the minimum, and a bit of a half-way house as I’ll show you in a moment. As we have seen with Sony’s “Mastered in 4K” Blu Ray releases, doubling the bitrate of the video stream reduces picture noise and increases fine detail even at the same resolution.

OPTION 2: Use a New Forward-Looking Disc Format

The second option would be to use a new disc format entirely. The idea of using fluorescent dyes in place of the current non-fluorescent variety used on current disc-based media has been an area of research for a number of years now.

Fluorescent dyes allow for many more layers on the disc for data storage without the laser pickup getting confused about which layer to focus on. Current UHD Blu Rays discs use 2 to 3 layers for 66GB and 100GB of storage respectively (although the data-storage focused BDXL can use 4 layers). Fluorescent dyes would allow us to use many more.

Folio Photonics is the company that has come closest to building and demonstrating a workable prototype. Their proposed discs have up to 16 layers per side with possibility of making both sides of the disc readable for up to 32 layers!! Each disc will be able to store 1TB of data initially, but they are confident they could make 10TB disc “cartridges” by 2030.

Since such sizes are overkill for 8K, unless we are thinking of closer to lossless quality, I think it would make sense to delay the format and future-proof it for at least 16K resolution, but ideally archival-quality 32K resolution.

Such a format could store uncompressed 1080p video, and possibly uncompressed 4K UHD video for archival purposes using H.266, which has an lossless archival mode, so we could reach the same place we have with audio: the exact bit-for-bit studio master could be stored on the discs.

For 8K, 16K and 32K video, lossy compression would be required, but would be possible to store them. 8K is about the resolution of a high-quality 32mm film negative. IMAX 70mm is around 8x the size of 32mm, so it would sit in the 18K resolution range, but possibly even larger with good scanning equipment.

So it is worth pursuing even 32K, even if for the purpose of archiving film media, and definitely for the purpose of future-proofing the disc format. 32K displays have been produced already for very specific use cases after all.

Possible Bitrates

We discussed the minimum bitrates needed to make 8K look great earlier on. Now let’s look at the bitrates we would need to maximise each resolution with H.266 and make it comparable to the best UHD 4K releases currently on the market:

Resolution	H.265	H.266	Disc Layers Required (H.266)*	Disc Spin Speed (H.266)
4K	50Mbps – 66GB	25Mbps – 33GB	1	1x
8K	200Mbps – 264GB	100 Mbps – 132GB	4	3x (2x with dual laser)
16K	800Mpbs – 1TB	400Mbps – 528GB	16 at 33GB per layer 8 at 66GB per layer	12x (8x with dual laser) 6x (4x with dual laser)
32K	N/A	1600Mpbs – 2TB	16 on both sides using 66GB per layer or 2 discs in a cartridge	50x (16x with triple laser) 25x (8x with triple laser)

Of course, compression isn’t quite as linear as that. There could be more bitrate savings at higher resolutions. The only problem with that theory is that H.266’s compression sweet-spot is 8K due to the prediction block sizes, so there wouldn’t be heaps of bitrate savings for the same quality above that resolution. To save bitrate, we could lower the bitstream quality for 16K and 32K content and still fit onto 16 layers. Of course, the compression in that case would be higher than today’s best UHD releases and would fall in the “average” category.

I also wanted to calculate the size requirements for lossless streams, so we could see if lossless video streams would be achievable. While these are estimates made with H.265 capability, it is possible that H.266 would do better than this. As we can see from the table, though, it is entirely possible to do lossless 4K streams, while near-lossless 8K is also within the realm of possibility. This would be great news for videophiles and would do for video what Blu Ray did for audio: no longer having to worry about whether compression is compromising image quality, and would give a massive leg up compared to streaming.

Resolution	Compressed Size (H.265)	Lossless Size (Approx x 13)
HD	20GB	260GB
4K	66GB	858GB
8K*	264GB	3TB (Possibly 1.5TB with H.266)

Using Data Enhancement Layers

We can see from the data in the previous section: 16K resolution would be achievable using around 8 layers and a 12x disc spin speed (1x for Blu Ray is defined as 32Mbps). This is well within achievability with current technology. The disc spin speed is a little too high and is actually the maximum current Blu Ray data drives tend to spin at. However, at this speed, the drive is very loud, so would need extra bracing and sound proofing within the player chassis. This would increase cost.

The better alternative would be to have a dual pick-up head, whereby two lasers read adjacent tracks or better yet – layers, so that the image could be reconstructed using dual data streams. The best way to achieve this would be with H.266’s multi-layer coding feature. This feature would allow the main stream such as a 4K stream to be stored on the first layer, while the 8K extension data to be stored on the next 2 layers. The 16K extension can then be stored on the remaining 5 layers.

This way, a 4K player would only need to support the reading of the first layer with a single laser head, while 8K and 16K players would need to support reading the adjacent layers with 2 or 3 laser pickups respectively.

This would also allow discs to support only the resolutions that make sense for the content. So for example movies that only had a 4K master (called a digital intermediate or DI) might only use 4K resolution on the first layer, while movies that had 8K or 16K masters could fill the other layers as well.

As an additional option, 4K discs could still use the enhancement layers to provide extra content or use extension data to provide a higher-quality or lossless 4K stream. This enhanced 4K stream might only be decoded by higher-end players as it would still require dual or triple laser pickup as the extension data would need to be read and decoded simultaneously.

Playback Device Tiers	Laser Pickup Per Side	Disc Sides Read At the Same Time	Description
Tier 1	1	1	Low cost players able to play back HD, 4K and 8K streams on the first layer only
Tier 2	2	1	Medium cost devices able to decode the first data enhancement layer
Tier 3	3	1	High-cost devices able to decode both data enhancement layers
Tier 4	3	2	High-cost devices intended for production and archival

We could also skip the 4K layer entirely and start with 8K layers at 3x spin speed, then use a second laser for the 16K enhancement layers. The 16K enhancement layers could be double-dense at 66GB per layer to allow for the drive to still only spin at 4x spin speed and laser tracking to be simplified. The 16K enhancement layer would be somewhat more compressed than the 8K layer but this would likely not matter at such high resolutions and further data layers could be used to provide a higher-quality 16K stream.

All the possible permutations are included in the table below. HD base layers were included because a lot of movies have been upscaled from 2K DIs for 4K releases. I think that this isn’t necessarily ideal for preserving the original master. A better option would be to store the lossless HD stream and allow the player to do the upscaling if necessary.

Option	Main Layer* (Tier 1 Players can decode)	Data Enhancement Layer 1 (Tier 2 Players can decode)	Data Enhancement Layer 2** (Tier 3 Players can decode)	Include In Format Specs
1	HD	HD (lossless)	4K	Yes
2	HD	4K	8K	Yes
3	4K	4K (high-bitrate)	4K (lossless)	Yes
4	4K	8K	High-Bitrate 8K	Yes
5	4K	8K	16K	Yes
6	8K	High-bitrate 8K	Near-Lossless 8K	Yes
7	8K	16K	High-Bitrate 16K	Optional
8	8K	16K	32K	No
9	16K	High-Bitrate 16K	Ultra-High Bitrate 16K	Tier 4 Players Only (Two sides read at once)
10	16K	Medium-Bitrate 32K	High-Bitrate 32K	Tier 4 Players Only (Two sides read at once)

I also think that HDR should be delivered as an additional data enhancement layer, on top of the base SDR BT.2020 data layer and stored alongside the main stream data. This is how Dolby Vision is coded on UHD 4K discs today. This way the studio would be required to have a properly authored SDR BT.2020 tone-mapped version and the format would fully support projection, which the current UHD disc format does not.

As discussed earlier, film grain could also be delivered as an additional data enhancement layer. Alternatively, it could be simulated at lower tiers but included in the lossless and high-bitrate enhancement layers.

A new disc format could also offer the AV1 codec as an option. AV1 is a royalty-free codec developed by the Alliance for Open Media (AO Media) Group. It is about 20% less efficient than H.266, so would need respectively higher bitrates. However, since it is royalty-free, it would be possible to use this for the main data stream only to reduce costs for lower-end players and discs even further, and only use H.266 for the enhancement layers, if this is technically feasible that is.

Maximising 8K

A lot of the movies on UHD 4K are still upscaled versions of 1080p movies with an HDR layer on top. This is because a lot of filmmakers embraced digital production and even if a movie was shot at 4K, the Digital Intermediate (DI) used for producing the visual effects and the final movie was done at 2K (essentially 1080p).

The second issue is that you need about 4x the target resolution to achieve maximum fidelity of a particular medium. This is due to the inherent characteristics of digital media: noise needs to be filtered to a point, but even too much high-frequency information will cause issues with mastering, so some of that also needs to be filtered. This can result in a loss of fine detail.

However, when this filtering is done at 4x the target resolution, the resulting video can still retain a lot of fine detail as it is downscaled to the target. This is why Blu Rays mastered at 4K then downsampled to 1080p look cleaner with more fine detail than Blu Rays made from a 1080p master.

So for us to maximise 4K and actually see what the format can do, movies will need to be mastered in 8K with an 8K digital intermediate! Only then will 4K look sensationally good.

In a way, this is a chicken and egg problem. We need 8K on the production side for 4K to look great, and we will likely need 16K on the production side for 8K to look impressive. We are still far from this goal, at least for movies that are very digital effects (CGI) heavy. However, for movies that have been or are being shot on optical film, this won’t be an issue, as scanning technology is now able to scan at 16K. This 16K content can then be transferred onto 8K – or preserved/archived at 16K if a future format is going to support that.

Conclusion

While an 8K disc format might take longer than usual to appear, and might only be useful for archival purposes, as opposed to consumer use, we could argue that 8K is here today: on YouTube with some specifically shot footage that is. However, YouTube content is very heavily compressed so we can also argue that it isn’t really 8K. Modern compression algorithms will selectively scale down part of the screen area dependent on the amount of detail and bitrate allowed for the content. So using AV1 at low bitrates like on YouTube is nowhere near good enough to show what 8K can really do.

8K movies will likely start being available on specialised streaming services first – such as Sony’s Bravia Core, with Netflix or Disney+ following once H.266 and AV1 decoding is more commonplace on streaming devices and 8K displays have a much larger market share.

I would predict Sony launching some limited 8K content with the PS5 Pro in early 2025. However, if previous release schedules are to be relied upon, then 2026 – 2028 is when 8K will start to be more commonplace and we might expect a potential 8K disc format to show up between 2027 – 2029. But make no mistake about it: 8K is coming!!!