Is 4K or 6K Enough.  Do we even need 8K or more?

A 65” OLED is better than a 55” right?  Surely a 300-mile range on an EV will always be better than a 200 mile range?  Then what about a 15” pizza being better than a 12” pizza?

The truth is that bigger may always seem like it is better (unless you have a small flat, can’t afford an expensive car or are trying to lose weight…), so what about the argument for shooting 6K as opposed to 4K?  Then what about 8K, as opposed to 6K?  When is more, well, less?

This article aims to explore that there are further considerations to make when considering the image size and CODEC, which would obviously require using more cards on location, but this also means more transfer time, more media resource for safety copies and in post, more render time and this will ultimately cost production more money.  Then there is also a sustainability argument which becomes relevant, since larger image sizes ultimately demand greater data storage, data movement, post-production rendering and FX, which in turn means more energy usage and thus a greater carbon footprint.

Bigger isn’t always better if the end user can’t see the difference, especially if the cost and the additional CO2 emissions don’t justify it, so maybe an argument can be made for sticking with 6K for all but the most epic productions.

This is an article about the hidden costs of shooting with the largest pixel count and the impact that shooting RAW can have, aside from just the number of cards needed on location.

 I think that resolution is one of those things that’s needed up until a point but can actually serve as a distraction in terms of resources. Obsessing over “more=better” can cause a lot of newer filmmakers to put all their budget into the highest resolution camera and then go with cheap lighting and lenses which means they’re not even making the most of the camera. Much like with cars, you can have all the brake horsepower you like in the engine, but if it doesn’t have good tyres, a good chassis and generate good downforce you won’t set a good time on the race track – no matter how high the top speed is! Might be a bit controversial but the way most films are consumed nowadays, even 1080p is perfectly fine so long as the film is good! I’d much rather watch a film in 1080p with great performances and interesting aesthetic choices over a film in 4K for the sake of it.

Guy Taylor, Gatton Films

When you consider that the latest IMAX epic, “Civil War” cost $50m and was still only shot in 6K using a Ronin 4D, you really have to question whether we are becoming fixated about shooting with ever higher resolutions, which cost more money, burn more energy without the viewer ever seeing the benefits.

Fantastic looking 4K can even look better than 8K at a given data rate, if images are fast moving, since compression artifacts will deteriorate super high res images at low data rates.  BBC research claims that on average sofas are 2.7-3m away from the TV in 3m x 3m rooms, so at this proximity to the screen, 8K images can’t possibly be appreciated by the viewer unless their screens are really huge, which in the UK, they tend not to be. Cinematographer, Kit Mackenzie commented that CODEC efficiency also deserves consideration too, since the higher the resolution, the more efficient the CODEC needs to be (in order to keep the bit rate low) and the quicker it can fall apart. It is interesting to note that a 1080p Bluray achieves a much higher bitrate than a 4k Netflix stream by way of example and even sites like YouTube and Vimeo don’t serve 4K streams at bitrates 4 x the 1080p file counterpart, so your data per pixel goes down – if that makes sense!

In our experience the budget and master deliverable should determine the codec and resolution.  Shooting at the resolution of the master deliverable makes for more efficient post production both in cost and resources.  There’s a limit as to how much you can enlarge an image. Even if the resolution of the recording is higher, the lenses used have a significant impact over how far you can enlarge your pictures without them looking soft. More important than resolution is shooting RAW. BMD and RED camera RAW codecs are absolute resource hogs and for most TV deliverables are overkill.

Peter Zacaroli, Managing Director and Senior Colourist, West Digital Post Production

Of course, down the line, consumers might be introduced to 8K images on their consumer sets but in March 2023, the EU put paid to that, by introducing new efficiency rules which effectively banned the sale of power-hungry 8K TV sets. Why, you ask?  Well, my 4K OLED set has an A++ efficiency rating, whereas the best 8K sets had an E efficiency rating… With everybody now being more aware of efficiency measures and reducing their carbon emissions, if we can’t deliver on more than 4K, the question really needs to be asked, why do we need to capture beyond 6K at all?

As Studios, Streamers and Broadcasters look to reduce their emissions, there has been a great push to reduce power consumption, yet little thought is seemingly being paid to consider when is enough in terms of quality. You wouldn’t expect a newspaper to print high quality images designed for billboards so why are we pushing for IMAX quality in our living rooms?

Changing the “more is better” narrative is a key factor in the transition towards more environmentally sustainable industry. While we are starting to see steps towards more efficient off-camera consumption in areas such as catering, travel and fuel usage, further consideration needs to be put into the area of shooting technology.

Jimmy Keeping, Director, Sustainable Film

You are still able to shoot on a Sony Venice 1 using the X-AVC I codec or even the older RED Raven with REDCODE 6:1 that produce crisp 4K+ images which will save a lot of your budget whilst using much less drive space. Netflix requirements are clear, since they test every camera and stipulate clearly which pass and which do not.  This clear demarcation over ‘true 4K’ versus, well, ‘not true 4K’ was always a real point of contention when they refused to qualify the original 4K ARRI Amira and Alexa Mini cameras, though this has been more than rectified with ARRI’s current Mini LF and Alexa 35 cameras.

It is quite obvious that Directors and DPs will always choose to shoot at the highest image quality possible they can, as this will more easily permit crop, zoom and pan-and-scan in post, which is very easy to achieve when shooting in 8k.  Also, more grading options are also available when acquiring in an uncompressed RAW codec too but the reality is that these are out of reach for many productions and do you really need to go to such lengths anyway?

Most of the video shot in the world is watched on home devices, and huge proportion of that is mobile phone viewing. Even HD resolutions are rarely necessary. But knowing that in the digital world your content has the potential to appear anywhere, producers will plan for the biggest and the best, just in case. Commercials are interesting case in point. They are often shot – to please the client – at the highest possible resolutions when everyone knows full well the results won’t be seen beyond a social media platform.

One thing sustainability requires is commitment. You can’t have it all ways, all the time. Production teams are going to have start being grown-up and making plans and decisions based on specific distribution commitments and requirements. If you know you’re shoot film isn’t going to make it to the big screen, then don’t shoot for the big screen. I also think that soon AI will allow us to create whatever display resolution we want, regardless of the source. I’ll leave it to groups like Greening of Streaming to figure out if it’s the most resourceful way of using energy or not, but conceivably, you could shoot content at relatively low resolutions and beef it up to a simulated higher resolution only when you need to. The gamble would be that the energy saved in shooting at lower resolution would be more than the energy used by your AI resolution boosting tech. Of course, the most sustainable footage is the footage you don’t shoot at all. Plan ahead. Take your time. As they say, “Measure twice, shoot once.”

Neal Romanek, Editorial Director, The Flint

On the subject of grading, HDR capture and delivery makes such a dramatic impact on productions, which in my opinion is more important than image resolution and the capability of HDR is entirely irrelevant on image capture resolution, as this is as a consequence of what LUT is applied when shooting.

Having spoken to many filmmakers, the overwhelming feeling seems to be that 6k can give more than enough capability for 4K mastering, so for example, using a lossless Sony codec, like the OCN-LT (the lowest Sony lossless codec), ought to be good enough for big screen cinema productions and offer sufficient latitude in post. Perhaps using this argument, 6k acquisition could almost be considered as a modern-day version of the old 35mm open gate, by offering sufficient leeway for cropping in post, if a C-stand leg or flag creeps its way into a shot.

Other pitfalls must be considered when filming at very high resolutions.  Full frame sensors can be very unforgiving on aged skin with bad lighting and visible makeup marks, so we strongly encourage that tests are performed on wardrobe, make up and filtration prior to any serious shoot.

The codecs for some of these 8K high-resolution projects hungrily fill data cards and eat hard drives in contrast to a whole short film being able to be filmed on a single 1Tb drive using the X-OCN LT codec at 6K.  Reshooting scenes unnecessarily also costs a lot of money.

4k is a little bit overkill! for human eyesight at close quarters, and 4k lighting is more than ample coverage for 1080 resolution, but it is acceptable and probably desirable for most productions. Lens quality is important for sharpness on our TVs, since fewer and fewer people are going to the cinema. The above is more than sufficient for HD TVs and even small room cinemas. If you intend filming for larger cinemas, then the bigger the cinema, means the higher the resolution required for the maximum eyesight perception, which I call (MEP) – no relation to our toothless rulers. However, for small cinema rooms, we are back to high quality lenses for 1080 for the HQ LED screens and HD TVs. 

Adrian Thurston, Technical and Creative Consultant and a founder member of Directors Guild of Great Britain

Thankfully, the debate of whether film is better than digital has long since been abandoned, with film increasingly being used today for artistic reasons – like Oppenheimer.  However, film, as a chemical/optical process, performs at any given resolution no matter what size frame shot on and importantly, this is not universally true for digital sensors, which are designed for optimal performance at given resolutions.  This is a really important factor because an Alexa Mini LF can only deliver proper 4K when shooting in FF.  If windowed (cropped) to S-35, perhaps because S-35 lenses are used, then it will have the same resolution as an Alexa Mini, which will no longer comply with Netflix 4K standards.  In contrast, the Alexa 35 delivers 4.6K in S-35 but can’t shoot in FF, as the sensor size is too small.  Cinematographers need to be aware of these specifications and limitations and choose resolutions and image sizes to the camera capabilities and lenses being used.

Sony’s higher resolution FF sensors used in the original 6K Sony Venice 1 and 8K Venice 2, allow cinematographers to discard unwanted pixels of the sensor, so they can acquire 4K in any aspect ratio that they desire.  The difficulty with this approach is that cinematographers are rarely happy cropping a large part of the sensor, as this limits the sensor size and look of the production.  For example, cinematographers wanting to capture 6K on a Venice 1 will capture FF but if they want to use a Venice 2, then they would need to crop the sensor (which would change the coverage of the lenses and thus the look), else acquire in 8K and then throw away pixels in post.  This adds time, effort, cost and energy for no benefit.

So with very high speed sensors now the norm, many of the original concerns regarding shooting interlaced/progressive, or with a global/rolling shutter are no issues, so cinematographers now just need to decide on an aspect ratio and resolution to capture, edit and deliver on.

I’ve never been particularly focused on resolutions as a DP. However, some decisions have prompted me to consider the vast array of choices available. Recently, I’ve been carefully selecting sensor sizes to match the lenses I use, as this combination significantly influences how images are rendered. Understanding the image circle of your lenses is crucial when selecting a sensor size and resolution. For instance, on the Alexa 35, I’ve been shooting in 4K 16:9 using vintage Cooke Speed Panchros. This resolution, to me, yields the most visually pleasing images. Likewise, for the documentary ‘Washing Away’ (Directed by Sophie Nielsen), pairing the FF Cooke Panchro Classics from VMI with the Venice 2 in 5.7K 16:9 allowed me to utilize most of the sensor. I can’t tell between the 4K and 5.7K image, but importantly, both these resolutions allow for those lovely aberrations to play on the outer edges of the frame contributing to the unique character of the images.

Jack Harrison, DP

So far, so good –  so why should this last element be a concern?

Manufacturers make sensors and the design optimizes resolution for a given sensor size and every manufacturer developed the Optical Low Pass Filter (OPLF) to produce very sharp images and avoid moiré patterning and each found their ‘sweet spot’ with the current crop of sensors. 

So given all of this info, where does that lead us?  Recognising that DPs generally want to use the entire sensor, so shooting 6K on an 8K sensor means windowing, since all productions will want to minimize rendering and avoid down-sampling.  Our position is to encourage productions to capture at 6K and not take the plunge to go 8K, unless they need to. 

How do we do this?  Simple-  we spec’d our Venice 2s to include native 6K sensors as standard, in place of the default 8K sensors.  This allows DPs to use the entire sensor and acquire in 6K avoiding the complication of down-sampling or windowing.  Obviously, if a major production needs an 8K capture, then we can replace the sensor for an 8K sensor for their production but, as we have hopefully demonstrated, there should be very few occasions when this is necessary.

In Natural History, windowing of sensor can be used to extend the apparent reach of the lens… windowing is also the only way to achieve higher frame rates when shooting RAW so there are other considerations we have. Actually, much of what you’ve written chimes with why we have gone down the Red Raptor VV 8K route for our workhorse Nat Hist camera.  When paired with the Canon CN20, the Raptor can only achieve 5 or 6K (depending on focal length) without vignetting.  This keeps data down and helps with all the other points you’ve raised about data but, if we need to shoot 8K we can either put in the 1.5x extender or use full frame glass and shoot 8k (where there’s a purpose… perhaps for VFX plates / people expecting extensive cropping in post etc).  We feel the Raptor VV is a better low light performer than it’s S35 sibling and for the higher frame rate of 4K 240fps – the VV’s bigger pixel pitch/light gathering means it produces a cleaner image overall.

Senior Producer BBC NHU

In conclusion all modern cameras deliver well above the streamers’ acceptable range for 4K delivery, as a culmination of years of research and development of designing sensors that perform at their peak.

Whilst 4K was out of reach for many manufacturers for a long time, this is now the norm and manufacturers like ARRI, Sony, Canon, RED and pretty much everyone else, delivers quality thresholds well above what was for many years, considered to be the best possible. (Star Wars Episode II for example, was shot on the very early Sony F900/3 in 1440×1080 and cropped to 1,440 x 720 in 3:1:1 colour space and still looked great!) – thank you to Kit Mackenzie for correcting me on Ep II and not Ep I!.

As we have seen from the industry through grassroots collectives like Filmmakers for Future: Wildlife, people are calling for more sustainability in production. Reducing the data captured is an easy win.”

Ruth Campbell, Producer Director and Co-organiser Filmmakers For Future: Wildlife

Blockbusters represent only a tiny number of productions, with the majority only ever being seen on the small screen, even including most films, courtesy of the streamers.

The quality debate can continue endlessly with manufacturers, DITs, scientists, and people with ‘pens in their top pockets’ all having their own opinion, much to the frustration of Directors, DPs, Producers and yes, let’s not forget the poor rental companies that must keep up with the trends.  The proof is in pictures and whether viewers are happy, so the consideration of whether something was shot in 4K, 6K, 8K or more, to me, isn’t even a consideration any more.

Leave a comment and share what you think.

Barry Bassett and Gerard Botha, April 2024

Large players in the film and TV production industry are signaling a shift towards more sustainable practices, as they face increasing regulation and consumer pressure to reduce their emissions. By prioritizing the planet, vendors and production companies have the opportunity to stand out from competitors and win new business, reduce costs, and improve their operations.

The SME Climate Hub provides small and medium-sized businesses across industries with the free tools and resources they need to take climate action. Last year, with industry knowledge from Netflix, the initiative developed a climate education course for production companies, expanding the resources available to businesses to phase out fossil fuels. In order to thrive in the entertainment sector and build long-term resilience, businesses must adopt begin taking action.

Lydia Elliot, Deputy Director, Supplier Decarbonization, SME Climate Hub