Equipment

Six Future Techs We Hope to See Become the Industry Standard in Photography and Videography

With WPPI, NAB, and the dozen other photo and videography shows going on over the last few months, we’ve been busy engulfed in new cameras, new technology, and new workflows for the coming year and beyond. We’ve certainly done our fair share of coverage on these shows, and we’ll know more when the products are actually released and tested. But the excitement got us thinking, what are some of the tech teased in recent years that we hope to see implemented into the standard? Various brands are all releasing their densely developed concepts, and we decided we wanted to take a moment and discuss the most innovative, and what we hope to see become a standard in the industry.

While the obvious always stands true. Everyone wants faster frame rates, more focus points, better dynamic range and longer battery life. So we’re using this as an opportunity to spotlight the tools that have been shown off but are different from the usual requests we want out of camera systems.

Canon 5d Mark IV’s Dual-Pixel RAW [photo]

At the launch of the Canon 5d Mark IV, they teased a focusing tool that teetered on the gimmicks put out by Lytro, where you could adjust focusing after the shot, making sure all of your photos are always in focus. The technology, admittedly, fell flat. Its failure was the result of no one really implementing it, on the software side. Still, at the time of writing this piece, the only way to take advantage of this technology is to use Canon’s awful imaging software. Those who are working with Photoshop, Lightroom or Capture One aren’t able to play with this tool, which was never fully adapted.

Canon 5d Mark IV Dual Pixel Raw

How it works is simple. It works by taking two photos using the right and left half of a single pixel, then filters out the information. Since this is capturing different data from each half of the pixel, such as flare, and sharpness, and then allows you to adjust them accordingly, while still maintaining the 30MP image size. The adjustments of this tool are admittingly subtle but could help considerably when you’ve taken a fantastic shot, only to see your subject is ever so slightly out of focus. While this isn’t as in-depth or as impressive at the Lytro Cameras from years past, it does have a practical purpose that we’d like to see implemented further.

Electronic Shutter of the Canon 1D [photo]

While not new tech, the original Canon 1D had an impressive piece of technology in it, that was quickly abandoned. And that technology was known as the electronic shutter. Whereas most cameras use a series of shutter blades to expose the sensor to light, the original Canon 1D did it by shutting off the electricity to the sensor, when exposing the sensor. It still had a mechanical shutter, but it was only used in the readout process. All of this was achieved because the Canon 1D used a CCD sensor, as opposed to the CMOS sensors typically found on modern systems. While as of right now, it seems impossible to get the same type of technology in a CMOS sensor, but we’d love to see developments to be made to give us the electronic shutter in a CMOS system.

A visualization of how your traditional shutter works, and why your flash sync speed has a maximum theoretical speed.

So what makes this special? For one, general reliability. The Canon 1D, through the use of an electronic shutter, had a shutter actuation rated for around 500K, nearly four times what most cameras promise. Secondly, for flash sync speed. Most cameras allow sync speeds anywhere from 1/160th of a second to 1/250th, where the Canon 1D had a flash sync speed of 1/500th of a second. This was because most cameras sync speeds are dependent on how quickly the shutter can fully open, trigger the light, and the begin the closing process. The theoretical limit to this is at 1/200th of a second for a full frame sensor and 1/250th on a crop sensor. Additionally, the Canon 1D allowed for shutter speeds as high as 1/16,000 of a second. Impractical, maybe, but impressive? Definitely. Of the technologies mentioned in this piece, this one seems likely to make a comeback, with it being shown off in many of the Fujifilm systems, such as the FujiFilm X-T1.

Pixel Shift Technology [photo]

Technology originally developed for the Pentax K-1, pixel shift technology is finally making its rounds to the mainstream, with it most notably being introduced to the Hasselblad H5D-200c MS and the Sony a7RIII. What pixel shifting does is essentially take multiple frames, with the sensor shifting a pixel with each shot. This allows for up to 4 times the resolutions and sharpness in your photos.


Pixel Shifting promises more than just increased sharpness though. The tool can also be used for removing moire and aliasing, as well as producing better color accuracy, as each pixel is being shot multiple times to assure accuracy on the sensor. While it sometimes feels like we’re hitting the ceiling of sharpness, this technology, as it grows and furthers its development, could be a unique tool to get uber resolutions and sharpness in smaller cameras and sensors. Let’s just hope that this will also eventually make its way into editing platforms like Lightroom and Capture One, as it’s not yet available to third-party editing platforms.

Multiple RAW Sizes [photo]

At this point, I expect to get some criticisms that I’m a Canon fanboy, though I certainly have my list of complaints with their platform. But there is one thing that they do right, that virtually no other brand does; multiple Raw sizes. What has seemingly been on Canon systems forever now (seriously, I had cameras ten years ago with this functionality), it’s still lacking on both the Nikon and Sony systems. Traditionally, Canon DSLRs are equipped with multiple Raw formats – sRaw, which is about 1/3rd the resolution of full Raw, mRaw, at around 2/3rds the resolution and Raw, which is, of course, the full file size.

So why is this useful? Well simply because you generally don’t need a full resolution all the time. With camera systems pushing 50mp, it’s rare that you actually need that resolution for every scenario. By implementing a smaller resolution, you’re able to create a faster workflow, save card and hard drive space, and churn out the images to your client faster. The image size I’d need for a 2-page advertising campaign is not the same resolution needed for my nephews birthday party, the having the ability to adjust the resolution as you shoot is a huge advantage, and it’s surprising to not see it on other platforms.

Dual Native ISO

Typically, a video camera has a single native ISO at which the sensor performs at its best. Cranking the ISO higher than the native setting will introduce more camera noise, and lowering the ISO below the native setting affects dynamic range. Cameras with dual native ISOs have two analog circuits per photo site, enabling an equal signal-to-noise ratio and dynamic range at two different ISO levels. Panasonic debuted this technology back in 2014 with the Varicam 35. Since then, it’s also been a feature of their Varicam LT, EVA-1, and GH5S. Recently, we’ve also seen other manufacturers adopt this design. RED’s new 5K Gemini sensor has dual native ISOs of 800 and 3200, and Blackmagic’s upcoming 4K pocket camera has dual native ISOs of 400 and 2500.

The flexibility afforded by dual native ISOs makes these cameras much more attractive to customers who frequently have to work in low-light fields such as documentary, news, weddings, or basically anything that isn’t a traditional studio shoot. Based on the popularity and quick adoption of this feature, we wouldn’t be surprised if it’s as common as internal ND filters in the next five years or so.

ProRes RAW

A joint venture between Apple and Atomos, ProRes RAW is being advertised as “The flexibility of RAW, with the simplicity of ProRes.” I don’t have the page space to talk about it in too much detail (you can read Apple’s white paper if you’re especially interested), the whole idea basically boils down to a compressed RAW format that puts off debayering until post-production. Since uncompressed RAW data rates are so high, and RAW transcoding is so processor-intensive, the RAW workflow isn’t really an option for shooters without exceptionally deep pockets and forgiving deadlines. ProRes RAW, on the other hand, takes up far less drive space than uncompressed RAW and should play back reliably on most recent computers, at least in applications that support it. While some manufacturers, like RED and Blackmagic, have introduced compressed RAW recording, but those are still proprietary formats limited to in-camera recording. Atomos and Apple are hoping that ProRes RAW becomes an industry-standard solution for external recording.

This is very promising technology that unfortunately has a good chance of being ruined by video industry infighting. If things go well, we could see ProRes RAW become an open format that allows more people to reap the benefits of a RAW workflow. If things go poorly, we could see Apple and Atomos close walls around the format and force out customers who aren’t a part of their specific ecosystems. Currently, Final Cut Pro X is the only NLE that supports ProRes RAW and Atomos recorders are the only ones that will record it. Here’s hoping that changes soon.

 

Certainly, with the various brands, we’ve missed an interesting piece of technology in the mix. Have something you think is worth talking about? Feel free to share your thoughts and opinions below in the comments.

 

Zach Sutton & Ryan Hill

Author: Zach Sutton

I’m Zach and I’m the editor and a frequent writer here at Lensrentals.com. I’m also an editorial and portrait photographer in Los Angeles, CA, and offer educational workshops on photography and lighting all over North America.

Posted in Equipment
  • Athanasius Kirchner

    I’m guessing that the theoretical limit was calculated based on current materials available, because, in practice, the full readout speed of a sensor can be measured in thousands of a second – that’s the fabled global electronic shutter.
    As for mechanical shutters, some leaf shutter units achieve x-sync at 1/4000th of a second. Vertical curtain units are indeed much slower, because the parts they need to move are relatively heavy. As of today, I think it’s far more likely that sensors will become quick enough to allow for a perfect electronic shutter, than it is that engineers will develop a better mechanical unit.

  • Fritz Rutz

    These camera shutters are indeed mechanical marvels (especially when considering how reliable they are). However, I still find it odd that a theoretical limit of 1/200s is postulated when in fact all professional cameras since mid of the 80’s are surpassing this limit by offering at least a 25% faster speed of 1/250s. Therefore I would expect that a theoretical limit would be faster than the possible 1/300s reached by existing cameras (e.g. when the acceleration/deceleration of the shutter would tend to be become infinite). With the introduction of DSLR’s the fastest sync speed became in my opinion also less important because with film most photographers tried to avoid exceeding ISO 400. Nowadays it’s no problem to use ISO3200 even with APS-C cameras, therefore the need for such fast sync speed has decreased for most applications (additionally most cameras also offer high sync speed for close objects). With my Alpha 900 I try to stay below ISO400 and rely heavily on flash usage while for the X-T1 I use the range up to ISO3200 and even have no flash for it.

  • Athanasius Kirchner

    The physics get a bit mad already at 1/200th of a second. Think about it – the blades open, the sensor is exposed for 1/200th of a second, and then needs to be closed almost instantly. This means that the upper curtain must travel down very, very fast, and do so without breaking anything, or causing too much vibration (it’d be possible to achieve much higher mechanical x-sync speeds with thick kevlar shutter blades, but photos would likely appear blurry). When manufacturers used cloth-based shutter curtains, speeds above 1/200th appeared impossible. Later, when steel shutters were introduced, maximum speeds of even 1/8000th of a second became common, because steel allowed for thinner, lighter, more rigid and stronger blades. Recently, electronic motors have become more precise, and materials science has also advanced, making the somewhat faster shutters that you point out more affordable.
    From what I understand, the Minolta 9xi’s (also used on the Alpha 9) shutter is the most advanced mechanical design ever produced, incorporating innovative materials in the blades. It was also bloody expensive, apparently, because it’s never been replicated or used again.

  • baron

    OK, now I remember that when Imacon came out with the 3020 model, they put together tech from at least one other company that they had acquired. I believe they where the first to have a single shot and a px shift camera in one. That was in 2001.

  • baron

    Franck, Agree with what your saying, except Hasselblad did not develop the Pixel Shift technology, Imacon did. Imacon was a medium format digital back mfg. that was bought by Hasselblad, that is how Hasselblad acquired the tech. I purchased a Imacon and used it for years. Great tech for its time.

  • T N Args

    Yo dude. I say depressurize, you say dePascalize.

  • Ryan Hill

    Everything’s made up, man. Stay woke.

  • Nemo Niemann

    I definitely hear you when you speak of the Canon electronic shutter and multiple RAW sizes. I have lamented the loss of 1/500 sync without resorting to associated trickery (Hi-Sync, HyperSync, High Speed Sync, ad nauseum) for many years. I never realized that it was abandoned (sadly) with the shift to CMOS sensors. Thanks for filling in the blanks. As for multi-sized RAW, I use these constantly with my 5Ds and even my 1Dx (usually when shooting something specifically for the web and I know I won’t need the full 50MP or 18MP RAW).

  • Fritz Rutz

    I’m a little bit surprised about the following statement regarding the flash sync speed: “This was because most cameras sync speeds are dependent on how quickly the shutter can fully open, trigger the light, and the begin the closing process. The theoretical limit to this is at 1/200th of a second for a full frame sensor and 1/250th on a crop sensor.”
    Many film/digital SLR’s in the professional league allowing a flash sync speed of 1/250s (e.g. Nikon F4, Canon EOS-1, Nikon D5, EOS-1D X Mark II) and the 1992 introduced Minolta 9xi was the first camera offering 1/300s (with a minimum shutter speed of 1/12000s). Therefore it would be interesting to know how the theoretical limit of 1/200s for FF has been ascertained when most professional FF bodies offer 1/250s (with full open curtain).

  • T N Args

    “Debayering” is one of those unhelpful words that replaces a descriptive term with the name of an inventor. “Demosaicing” is a much more useful, descriptive and self-explanatory term.

    Or avoid made-up terms altogether, although doing so might add 1% to the length of an article, and some writers seem to think that would be an insufferable burden.

  • CheshireCat

    Unfortunately, I don’t know where it’s going. I only know where it _should_ be going 😉
    Where there are no official standards, there are de-facto standards formed by software and user-adoption.
    Unfortunately, my crystal ball is a bit opaque as far as the future of professional NLE is concerned; however, generic common sense applies.
    Companies should avoid reinventing the wheel, especially if the new wheel is worse than the old one. Otherwise they are only contributing to chaos.
    Real innovations, especially by means of open standards, do reduce chaos as developers and users converge to the best technology.
    In this case, Apple should rather focus their energies into adding full native support for CinemaDNG in Final Cut. This is the lowest hanging fruit.
    Pro users are different: you don’t tell them what to do, you listen to them and provide what they want.

  • Ryan Hill

    Totally fair points, but I don’t think HEVC has the same wide-adoption potential as ProRes RAW as a capture format (as opposed to a delivery format), if only because of the possibility that more recorders will support it. Atomos being on board already gives the format a pretty large market share. If it catches on I’d imagine more manufacturers will follow.

    CinemaDNG 3:1 or 4:1 are also great, and so far better-supported in NLEs, but also not available everywhere.

    I guess what I’m saying, and I could have made this clearer in the article, is that I want a standardized compressed RAW format available across as many cameras, recorders, and NLEs as possible, and I don’t really care what that format is. ProRes RAW isn’t necessarily the only option from a technical standpoint, but with Apple and Atomos promoting it and getting more and more other companies on board, I think it has the greatest potential to become that standardized format.

    Thanks for reading and, since you’re clearly familiar with this whole format mess, I’m curious where you think it’s going. Do you anticipate HEVC or compressed CinemaDNG becoming a standard, or do you think we’re going to be stuck with competing file formats until the end of time? I hate to be pessimistic, but I’m starting to feel like it’s the latter.

  • David Bateman

    Zach,
    I think you have to stop using Canon and start using Panasonic cameras. Most of your wish list is solved

  • mohammad mehrzad

    don’t want to start a canon Nikon war, just something that was missed in Zach’s article.

  • CheshireCat

    Disagree about ProRes RAW.
    There is much better technology, such as HEVC (H.265) All-I, which can be repurposed to encode the single Bayer planes before interpolation.
    Also, CinemaDNG is a relatively low-complexity open format available since 2008.
    The only reason Atomos/Apple are pushing it is because it allows for cheaper hardware and higher profit margins.
    Note how Apple’s whitepaper (where’s the “open” specification ?) fails to compare ProRes RAW to anything but Apple’s own inefficient codecs.

  • Paul Trantow

    Yep. And we Nikon shooters are STOKED about it. Research!

  • grubernd

    There’s too much near miss in the technical correctness (raw sizes, electronic shutter, ..) in this article for my taste. Keep up the quality, LensRentals, please.
    And except for the fenced-community codec thingy none of this is new at all.

  • Franck Mée

    Hmmmm… No.
    Hasselblad does the exact same thing as Pentax : take four Bayer shots, shifting the sensor one pixel at a time. So they also get 2 green, 1 blue and 1 red sample for every pixel, thus not needing any demosaicing.

    Olympus takes 8 Bayer shots. Getting full samples (2G, 1B, 1R) for a quadrupled resolution would need 16 shots and they only take 8, so there is some amount of interpolation in the process.
    According to this site: http://www.wrotniak.net/photo/m43/em1.2-hires.html, they take 4 shots with 1 px shifts (the same way as Hasselblad and Pentax) and 4 other shots also with 1 px shifts, the two series being separated by a diagonal 1/2 px shift. This way, they get full color samples at each position, twice the number of positions, and end up with a checker of full-color pixels. They they interpolate to fill the empty positions, just the way Fujifilm did with their diagonal-mosaic sensors (Super CCD HR and EXR).

    But the Bayer mosaic is circumvented during the shot, just like for Hasselblad and Pentax (oh, and Sony, I saw they started to follow the trend too).

  • mohammad mehrzad

    Nikon D850 also has multiple raw sizes

  • TinusVerdino

    Olympus uses stacking of full RGB images, Pentax uses stacking of a R + G + B image (with a full RGB image to cover for motion artifacts). Different methods. Also different from Hasselblad. Pentax is the only one that circumvents debayering.

  • Miguel Ángel Roldán

    Pentax k3 II, 2015

  • Miguel Ángel Roldán

    H4D
    In 2009, Hasselblad introduced the H4D product line.[31] The current H4D products include H4D-31, H4D-40, H4D-50, H4D-50MS, H4D-60 and H4D-200MS. 5th generation integrated DSLR.

    Model Sensor ISO range ISO range
    (with Phocus) Capture speed HC lens factor Eq. focal length Display Storage Video recording
    H4D-50 36.8 mm × 49.1 mm, 50 megapixels, 16 bit
    50–400 50–800 1.1 second 1.1 28 mm 3″ CF
    H4D-60 40.2 mm × 53.7 mm, 60 megapixels, 16 bit
    50–400 50–800 1.1 second 1.0 28 mm 3″ CF
    H4D-200MS 36.7 mm × 49.1 mm, 50 megapixels, 16 bit
    200 megapixels in multishot mode
    50–400 50–800 1.1 second 1.0 28 mm 3″ CF None
    H4X

  • DrJon

    I think you mean “Electronic Global Shutter” as the World is full of CMOS cameras with Electronic shutters allowing very high shutter speeds, they just don’t work well with flash due to the readout speed.

    A lot of ILC cameras sensors are dual-ISO too. You can spot it as the DR curve has an upward kink somewhere near ISO 400 (a kink at ISO 12800 is just them applying noise reduction to the Raws, so cooling them somewhat).
    Examples:
    http://www.photonstophotos.net/Charts/PDR.htm#FujiFilm%20X-H1,Nikon%20D7500,Nikon%20D850,Sony%20ILCE-6500,Sony%20ILCE-7M3,Sony%20ILCE-7RM3,Sony%20ILCE-9
    (You may need to disable a lot to see clearly, plus not exhaustive.)

  • Franck Mée

    Hi Zach,
    as much as I love Pentax being credited, they actually did not originally develop Pixel Shift. Hasselblad’s Multi-Shot is the exact same thing and predates Pixel Shift Resolution by a few years ; Olympus also launched a High Resolution mode on their E-M5 II, one year before the K-1, with a similar logic but working on half-pixel steps (8 shots).
    Also, Pixel Shift is actually supported in Adobe Lightroom, PhaseOne Capture One, RawTherapee… I don’t know about the paying ones, but RawTherapee supports using all four images or extracting only one (and you can select the one you want), and the effect on moire, aliasing and noise is really noticeable.

Follow on Feedly