What we know about Red’s Hydrogen phone

At the end of last year, chip manufacturer Qualcomm launched its new Snapdragon 845 device, intended for cellphones, at a technology summit in Hawaii.

While handheld devices are an important way of delivering moving pictures to viewers, the release of a new chip might have seemed worthy of little more than a footnote in the world of film and TV.

However, this happened at more or less the time that interest in Red’s Hydrogen smartphone began to peak ahead of the first quarter of 2018, when the company had anticipated releasing the phone. That release date is not set in stone, with the company stating that “specs and delivery dates can change anytime and for any reason,” and Red founder Jim Jannard telling users of the Red Forum that there had been “unprecedented” carrier support, but that in turn had resulted in prolonged testing. As such, he said the ship date for carrier phones is likely to be “sometime in the summer”, while unlocked pre-orders will ship before that.

Whether or not Red actually make their first-quarter release is a matter of conjecture at this point, but what’s certainly true is that the phone is run by a device that the company calls the 835x. The x-suffix isn’t part of Qualcomm’s standard nomenclature, but we might reasonably assume the Hydrogen phone’s core electronics are fundamentally based on the previous version of Snapdragon technology.

That, in itself, is entirely reasonable. The people who use Qualcomm’s products to make phones would have received advance notice of the new device – even samples – long before the public announcement. Even so, we wouldn’t expect that a fundamental technology like this announced in one quarter to be available at retail during the following quarter, especially in a device such as Red’s phone, which is likely to be manufactured in fairly small numbers in smartphone terms. However, the feature set of the Snapdragon 845 throws some interesting light on Red’s phone and how it compares to what the rest of the mobile industry is doing.

The 845 is intended to be about 30% faster than previous generations while consuming 30% less power. These are general improvements which don’t tell us anything about specific feature intent. More specifically, the promotional material talks a lot about features for video recording and playback, including features intended to support 10-bit, high dynamic range pictures, 4K, high frame rates, and, when used in a headset, very capable rendering of VR environments.

That doesn’t mean that Red’s phone will be significantly less able to do any of these things than something like a Samsung Galaxy S9, which was on show at Mobile World Congress and boasts a Snapdragon 845.

The Snapdragon devices include a lot of traditionally-separate devices integrated onto one chip, and the eight CPU cores on the 835 and 845 are similar in performance. It’s naturally possible to include additional hardware external to the Snapdragon system-on-chip to add more capability and we don’t yet know what that x-suffix means. What this does suggest is that some of the VR, AR and imaging applications that Red seem to have in mind for Hydrogen are things that the mainstream mobile industry has also been thinking about for some time.

Display tech

The unique features of the Hydrogen phone are clearly its display, which we’ve previously discussed in as much detail as can be gleaned from historical press materials and patents, and the modular expansion system.

To recap briefly, we might cautiously describe the display as autostereoscopic, in as much as it’s designed to display images which stimulate human stereoscopic depth perception without requiring special headgear on the user, as well as providing some degree of variability in the point of view. You’ll probably be able to move around the display and look at the displayed object from different angles, within certain limits. It has been described as holographic, although the word “hologram” has historically been used to refer to a recorded pattern of optical interference fringes. Based on available information, which is not complete, the display actually seems intended to produce a sparse lightfield somewhat like the ones recorded by multi-camera arrays, though probably of higher angular resolution.

This is likely to lead to at least some small visible discontinuity – steppiness – when the observer moves slowly relative to the display, but it is unique. It is also likely to be better, in some ways, than eye-tracking approaches. These alter the angle of view based on the relative position of the observer and the display by using a camera to track the observer’s position using image recognition. That works, and avoids visible steps when performing real time 3D rendering, but it only handles one viewer at once, and tracking based on image recognition may not be entirely reliable.

Modular approach

The other feature that’s interesting, though not entirely a new idea, is the modularity.

There is so far talk of a “cinema camera” module including both the imaging sensor and hardware to record the codec used by its existing cameras. That isn’t something that even the mightiest Qualcomm device could handle on the fly, but adding new hardware would make it feasible, creating something perhaps a little like a higher resolution version of Blackmagic’s Pocket Cinema Camera. Whether a cellphone, even one with a comparatively mighty 4.5 ampere-hour battery, is really an ideal platform for serious camerawork remains to be seen, and we have no details of the camera. Still, it may be among the first phones to really implement a modular expansion system, assuming a worthwhile ecosystem of modules becomes available.

Strategic intention

The phone itself is one thing. The real question is Red’s strategic intention. With a price tag well over $1000 for either the aluminium or titanium versions, they’re clearly not making a grab for the mass market, any more than the similarly-priced 256GB version of the iPhone X.

What’s crucial is not so much the display but the ability of the camera array on the phone (or any camera array) to shoot material for that display, the ability of editing applications to handle that material, and how easy it is to distribute. After all, even the early Leia displays would require something like 64 times the data of a conventional 2D video stream for equivalent resolution, and there are limits on how much clever codec mathematics can help.

It is also hard to avoid the reality that while the display is novel, 3D display technology has been repeatedly tried, over decades, and has repeatedly struggled.

Improved technology has been just as repeatedly mooted as a solution to this problem and, while the level of novelty in the Leia display is high, history suggests that improving the technology does not significantly affect the level of demand. The stereo 3D technique used in feature films of today is vastly superior to that used in the mid twentieth century and it has still been only moderately successful. At launch, with exactly one device capable of displaying the material in question, we can’t expect a huge market for Red’s new 3D format, genuinely new as it is.

And for that reason only, it’s reasonable to assume that Hydrogen is intended to be the beginning of something aimed much more squarely at mass adoption. One expensive, high-end, limited-market phone capable of shooting and displaying autostereoscopic material might be described as a curiosity at best. To be anything other than a toy, Hydrogen really can’t be the beginning and the end of the matter. Whether or not Red itself has any fixed idea of what the future holds is unknown, but it wouldn’t be surprising to see compatible display and acquisition devices emerge too.