BLOG

Apple: 3DS VCSEL Learning Curve Edge

November, 2018

While fibeReality continues to see downsides for the 3D-Sensing (3DS) VCSEL component suppliers, themselves, Apple’s two-year head-start over any other company in the world in developing its own IP with these solutions gets an inadequate amount of attention, not only for the iPhone product line, but as a probable cornerstone technology in the future for other new types of offerings that it undoubtedly has on internal drawing boards. The vendor already has three different VCSEL chips being used in its high-end phones: the Dot Projector (DP), operating only towards the face; the illuminator, and the proximity sensor. Apple will likely put in another chip on the other side to do a Time of Flight (ToF) measurement for objects that are further away, just about assuredly, a second DP. Moreover, fibeReality is starting to have doubts about any kind of widespread use of Edge Emitters (EEMs) in smartphones because of the inherent advantages with the VCSEL approach.

Naturally, with 3DS VCSEL arrays just beginning as a new market, no one can even come close to knowing exactly how it is going to develop, and it is prudent to be careful about hyping later, anticipated applications, as the hurdles in creating such products still need to be overcome.  Of course, for the extremely long term, there has been wide discussion that the use of these sensors will enable driverless cars. However, we have cautioned against overly aggressive prognostications for the arrival of this app in any kind of ubiquitous manner.

Nevertheless, we believe that Apple is betting that the use of these VCSELs for facial recognition is only a quite conservative first step (and now ostensibly, in an even more moderate way) in later employing them for cheap, very precise 3D machine vision, as part of an entire, potential ecosystem. For example, we foresee Apple offering robots, which will roam around homes, using the same technology.

Moreover, in fibeReality’s opinion, Apple has an opportunity to get even further out on the learning curve with 3DS, because, in general, the other smartphone competitors mainly seem clueless and lost as to how they will enter the market, apparently being stuck in the mode of just sitting back and watching for now. We think that at least some of them have just putting a camera on the phone under consideration to recognize faces, and it may be only good enough for 1 in 500 or in 5,000 people being able to open it up, instead of one in ten million, but with the hope that their customers would not realize the difference. Obviously, the face ID offered by Apple feeds into buyers putting everything on Apple Pay, and it is more secure, along with avoiding the need for a password, which cannot really be done with a regular camera.

While the other smartphone providers are interested in ToF, they are unable to go in the same direction as Apple because they cannot from an IP perspective. Any engineer who has an intimate understanding of the technology knows how hard it is to just begin walking into this particular area. The depth of expertise and the large amount of vertical integration required does not lend itself well to the distributed nature of most cell phone manufacturers.

We have talked about AMS being well-positioned to be competitive in this space with the Android makers, because of the experience of the former with lasers integrated with some fairly complex optics. Yet, although the other smartphone vendors have engaged in some work with DP technology, they evidently are not sure about the exact direction.

Concerning EEMs, our latest intelligence indicates that they do not have all the advantages of power consumption and use of space that Apple has with its current technology. (In fact, during its last and next-to-last quarterly earnings conferences, Lumentum did not mention EEMs at all, and only talked about shipping VCSEL arrays to Android customers during the call this month.) Also, Apple’s method can use a rolling shutter camera, which is relatively slow and low-cost, while the other devices require global shutter camera (CMOS), which is a little more expensive. Although in the long term, it should be noted that this last benefit may not matter.

As always, fibeReality does not recommend any securities, and this writer does not invest in any companies being analyzed by us.

To follow us on our totally separate, quick update company blog, which is exclusively on fibeReality’s LinkedIn page, please click here.

[written by Mark Lutkowitz]

 

Share Article

Add Comment

 

Back

 

Read Comments