There can be little doubt that most of the strategic focus of the AOC manufacturers will continue to be on the data-center market for the next few years. Here they will find a relatively mature market and a fairly clear path for further penetration of the market. Thus we expect to see a growing range of AOCs equipped with the appropriate IB, Ethernet, and Fibre Channel connectors.
AOCs Outside the Data Center: Market Potential?
Meanwhile, the market for AOC sales outside of the data center is looking increasingly attractive for a number of reasons.
Higher data rates needed outside the data center: The personal computing, consumer electronics and digital signage markets are all moving down the road towards higher data rate interface requirements in a manner similar to that of the data center. The reasons for this in both the data center and these other markets are all the usual ones that have driven data rate requirements up for several decades. For these newer areas that are being targeted by AOC vendors (or soon will be) a subsidiary factor is that as video shifts to HDTV, UHDTV and 3D the data rates for uncompressed formats become harder to handle using copper interconnection.
Addressable markets are very large: The addressable markets for AOCs in this space are potentially huge. The numbers of PCs, computer monitors and televisions installed/shipped are higher by orders of magnitude than the number of servers shipped.
AOCs are user-friendly fiber optics: Both these drivers suggest that fiber optics could have a growing role outside the data center. Such a fiberization could be delivered in a number of ways, but the users of products in the personal computing, consumer electronics and signage sectors are likely to find the user-friendly nature of AOCs highly attractive.
Three Reasons for Caution in the Non-Data Center AOC Market
While this all seems fairly clear cut, we think that it will be quite easy for AOC vendors to overestimate how many links can be sold in these new markets. But, in fact, there are plenty of reasons for caution.
Market is not entirely novel: AOCs have been sold into these non-data center markets for quite a few years and have never moved beyond niche status. AOCs are, for example, hardly ever found in consumer electronics stores—they are more likely to be bought on the Internet or through specialist stores serving videophiles.
In this context, the AOCs that we are interested in in the context of this report are typically sold as optical extenders, although not all optical extenders are AOC-like. Optical extenders of any kind are used in quite a small minority of installations at the present time.
Quality of suppliers: For the most part, AOC suppliers that have started supplying to non-data center applications have been second-tier firms and the most common product currently available for these new markets is the HDMI extender AOC.
We suspect that these and other similar AOC products will start to become mainstreamed in the very near future. The arrival of Corning in this space with AOCs for USB and Thunderbolt may turn out to be a sign of times. Also, while just a few years ago, AOCs for HDMI and similar products tended to come from small Chinese companies, some of these companies themselves are now growing up into “serious” firms with a noticeable impact on the market.
The copper challenge to AOCs will not go away any time soon: As in almost every sector of the data communications market, there is no chance that AOCs (or other types of optical link) can take a really large share of the connectivity market for consumer electronics, personal computing and digital signage. Indeed, the arrival of USB 3 would seem to keep things firmly in the copper sphere. Similarly, we note that Intel dropped the idea of a optical version of Thunderbolt a few years back.
Despite all of the above, we do expect the market for consumer/personal computing AOCs to grow, because, as the new high bandwidth video and data standards become more ubiquitous, it will reduce the distance at which fiber optics makes sense. In this regard, the non-data center market for AOCs resembles the data center market—when data centers were all 10 Gbps and 1 Gbps there wasn’t much need for fiber optics. Now that 40 Gbps and 100 Gbps have been mainstreamed, fiber has got many uses in the data center once one gets over a few meters,
Product Differentiation and Pricing in the Non-Data Center AOC Market
The analysis above suggests that the non-data center AOC market is in a state of flux and is being pulled in a number of different directions by several countervailing forces. However, an open question is how AOC suppliers can make money in this space; it is hard not to make AOCs that are in some sense a generic product.
One way to distinguish AOCs is through better technology, but this may be easier said than done. Luxtera built AOCs based on SMF fiber and CW lasers, before this line was sold to Molex. But most other AOCs that are likely to be sold into the markets with which we are concerned in this report are going to consist of standard interfaces such as HDMI or PCIe attached to SMF.
In our opinion, the only comprehensive way that AOCs can be distinguished in the marketplace is through appropriate marketing and pricing strategies. The pricing aspect of all this is quite easy to understand—both personal computing and consumer electronics are markets where pricing matters. This is true in the data center too, but here the quality and performance might trump low prices. In any case, no supplier really wants to compete on price alone and the obvious way to reduce the pricing issue is to introduce a strong branding strategy.
In the context of the markets considered in this report, branding may take several forms. One option would be for the emerging Asian firms to adopt and build brands which add perceived value to all their cables—not just the AOCs. Another option is for AOCs to adopt the brands of famous names in the computing and consumer electronics spaces. One can thus imagine Dell AOCs or Samsung AOCs becoming available (perhaps at a premium) through the usual consumer electronics retail outlets.
This also suggests another part of the marketing plans for AOC makers targeting the consumer electronics and computing sectors. They will have to build strong supply chains that end with the big consumer electronics retailers—online and off line. There is nothing magical about this. It simply takes lots of negotiations, which in turn takes lots of time!
The bottom line then is that CIR sees the sales of AOCs into the consumer electronics, personal computing and digital signage markets as considerable opportunities for the future. However, our expectation is that penetration of AOCs in these space will grow slowly but steadily as new generations of high-quality video and high-speed data interfaces make the cable reaches that can be achieved without active cabling shorter.