Moore's law says that computer processors will double in complexity every two years. That has largely been interpreted as "the number of transistors per inch on a chip will double every two years."
Although this is an overly simplification of what Mr. Moore said back in 1965, it has been true for the last 50 years or so. However, a possible end to this trend may be in sight when 1nm node manufacturing becomes available around 2028.
Although science may find a way around this barrier with nano-tubes or quantum computing, a prudent company like Intel (INTC) has to take seriously into consideration hedging against the possibility of a not-so-bright future.
"But why?" you may ask, "Intel can just keep using the same technology over a longer period of time." Although this is possible it is not a good scenario for Intel. If transistor manufacturing stops advancing, or new techniques become too expensive to implement, then sooner or later everyone will catch up with Intel and CPUs will become a highly commoditized business with low prices and extremely low profit margins.
And that is the main reason that Intel is constantly trying to expand in other seemingly unrelated markets. Its latest venture seems to be an attempt to revolutionize the TV industry.
What we know for this venture is that Intel will make a feature-rich set-top box for consumers that will complement an internet based TV subscription service.
From what we know the set-top box is likely to have an interface similar to Comcast's X1 service that Intel helped to create and is way ahead of the competition. Furthermore the device will have a camera on-board that will allow it to identify who is using the TV and make appropriate suggestions based on his/her taste. It will also be handy for Skype calls from the living room with the whole family, gesture or eye movement recognition and many other things. Finally to avoid people freaking out about privacy Intel will give consumers the option to disable the camera and\or use the embedded physical shutter to cover it too.
The service will be accessed through the internet and the consumer will be able to use it across all of their devices, from phones to tablets to regular PCs and TVs. Moreover Intel claims that the channel bundles that will be offered will be more flexible than the ones currently offered by cable providers. This may be implemented by allowing the consumer to manage a portion of the bundle they choose adding and/or subtracting the channels they want.
Until Intel actually launches this service (within 2013 according to the company) we won't really know anything for sure. However, according to adage.com test users that have used the set-top box say it is:
"a significant advance over any existing cable or satellite platform."
"the kind of industrial design that makes you not want to hide it in a cabinet."
Another feature that the service will have is a 7-day program archive across all channels that will make possible to watch a previous day's program as if it was broadcast today. This feature will also help in creating a moat around Intel's service. This is because of the massive scale that is needed for one to keep all the content properly archived and stored.
If Intel manages to crack the TV market that many others have failed to crack it stands to benefit from a wide variety of things. It will benefit from the chips and the hard-drives (SSD maybe?) that will be inside the device and the subscriptions for the service. I'd like to point out here that Intel won't compete as a cheap-priced provider but as a quality driven one.
Summing up, if Intel pulls this through it will have succeeded in starting to diversify against a possible demise of its main business while increasing substantially its revenue and profits. There are very few companies out there that have this kind of long-term thinking, making them excellent investments for long-term shareholders.