Broadcom has recently demonstrated that its smartphone modem can communicate with a “live” 300 Mbps LTE commercial network. Although demonstrations using a similar technology have been done in the past, Broadcom says that it is the first to do so on a “live” network (in Finland) which means that a cell tower used to serve actual wireless customers was upgraded and used for the demonstration. The test device itself was only yards away from the tower to achieve the best speeds, but this is not unusual for that kind of demonstration.
I recently met with Scott McGregor (Broadcom’s CEO) and for him, this milestone shows that Broadcom has basically caught up to Qualcomm over the past two years since both companies have now announced their Cat6 LTE support within a week from one another. At this point in time, these are the only two companies that I know of that have reached this stage of LTE support.
The modem is the key element in the phone processor business
Broadcom knows that having a great and easy to integrate LTE solution is key to success. In fact, it is why Qualcomm is enjoying continued success at the moment, and there’s no way around it: no matter what one’s CPU or graphics performance is, the modem will drive that business – at least for the time being.
When a device manufacturer launches a new handset, being able to hit as many countries/markets/carriers as possible within the shortest possible amount of time is critical.Failure to do so means that the marketing efforts will be diluted and a late arrival means that a competitor will show up with better hardware and take customers away. Given that we are on a yearly upgrade cycle, with new handsets arriving every quarter (if not every month) the clock is ticking.
To wins big contracts, modems need to be qualified worldwide
To achieve the feat of launching worldwide within the span of a few weeks means that whatever modem is used in the handsets has to be tested and qualified by the carriers and the wirel
To achieve the feat of launching worldwide within the span of a few weeks means that whatever modem is used in the handsets has to be tested and qualified by the carriers and the wireless regulatory bodies of each market. Short of that, the only thing that is harder is to have a stable product in the streets. At the moment Broadcom says that its LTE products are qualified in 20 countries and 40 carriers, and they expect this to grow further and faster during 2014. Since it has had success with its 3G lineup (see above), Broadcom hopes to do the same, and more, with its LTE products.
When I asked Scott McGregor if he thought that Broadcom could support a partner launching a handset in 80 countries by the end of 2014, he answered “no problem”, and added that his company already secured the qualification with the two most demanding carriers in the world: AT&T and NTT DoCoMo.
It’s interesting to see how Broadcom thinks about the “high-end market” in terms of modem while end-users may think about CPU & graphics speed. However, Broadcom also looks at this from a “wireless efficiency” point of view: Carriers are very eager to switch to faster LTE networks because those are much more radio-band efficient. This means that using the same spectrum they already own, carriers can serve more customers at higher speeds. This opens an opportunity for Broadcom to compete mostly based on two things: 1/ modem performance 2/ platform ease of integration.
Make it easy for customers to deploy
Broadcom has been very busy building reference phone designs that partners can use to get to market very quickly, and there’s nothing like solving someone else’s problems to make money. This is a good way to make quick inroads into the volume smartphone business which is operated by companies that want a great platform that can be used virtually “as is” without having to invest heavily in hardware and software engineering.
Broadcom is gathering strength and contracts, especially in emerging markets: a lot of Samsung mid-range phones are powered by Broadcom which predicts that the trend will continue in 2014. In the west, the question is whether or not Broadcom will end up competing on the very high-end SoC business with Qualcomm. It’s not a priority at the moment, and Broadcom does not yet have the graphics and CPU performance to do so, but if Broadcom is successful in its mid-range business, that would be the logical next step.
It twill be interesting to see how Intel and NVIDIA will manage their LTE business in 2014 and beyond. Intel has a very compelling story with its Infineon subsidiary releasing LTE products, and NVIDIA has a programmable modem that looks great on paper, but is yet unproven in worldwide markets.