Airgoing: More Vowel Movements

This week, Airgo had another one of its quarterly vowel movements (with a "V") saying that, when it comes to Wi-Fi,  "it's all about sustaining high performance....the game has changed," Airgo said.

The Wi-Fi game has never changed. It has always been (and always will be) about reliable performance.  We're building a big business around making Wi-Fi predictable. 

Users really don't care about True MIMO, False MIMO, silicon-injected MIMO, smart antennas, spatial multiplexing, orthogonal frequency division multiplexing or 802.11(pick a letter and put it here). They care about their applications working anywhere, anytime - dependably.

Sure they want to go far and fast, but when it comes to multimedia, people just won't tolerate wild fluctuations in performance in their music, video or phone calls - neither will the providers delivering such products or services. 

Check out the chart on the left (click it to expand).  This picture is worth more than a thousand words (but it's based on our own tests so be somewhat suspicious).

Making Wi-Fi predictable is an incredibly difficult thing to do - it's rocket science stuff.  And something that, quite frankly, can't happen with just more gates (i.e. chips). It really takes a systems approach.  The best way to ensure predictability is to control the multi-path and mitigate interference, whether it comes from the neighbors, the microwave or the cordless phone.  This is exactly what we do but without having to purchase expensive and not-ready-for-prime-time chips.

That's why we decided, long ago, to take a more flexible approach to making Wi-Fi better by developing technologies that add predictability on top of ANY Wi-Fi chipset. We chose 802.11a/g first because it's real, and it works.  And when we get to 802.11n, watch out (we say hoping, hoping, hoping).

And 802.11n is going to be great - going to be.  The problem is that the emerging standard is ...well....emerging. The IEEE 802.11n standard is "still in the air" as the task group received 12,000 comments on the Draft 1.0 proposal that was accepted as a working draft in March. And early tests of pre-N products have delivered miserable results.

Think about it. Why would anyone want to purchase Airgo chips now when they will be thrown away soon?  For Airgo, the conundrum is that carriers’ margins can't support expensive silicon (an Airgo-based router is typically 3 to 4X the bill of materials of a standard 802.11G product) not to mention that carriers MUST provide standards-based solutions. 

This leaves consumers and reviewers, who advocate the biggest throughput number, forcing Airgo's application-unaware silicon into unstable but heroic (at very short distances) data rates.

Today, Airgo gets most of its "gains" from transmit power not from MIMO. And even with more power, they can't sustain enough throughput for an HD video stream throughout a home.

George Ou had some excellent insights on this whole topic (and we're not sure he actually even likes us).  Think of it this way.  We're the tires, handling and suspension of a car. The chips are the engine (and the bulk of the cost). 

So what's our long-winded point?

Adapting and controlling the wireless environment - an environment that by definition is subject to constant change - is what's most important. Adapting to this constant change is best done in smart software that sits up the stack.