Gogo 2Ku media flight drives discussion on Bits, Bytes, Benjamins

When it comes to building the fastest inflight Internet service there is more at play than just the technology. This reality was thrown into stark relief on 12 November as Gogo put its new 2Ku system on display for the media, inviting a couple dozen people to Chicago to fly on the Boeing 737-500 test aircraft and put the service through its paces.

Following two test flights the general consensus was that the service was massively better than the ATG options the company has installed on 2000+ commercial aircraft today, but performance was inconsistent, with travelers seeing wide variation in speed test results and actual application/browsing performance in-flight. Some identified as having a “great” or “good” #PaxEx; others cited a “mixed” #PaxEx.

Following the test flight RGN spoke with several executives about the performance of 2Ku to gauge the company’s view of the results. CEO Michael Small remarked that the test flight consumed nearly 11 GB of data, more than double a typical transcon flight and with far fewer passengers on board. That is an impressive haul, but he also acknowledged that the limiting factor on the transcons is typically the capacity available to the aircraft, not the demand from passengers.

Kontron NBAAStill, some back of napkin math suggests that the 11 GB number represents approximately a 25Mbit/s consumption rate, far lower than the 70Mbit/s number the company has advertised. That seemed to be a tremendous miss in terms of performance until Small later noted that the link for the test flight was capped at that rate.

The test aircraft was only allocated 25Mbit/s for the demo flights and it used every single bit of that data pipe.

“We didn’t do anything to try and shape traffic in any way and we kept the configuration at production level to see what happens when you bring some of the most discerning journalists in the world to get on a plane to try and ‘break’ the network. We see this as extreme focus group testing and we can say we were very pleased with the results,” said Small in a follow-up interview with RGN contributing editor John Walton.

“Just like any network, you’re going to see variations depending on what people are doing at the time of the tests. On the flight, we had 50 devices, which pulled in close to 11 GBs of data, which is what we expected given the way the system is engineered today. Our observations were that people were able to stream high definition movies and conduct a lot of really high bandwidth operations. We even understand that someone was downloading a file from BitTorrent in excess of a GB of data while simultaneously video conferencing.”

When pressed further on why the system was handicapped for the demo, Small brought up the obvious but oft-overlooked factor: cost. Providing that 70Mbit/s of capacity to the 500+ planes expecting 2Ku in the next few years is a proposition which is unlikely to be amenable to the airlines and to Gogo relative to the costs of delivering that service vis a vis transponder time and passenger pay rates. And so the test flight was set to emulate something much more akin to what Gogo expects travelers will see on board as 2Ku enters commercial service.

“We are unbelievably excited about where 2Ku is at right now,” Small told RGN. “Our tests and the media event have showed us we have a 10x improvement from where we were. That makes us happy and we think it will make our airline partners happy, too. No other global solution comes close. And, 2Ku will only get better with the advent of [high throughput] HTS satellites and other optimizations.  This is the beginning with a lot of runway ahead on this technology.”

Separately, there were discussions about other technologies in play on the 2Ku service, including future upgrades to HTS satellites and on-board technologies like the modems and ACPUs. Each of these improvements will offer incremental benefits in performance and cost, while the initial 2Ku launch is a step change in the industry. But given the Ku capacity available to the aero market today (and the costs of that capacity) it seems that expecting 70Mbit/s to every 2Ku plane in-flight is, at least for now, not something passengers should expect.

4 Comments

  1. Jack

    10 times improvement to where we were (assuming ATG4 or current Ku), that would imply >100Mbps, which is definitely not the case based on this report.

    • Seth Miller

      Agreed. The 25-30Mbit/s number would be 10x ATG which is a huge improvement but it comes across as less than the promises/hype leading up to the flight.

    • Seth Miller

      ATG is ~3MBit, ATG4 is ~10MBit. The test flight was allocated at 25MBit.

      Everyone I spoke with indicated it could have been more but the company chose to stay with the lower number.