1. 11

  2. 21

    The Air Force runs a newer version of the software known as “Block 3i” on its F-35s, and gets roughly half the time between significant software glitches—though F-35 program chief Lt. Gen. Christopher Bogdan recently told reporters that a new version of Block 3i software appears to have tripled in stability during tests, going up to 15 hours without a serious software issue.

    Shouldn’t the stability of software on a fighter aircraft be measured in units longer than hours?

    1. 9

      You go to war, I mean measure, with the units you have, not the units you want.

      1. 3

        Perhaps? However some airborne radars have MBTF’s in the single to double digit ranges.

        Honestly for as much of a boondoggle as the F35 may be, I think people forget this isn’t uncommon. Example, the P38 Lightning had 2 engines, but the first models only had one generator. What happens when that fails? Well the engine that had the generator would get its propellor feathered and you switch to batteries for the other. So what happens when the batteries die? Well, most of the time the plane crashed.

        This is just one example from another fighter aircraft that people may know of. I knew an airline pilot and he just noted its pretty common to be rebooting things mid-flight (like the radar). I suppose this whole reply amounts to: what should the stability of flight software be when its still being built?

        1. 2

          I think these are somewhat separate issues. MTBF for a single engine failure on a P-38 was at least days, if not months or years.

          Modern airliners have to reboot software such as weather radar, but generally this does not interrupt the flight itself. A modern fighter jet relies much more on its software for survival, from radar to IFF detection, to the fly-by-wire system (generally more complicated than an airliner’s). The reliability of these components could be the difference between life and death.

        2. 1

          I wonder how many non-serious issues (whatever it means) they had in those 15hrs

        3. 10

          It’s almost as if the F-35 program is a highly inefficient, totally crooked Keynesian stimulus program disguised as a mismanaged software engineering deathmarch.

          1. 11

            I wish our inefficient, crooked Keynesian stimulus programs could produce something even slightly useful, rather than dysfunctional weapons for a style of warfare that’s gone and by all accounts never coming back.

            1. 3

              a style of warfare that’s gone and by all accounts never coming back.

              I’m not sure that this is an accurate or correct statement.

            2. 5

              Almost. F-35 components are only produced in 45 of the 50 states.

              1. 4

                There’s nothing Keynesian about the F-35 program.