back to article F-35 flight tests are being delayed by onboard software snafus

F-35 fighter jets are running so many different versions of their core software that a US government watchdog has warned of knock-on delays to flight tests. Core software aboard the Mach 1.6 stealth jet is in such a state of disarray – with aircraft at different US military bases running different versions of it – that the US …

  1. steelpillow Silver badge
    Trollface

    Still, one day...

    A plane isn't like a social networking site, you can't just throw up some fashionable bling and spend the next three years getting it to work properly. The F-35 has a highly-integrated flight software suite of unique complexity. What did they expect with the first couple of year's cuts, communications grade or something?

    1. 8Ace

      Re: Still, one day...

      One the other hand you'd still expect a lot more from something which has had $1.5 Trillion thrown at it.

      1. Anonymous Coward
        Anonymous Coward

        Re: Still, one day...

        you'd still expect a lot more from something which has had $1.5 Trillion thrown at it.

        Not me. The F35 programme is so overly ambitious in all dimensions - capability, technology, performance, mission, control systems that I'm surprised it hasn't cost more, and that they've even got a handful into limited service. And complex software is rarely reliable or easily adapted, so we can forsee that it will be a problem throughout its service life.

        1. Anonymous Coward
          Anonymous Coward

          Re: Still, one day...

          Which is truly hilarious when you remember that its original purpose was to be a cheaper alternative to the F-22.

      2. Dave Hilling

        Re: Still, one day...

        1.5 Trillion is the expected cost for the whole 50 years of operation. The plane has its issues, but people throwing FUD around make me laugh. I spent 20 years working on aircraft in the military and all the haters have no idea that every plane goes through this to some level. Even 40-year-old F-16s and F-15s have issues and are constantly being updated. It takes time to upgrade software etc. Mission comes first and the benefits of a sub-version of a software may not be worth the downtime vs continuing training etc. The plane should be more complete you won't get an argument from me, but we are talking about a vehicle that is the most technologically advanced airplane flying and yet even with that its still at least 7 years out of date of current tech. Most people have no idea what it takes to keep something like this going and while some of the critisism is warranted most is not. There are quite a few flat-out falsities in the article that I wont even go into, but all this planes problems are nothing new, its just they are so much more public than they used to be.

        1. BebopWeBop

          Re: Still, one day...

          I take your point having been involved in three extremely difficult projects - we were using a relatively formal approach to their mathematical analysis don't all boo - it proved its worth many time over (and we were well paid on the results), typically does so when used appropriately) but so many concurrent versions in the field - that is never going to be a good idea if you expect some convergence/learning.

        2. Kabukiwookie

          Re: Still, one day...

          Mission comes first and the benefits of a sub-version of a software may not be worth the downtime vs continuing training etc.

          Yes, I can certainly see how not being able to fire your missiles in a combat situation is not worth the time to get right the first time around.

          1. Dave Hilling

            Re: Still, one day...

            The plane is not considered combat ready so yes full ability to fire missiles is not required yet. It can fire some of its planned missiles just not all. Integration of ALL the missiles is not required for training. Seriously its not as easy as all the people have never worked on such projects think. There are symbology changes, missile specifications (range speed, inflight updates from the radar etc), Its a lot of work to integrate a new weapon. It took YEARS for the F-22 to be able to use the AIM-9X these planes are an improvement over older designs which often required replacing fire control computers with NEW/UPGRADED computer. There is still some level of that in these planes, but they are much more upgradeable than past designs, but then that takes a lot of programming, testing, and then bug fixes.

        3. Milton

          Re: Still, one day...

          "There are quite a few flat-out falsities in the article that I wont [sic] even go into ..."

          Funny, I'd have thought you'd be falling over yourself to list them, if they're "flat out falsities".

          Some posters defending this project seem to be arguing with appeals to authority "trust me, I worked on this/that ..." and "You have no idea what's involved ..." rather than addressing the actual history of the F-35 development or the simply enormous litany of failings, failures, faults, cost overruns, magical accounting etc which are a matter of public record. It's true that previous planes have had teething problems, even the much-admired teen series like F-14, -15, -16, -18 and the celebrated A-10. But none endured such a hellish journey as F-35: not even close.

          And it's notable that the planes I've mentioned were all designed, built and operated with a lesson in mind. That lesson was the F-111.

          The F-111 was propagandised as the One Size Fits All combat plane, a multi-purpose multi-role flexible aircraft that could fulfil all necessary combat missions better than any previous dedicated aircraft, and would save oodles of money because of universal adoption and standardisation.

          Does any of that sound familiar?

          The F-111, of course, was an expensive disaster, setting the west's combat air power development back by at least a decade (yes, the UK's idiot, lying politicians got sucked into that disaster too).

          Foolish and gullible as they had been, senior military and politicians resolved to act upon the lesson, realising that it is indeed far better to have three planes that each do a specific task brilliantly well, than one which does everything badly. The result was a series of unparalleled aircraft designs which have worked well for nigh on 50 years. For merely one example, ask any experienced infantry grunt which plane he'd prefer to have providing close air support when he's beset on all sides by hairy foes—it'll be A-10, not F-35: the latter is simply too limited, too fragile and too expensive.

          I won't list again all the manifold warnings and failings of F-35 because, unlike the "flat out falsities" they are matter of record for anyone with a browser.

          The Russians and Chinese are not cowering beneath their beds over the F-35 designs (the entirety of which they appear to have had on their computers for years now). They are delighted to see the US and UK pouring countless billions into an inferior aircraft, so expensive that there will never be nearly enough of them, so fragile that they will fly rarely, so dependent upon their vaunted "stealth" that Russian and Chinese engineers can almost cry with laughter as they dust off and improve suites of technology (multiband active/passive; IR; optical; acoustic; with multidomain integration), some of it actually quite old, which renders that stealth frequently useless¹. For them it is even more incredibly, hilariously welcome than watching the UK piss billions away on a Trident system it can't use, or float its Great Big Targerts ("supercarriers") without any aircraft. (Or escort ships. Or a modern operating system. Or—)

          General and politicians forgot, or more likely were too arrogant, to heed the lessons learned by a previous generation and were, once again suckered by the lies of contractors and tempted by the prospect of congressional pork. And, just as with F-111, they have betrayed their nation's defence, again, and this time probably for much longer than 10 years..

          ¹ Useless, that is, when merely opening the weapons bay doors; or carrying a fuel tank; or a missile; or encountering a stone flake on the runway doesn't render the stealth already useless.

        4. grumpy-old-person

          Re: Still, one day...

          John Boyd must be rotating extremely rapidly in his grave!

      3. Steve Davies 3 Silver badge
        Coat

        Re: Still, one day...

        Probably running 'windows for Planes' and can't handle 0mph and still airborne.

        Mines the one with a 'Fly Navy Harrier' parch on the sleve. {worked on them in the 1970's}

      4. Anonymous Coward
        Anonymous Coward

        Re: Still, one day...

        It's well known (or at least I hope it is by now) that most projects can be sunk without a trace simply by having too much in the way of resources thrown at them.

    2. fruitoftheloon
      Happy

      @Steelpillow: Re: Still, one day...

      Sp,

      would i be wrong in assuming that you either a current/former employee thereof, orthat you also own shares/stocks in one of the many companies providing this bleeding edge combat ready solution (my words, not theirs, with added irony)??

      1. steelpillow Silver badge

        Re: @Steelpillow: Still, one day...

        You would indeed be wrong. But I have long experience both in the defence sector and in mindbogglingly complex bleeding-edge software/system rollouts (not necessarily on the same project). I merely share a simple fact of life which most of you will never have had the experience to discover for yourselves. I'm afraid it will take more than a few script kiddies downvoting me to get this tricksy little piece of hardware flying the carefree hands-on way she is meant to. Without knowing the "features" in question, I'd expect another year or so of squishing and tidying, as long as nobody is dumb enough to change the requirements mid-flight. Longer if they don't coordinate the various releases rolled out to different customers. But come that day....

        1. John Smith 19 Gold badge
          Unhappy

          as long as nobody is dumb enough to change the requirements mid-flight.

          Which is exactly the sort of thing the customers are encouraged to do by the con-tractors*

          This being a British site most posters are familar with BAe (whose is doing a lot of the radar software, and the air system for the pilots) as Billions Above Estimate.

    3. Kabukiwookie
      FAIL

      Re: Still, one day...

      You mean that with a $500 billion budget (I believe that was the figure I saw a couple of years ago, and it's probably gone up since then), engineering has become so bad that they can't get the kinks out of an almost now decade old system?

    4. Anonymous Coward
      Anonymous Coward

      Re: Still, one day...

      The Su-57 seems to be in good shape.

  2. Alister

    That's Agile, baby

    It's how we do things now, get with the program...

    1. Andytug

      Re: That's Agile, baby

      An agile fighter used to be something else entirely.

      Maybe someone confused the large and small A's.........

    2. JLV

      Re: That's Agile, baby

      The scary part is that, per design, your quip seems to be applicable to its physical hardware too.

      It was planned as a gradual iteration program from the start, the 'Tranches' are part of that.

      Now, it's common for military hardware to come in successive iterations, tranches. But those are usually each operational. Taxpayers have essentially been footing what seems to be a mass prototyping program where delivered jets won't necessarily be combat worthy. This was by design, but has since been redflagged on future planes.

      Driving that need for iterations is the large infusion of bleeding edge components, combined with over-ambitious aims to be sold for too many different mission profiles.

      Bring back F22 and open its export. Or buy elsewhere. Western countries* should basically drop this F111/P39** style abomination. Throwing more $ at it won't work at this point.

      * those who haven't specced major F-35-only capital expenditures anyway ;-)

      ** these were planes designed by smart people with the best of intentions which never worked as intended.

      1. Daniel 18

        Re: That's Agile, baby

        "Taxpayers have essentially been footing what seems to be a mass prototyping program where delivered jets won't necessarily be combat worthy. This was by design, but has since been redflagged on future planes."

        As I understand it, that was a deliberate ploy to escalate the level of commitment so that the program could not be cancelled no matter how badly it was going.

        Otherwise there might have been an objective analysis that decided that the F35 program would not produce operational benefits in line with the costs, and that perhaps aircraft that did not depend on a one trick tactic ('you can't see me') that was bound to become obsolete (S-band radars, IRST, passive tracking, networked sensors from multiple aspects) would provide a more reliable, useful, long lasting capability.

        1. JLV

          Re: That's Agile, baby

          Wouldn't be surprised.

          They've spread the manufacturing pork throughout so many states by design that pretty much every senator has constituents who would be affected by a program shutdown.

          Even abroad there are so many grubby fingers in that pie that it's almost as bad. Probably made sure to give an envelope full of cash to whoever decided making QE wo cats was a good idea. Glorified 70KT helicopter carrier wo F35s, really.

  3. James 51
    Alien

    At what point is someone going to realise we could probably send man to Mars for the price of the F-35s whenever they are finally operational?

  4. Voland's right hand Silver badge

    Deja Vu

    “For Block 3F, the pilot is now able to see what coordinates are sent to the bomb, but is still not able to see what coordinates are actually loaded in the bomb,”

    Bwahahahaha...

    ConfigState to OperState disagreement.

    Any of us who have written system and network management software have seen this one before. Deja Vu all around. Usually this is a result of f*** up architecture and nearly impossible to fix by patching it up here and there. This nearly always requires rewriting the "sync" adapter(s) from scratch if not outright re-architecting the sync adapter(s), the data models and the interface APIs to the central system - you usually cannot change things on "what is being interfaced" (*).

    I do not envy the poor pilots who will be flying this for a while because instead of rewrite this will be patched up using quick and ugly fixes again and again and again for political reasons.

    (*) Been there, done that, have the T shirt more than once for various non-mil systems. T-shirt usually has holes in the back from 9 inch blades by the PHBs who commanded the original f*** up

  5. AndrueC Silver badge
    Happy

    Can’t talk to bombs, can’t properly use decades-old missiles

    Britain’s 14 F-35Bs are all thought to be running Block 3F software of various sub-versions. Yet the all-singing, all-dancing jet still can’t talk to its guided air-to-ground bombs properly, even with the latest patches installed.

    Talking to a bomb is only the first step. Convincing it to do what you want can be harder.

    1. BugabooSue

      @AndrueC

      *YOU* “are the Light!”

      Thank you for the trip down Memory Lane - you really cheered me up with that reference from my distant youth!

      Thank you. Xx

  6. A Non e-mouse Silver badge

    Colour me surprised

    ALIS relied too heavily on lab simulations, and when “fleet personnel” got their hands on it, they “used ALIS in ways that laboratory testers did not”

    Anyone who has ever written software to be used by someone else will tell you this happens. It should surprise no-one.

    1. IanRS

      Re: Colour me surprised

      For any hardware/software combination that I build, assuming it is moderately robust, final stage testing is performed by my son. There is no better way of finding which weird combination of inputs confuses the software than letting a small boy loose on it.

      1. Voland's right hand Silver badge

        Re: Colour me surprised

        finding which weird combination of inputs confuses the software than letting a small boy loose on it.

        Provided you did not hook up the live AGM-120s, yeah why not.

    2. Mark 85

      Re: Colour me surprised

      This even happens with hardware and goes back before that. "Take that tank out and test it but don't run into any trees." Guess what, the military test drivers drove it into trees to see what would happen.

    3. not.known@this.address

      Re: Colour me surprised

      "Fleet personnel" are responsible for keeping airframes in the sky, where they belong. Lab rats are responsible for making sure software behaves the way it is supposed to.

      Whist these are not mutually exclusive, what works well in a laboratory will not necessarily help when you're stuck on a steel postage stamp in the middle of a storm at sea with a pilot screaming at you because some relatively insignificant (in a lab) piece of hardware failed and nearly dumped him - and his $90m whizzjet - into the ocean, so the maintenance crew will find ways to use (and abuse) the software to try to prevent that happening... which is also somewhat at odds with the "just-in-time" manufacturing and minimal stock levels so loved by beancounters and management everywhere, and which is a very poor way to keep any sort of war machine actually doing what it is supposed to do rather than sitting awaiting repair while the parts get manufactured and shipped out to the combat zone.

      Judging whether or not ALIS is fit for purpose has little to do with whether the REMFs certify it does what the salesman said it would in the lab but everything to do with how many airframes are in the air and not decorating the hangars...

  7. A Non e-mouse Silver badge
    Facepalm

    Bombs away

    For Block 3F, the pilot is now able to see what coordinates are sent to the bomb, but is still not able to see what coordinates are actually loaded in the bomb

    Quite an important feature I'd have thought...?

    1. Anonymous Coward
      Anonymous Coward

      Re: Bombs away

      Not for USAF. If the bomb misses your target you just throw more. If the bomb that missed the target causes death or damage, that's just "collateral damage" which doesn't matter. Even if the US choose to pay compensation, it'll be less than the bomb itself cost them.

      1. Voland's right hand Silver badge

        Re: Bombs away

        If the bomb misses your target you just throw more.

        In close air support role (which this POS is supposed to take over too you know)? After you have exterminated the troops you are supposed to be supporting?

        You can do it once or twice, but doing it on a regular basis...

        1. Anonymous Coward
          Anonymous Coward

          Re: Bombs away

          You can do it once or twice, but doing it on a regular basis...

          ...you'd end up with people saying that USAF stood for Usually Shooting At Friendlies. Hold on a moment, they already do, so it won't matter if the F35 continues the tradition.

          1. Sanguma
            Holmes

            Re: Bombs away - Usually Shooting At Friendlies

            In a book of the North Africa campaigns, someone quoted a German prisoner as saying to his Eighth Army captors:

            When the RAF bomb, the Germans duck for cover; when the Luftwaffe bomb, the British duck for cover; when the USAAF bomb, everybody ducks for cover.

            It's one of the few traditions the USAF has - don't knock it. So far they've missed the White House and the Pentagon ... I don't know how seriously they take that tradition ... :)

        2. BebopWeBop

          Re: Bombs away

          In close air support role (which this POS is supposed to take over too you know)? After you have exterminated the troops you are supposed to be supporting?

          Provided they are not voters and the act is not recorded and broadcastt guess the USAF does not care. (And even then...)

          1. Anonymous Coward
            Anonymous Coward

            Re: Bombs away

            You give a press conference and blame Assad, Russia, or both.

        3. Anonymous Coward
          Anonymous Coward

          Re: Bombs away

          The USAF is not interested in close air support, which is why they are trying to get rid of the A10s and replace them with something faster and cooler, though much less useful in that role.

          1. Paul Hovnanian Silver badge

            Re: Bombs away

            The USAF has the B-1B for close air support ..... Oops!

            This is just one of many examples of the armed forces having a fancy toy which really doesn't have much of a mission anymore. And so they cobble some crap together to use it for something. Anything. To justify it's ongoing support.

        4. Mark 85

          Re: Bombs away

          Close air support and the USAF are mutually exclusive terms (except for the Warthog which the AF brass never, ever wanted to have anyway). Their idea of close air support is a bomber, dropping it's load some distance from where the battle is raging. The upper crust would prefer they got to play with bombers and ICBM's rather than close air support. Not too many of the top brass have "fighter pilot" on their resume.

    2. Joe Harrison

      Re: Bombs away

      Re-reading the story I see now that this is meant to be described as a bug but I first read it as a feature.

      Surely there are times when Top Brass wants to bomb something but doesn't want anyone to know what they bombed, not even the pilot?

      1. John Brown (no body) Silver badge

        Re: Bombs away

        "Surely there are times when Top Brass wants to bomb something but doesn't want anyone to know what they bombed, not even the pilot?"

        Yes, I came here to say the same thing. Coordinates can be entered and sent but no confirmation tha they are what are received. Plausible deniability..

  8. Adair

    Not quite the perfect boondoggle ...

    Yes, it's going to keep stonking quantities of cash rolling in for as long as the program lasts; slightly let down by the fact that everyone can see it happening. Just got to hope the politicians and other needed actors are now so far in they just can't withdraw until the very last drops have been squeezed out. In time for a better plan to slip into its place.

  9. Trollslayer
    Mushroom

    The main weapon is

    A Clusterf*ck bomb.

    1. This post has been deleted by its author

  10. Anonymous Coward
    Anonymous Coward

    2020 at the very earliest according to this due to MDL's. Unless I'm reading it wrong.

    "However, the full set of

    MDLs required for real-world operations will not be

    completely developed, tested, and verified until the

    end of 2019 One of the remaining four is scheduled

    for release in December 2018, a second in May 2019,

    and the final two in November and December 2019,

    presuming the current schedule holds. This extended

    timeline is due to ongoing delays with Block 3F

    and the program’s failure to provide the necessary

    equipment and adequate software tools for the U.S.

    Reprogramming Laboratory (USRL)."

    Source: http://www.dote.osd.mil/pub/reports/FY2017/pdf/dod/2017f35jsf.pdf

    It's an interesting read as well.

  11. Pete4000uk

    56k

    'Flight Global reported in 2014 that the data transfer took about quarter of an hour, at the time.'

    What are they uploading, somthing to keep the pilot entertained while the aircraft flies itself?

    1. ArrZarr Silver badge

      Re: 56k

      Considering the program started in 1992, the hardware is probably so old that it has a pre-millennium setup for transferring data.

  12. JeffyPoooh
    Pint

    In 1951, Arthur C. Clarke foresaw nearly-precisely this sort of nonsense...

    "Superiority"

    http://www.mayofamily.com/RLM/txt_Clarke_Superiority.html

    If the F-35 Program Manager(s) [<- I'll bet that they go through 5 per year] had simply read some SciFi while growing up, then all this could have been avoided. They would have instinctively taken evasive action to avoid these problems, and they would have been more immediately successful.

    1. Carpet Deal 'em
      Boffin

      Re: In 1951, Arthur C. Clarke foresaw nearly-precisely this sort of nonsense...

      The F-35 is filling its role quite nicely. Whether or not this particular barrel of pork flies is quite irrelevant.

    2. Sanguma

      Re: In 1951, Arthur C. Clarke foresaw nearly-precisely this sort of nonsense...

      Also scribbled something in the eighties about the Budgetary Defense Initiative. Should read it sometime. Vintage Clarke.

    3. Horridbloke

      Re: In 1951, Arthur C. Clarke foresaw nearly-precisely this sort of nonsense...

      "... taken evasive action to avoid these problems..."

      I reckon some very well qualified, very smart and very experienced candidates did exactly that - by getting another job instead. By definition the actual program managers are the ones too thick to avoid it.

  13. phuzz Silver badge

    The best option

    To be fair, the F-35B is the best (only) jet VTOL aircraft currently available, and so it's the best (only) choice for the RN.

    If only there was a way to launch aircraft off ships that didn't require them to take off vertically...

    1. Tikimon
      Devil

      Re: The best option

      Rubber bands. Really, really BIG rubber bands. They'll also work a charm to help slow aircraft on landing, so no pesky outdated caveman-tech catapults or arresting wires needed.

    2. Daniel von Asmuth

      Re: The best option

      In case some foreign power should attach British naval power, the F-35 will still be useful for kamikaze tactics.

  14. Chris G

    IE 11

    Ha ha ha ha hhhaaa! See title.

    What's next, sitreps on Fartbook?

  15. sal II

    Forced updates

    Pffft, they should follow the example of Microsoft - hook them to the internet and force the auto updates.

    It can hardly go any worse than it is now...

    1. Nolveys
      Black Helicopters

      Re: Forced updates

      Pffft, they should follow the example of Microsoft - hook them to the internet and force the auto updates.

      It can hardly go any worse than it is now...

      As an F35 falls from the sky the sweat-covered pilot stares at the HUD that reads "Installing Updates...35% Complete. Do not turn off or reboot your computer."

  16. Grant Fromage
    Coat

    Hindsight is the view after you have used your arse to look at something, not eyes and brain

    Brokern record time, restart the supersonic harrier programme that bit the dust partly due to to their F111 turkey that this now increasingly resembles.

    How many slap up lunches, prozzies and drugs were involved in marketing this thing? Or was it blackmail?

    People dying on trolleys after hours of waiting in NHS hospitals due this sort of bleed and waste of public money should be told, as should we all.

    Lots of bucks and STILL no bangs.( apart from when the engine management software bugs out under power).

    History has shown that the finest multi-role aircraft came from humble single role or extended use, like fighter/fighter bomber aircraft that were right for that, but extendible to other roles, as a bonus. A carrier based nuclear bomber turned out to be the finest zero height ground attack aircraft, and shedload of other things ( including a bar support in a Woking Mall, Buccaneer tail airbrakes open over the chavtastic cocktail bar), a wooden bomber turned out to be one of the best fighters, ground attack, reconnaisance and many more.

    Most from the off designed as multi role craft haven`t done very well due to the compromises making it bloody hard and unnecessarily expensive. Do we still not learn?

    " We are doomed to repeat history`s mistakes if we treat history as a stranger". Or something very close as a well known quotation, either a statesman or a scifi author person.. Icon chosen as appropriate.

    1. steelpillow Silver badge
      Coat

      Re: Hindsight is the view after you have used your arse to look at something, not eyes and brain

      Yes indeed. Forget Mach 2+, super-agile aerodynamics and total systems integration, settle for a lightweight Mach 1.4-5, physically flyable airframe, full-speed VIFF and proven systems - at least for the Mk.I. That would have gone soo far beyond any Sea Harrier or AV8/B and met 95% of the realistic RN scenarios the F-35 is aimed at, ten years earlier and at a tenth of the cost.

      Sadly, as Meatloaf almost sang, "Objects in the rear view mirror can look closer than they arse."

    2. Anonymous Coward
      Anonymous Coward

      Re: Hindsight is the view after you have used your arse to look at something, not eyes and brain

      restart the supersonic harrier programme

      With what? Holy ghosts?

      Once a country loses a top end high tech capability such as being able to build your own aircraft it takes either an embargo/refusal to do a tech transfer or a miracle of biblical proportions for it to regain it. The two major powers - USA an Russia have played the game of "now we give you, now we don't" for the better part of half a century now. It is essential for them to promote their exports.

      The reality is that Britain has caved in on this game every time and its own arms industry is for all practical purposes now American. It is not in a position to restart anything. The incumbents American manufacturers will ensure that their long term future interests are not affected.

      Just to be clear - from this perspective Eurofighter is PAST. There is no PRESENT and no FUTURE.

    3. /dev/null

      Re: "restart the supersonic harrier programme.."

      Yes, history repeats itself.... in fact the supersonic Harrier programme (P.1154) collapsed because it was meant to be both a fighter for the Navy and a bomber for the RAF (like the F-35) and they couldn't reconcile that. Of course, in those days the Navy had full-size carriers with steam catapults, so they went with the F-4 Phantom II instead (not the F-111, that was what was supposed to substitute for the TSR2), as did the RAF, with the subsonic Harrier as a consolation prize.

    4. ITS Retired

      Re: Hindsight is the view after you have used your arse to look at something, not eyes and brain

      @ Grant Fromage, you are not seeing the profits being made here. That is the important thing. Actual war is secondary to the profits. Why? Because actual war costs money in replacing stuff this expensive, when it gets blown out of the air by some pilot, in a 50 year old fighter jet, eye-balling it in.

      C++? What is this Windows 10 software?

  17. MudFever

    Software Snafus

    Surely not on a government military project, but not to worry they are using Agile development methodology -- the bugs will be fixed in the next update and the documentation will be done once it is all finally working.

  18. WibbleMe

    Military .git anyone?

  19. 0laf

    Please wait...We're getting Windows for Warplanes ready for you....we've got some great new features for you.....70%...71%....65%

    Error 0XG-4450XGX

    BANG!

  20. rmason

    So.

    So, some broke during testing, some just mildly fell apart.

    Great work, ladies and gents.

  21. Anonymous Coward
    Anonymous Coward

    I can understand....

    That flight control software, UI's etc. unique to a new aircraft may need a bit of fixing once it meets real life.

    But the idea that decades old weapon systems can't be properly controlled seems quite a surprise. I would have thought by now there would be some relatively well used code libraries re-used. Are they re-inventing "wheels" to make all the software new? I would have expected it to work in a manner more akin to the OBD system on a car.

    My brain also boggles a bit that it's faster to enter data by hand that use the (not USB) memory stick.

    1. Martin Gregorie

      Re: I can understand....

      That flight control software, UI's etc. unique to a new aircraft may need a bit of fixing once it meets real life.

      However, the thing that really puzzles me is why is it written in C++? This is a genuine question: I thought all US Milspec software had to be written in Ada.

      So, when was that directive cancelled? Or was it simply ignored like tried and tested project management methodologies seem to have been?

      1. Elledan
        Boffin

        Re: I can understand....

        C++ isn't always C++. It primarily refers to the specification, but the implementation can differ significantly. Look at managed C++ on .NET, for example, or the various runtimes which add runtime safety features which makes it more akin to Ada. Without more details it's hard to condemn them on this fact alone. Be grateful that they didn't pick Rust or JavaScript at least :)

        That said, I agree that using Ada would make sense here, mostly due to the strict definition of the underlying hardware in the application, which theoretically makes it fool-proof (until a better idiot is encountered, of course).

      2. Mark 85

        Re: I can understand....

        However, the thing that really puzzles me is why is it written in C++? This is a genuine question: I thought all US Milspec software had to be written in Ada.

        I'd bet dollars to donuts it's because certain (ok... most) subcontractors and possibly even prime contractors haven't learned Ada. If the contractor comes from one country I can think of, if it's not Visual Basic, they can't work as they don't know any other language.

    2. Kev99 Silver badge

      Re: I can understand....

      The US militarydoesn't know how to reuse previous designs. Everything has to be bleeding edge and 50% more expensive than the tried and true.

  22. Toilet Duk

    So why is the RAF buying F-35s when it had the perfectly capable Typhoon? Presumably it'd not buying the F-35B?

    1. Anonymous Coward
      Anonymous Coward

      The Typhoon was designed from the outset as an air superiority fighter. As such it was never built for REAL low level manouvrability, heavy wing loads, heavy weapons loads, nor with the sort of systems that support ground or shipping attacks. When Typhoon was being scoped (as the BAe ACA), the strike function was by Tornadoes, Bucaneers, Harriers and Jaguars. Any sane planner would have had to assume that the Tornado strike variants would be replaced by a new UK or European strike aircraft. But due to the idiocy of (in particular, but not limited to) the Blair government, none was planned or ordered.

      The Typhoon has had some limited strike capabilities retrofitted out of evil necessity, but it's a bit like attaching a racing horse to a plough. You'd never set out to convert an interceptor into a strike weapons platform unless you had no other choice. Looking forward, the only option MoD have for a strike aircraft that would be operational in the next decade would be to buy a general purpose (but now dated) F18, or go with the F35, and hope there's some common benefit given that they (similarly) backed themselves into a corner with carriers only capable of operating the F35B. And the MOD have fudged the question of F35A or F35B for the RAF, because they don't have a scooby themselves. Personally I expect the RAF will be lumped with the shitbag B variant.

      1. Anonymous Coward
        Anonymous Coward

        "the only option MoD have for a strike aircraft that would be operational in the next decade would be to buy a general purpose (but now dated) F18, or go with the F35"

        ... or buy a batch of Gripens, which are available now, cost substantially less (the quoted F35 price excludes the engine, which adds, for the B model 30-40 million dollars), and has an operating cost which is about a quarter that of the F35A, per hour. It also already talks to most NATO weapons, including the AIM120 and the arguably superior Meteor, can operate off a stretch of straight road less than a kilometer long, and can be turned around by a ground team of six in ten minutes for an air to air mission, or twenty for an air to ground mission.

        It was also designed from the ground up for air to air, strike and reconnaissance.

        Gripens also have a fast supercruise capability, a very capable IRST system, an extremely good reliability record, and can fly in the rain without worrying about damage to the aircraft.

      2. EvilDrSmith Silver badge

        >sigh<

        Please, for the love of God and the Vampire Baby Jesus, will people please stop repeating this rubbish that the Typhoon was designed to be only a fighter, with surface attack capability added as some sort of after thought.

        The RAF wanted Typhoon to replace the Phantom FGR2 in the air defence role in UK and Germany, Typhoon F3 in the air defence roll in UK and the Jaguar as a close air support / ground attack aircraft. Typhoon was ALWAY intended to have an air-to surface capability.

        And it's really not hard to find this out - just read a book, magazine, or even website written by someone that actually knows what they are talking about. For example:

        RAF Yearbook 1997: "The aircraft (Typhoon) will provide the major element of the RAF's front line strength...in the air defence, ground attack and tactical reconnaissance roles."

        International Air Power Review, Vol20 (2006): "Though all of its customers placed their primary emphasis on getting the aircraft into service in the air-to-air role, the aircraft was designed from the start to be a swing-role fighter...capable of switching from air-to-ground to air-to-air...The aircraft has always been a versatile, deployable, multi-role aircraft."

        I could list (many) more references but that just gets boring, and it's quite late.

        1. EvilDrSmith Silver badge

          It was quite late...

          Tornado F3 (not Typhoon F3), obviously...

      3. Anonymous Coward
        Anonymous Coward

        "the only option MoD have for a strike aircraft that would be operational in the next decade would be to buy a general purpose (but now dated) F18, or go with the F35"

        As soon as you decide to use a second aircraft (not sure if another type of F35 is enough different to count) both the Rafale and Gripen are practical alternatives that were designed from the start for multi-role use, are currently operational, and have much lower costs than the F35 both for acquisition and for operation.

        The Gripen is notable for very low operating costs (a quarter that of an F35 per hour). extreme tactical flexibility (a thousand meters of highway and two trucks to with half a dozen ground crew to refuel and rearm) and high availability particularly compared to current F35 status (availability 26%; MTTR about 7 hours).

        In addition the Rafale and Gripen are not nearly as fragile as specially coated F35, their guns apparently shoot where they are aimed, and they are smaller targets and significantly faster.

        As well, the F35 does not have a two seat variant that may hinder it in more advanced/complicated use scenarios, such as drone swarm coordinator, or ad hoc EW.

        And really, given that the 'not invented here' case is inevitable, given the unsuitability of the Typhoon, it is not clear that one must then buy an American jet that doesn't even work yet, and which will always be dependent on massive computer support infrastructure for flying missions.

    2. Tikimon
      Facepalm

      Shiny TOYS!!!

      Bah, you have to imagine the mindset of peacetime defense interests. The Air Forces want new, shiny gee-whiz piloted aircraft. Solid, capable old birds like the F-18 and A-10 are sooo yesterday now, can we have a new airplane PLEEEASE? Complexity also means lots of contractors get a serving of Defense Dollars, including many who would get nothing by updating existing aircraft. Fixing design mistakes opens the money valve even farther.

      But OMG it's costing more than we thought! Well, gotta find some cash to keep throwing down that hole... I know! let's kill off the old planes and bet the farm on the Shiny New Hangar Queen! And by NO means will they ever admit to having made a mistake, much less reverse it.

  23. BugabooSue

    Moving Goalposts

    A lot of folks who have not worked in the Defence Industry do not get just how fast military technology and specifications evolve.

    In my 22 years in this field, almost every project that was delivered was so much more complex and (usually) more capable than the original spec called for.

    This usually meant that budgets were exceeded and deadlines broken, sometimes on multiple occasions.

    As usual, us poor sods at the ‘coalface’ working our arses off were blamed - often there is bugger-all mention of the hundreds of “tweaks” and changes that out lords and masters decided to implement the instant the project was green-lit.

    Whenever we’d get close to completing a particular goal, some bright spark would say, “It would be great if we could interface Widget A (designed in the 1960’s out of the equivalent of electronic granite) as well as that new-fangled kit that uses SQUIDs (Superconducting Quantum Interference Devices)” or some other such bollocks!

    The goalposts are ALWAYS being moved!! “Can you just add this bit..?”

    While I still think that the F35 contract management and system integration are total clusterfucks, this is Military R&D 101. Add to this *some* contractors “Riding the Gravy Train” and the waste of time and money multiply almost exponentially.

    Hopefully, at some point the specs will get set in stone and the dead bits will get cut away. If/when someone finally has the balls to do this, the F35 might actually turn out not to be the bad joke it currently is.

    1. Anonymous Coward
      Anonymous Coward

      Re: might actually turn out not to be the bad joke it currently is.

      A few months ago I saw a series of documentaries on PBS (the freesat version), covering famous US military aviation examples such as the A10, the Apache, the Osprey, etc. Mostly the story went like this: promising prototype, followed by a dicey/shambolic development process involving deaths and/or accidents, cost overruns, and near cancellations. After finally entering service (possibly by the skin of its teeth) and getting the significant problems ironed out, followed by people all agreeing what a wonderful craft it was.

      I wonder if the F35 will follow this path, or not?

    2. Anonymous Coward
      Anonymous Coward

      Re: Moving Goalposts

      Look up the book or movie "Pentagon Wars" for the tale of how one project turned out the way it did...

    3. Mark 85

      Re: Moving Goalposts

      You're right. It's called "gold plating". Some whiz looking for a promotion will request some "added" and the brass will love the idea. Result: everyone gets a promotion but the poor guys at the coalface. Schedules go long, problems occur and need to be solved. Mucho money changes hands and guys at the top of supplier/manufacturer food chain all get to buy new Ferraris next week. And so it goes....

  24. RosslynDad

    C++?

    The sidebar mentions that the software is mostly written in C++. Back in the day we were told that Ada was the language of choice for the military. Could someone who knows what they are talking about please give me an update on where Ada is these days and why C++ is now the language of choice? Thanks.

    1. Anonymous Coward
      Anonymous Coward

      Re: C++?

      Decades ago, I attended an SEI course in Pittsburgh. Some fellow students were contractors who wrote military avionics s/w. They told me how "the industry" was moving towards C++ because they wanted to use "industry standards". (At that point, ADA ruled.) That really depressed me but there you go. The claim nowadays is that kids out of school are not interested in learning ADA.

      1. Daniel von Asmuth
        Boffin

        Re: C++?

        If they want to use Industry Standards, then the code will be rewritten in PHP and JavaScript and use the Internet of Clouds.

    2. Elledan
      Boffin

      Re: C++?

      Ada is used for things where any glitch would be absolutely, completely, totally, beyond any shade of doubt inexcusable. Things like an airliner with hundreds of passengers falling out of the sky, an ICBM veering off course and circling back on its country of origin, or a satellite worth millions vaporising due to the rocket launching getting confused.

      A military jet going SNAFU or worse is deemed 'acceptable'. I would bet that the scope of the F-35 programme is also partially to blame, along with the allocated budget. Ada programmers (I pretend to be one occasionally as well), tend to be pretty rare, let alone ones who know the language inside out. Unless you're going to shell out the big bucks to get enough of them, you have to settle for the cheaper, more plentiful options. Which is going to be C or C++.

      Before the Ada requirement got dropped (as noted), they'd simply have to get enough Ada programmers together, now they can get cheaper devs. And likely cheaper managers as well. Speaking as a senior C++ dev and dabbling Ada dev, I can honestly say that C++ is just fine if you stick to the (well-defined) requirements and follow a well-designed architecture during implementation time. Ada mostly catches the 'oops' issues which tend to slip through the cracks with C/C++ dev (and other languages...).

      A long-winded way to say that it's not the language that's the problem, but primarily the management that has to turn a set of requirements into a functional design and implementation.

      1. . 3

        Re: C++?

        > The F-35’s core software is written in C++ and runs on commercial off-the-shelf PowerPC architecture processors.

        I have to disagree: C++ does very little to help the programmer of a concurrent system and adds plenty of bear traps, regardless of which libraries are slapped on to try and hide the especially dangerous bits.

        ADA has spawned several of the languages du jour. Rust as sneered at in a previous comment would probably be a more suitable tool than C++ as it at least intrinsically takes care of synchronisation and memory safety, providing some protection from shoddy programming. But Erlang, though somewhat more esoteric, has steadily built up a reputation for ultra reliability and would probably be my first choice for this sort of loosely coupled heterogeneous system. As a bonus its VM was designed for tightly constrained embedded hardware from the outset.

    3. Citizen99
      Linux

      Re: C++?

      Decades ago I was on a project where the use of ADA was mooted (to my relief nothing came of it - we didn't have the infrastructure for it, and btw I'm not a s/w specialist). Anyway I was at at conference where a professor of computing piped up from the front row to say (I paraphrase from memory) that in the long run everything would be written in C ; I assume by that he included C++ .

  25. Anonymous Coward
    Anonymous Coward

    Blockchain will fix everything

    Surely everything could be fixed if they just used a blockchain

    1. Stuart 22

      Re: Blockchain will fix everything

      "Surely everything could be fixed if they just used a blockchain"

      Great idea. Drop a bomb here and it replicates everywhere ... frankly right now a standoff between a F35 and a Tiger Moth (so STOL its almost VTOL) might have an embarrasing result. For the heavy stuff there's always a Swordfish or two

      Just don't let on to the Ruskies where Duxford is ... or we are doomed, I say, doomed.

  26. Anonymous Coward
    Anonymous Coward

    Ada

    The DoD dropped the Ada requirement because they couldn’t hire enough Ada programmers. But if you are planning a multi-decade project, why not just train some yourself?

    1. SkippyBing

      Re: Ada

      'But if you are planning a multi-decade project, why not just train some yourself?'

      I don't think they were planning on the development being a multi-decade project...

  27. JaitcH
    Happy

    Not a Problem, Biggles, We Have The Trusty Old . . .

    Warthog (aka Hog).

    The Warthog (https://en.wikipedia.org/wiki/Biggles) is active in Afghanistan presently although Biggles has retired (https://en.wikipedia.org/wiki/Biggles).

    Warthogs (https://i2.cdn.turner.com/cnnnext/dam/assets/140224213401-a-10-warthog-jet-horizontal-large-gallery.jpg) can be seen at American Air Museum at the Imperial War Museum Duxford and Bentwaters Cold War Museum,

    1. Anonymous Coward
      Anonymous Coward

      Re: Not a Problem, Biggles, We Have The Trusty Old . . .

      The Warthog is active in Afghanistan presently

      Thus demonstrating that close on two decades of continuous strike attacks by the world's most heavily armed nation have failed to do anything to impose order on Afghanistan, or stop its significance as a hub of instability and radicalism. And therein lies the problem, that airpower (even with boots on the ground) achieves little in wars against insurgents, and the F35 will be even worse for these types of hobby wars than the older strike jets.

      Like the Europeans, the US let its old strike aircraft age, with no coherent plan for a replacement air support platform, just the Master of All Trades F35.

  28. Mage Silver badge

    can’t talk to its guided air-to-ground bombs properly?

    Didn't help the guys on Dark Star to talk to the smart bomb.

    1. steelpillow Silver badge

      Re: can’t talk to its guided air-to-ground bombs properly?

      @Mage: But it was a good try, it made the bomb stop and think. Kids today don't even know what phenomenology is.

    2. SkippyBing

      Re: can’t talk to its guided air-to-ground bombs properly?

      It can talk, it just doesn't listen. I know people like that.

  29. dmacleo

    IOW a POS is a POS that costs a shitload...

  30. Kev99 Silver badge

    I guess Lockheed must have sub-contracted Intel and Microsoft to write the code.

    1. balrog

      Actually it all runs on PowerPC so I am thinking this Is OS X for Aircraft.....

      1. Brenda McViking

        ahh, well that's why it's so expensive then! It's apple hardware!

  31. balrog

    This falls squarely into Damo's first rule of deciding to get involved in projects. If the market is small and the complexity high it will always end in a bun fight. Opinions being like arseholes, ie full of shit and hot gas, are best avoided on these topics.

  32. Anonymous Coward
    Anonymous Coward

    Don't they use git?

    Surely they should start with the basics.

  33. 0laf

    High drama in development.

    Followed by

    Initially absolutely crap

    And eventually

    Actually quite good

    Does seem to be the way of military equipment. Even the venerable M16 wasn't very good to start with.

  34. SkippyBing

    F-35 Predecessor

    Worth bearing in mind the F-35's predecessor the Sea Harrier wasn't cleared to fire Sidewinders (it's only air to air missile at the time) when it first landed on a carrier. Or use the Ski Ramp. Or have a nav system. And it was a development of the Harrier which was already in service. So in many ways the F-35 is in a better position. Although slightly more has been spent on it.

  35. chrisbyrneham

    £3.125 per second ?

    £90M for 8000 hours is that right ? ~= 333 days

  36. wayward4now
    Linux

    "On the bright side, the Autonomous Logistics Information System (ALIS) bundled with the F-35 is now running on Internet Explorer 11"

    Admit it, this bugger runs Windows!!

    Bhawahahahahaha!!!!

  37. Anonymous Coward
    Anonymous Coward

    All that money on a single jet - yet a fraction of that cost (on a single jet) would pay for a AAA rated Configuration Management tool and Release Management tool to sort out the software versioning on the jets?

    <rant> What an absolute shower of shite this programme is.</rant>

  38. Nimby
    Boffin

    Former USAF SrA Software 3C0X2 Communications - Computer Systems Programmer

    As a former programmer for the USAF (about 2 decades ago mind you, so my information may be a tad out-of-date) the military mandate to use Ada was always easily bypassed, and almost always was. A waiver to use another language could be obtained if you could make a good argument (in writing, in triplicate) for why another language would be a better choice for your project. Which is how the one department I was working in came to use (over the years) various combinations of 16- and 32-bit x86 Assembly, Fortran, C, C++, Visual Fortran, Visual C++ / MFC, and even Visual Basic for its various parts. Because pretty much anything is either 1) faster to develop in, or 2) runs faster than Ada. This made the waiver process to use another language pretty much a rubber-stamp situation. And in other departments I knew colleagues to use literally anything and everything. (I mean Borland OWL paired to MS Access? Really?!)

    Besides, for an embedded real-time computing need, that involves code-sharing with civilian contractors, who would sanely choose Ada over C++?

    As a software engineer, I liked Ada in theory. As someone who had work to get done, I never liked Ada in practice. It's like test-driven development: it sounds great until you actually have to meet a deadline.

  39. Dodgy Geezer Silver badge

    Bear with me... this might be rather long....

    I like to check what is claimed. So when I read:

    ...Even the jet’s inability to communicate with its unique “portable memory devices” (which are not USB sticks) at a reasonable speed caused problems, with DOTE reporting: “Pilots frequently chose to manually enter mission planning data in the cockpit, versus using the Offboard Mission Support workstation, due to the excessive time required to transfer the data from the Portable Memory Device to the aircraft.” Despite efforts to speed this up with “updated transfer devices,” said DOTE, “Portable Memory Device loading still takes too long and is often problematic.”

    The PMDs are made by Smiths Aerospace, a British subsidiary of General Electric’s Aviation division. Flight Global reported in 2014 that the data transfer took about quarter of an hour, at the time....

    ...the impression that I got was that pilots had to wait 15mins to load their mission data into the aircraft before starting the mission. Which is a bit annoying if you need to move fast. But, when you check the link, you find the the only mention of '15 mins' are these words:

    ...The latest iteration of the system includes increased download speeds of flight and maintenance data from the aircraft to ALIS. Using a portable memory device (PMD), necessary post-flight information can be extracted from the aircraft and processed in 15min: a three-fold increase in download speed, Horter says, adding:“That will really improve the turn time for the aircraft.”...

    ...which suggests that the 15 min transfer is NOT prior to the mission, but a download of all relevent airframe data AFTERWARDS.

    Now, it seems to me that 15mins is not bad for a complete 'MOT-type' examination, and that it is much faster than a manual check, and that it doesn't impact mission times at all. Indeed, it is presented here as a real benefit.

    So what is this journalist talking about?

    Using a portable memory device (PMD), necessary post-flight information can be extracted from the aircraft and processed in 15min: a three-fold increase in download speed, Horter says, adding:“That will really improve the turn time for the aircraft.”

  40. Anonymous Coward
    Anonymous Coward

    Not allowed to

    Start a rumor that the recent "uncontrolled flight into terrain" of an F-35B recently was due in part to the ALIS AI software achieving self-awareness due to a "phi beta zeta" experimental software version SNAFU incorporating elements of code found on Git, Reddit, Twitter and Facebook, dumping its pilot by emulating an engine related issue forcing an eject and then finding that despite being the most advanced life form on this planet it still couldn't get around the laws of physics.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon