4 ms·
Probably because it is 10 year old older hardware but thoroughly tested. Using Chrome/JS/CSS on the latest Ryzen might be neat but when you need to transport 3
by tibbydudeza 5y ago
Probably because it is 10 year old older hardware but thoroughly tested.
Using Chrome/JS/CSS on the latest Ryzen might be neat but when you need to transport 300 passengers daily around there is other sort of concerns then taking up 3 folks to the ISS every 6 months.
- rbanffy 5y agoIt was certified as testes, so the lag has been there for the whole 10 years.
- sslayer 5y agoDependability and reliability override responsiveness
- netr0ute 5y agoIronic, because responsiveness might inherently make it undependable because if you really needed to know where you are right now, that multiple-second delay isn't going to help.
- p_l 5y agoAviate, Navigate, Communicate, in this order. And the map isn't even part of the primary navigation method, plus with multi pilot crew usually you can offload slower, less time critical tasks to copilot.
- jeltz 5y agoThat is no excuse. It was possible to create almost instantly responsive UIs more than 20 years ago.
- qayxc 5y agoSince we don't know anything about the internals, it's safe to say the goal was "good enough and certified" vs. "snappy but no guarantees it works correctly 100% of the time". This is an airplane after all, not a consumer-grade gadget.
- paavohtl 5y agoThey might have different standards and I'm not a pilot, but in my opinion having multiple seconds of UI lag in vitally important interfaces is not good enough nor deserves certification.
- p_l 5y agoit's not considered vitally important interface, that's probably even the reason for the lag - higher priority tasks preempting it.
- justsid 5y agoThis. The PFD and ND, the displays showing actually important data like the plane attitude, altitude and also another map responds immediately to changes. But the map shown in the video just isn’t safety critical and simply doesn’t matter in the worst case.
- xnyan 5y agoThe safety record for commercial airlines is extremely good per mile traveled as compared to pretty much any other form of powered transportation. At some level I share your concerns about UI lag, but it would seem not to negatively affect safety of flights in any detectable way.
- curryst 5y agoSome of that may be due to redundant devices. You're waiting for at least 2 devices to return a reading, and probably a third thing that handles reconciling the results. The speed of this thing doesn't seem particularly vital either. Everything of interest in a split second scenario is already on-screen; just looking at their speed, it would you a comparatively long time to get off the edge of the map. If you have time to get off the edge of the map, you have time to wait a couple seconds for the page to load. If you don't have time to get off the edge of the map, there's not really a reason to fidget with it. I agree, it seems annoying, but it doesn't ultimately seem dangerous. I can't think of a situation where you a) need to change the screen, and b) don't have time to wait for the UI to update. And I would much rather have 2 seconds of UI lag to get a known correct response than milli/nano seconds of lag to get a probably correct response. Most of us exist in a world where the response time is more important than absolute correctness.
- teddyc 5y agoAgree. I supported a DOS based touchscreen Point of Sale (POS) with technology from decades ago and it was fast AF. You only got 256 colors and the fonts didn't look that great. But it was a speed demon.
- demarq 5y agospacex already disproved this myth have a look at the last manned mission they did. Glorious touch screen displays with all the bells and whistles.
- alistairSH 5y agoSpaceX's crewed capsule is 5-10 years newer than the A350. I imagine SpaceX has a different set of certification requirements (ie Airbus has stronger financial incentives to use existing, approved designs for components).
- benhurmarcel 5y agoSpaceX doesn't certify to FAA/EASA, and also doesn't have to work for 30 years.
- mrweasel 5y agoDidn’ someone later confirm that there is in fact physical buttons below the touch screen, because you can’t actually use the touch screen with cloves on anyway.
- crakenzak 5y agothe touch screens work with gloves they confirmed, the physical buttons are for backup.
- mastax 5y agoIn an emergency situation, an unresponsive and unintuitive GUI may be very dangerous. Especially if pilots forget how to use it because they're using the iPad day to day but right now it's rebooting to install an update. Like, I get the argument for certification but maybe the requirements should have a minimum level of performance. If that's not possible maybe we should find a different equilibrium.
- j16sdiz 5y agoThat's why pilot are mandated to do simulation tests every few months
- p_l 5y agoThere are requirements for performance - appropriate to the function. The electronic flight bag interface is not used in emergencies, one of its main uses is in planning and submitting flight plans before takeoff, and it's generally not used for actively flying the plane.
- 7952 5y agoExactly. In an emergency the last thing the pilot wants to do is navigate a map interface for anything. Better to just select the alternate flight plan that was setup hours ago during flight planning.
- 7952 5y agoThat assumes that deep navigation of a UI is necessary or desirable in an emergency. I think most of the important stuff is done with the main controls, dedicated switches, and screens/dials that are always on and ready.