Not to mention, a 90% successful autopilot isn't something to brag about these days.

It is when we're talking about doing considerably more than just holding heading, speed, and altitude. This is more like autopilot on steroids, and as the technology matures, the success rate will increase.

Forgetting about 'does xplane have a better FM to do this' because it doesnt need to, it is an interesting application. There is very little information about it _but_ if it is using the gps and can land itself on runways that dont support ILS then it is doing pretty well I would think.

How landmark? .. what "Ultimate Traffic, Traffic X, and others dont count?

Hahahahahahahaha, have you SEEN the drones of Ultimate Traffic in action?! Good for lawnmowing I suppose. They are far too much on rails to trust

Hahahahahahahaha, have you SEEN the drones of Ultimate Traffic in action?! Good for lawnmowing I suppose. They are far too much on rails to trust

When an A/P flys like it's on rails.................then that's the type to trust.L.Adamson

As someone who has actually written an autopilot (both lateral and vertical control)... I can tell you that 90% isn't bragging rights. Seriously.It has to be 100%, time after time after time after time... etc. A Boeing 767 doesn't autoland correctly 90% of the time.A spiral descent to a TDZ isn't amazing math, it's good math... but not amazing.And yes, if I'm in a plane with that system... I want 100%. A normal autopilot is backed up by a human... this system is designed to be a human's backup. Because of that 100% is the lowest level of accuracy that it can be expected to accomplish.

When an A/P flys like it's on rails.................then that's the type to trust.L.Adamson

Sorry, I guess I wasn't clear enough. By rails, I meant 'using a very simplistic lookup table that animates far too consistent approaches/take offs', not 'An ai routine so superior that it makes real aircraft look unstable'.

"The next step – to incorporate this technology into a fully-fledged autopilot system – is already underway. “

You can have an engine failure or a pilot failure,

” explains Meyer, “hit this button and the airplane will use artificial intelligence to land itself. In X-Plane, I’m achieving about a 90% success ratio."

Ed, you're saying that an aircraft is expected to land safely, after a spiral descent, 100% of the time with an engine failure?WOW. I don't know a pilot (and I know quite a few) that would trust the autopilot to land an aircraft, with one engine out, on a straight in approach...let alone on a spiral descent.

It's a nice little feature article.The author has also written about Star Wars Galaxies, Ghost Recon Online, and the top five RPG Maker games.I'd be more interested to hear what Flying magazine or AOPA Pilot has to say about this system. And I'll be glad to read more about X-Plane in PC Pilot, or in the formal reviews on Avsim. If, say, Chris Frishmuth has anything to say about it at Sim HQ, I'd pay attention to that, too.Real coverage by qualified people, please.

Goran, how many autopilots have you written?

None. What's your point? Going from what you have said, there's really nothing to it anyway.

Goran, how many autopilots have you written?

Hey Ed, the link to your blog looks dead. Like this thread again.Oh, and I can tell a good cruise control in a car compared to an old 80's one. But I don't have to have written its logic, you can just tell.

Incredible, every single useful post that somehow compliments x plane gets degraded to a pointless "That's not a big deal" post or a "FSX can do that too" post by the exact same people.

My only point is that if the same author had trashed the X-Plane flight model and called FSX "hyperrealistic," the militant X-Plane supporters would be all over him. It cuts both ways. I'm happy to see X-Plane praised. It'd just be more meaningful to me if their credentials qualified them better to render an opinion. And I hope that if a credentialed, qualified reviewer said something positive about FSX, that'd be taken seriously by the X-Plane militants, even if they ultimately disagreeed with it. That's all.

AlanI'm sorry, but I would have to disagree with you there. Firstly, the author didn't trash FSX.All serious flight sim users have read PC Pilot or Computer Pilot at least a few times in their simming life. Those magazines promote FSX and it's add ons ad nauseam. I don't know of any x plane user voicing their disappointment over it. I see X Plane users disappointed there isn't more coverage of x plane, but other than that, there is no malice held towards the pro fsx reviewers. It is what it is and it's seriously, water off a ducks back.What X Plane users DON'T LIKE is being told that x plane is a joke, filled with false hype, and far from realistic. That's when the retaliations start coming out.

"filled with false hype" is not that inaccurate... but then, that's what they call marketing.However... my point in this thread has nothing to do with X-Plane... and aside from an unrealistic "hyper realistic flight simulator" statement... that article isn't about X-Plane either. I was speaking of the accuracy of the software being discussed in the article. It involves math and such... and as one who's written autopilot code, I'm saying that 90% success isn't good enough.