A little Heads-Up on the NGX...

[yos233 1]

Thats awsome. Is there no end to the level of detail you guys are going too?. I would hate to see how much code it takes to model an aircraft to that level of detail. One thing im wondering while we are on the faliure subject is will we be able to select time between fails. I mean the MD-11 and 747-400 are stuck at 10 hours and cant be changed. Id like to be able to set it to a diffrent amount like 2 Hours or 2 days. With the Leonardo MD80 maddog you can set the fail ratefrom minutes to months if you want. Would be nice but I understand if not.

You can do that, as far apart as 1 per 20 hours. But you can increase the number to like 10 failures per 20 hour period so thats 1 per 2 hour period.

[Roadburner426 19]

Very interesting details. Obviously people won't be flying 12 hours at a time (well, some of us might in the beginning). So this means that these hours are cummulative across saved flights ,or in a special PMDG module that tracks hours across a particular frame? Suppose these are highly detailed questions that won't be answered till later, but just what is on my mind at the moment after reading the above. Certaintly interesting that you model fluid foaming, and it tricking the reservior high levels/cavitation of the hyd pumps. Must be some really complicated math in there across all the different fluid/pneumatic types across all the temperature ranges.. Look forward to breaking her.

[michal 44]

Obviously people won't be flying 12 hours at a time (well, some of us might in the beginning). So this means that these hours are cummulative across saved flights

I can't speak for PMDG, I don't write software for them but the best implementation would not need to get into such things as "flying 12 hrs at a time" or adding total flight hours, etc. Totally unnecessary. This is all about probability, Poisson distribution, etc. The same way you can walk into a casino, play only for 10 mins and hit double-zero twice in a row on a roulette table even though according to math you should have spent at least 10 yrs in a casino to have reasonable guarantee of such a win. Failures happen totally randomly and not based on some elapsed hours or what happened or did not happen in the past.

[sddjd 117]

...Failures happen totally randomly and not based on some elapsed hours or what happened or did not happen in the past.

+1 Would love to see this implemented; same as a coin toss- you're odds aren't based on how many heads or tails you've flipped already, they "reset" every instant.One question, are failures related in that a failed system might stress other systems, leading to the infamous cascades of failures?Dan D.

[yycvor 5]

And what if you DID select this failure, or the computer selects this failure for you as a random failure- how long will it take to show itself?Somewhere between now... and twelve hours from now. The tanks do a good job of holding pressure...So there is just one example of how the NGX might surprise you!

So with that said, Im not one to ever pre-select a failure for when I’m simulating a typical commercial flight. Unless I feel like brushing my skills on a "training flight". This new failure model excites me very much. Is it safe to assume there is certain dispatch reliability built in to the model? And am I going to experience these random failures at a real world realistic rate of hours flown vs tech problems. Obviously its difficult to predict when things break, but Id like to know what are the chances of ever actually experiencing a random engine failure. I've seen in an company internal SMS video stating that pilots will probably never experience in their entire ATP carrer, given today’s engine reliability, and in particular the CFMs on the NG.

[michal 44]

Not sure what you mean but it is safe to assume that you will have an option to disable all failures. As to "random failures at a real world realistic rate of hours" - such a thing would mean that you might have to fly for literally hundreds/thousands of hours to experience a single failure hence user selectable time period is really the most sensible option.

[Tabs 2,790]

I'll let RSR explain the failure system later on, but you guys don't need to worry, I think it's going to exceed everyone's expectations.

I demand a failure that turns the 737NG into a Woodpigeon! :(

Hilariously enough, a hidden button that turned the external model into one was actually discussed haha.

[yycvor 5]

Not sure what you mean but it is safe to assume that you will have an option to disable all failures. As to "random failures at a real world realistic rate of hours" - such a thing would mean that you might have to fly for literally hundreds/thousands of hours to experience a single failure hence user selectable time period is really the most sensible option.

I dont know about you but I do plan on putting thousands of hours on the NGX. With that said, I'm am indeed hoping that I would have to fly for literally hundreds/thousands of hours to experience a single failure rather then the "sensable" user selectable time period. With the strive is for realism at every detail, it seems silly to have a aircraft breaking at a rate 10x faster then the real world. Im a customer either way obviously :)

[michal 44]

it seems silly to have a aircraft breaking at a rate 10x faster then the real world.

I don't think it is silly. In commercial flight simulators they cause failures "on demand" so pilots can actually train to cope with them. A typical simulator session means pilot has multiple failures thrown at him/her. By the way, this is actually one of the most obvious benefits of having simulators - ability to practice events that can be exceedingly rare in real life.

[ExNusquam 11]

Hilariously enough, a hidden button that turned the external model into one was actually discussed haha.

Did it make it into the final product? :(

[Tabs 2,790]

I dont know about you but I do plan on putting thousands of hours on the NGX. With that said, I'm am indeed hoping that I would have to fly for literally hundreds/thousands of hours to experience a single failure rather then the "sensable" user selectable time period. With the strive is for realism at every detail, it seems silly to have a aircraft breaking at a rate 10x faster then the real world. Im a customer either way obviously :)

Both a realistic rate based on real life failure probability per componet and user selectable rates will be there guys, like I said don't worry.

Did it make it into the final product? :(

No, the model's already the most complex ever made for FSX and we didn't want to add a ton more by creating a second hidden aircraft haha.

[yycvor 5]

I don't think it is silly. In commercial flight simulators they cause failures "on demand" so pilots can actually train to cope with them. A typical simulator session means pilot has multiple failures thrown at him/her. By the way, this is actually one of the most obvious benefits of having simulators - ability to practice events that can be exceedingly rare in real life.

Im all for "on demand" For when I feel like doing a "training flight"Don’t take this the wrong way but I fully understand the purpose of real world flight simulators, WestJet's 3 CAE 737NG full motion simulators are about 100 feet from where I'm currently sitting. I’ve got a few hours on them too, and I know exactly what and why things are done the way they are in there. I’m also more then just a desktop flyer, so what you’re saying is not new to me or most in this forum for that matter....With that said, I don’t fly a NG in real life, I’m not in MSFS to prove I can still hold my type rating when *hit hits the fan with lives on the line. Rather I enjoy replicating typical airline operations, and I think it’s far more enjoyable to know that the minor wear and tear issues could sneak up on you at a rate comparable what’s happening in the real industry. Hitting a button then knowing deep inside that’s something is going to go wrong in the next XX hours is a very simple way to failure models if you ask me. I prefer my aircraft to make it to the destination I pointed it at, little tech issues added randomly and at the proper rate will just add to the already insane realism. I think a mix of both is the way to go, which as Ryan stated appears to be the way PMDG is moving. Being able to select major and minor failures is great for when I want to learn the process of dealing with what ever issue was selected is nice, But think of how enjoyable and rewarding it would be If you ended up experiencing something major randomly on a typical point A to B flight weeks, months or years after you’ve gotten complacent with the NGX.To each his own...Cheers Besides, Ryan has stated RSR will clarify things in the future. Edited November 30, 2010 by yycvor

[NewChucky 0]

Now I'm really starting to get interested. A 'realistic' failure model that lets me complete my flights without too much drama but actually makes me keep an eye on the diagnostics is exactly what I've been wanting for a long time. I have a couple of questions about it though. Is there any way it would be able to interact with the FSX mission structure? It might be fun to have people put together training flights so that we don't know what's coming, but have something a little bit more structured than a random generator. Another minor feature that I would love is a little bit of startup randomization - switch positions and things like that, which I guess aren't outright failures. My experience flying real aircraft is that you never know how the last pilot left the aircraft - maybe a transponder left to the old squawk code, maybe the autopilot switch still on (that's always good for a laugh!), maybe the battery switch left on overnight.... I find in FSX after a certain number of flights I start doing my startup to a given panel state since I know which switches I don't even need to check. These are pretty minor issues, I guess.Of course, none of this will likely matter to me, since it looks more and more like you're going to manage to release it right when I'm in the middle of writing up my thesis. :( Oh, well. - Charles

[PUP4ORD 1]

In the Level D simulators, there's a multiude of different senarios that could happen during flight. The main key to all of that training is be prepared and to be aware.

[ahsmatt7 1,213]

Now I'm really starting to get interested. A 'realistic' failure model that lets me complete my flights without too much drama but actually makes me keep an eye on the diagnostics is exactly what I've been wanting for a long time. I have a couple of questions about it though. Is there any way it would be able to interact with the FSX mission structure? It might be fun to have people put together training flights so that we don't know what's coming, but have something a little bit more structured than a random generator. Another minor feature that I would love is a little bit of startup randomization - switch positions and things like that, which I guess aren't outright failures. My experience flying real aircraft is that you never know how the last pilot left the aircraft - maybe a transponder left to the old squawk code, maybe the autopilot switch still on (that's always good for a laugh!), maybe the battery switch left on overnight.... I find in FSX after a certain number of flights I start doing my startup to a given panel state since I know which switches I don't even need to check. These are pretty minor issues, I guess.Of course, none of this will likely matter to me, since it looks more and more like you're going to manage to release it right when I'm in the middle of writing up my thesis. :( Oh, well. - Charles

In the airline world, everything is done to the checklist. At least it should be. In most cases, in line with company SOPs, the aircraft will usually be in the same state every time you arrive to it. Provided there aren't any special circumstances. That's why there are shut-down and securing checklists. Since these should be done at the end of every flight. The next flight crew will usually see the aircraft in the same exact state as the last flight they flew when they arrived. Once you arrive to the aircraft, you usually have to do a safety flow and sometimes a power-up flow. After years of experience in a certain plane, you will know the flow and aircraft by heart and then you will really notice when something is out of place. If it isn't, then you wont notice anything.