Flight model comparison - will it ever happen?

[n4gix 4,947]

On 10/5/2020 at 11:05 AM, eslader said:

Never mind that anyone who seriously thinks a "professional flight training system" that includes default aircraft such as the F-22 which no real pilot will train on in P3d, as well as weird stuff like a "flyable" nuclear missile submarine is actually a professional flight training system is drinking way too much of the Kool-Aid. 

In reality, none of L-M's clients use the default aircraft, having commissioned custom aircraft and panel/gauge systems from professional developers.

See: https://www.prepar3d.com/case-studies/multi-function-training-aid/

Also: https://www.prepar3d.com/case-studies/

[Noel 4,502]

2 hours ago, Chock said:

Nah, speaking as someone who actually died (and was revived obviously) following a motorcycle crash, and who had to have extensive surgery to sort me out including several metal plates in my skull and some reconstructive surgery, I can tell you that motorcycling is ALWAYS dangerous! 🤣

That motorcycle crash wasn't my fault by the way; it was caused by a private hire taxi driver pulling out on me - he buggered off from the scene of the crash, and the police never found him despite trying to. 😠

 

Nah?  Nah what, the actual data, or your one anecdote?

[HighBypass 2,339]

On 10/5/2020 at 4:30 PM, Noel said:

and right about the time the rider has gotten comfy w/ the basics of operation crash risk goes up so the data shows

I don't think that's in doubt, but even the most experienced rider cannot stop the attack of the stupids, although they can mitigate a lot of those instances due to experience.

Any driver or rider with a significant amount of time spent operating their machine develops the ability to, if not predict the future, to second guess what might happen - we've all felt this - that gut feeling: "That car's not looking my way.. I wonder if.." That truck might be wanting to run wide here to turn left into that industrial estate up ahead" "that jaywalking pedestrian with the headphones on will not even check before crossing that road junction" etc.

So, yes motorcycling is ALWAYS more risky than driving a vehicle 😉

Landing a Blackburn Beverley at Courchevel is ALWAYS possible in the sim! 

[ErichB 3,643]

After many years of flightsimming, I have come to the conclusion that there is no such thing as a realistic flight model on a home based sim.   And let's be honest, most people wouldn't know what a 'realistic'  flight model is anyway.  People assume realism based on what they 'think' an aircraft should fly like.  Most here won't even reference the numbers .  To most, it will either feel plausible (i.e. feel realistic) or it doesn't.

Discounting obviously flawed flight models on certain badly finished aircraft, you need to add to that the multitude of controller variables (crappy to solid/high quality) which  will also majorly influence a sense of what is feels 'real' on a PC.  You could probably install  a NASA flight model in a sim and fly it with a cheap £40 controller and it would still 'feel' unrealistic. Invariably, you'd get someone complaining about the flight model instead of addressing the cheap interface controller infront of them, or  badly adjusted controller settings.

Instead of chasing the red herring of a realistic flight model, people should instead be more holistic about their flightsimming and assess what combination of variables (including a well modelled and designed aircraft) gives them the best level of immersion.     The human brain is very good at filling in the blanks of things which are absent to give you a sense of immersion.  But that is also highly subjective and dependent on what any individual is willing to settle on.

In flightsimming terms, one size has never fitted all.

 

Edited October 7, 2020 by ErichB

[KenG 492]

14 hours ago, n4gix said:

In reality, none of L-M's clients use the default aircraft, having commissioned custom aircraft and panel/gauge systems from professional developers.

I had not considered none because at the BATD level they could use the default model as long as it met the very low fidelity requirements by the FAA.

The list of approved BATD and AATDs are small. https://www.faa.gov/about/office_org/headquarters_offices/avs/offices/afx/afs/afs800/afs810/media/FAA_Approved_Airplane_ATDs.pdf

Companies listed with an asterisk (*) do not sell devices.

Under two dozen companies sell devices and only about a half dozen even sell BATDs. Out of them I can only think of a few that even use LM P3D. 

I guess even at the $8000 price point you are getting a bit more than a vanilla copy of P3D with a BATD.

[Noel 4,502]

1 hour ago, HighBypass said:

So, yes motorcycling is ALWAYS more risky than driving a vehicle 😉

 

So, yes I don't think that's EVER been in doubt either.  

[eslader 1,783]

15 hours ago, n4gix said:

In reality, none of L-M's clients use the default aircraft, having commissioned custom aircraft and panel/gauge systems from professional developers.

Yeah, that's another point I was hinting at about the absurdity of some of the reactions. MSFS is just an Xbox game because 3 minutes after launch the jets didn't work very well, but P3d is an absolutely 100% almost level-D sim even though everyone who does any real training on it uses 3pd content to pull it off.

 

[RaptyrOne 1,412]

1 hour ago, ErichB said:

After many years of flightsimming, I have come to the conclusion that there is no such thing as a realistic flight model on a home based sim.   And let's be honest, most people wouldn't know what a 'realistic'  flight model is anyway.

I think it is very possible to get an accurate, realistic flight model. The bigger issue is controllers. No matter how realistic or plausible the flight model, using a basic joystick on an aircraft that uses a yoke, or an Xbox controller on an aircraft that has a stick (just as examples), would make the whole discussion less relevant. These controllers have an overriding influence on what is realistic in a PC based sim.

Edited October 7, 2020 by RaptyrOne

[eslader 1,783]

Controllers are important, but aren't really relevant to discussions of a flight model. The yardstick for flight model quality should essentially begin and end with "whatever position the control surfaces/throttle/etc are in, does the virtual airplane react in a similar way to its real-world counterpart flying in identical atmospheric conditions and at identical speed?" 

Real airplanes can have differing methods of control inputs. Paraplegics fly with hand-controlled rudders. That doesn't mean their airplane isn't realistic, it only means the controls are different than an unmodified plane. Same story for whatever controllers we're using. Whether you're using an Xbox controller, a Warthog, or even just a keyboard, if you deflect the aileron 50% the plane should do whatever the real plane would do at 50% aileron deflection. If it does, that's a point in favor of the flight model. If it doesn't, that's a point against.

 

 

Edited October 7, 2020 by eslader

[KenG 492]

24 minutes ago, eslader said:

Controllers are important, but aren't really relevant to discussions of a flight model. The yardstick for flight model quality should essentially begin and end with "whatever position the control surfaces/throttle/etc are in, does the virtual airplane react in a similar way to its real-world counterpart flying in identical atmospheric conditions and at identical speed?" 

This yardstick is not available to your average developer unless they are going to spend lots of money for the 60.13 data. You are trying to measure against something that has no objective criteria so then you only have subjective criteria in which the quality of the controller does matter.

There is a reason we don’t put cheap game controllers on Full Flight Simulators. In fact the controls on an FFS cost way more then most AATDs. 

[eslader 1,783]

47 minutes ago, KenG said:

There is a reason we don’t put cheap game controllers on Full Flight Simulators. In fact the controls on an FFS cost way more then most AATDs. 

But there's also a reason you don't put expensive controls on an FFS running Airline Simulator 2. FFS designers work their butts off getting that simulated plane to behave as close as possible to the real thing. And yes, I know that one 747 will behave slightly differently than another, but that's a red herring. All 747s operate within an envelope of flight characteristics. You will not see a tubeliner flying like a Cub, ever, except in the Grand Theft Auto series, and that's why those games aren't considered to be good flight simulators. 

While most of us would love to raid a boneyard and come home with a full airliner cockpit for better controls, we can't afford to. We'll always be flying with a joystick, or maybe a Honeycomb yoke, and it's not gonna behave like the yoke in the 747, but at least the virtual 747 can behave like a 747. In other words, home sims should concentrate on the things they can control, and one of them is making the pretend plane behave within the envelope of behavior exhibited by the real thing. I absolutely expect the developers charging $100+ for a single plane to get access to whatever information they need to make that thing fly as realistically as possible. For the cheaper developers, I'm fine with just a pretty-good approximation. In other words, if I'm able to get my Icon up to FL400 and cross the Atlantic in 5 hours, that's a problem. 😉  More seriously, you don't need access to the high end data to find the fuel burn, service ceiling, etc of a given airplane. And if you can't afford the 60.13 data, you can at least afford to go down to the local airport and rent a C172 to fly around to see how close it behaves to the C172 you're developing for a sim.

 

 

Edited October 7, 2020 by eslader

[Janov 2,383]

I think the verifiable flightmodel in a desktop sim boils down to things like:

• Does the aircraft perform realistically (climb rates, glide distance, maximum speed, service ceiling, takeoff run at weight X)?
• Does the aircraft exhibit correct flight-control response (roll rate at certain speeds, elevator authority, rudder authority)
• Does the aircraft show correct pitch/power/airspeed performance (i.e. flies level at 100kts IAS with X degrees nose up)

I think most of these things are "fairly" easy to evaluate with the help of a real plane and a stopwatch and a POH. Anything beyond that "the plane feels too nimble, the plane feels too stable, the plane just feels right" is drifting off into the personal hear-say anecdotal range and you won´t get two pilots to agree here unless something is seriously wrong.

Now wether a plane in a desktop simulator gets these things right is dependent on two things - A.) the capability of the desktop simulator and B.) the capability of the aircraft model designer. If the flightmodel is wrong, either (or both) of those could be at fault.

It is therefore difficult to judge the capability of the simulator by the aircraft it includes. If an airplane gets the above stuff right, the simulator is good. If it doesn´t the simulator may still be good (but the aircraft designer inept).

As time goes by and more knowledgable and patient designers start making aircraft for a simulator the true limitations of it become more obvious.

Edited October 7, 2020 by Janov

[bean_sprout 1,201]

Regarding controllers - XBOX users will be using the (some call it awesome) XBOX wireless controller.  So once again we are back to the arcade theme.  Some jaded simmers say that it is not bad at all.

As some have indicated a computer flight sim is not equivalent to a commercial simulator / real aircraft.  However it does approximate the real thing and with the great FS2020 environment it does that really well.  What more can we ask for?

Cheers

bs  

[Tierborn 188]

9 hours ago, ErichB said:

After many years of flightsimming, I have come to the conclusion that there is no such thing as a realistic flight model on a home based sim.   And let's be honest, most people wouldn't know what a 'realistic'  flight model is anyway.  People assume realism based on what they 'think' an aircraft should fly like.  Most here won't even reference the numbers .  To most, it will either feel plausible (i.e. feel realistic) or it doesn't.

I would go one step further, most people make conclusions on flight model based on the companies branding of its flight model.  Xplane spent tons of time and money "branding" its flight model so many were influenced by the branding and believe that Xplane has the best flight modeling... MSFS20 branding of its flight model will eventually catch hold and people will then believe that MSFS has the best flight model.   For most non-real world pilots their judgement of quality of flight model is based on "branding".

[n4gix 4,947]

4 hours ago, bean_sprout said:

Regarding controllers - XBOX users will be using the (some call it awesome) XBOX wireless controller. 

I have a wireless XBox controller, but all I use it for is to control the drone, which frankly doesn't work well with yoke and rudder pedals! 😅