alpha testers are becoming more vocal

[robert young 2,897]

Regarding Flight Models, all the stuff about multiple aerodynamic points on the wings is frankly Public Relations and is unconnected with whether aircraft fly "accurately". Accuracy in this context is not so much about how many coefficient centres of lift are present, but how the overall connection between the stick and aircraft behaviour plays out. You can have ONE point of lift or aerodynamic wing behaviour that could work beautifully or you could have a hundred of them and if you tune them badly you might as well have none.

The thing most people do not grasp about flight modelling is that you have two rather conflicting pieces of information. One is the behaviour of aircraft in STILL air and the other is the behaviour of aircraft in a turbulent, moving airstream. The two are almost completely different. I don't have that many hours but over the course of my flight model tuning career I have been invited to fly a large number of different aircraft, or at least be in a right hand seat while someone else flies. What comes over strongly is that the tiniest little aircraft CAN fly incredibly elegantly without any turbulence.

I've been in many gliders, tail draggers and club aircraft that can fly as smoothly as an A380, provided there is no turbulence. The accusation that FSX and P3d aircraft "FLY ON RAILS" is a complete myth. Designing a sim aircraft that bucks and weaves in zero wind is not authentic. It depends how much turbulence, wind changes and pressure changes over the wings are present. In still air almost any decently designed aircraft WILL fly "on rails", in the sense that every stick input is entirely predictable. I've flown gliders in early morning and late evening that might as well have been a glorious flight in an A330 in a benign approach to a balmy mediterranean paradise, but I've also witnessed a 737 approach to Kefalonia that was a nightmare of frightening proportions. Both are possible depending on turbulence, NOT the flight model.

The other factor is stick force. Anyone who has piloted a winch launch in an ASK 21 glider or similar will know the large back stick forces required to fight the winch cable's desire to keep the nose down. This is the most regrettable part of flight simming for small aircraft: Force feedback is now not the fashion which means sim flyers can yank back a stick to full deflection effortlessly and then complain that so-and-so aircraft can get to 6g too easily. Without stick force restraints, many sim aircraft can become very twitchy in pitch and regrettably I have already seen from videos how twitchy the FS2020 aircraft are. The tell-tails are the tiny movements alpha-programme pilots are apparently inputting and the over-reaction I have seen from the flight models. The fact is that a real aircraft does not act like an uncontrolled pendulum. Nor do they buck up and down to modest inputs.

I once was taken up in an Extra 300 owned by the late Mungo Amyatt-Leir, at the time owner of Flight 1 Europe. We flew on a glorious day over the beautiful Sussex coastline and he generously let me fly a good part of the flight. I was amazed by the stability of the Extra. Despite having enormous ailerons you could fly it like a huge airliner with subtle inputs, all of which were perfectly weighted and balanced. There was no instability, and the stick forces were so perfect that you could elect to fly a passenger-safe elegant maneouvre with perfect harmony or you could choose a savagely quick roll or loop with incredible speed. Both were possible because the Extra has sublime characteristics. There was no hint of bucking or weaving. There was no ridiculously unstable pitch. There was no elastic-band spring-like reaction. The nose stayed where you put it.

Coming back to FS2020, I think you will find that the flight dynamics will be rather conforming to claimed mathematical or PR-lead publicity. This will not mean the core flight model will be either good or bad, but open to interpretation by others. But for now, be sceptical about those "multiple lift points". The visuals look stunning and I'm concentrating on those. Flight modelling is not a function of PR or announcements in promotional videos. I am pretty sure that the flight modelling will be as flawed as it ever was in previous versions.And NO, X plane is not really any better. Large, high inertia aircraft are quite easy to model. The litmus test is in small aircraft.

Edited May 20, 2020 by robert young
typo

[Claviateur 789]

1 hour ago, robert young said:

Accuracy in this context is not so much about how many coefficient centres of lift are present, but how the overall connection between the stick and aircraft behaviour plays out.

I do not know if I understood this correctly but if I did, I agree that if we focus on the numbers and the data or techniques when it comes to simulating flight models but at the end of the day, we forget that the virtual behavior of a virtual aircraft can simply act differently depending on the controllers we have or get completely ruined if the controllers are cheap, or uncalibrated in terms of precision and dead zones...

We must not forget the frame rate as well...

These are obviously impossible to control or predict or be part of any technique or data behind any flight dynamics no matter how well thought it is... 

For this reason, I learned through the years of flight simulation that many complaints regarding flights models are linked to these factors rather than the model itself...

So I think we must expect the same scenario will also happen in MSFS. This means the same virtual aircraft can have as many behaviors as the type and quality of controllers users use, their calibrations and dead zones settings as well as the simulator's fps...

Edited May 20, 2020 by Claviateur

[HiFlyer 9,359]

Or....... Perhaps we simply are incorrect to expect commercial-level simulation fidelity in our dens and living rooms.......

[fogboundturtle 2,708]

41 minutes ago, Claviateur said:

I do not know if I understood this correctly but if I did, I agree that if we focus on the numbers and the data or techniques when it comes to simulating flight models but at the end of the day, we forget that the virtual behavior of a virtual aircraft can simply act differently depending on the controllers we have or get completely ruined if the controllers are cheap, or uncalibrated in terms of precision and dead zones...

We must not forget the frame rate as well...

These are obviously impossible to control or predict or be part of any technique or data behind any flight dynamics no matter how well thought it is... 

For this reason, I learned through the years of flight simulation that many complaints regarding flights models are linked to these factors rather than the model itself...

So I think we must expect the same scenario will also happen in MSFS. This means the same virtual aircraft can have as many behaviors as the type and quality of controllers users use, their calibrations and dead zones settings as well as the simulator's fps...

you can expect the same flight sim guru who actually never flew a plane in their lifetime to complain about the accuracy of a flight simulator.

[Murmur 3,941]

Finite element flight models (like in the new MSFS, or X-Plane, or pretty much every other modern simulator) allow the modeling of some aerodynamic effects that are difficult or impossible to model in an FSX-like flight model.

First example that comes to my mind, the gradient effects a glider encounters when entering or flying near a thermal (i.e. updraft asymmetry between nose/tail or left/right wing). One could find more examples.

Of course, they also have their own idiosyncrasies or weak points, e.g. more difficult tuning to match real world performance/handling data, and often a tendency to be more twitchy compared to real life.

I for once am glad MSFS is switching to a finite element flight model, although most of its details are unknown. So it's still unknown how good or bad it will actually be, how easy will it be to create new models, etc.

[robert young 2,897]

14 minutes ago, fogboundturtle said:

you can expect the same flight sim guru who actually never flew a plane in their lifetime to complain about the accuracy of a flight simulator.

Not sure who you are talking about here. I got my PPL in 1973 and have flown around 30-35 aircraft types up to when I retired from powered flying in 2006 but then took up gliding and flew a further 10-15 different types of glider.

Edited May 20, 2020 by robert young

[fogboundturtle 2,708]

1 hour ago, robert young said:

Not sure who you are talking about here. I got my PPL in 1973 and have flown around 30-35 aircraft types up to when I retired from powered flying in 2006 but then took up gliding and flew a further 10-15 different types of glider.

I wasn't talking about you.

[robert young 2,897]

4 hours ago, fogboundturtle said:

I wasn't talking about you.

OK thanks. Best to say who you are talking about to save confusions!

[robert young 2,897]

13 hours ago, KenG said:

 The reality is pilots are a bunch of type A guys and gals who speak with little regards to someone's feelings. From the start of their pilot career their every action and word was critiqued bluntly. All pilot's have to have a thick skin to survive training as most briefings focus not on what you did right, but what you screwed up and could have done better. The reality of flight training is it is not for the sensitive. When the pilot becomes an Instructor the direct communication becomes reinforced.  It was not until I started instructing for a major civilian training corporation that I was sent to charm school. 

That's very true, and different instructors have a different opinion, withering or otherwise, for the same student. I remember working to re-solo a glider way back when I'd had a break of a few years. I had two instructors the day I was sent off. The first tore my flying and circuit to shreds. A few minutes later I went up with a CFI who remained utterly silent for the whole flight and I thought "here goes - another disaster". I flew the exact same profile, speeds and approach. Upon landing I expected another list of put-downs, but he said: "that was near perfect - off you go!"

Yet another instructor criticised my downwind speeds, complaining that I was incapable of maintaining exactly 60 knots. I impertinently suggested maybe the trend was more important than the second by second speed, and asked him to demonstrate. His speeds were all over the place, and far worse than mine. You have to be lucky to find a decent instructor who isn't programmed to rip you apart at every opportunity, but maybe things have got better nowadays.

[Dominique_K 10,147]

13 hours ago, robert young said:

 But for now, be sceptical about those "multiple lift points".

Glad to see you poking your nose in this forum ! I don’t know whether you are in the alpha but I wish you were. They need people like you.

My philosophy is that being sceptical and looking forward to it, is not incompatible 😉.

Chances are that the new flight architecture will take some time to mature in the hands  (so to speak) of modelers anyway. And could bring some improvements in the modeling.

As you say , still and turbulent weathers are two different things. What seems different with the new sim is a more refined weather model and air dynamics with downdrafts/updrafts generated by the relief (including buildings). That would make landing at the edge of a cliff or over a row of large trees more interesting. The present ESP architecture cannot take advantage of that or so it seems.

Also, wouldn’t the new multiple points architecture allow to refine the stall behaviour at the edge of the flight enveloppe like one wing stalling ? 

 

Edited May 21, 2020 by Dominique_K

[robert young 2,897]

11 minutes ago, Dominique_K said:

Also, wouldn’t the new multiple points architecture allow to refine the stall behaviour at the edge of the flight enveloppe like one wing stalling ? 

 

Not sure about that, but it should have better potential than the current single wing lift in FSX and P3d which you have to cheat in order to promote a wing drop or spin. The problem is often however that the more complicated the core flight model the more fiddly it is to tune because you might have many basic params that clash with each other. That's why I'm sceptical about the claimed multiple lift points. It might end up being better to have just two on each wing - outer and inner, for the sake of elegant simplicity.

Edited May 21, 2020 by robert young
typo

[Dominique_K 10,147]

3 hours ago, robert young said:

It might end up being better to have just two on each wing - outer and inner, for the sake of elegant simplicity.

It might be wrong but I understand that the 1000 points are an upper limit.

[Murmur 3,941]

6 hours ago, robert young said:

That's why I'm sceptical about the claimed multiple lift points. It might end up being better to have just two on each wing - outer and inner, for the sake of elegant simplicity.

Good idea, that would be an elegant compromise. Outer with ailerons, inner with flaps. Same for other surfaces like stabilizers. That would probably result in a flight model that is easier to tune. I'm very curious to see how this new MSFS flight model will be implemented and how flexible will it actually be.

[HiFlyer 9,359]

2 minutes ago, Murmur said:

Good idea, that would be an elegant compromise. Outer with ailerons, inner with flaps. Same for other surfaces like stabilizers. That would probably result in a flight model that is easier to tune. I'm very curious to see how this new MSFS flight model will be implemented and how flexible will it actually be.

With all the resources they have available, they could probably literally run the flight model through a supercomputer to get the proper parameters for a given aircraft.......

[KenG 492]

12 hours ago, robert young said:

That's very true, and different instructors have a different opinion, withering or otherwise, for the same student. I remember working to re-solo a glider way back when I'd had a break of a few years. I had two instructors the day I was sent off. The first tore my flying and circuit to shreds. A few minutes later I went up with a CFI who remained utterly silent for the whole flight and I thought "here goes - another disaster". I flew the exact same profile, speeds and approach. Upon landing I expected another list of put-downs, but he said: "that was near perfect - off you go!"

Yet another instructor criticised my downwind speeds, complaining that I was incapable of maintaining exactly 60 knots. I impertinently suggested maybe the trend was more important than the second by second speed, and asked him to demonstrate. His speeds were all over the place, and far worse than mine. You have to be lucky to find a decent instructor who isn't programmed to rip you apart at every opportunity, but maybe things have got better nowadays.

Ouch, well I will not backseat critique another instructor having not been there and basing it solely on the observations of the pilot under instruction. Hopefully there was a Chief Flight Instructor, Quality Assurance Instructor, or Chief Pilot you could raise your concerns with.

As someone who teaches only professional pilots to a type rating standard all of my critiques strictly follow the guidance of the ATP ACS or the procedures in the AFM. So if I say you need to tighten up airspeed control, I base it on the expectations of written requirements. I'll admit there are some examiner expectations and I will caution pilots of those as well. There is usually some level of interpretation of the guidance or the FAA / local FSDO has put something out. However, I have gone toe to toe with the center examiner when they expected performance outside of written guidance. 

In any group, especially the larger the group gets, you are going to have those who act highly professional and those who are less than professional. There are plenty of videos of professional pilot acting juvenile with expensive turbojet aircraft. The problem is you have Hollywood who goes out of the way to heroize the bad boy image. Who can forget Maverick buzzing the tower in Top Gun. Who also can forget USAF Lt Col Holland who in 1994 decided the rules did not apply to him.

With that lets cue in the OT police.