Flight Dynamics

[airtran411]

And let me take back what I said about Sanmalav's landings, lol I was just agreeing, but honestly I see Sanmalav's landings as highly realistic besides the low crossing, you would have honestly felt your were watching a real life landing when it comes to his videos.

 

That's how it's really done in my opinion, no re-climbing or bouncing back up, his are sweet perfect landings in my honest opinion!!!

[WarpD]

His landings would result in stall after stall after stall in the real world. Not good landings at all.

 

You seem to believe that aircraft can just come down, flare and hang... well, not so. When flown correctly, aircraft should touch down very shortly after flair. If the aircraft 'floats'... the airspeed is too high, bad landing. If the aircraft 'drops'... the airspeed is too low, bad landing.

 

What you think is a great landing, is actually a horrible, horrible landing. Seriously.

[jcomm]

And... remind me because I no longer recall...

 

In the cut made to some of the FM regs, I believe there was no harm done to ground effect tables. I do believe that propwash effects on the tail surfaces did receive a cut between fs9 and fsx / esp though... but it's been a long time since I was able to debate this with that great man, simmer, and expert Ron Freimuth "was"...

[JB3DG]

Ground effect has been simulated correctly in Bernt Stolle's FSX FDEs for ages now along with propwash effects on the tail surfaces. The only thing missing from FSX AFAIK is the additional lift generated on twin engine props by the prop thrust. Which can be simulated with a little trickery....

[jcomm]

Ah, ok Jonathan, thx for making it clear :-)

 

Yes, I've always heard the best comments about Bernt's models, although I never got the chance to try one :-/

[ZG862]

When flown correctly, aircraft should touch down very shortly after flair [sic]. If the aircraft 'floats'... the airspeed is too high, bad landing. If the aircraft 'drops'... the airspeed is too low, bad landing.

 

 

Ed, would it not be fair to say that this differs a little by aircraft type? Whilst I don't doubt that the above may well be true for an airliner, when flying a 152, I was always instructed to fly a little above the PAPI path and have the stall warner sounding a couple of seconds before the main wheels gently kiss the tarmac with the control column fully back against the stop. (Incidentally, the 172 has a well-known tendency to want to float if the CoG is forward (i.e. no passengers); I was told to let it do so rather make any particular adjustment to landing technique.)

 

Maybe it has much to do with my control setup (which presently includes an unsophisticated Saitek AV8R stick) or my fuel load (which is probably generally higher than I'd have in real life) but I can never get FSX anywhere close to this behaviour in these aircraft types. The flare simply does not appear to induce the amount of additional drag I would expect and trying to get the control column full aft always generates a pitch up that defies aerodynamics. I've only done a very few landings on P3D and on an aircraft type with which I have no RW experience so I couldn't comment on whether anything has been changed that would alter this behaviour.

 

Is it the case that P3D's FDE is inherently more realistic with airliners or is the behaviour I describe above a more general FSX flaw that has addressed? Or, do I just need better control hardware? B)  

 

Z

[jcomm]

Interesting topic :-)

 

As far as airliners go, in FSX, I had a good experience with the PMDG NGX and their 777 initial release. As a matter of fact, the 777 nailed the landing pitch ( tipically low ), and smoothness under good weather conditions, that I could observe in some RW videos, inside and outside cockpit.

 

I fly gliders IRL, and it's curious that this "ground effect" subject has been brought here because, one of these days I was commenting with one of my colleagues that I always found ground effect overmodeled on most sims compared to my own experience, and a glider, by it's inherent aerodynamical characteristis, should be prone to exhibit extra ground effect behavior... Fact is, we land with our "lift destrutor" ( read spoilers ) out, and it does a brilliant job at killing that effect :-)

[JoshuaC]

Just my 2 cents on the subject of FDE realism,

Having worked on at least a major airliner for FS and having talked to Real World Pilots, Level D Sim guys and some not so public sources, the conclusion reached was similar:

Sim Trainers, or any desktop PC simulation are meant to fly the numbers, nothing more. What this simply means, anything related to the word "feel" and such are highly subjective. No doubt, there will be countless claims of super accurate feel in some products description but what it simply is - flying the numbers. While flying the numbers does not mean its not realistic, it does means everything is on rails, no matter how you put it. 

No doubt, I don't claim to be a expert, and am not one, but well, a forum it is.

[Gregg_Seipp]

 

 

Ah, ok Jonathan, thx for making it clear :-)



Yes, I've always heard the best comments about Bernt's models, although I never got the chance to try one :-/

 

Guys, we found him...the one person who's never tried a Bernt Stolle FDE!

 

 

it does means everything is on rails, no matter how you put it.

 

Umm...well, it probably shouldn't be but I don't think that's the fault of the FDE developers...more the issue with FSX/P3D air modelling.  Watching hundreds of videos, if you watch the pilot's hands, watch the airplane get bumped and slightly rolled by even the slightest change in air, watching the speed tape jostle up and down.  Turbulence and air movement in the sim is just not modelled right at all.  Boy, does that need a makeover.

[TrafficPilot]

Real Air Simulations, Level-D and PMDG have shown that it is possible to produce aircraft addons with accurate flight dynamics.

 

I look forward to seeing more of them in P3D. Hopefully!

[jcomm]

 

 

Guys, we found him...the one person who's never tried a Bernt Stolle FDE!

 

:blush:

 

LOL !

[ZG862]

Guys,

 

Is it only me who is finding the terminology here really confusing? The OP wanted to know which bits of the FDE needed attention and yet we seem to be talking about add-in aircraft.

 

I thought that FDE meant flight dynamics engine - i.e. the software component in the sim responsible for taking a bunch of rules that have lots to do with physics and aerodynamics and applying them to an aircraft model which in the case of FSX/P3D is expressed in an .air file and an aircraft.cfg file in order to calculate the attitude and acceleration rates at any given point in the flight. People seem to talk about an FDE as though it is something that ships with an add-in aircraft - something that I understood to be technically possible but quite unusual. For example, I understood that the Sim Skunk Works Harrier ships with a dynamic link library that takes over from the FSX FDE at certain points when the nozzles go down.

 

Can someone enlighten me?

 

Confused of UK.

[jcomm]

Z,

 

you're right about it, and indeed the core FDE is where, in ultimate place, an MSFS aircraft will get it's flight dynamics, power, performance and effects reproduced. The other component is the AIR  file and the CFG file. Both are used by the FDE for the specific data pertaining to the modeled aircraft.

 

The use of external programming can indeed allow the devs to tailor some of the characteristics in order to better match the RW data. For instance, they can model flap retraction speeds that aren't standard, slats in the leading edge, tweak the instruments to hide internal limitations of the prop and jet engines in MSFS, like the well known leaning bug which cause FF to increase initially when you'r flying above 3000', and the higher you go the worst the effect, when you lean the mixture, when IRL it should always decrease promptly, or tweak the FF readings on a free-running turbine so that at constant altitude and condition / throttle if you vary your Prop RPM within reasonable limits the FF will stay constant, and many many other things, like manipulating hidden lift / drag generation surfaces, etc...

 

But even when this sort of deep programming is used, the last word will be from MSFS's FDE. The only way to do it differently will be to follow an approach similar to that of Majestic Software with their Dash 800, and model the systems and flight dynamics completely out of MSFS / P3D / ESP...

 

One of my simulators, ELITE, has at least one product ( FNPT ) using exactly this same approach with ESP for the visuals. This is the only possible way of doing "miracles" with what can't really be done with MSFS / ESP / P3D FD and Systems modeling engines.

[ZG862]

Thanks J.

 

So if I have it straight, what is being described by many of the people above is the Flight Dynamics Model for a specific aircraft and not the behaviour of the underlying FDE. However, since both work in conjunction, an improvement in the FDE might actually result in a worsening of a particular aircraft model's dynamic behaviour since the black art wizardry injected by the aircraft developer to overcome the original issue would most probably introduce a new undesired effect.

 

Ideally, every time the FDE is enhanced, 3rd part developers would release a patched .air file and/or aircraft.cfg - assuming of course that the 3rd party developers are well-informed about the nature of the changes being made.

 

Does this actually happen?

 

Z

[jcomm]

Ideally, every time the FDE is enhanced, 3rd part developers would release a patched .air file and/or aircraft.cfg - assuming of course that the 3rd party developers are well-informed about the nature of the changes being made.

 

Does this actually happen?

 

I think that would really be required if there were actually major updates to the FDE.

 

When I was using FSX, I bought a Robin DR-400, and was astonished to find out that the aircraft did present roll due to torque a little bit overdone, but at least it was among the many aircraft I had tried for FSX probably the best at reproducing the slipstream/propwash effects on yaw at high power / AoA situations. I did have to use rudder on a high power climb, just as I've seen being continuously used in real life on that same aircraft model under similar circumstances.

 

I soon found out the developers had cleverly associated throttle with rudder, and whenever we added power, and the AoA was within a given range, rudder was applied to create the yaw that MSFS's FDE would never be able to generate natively.... It's a bit like what talented developers will have to do in X-Plane 10 now that Austin gave them 6 new variables ( called datares in XP parlance... ) to freely create autonoumus forces or moments applied in different points of the aircraft, allowing for fine tuning of, for instance, the prop effects...