Real 737 pilot praises Zibo flight model.

[Murmur 3,941]

1 hour ago, SAS443 said:

I support your notion that the default C172 is incorrect in behaviour..

Observe the slip/skid indicator during normal climb attitude (image 1) and near upset attitude (image 2). Both at full power using zero rudder input.

Last image is just not right. Don't believe me? (after all I'm only a dumb private pilot who flies the C172S)....Just read this snippet from Aviation Safety article "Proper Rudder Usage" 
(bolded emphasize mine)

I tried a full power stall on the default C172 in XP11 (both with analog gauges and the G1000), and the ball/slip indicator is definitely to the right when on the verge of the stall (and even before):

True, apparently it should be even more to the right. Still much better than the competition though, where there's nothing at all!

 

[JRBarrett 3,070]

6 hours ago, jarmstro said:

Is that the FSX lady in the vid? What was her name now....I can't remember but I miss her.

Martha King

[Bjoern 1,157]

The default C172 isn't perfect. Otherwise, this wouldn't exist: https://store.x-plane.org/C-172S-C-172R-C-172M-Enhanced-Flight-Models_p_1200.html

[jcomm 5,920]

I use the Simcoders REP and the Airfoilabs versions. I do prefer the REP version.

Real 172S / R pilots have commented that it really depends a LOT on the particular aircrfat & rigging.

Wasn't aware of that EFM variant...

 

Edited May 20, 2022 by jcomm

[Murmur 3,941]

1 hour ago, Bjoern said:

The default C172 isn't perfect. Otherwise, this wouldn't exist: https://store.x-plane.org/C-172S-C-172R-C-172M-Enhanced-Flight-Models_p_1200.html

Never said it was, for me the exaggerated pitch change with flaps is the most annoying flaw.

Anyway, that EFM C172 looks like a great product! It shows how powerful XP flight model is, matching the performance of the RL aircraft very closely, without using an external flight model.

 

[jarmstro 2,402]

1 hour ago, Murmur said:

matching the performance of the RL aircraft very closely,

Closely. But not accurately. Because it can't without accurate atmospheric modelling which is impossible. Without which it is not realistic,

Edited May 20, 2022 by jarmstro

[blingthinger 649]

On 5/19/2022 at 2:18 AM, 2reds2whites said:

your opinions are embarrassingly wrong

MY opinions? I'm quoting a pilot. And the SDK, but we'll get to that in a sec. Let's look beyond the fact that your ego...actually, let's first pretend that you actually are a pilot. If you are in one of the $$400000/hr 737/738 sims sitting in your living room (you know, the ones you were bragging about), what happens when you stomp on the rudder? Does this happen?

https://youtu.be/oq_X6I8DiDU?t=12036

Yes or no. It's a simple question.

As of now, it doesn't appear that there are any real 737 pilots out there willing to do a back to back comparisons like flightdeck2sim has. And even if you do answer, a great many internet readers/watchers will have no way of believing you to the level that flightdeck2sim is trusted. Do you have a youtube site? Do you have any verification? He has years of history making very public opinions that rarely, if ever, get challenged. Furthermore, that is but one of many of his "opinions" about the PMDG flight model, which did leave him "pleasantly surprised".

 

On 5/19/2022 at 2:18 AM, 2reds2whites said:

'read a few SDK's.'

https://docs.flightsimulator.com/html/Samples_And_Tutorials/Primers/Flight_Model_Physics.htm

Given that you look down upon those who read SDKs, I'll quote it for you:

"To this end, an original normalization algorithm has been developed that helps redistribute global aerodynamics coefficients and tables - as provided by user - across all surface elements so that the final forces and moments match the ones computed by FSX. Of course, this does not mean we are finally equivalent to FSX modelling, as forces are now distributed over all the aircraft geometry. Instead it means that - when summing up all contributions and thus losing geometrical distribution information - our more general model can be reduced down to an approximation of the FSX legacy model. This makes our model an extension of the legacy model, which is in accordance with the objective of retro-compatibility."

Firstly, the person(s) at Asobo who that page deserve a pat on the back and a raise. It's generally well written. 

Second, it's clear that Asobo's hands were tied by MS to retain "retro-compatibility". The FSX model remains the cement block on the feet of their 'force element' model, or whatever they are calling it these days. I recall early on they were declaring it a blade element model: clearly an attack on LR. But they've since backed down from that marketing tag line. And let's be honest here: Austin has been publicly lambasting MS flight sims for many years. "Eye for an eye" is the world we live in.

What's more, every single aircraft in MSFS, internally looks like this:

Though, the swept wings in this picture are incongruent with their previous description on the same page:

It very much appears that the SDK is saying that everything: biplanes, airliners, twin booms, single engine pistons, my grandmother with wings, etc., gets boiled down to the "Planform Model". This type of modeling gets you to 'good enough'. Airliners and autopilots. 1D...actually more like 1.5D here. The force element method does give it more dynamic response. But you still end up with developers complaining about this:

https://devsupport.flightsimulator.com/questions/6043/is-there-anything-which-can-control-behaviour-of-t.html

"By 'tuning' the deflection angles, do you mean moving away from the correct, documented angles published in the manufacturer's maintenance manuals?"

Developers are having to significantly deviate from REAL LIFE to make this work. I know everyone is aware of this at this point. I'm also not claiming that XP's model doesn't need hacks now and then: MS/Asobo even take a couple shots at LR in the SDK. Some of them are true. But XP is fundamentally superior in terms of flight modeling based simply upon what is available in the geometric model capabilities, not to mention the explicit reduction in aerodynamic parameters needed for MSFS (SDK verbage). So, yeah, I look forward to XP12's additional 'aerodynamic goodies'.

Reading through various sections of the SDK gave 'little old me' more respect for Asobo but it also occasionally made. me. laugh. Out loud at times. Jokes also make me laugh.

 

[blingthinger 649]

On 5/19/2022 at 11:40 AM, rka said:

I get it now, thanks.

Good for you. It's important to include many data sources when making life decisions: in this case his history of very public youtube content that continuously goes unchallenged. 

[blingthinger 649]

On 5/18/2022 at 10:25 PM, mjrhealth said:

I guess you where not around when they tried to flog there lovely was i the DC6. Nice plane, that was there test in the xplane market. I think it was too much Criticizm, and so they left for an easier one. Really dont care if you believe me or not. You asked I told you.

I'm just asking for proof. From anyone. I don't care if it's you or not. The fact that you say "I think it was too much Criticizm" certainly gives me no reason to trust your statements.

[blingthinger 649]

On 5/19/2022 at 3:07 PM, Murmur said:

I tried a full power stall on the default C172 in XP11 (both with analog gauges and the G1000), and the ball/slip indicator is definitely to the right when on the verge of the stall (and even before):

 

True, apparently it should be even more to the right. Still much better than the competition though, where there's nothing at all!

These variables in the MSFS SDK are interesting:

presspt_fwd_AlphaStall_pMAC, presspt_fwd_AlphaHiStall_pMAC

https://docs.flightsimulator.com/html/Content_Configuration/SimObjects/Aircraft_SimO/Flight_Model_Definition.htm

 

It looks like developers might have to literally move the wing along the aircraft axis when near stall, in order to get the dynamics closer to the real thing (effectively shifting center of pressure). Though this doesn't impact the slip dynamics per se.

In fact, I wonder if their latest CFD capability has helped with this behavior. The only thing my imagination can attribute it to is a lack of prop wash on the tail which is supposedly present now??

[Murmur 3,941]

26 minutes ago, blingthinger said:

what happens when you stomp on the rudder? Does this happen?

https://youtu.be/oq_X6I8DiDU?t=12036

Yes or no. It's a simple question.

I missed that detail! Yes, most XP aircraft have the typical strong yaw/roll coupling (dihedral effect), especially on airliners and other swept wings, which was always missing in the competition.

Many years ago, there were several 737 crashes due to rudder runaway, the aircraft rolling out of control just like shown in the video with the Zibo. Looks like the payware offering of the competition is not up to par on these details of the flight model...

[blingthinger 649]

20 minutes ago, Murmur said:

yaw/roll coupling (dihedral effect), especially on airliners and other swept wings

Hold up...I can see how positive dihedral (which normally stabilizes roll moments) would contribute to this coupling: yaw accelerates one wing, causing additional lift, that also happens to be angled into the rolling direction, right? But how does wing sweep contribute to this?

[rka 916]

1 hour ago, blingthinger said:

Good for you. It's important to include many data sources when making life decisions: in this case his history of very public youtube content that continuously goes unchallenged. 

It's very kind of you to spread your wisdom here, and for free 😉

[blingthinger 649]

1 minute ago, rka said:

and for free

Free? You should ask about the membership upgrades on this site. I get internet points. It makes me feel good about myself.

[Murmur 3,941]

20 minutes ago, blingthinger said:

Hold up...I can see how positive dihedral (which normally stabilizes roll moments) would contribute to this coupling: yaw accelerates one wing, causing additional lift, that also happens to be angled into the rolling direction, right? But how does wing sweep contribute to this?

The more a wing is swept, the less its lift slope (at first approximation, only the velocity perpendicular to the c/4 line produces lift). So if there's e.g. right yaw, the left wing will be less swept and produce more lift, and viceversa for the right wing, hence a right rolling moment.