Cheers, @simbol! That was a very good explanation and it helps me appreciate even more the hard work and headaches you devs go through to give us the planes we love.

So was perusing the iniBuilds discord as the action and anticipation there is heating up due to the coming A310.  And the topic of their recently announced payware P-40 came up and their announce video at this point https://youtu.be/FhLXEV5ytXA?t=21 where "comprehensive use of CFD flight model technology" is mentioned. Good to see a dev powerhouse like iniBuilds starting to embrace this new aerodynamics technology in MSFS and which means it'll be used in their future payware/freeware birds.

So the iniBuilds devs chimed in further (the user names in green) with their thoughts on MSFS's CFD in FMs:

https://discord.com/channels/535246634448191499/535249224254619648/1038457125971693578
 

MJ — 05/11/2022

     "it’s our own plane"

iamdumdum1234 — 05/11/2022

     "ah sweet, CFD physics"

Josh — 05/11/2022

     "Yeah it's a really cool addition to the sim's flight model from Asobo and has received a lot of work from them for helis and gliders"

iamdumdum1234 — 05/11/2022

     "this is really gonna improve msfs's physics by alot"

Josh — 05/11/2022

     "The C172 with CFD flies really nice"
 

Edited November 10, 20223 yr by lwt1971

 

9 hours ago, simbol said:

After the first step is completed, you need then enable CFD, as a result the first flight model gets "de-calibrated"

Fascinating. Thank you for the designer's perspective! 3rd party devs are the patient magicians of the community.

Can you elaborate on "de-calibrated"? What aspects of the baseline flight model generally need to be adjusted afterwards?

 

9 hours ago, simbol said:

Any CFD enabled airplane will behave much closer to real world than those that are not

What is it about the force element model that necessitates the CFD? It sounds like what you're saying is that the force element model is missing aerodynamic parameters that would otherwise add to the realism of the flight model, correct?

9 hours ago, mrueedi said:

If it is not clear to you that forces are calculated per surface after reading the SDK document, I might direct you to a youtube video telling you so: click here

I can only quote the SDK:

"We then initialize the algorithm by attributing - to each surface element - local lift, drag and moments coefficients, simply by uniformly distributing aerodynamics coefficients and tables provided by the user over all surface elements with a basic surface-area-based weighting. We then iteratively compute the same 8 global coefficients we have retrieved from FSX as explained above, but using the new surface elements model."

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

By all means, help us understand this, then. Firstly, I don't see any mention of the word "force" here. I do see "coefficient" quite a few times. Coefficients seem to be distributed (actually scaled by area-weighting function, based on element size), not forces. I also see "surface elements model". But not "force elements".

In fact, "force element" doesn't show up anywhere in the 8000 words. Not once. I realize the videos use that term, but why then is the SDK different?

I do also see reference to FSX many many times: Use of the the FSX tables. Normalizing to match the FSX tables. Etc.

Help, mrueedi. SDK says one thing. You say another. So confused here.

 

11 hours ago, Krakin said:

I think including the legacy FM is more trouble than it's worth.

100% agree. If someone in that corporate mix hadn't demanded backwards compatibility with FSX products (does it ever get used??), the model could have been so much better. Based on the SDK, FSX remains at the core of the force element model.

So sad.

27 minutes ago, blingthinger said:

100% agree. If someone in that corporate mix hadn't demanded backwards compatibility with FSX products (does it ever get used??), the model could have been so much better. Based on the SDK, FSX remains at the core of the force element model.

So sad.

Like mrueedi called out, you keep going on about the "force element model" without explaining what in the world you're talking about. In which videos and what exact timestamps do Asobo mention this "force element model"??

The modern flight model is separate from the legacy flight model, as has been stated a gazillion times. And take note here, the underlined terms are the terms that MSFS actually uses.

So help us decipher this sentence of yours please: "Based on the SDK, FSX remains at the core of the force element model.", whatever do you mean by this? That FSX remains at the core of the legacy flight model? If so then ya obvs. Are you using the term "force element model" to mean the legacy flight model? As also has been shown to you and then you youself realized above, the SDK makes no mention of "force element model".. so what do you mean by "Based on the SDK" lol? Like mrueedi astutely noticed, you seem to be throwing out invented terms and concepts without having any clue about the MSFS SDK (and you've also avoided answering the earlier questions about if you actually use or have used MSFS)
 

Edited November 10, 20223 yr by lwt1971

I'm no engineer, nor do I regularly stay at a holiday Inn, but from reading the sdk doc, my interpretation is that there are two models being discussed, and how the new modern model can still be used by aircraft modeled in fsx. The reference to the legacy fsx version seems to indicate that the complex calculation results of the modern flight model are simply aggregated and "averaged" to provide simpler input numbers when used by models that were designed for them. 

This was probably a method that would allow simpler porting of previous models which was more important at the release of the sim to get some 3rd party products introduced quickly. 

28 minutes ago, OneOfMany said:

I'm no engineer, nor do I regularly stay at a holiday Inn, but from reading the sdk doc, my interpretation is that there are two models being discussed, and how the new modern model can still be used by aircraft modeled in fsx. The reference to the legacy fsx version seems to indicate that the complex calculation results of the modern flight model are simply aggregated and "averaged" to provide simpler input numbers when used by models that were designed for them. 

This was probably a method that would allow simpler porting of previous models which was more important at the release of the sim to get some 3rd party products introduced quickly. 

Exactly, and the key here is that there are two *separate* core flight models (modern and legacy). And absolutely yes the reasoning to include the legacy model was to ease porting of FSX aircraft.

Matt Nischan of Working Title and now MSFS dev team wrote up a great explanation of the core modern flight model here (this is without CFD): https://www.avsim.com/forums/topic/601526-msfs-has-the-most-advanced-flight-model/page/12/?tab=comments#comment-4549236

MSFS also starts with a base geometrically defined lifting surface, but then goes a completely different direction and discretizes the lifting surface into a large number (comparatively) of grid samples. Each individual grid sample receives its own airflow simulation that gets input from the airflow model in true 3d space: i.e. the atmospheric model is also 3d and thus the air itself is not a just a single scalar contribution but instead a varying 3d contribution across each grid sample where the atmospheric model and grid intersect. This means that each grid sample on any lifting surface contributes its forces individually and is also affected by a 3d atmospheric model individually.
Whether or not one believes the current aircraft flight model configurations use this well or whether enough parameters are exposed, the base grid sampling of the MSFS flight model is of a much higher resolution and the atmospheric contribution in 3d is a consumer sim first (to my knowledge, anyway). It also has the benefit of generating different lift values across the surface from front to back, which can be critical value differences at the flight envelope edges.

 

Edited November 10, 20223 yr by lwt1971

30 minutes ago, lwt1971 said:

concepts without having any clue about the MSFS SDK

You mean aside from quoting it? mmmmok. This seems to bother you folks a great deal. Why so emotional when the SDK gets literally quoted? I didn't even use any $$!

That's exactly why I've asked mrueedi to explain. "force element" model is used regularly. I'll keep my peepers open and let you know when I see it again.

In the videos, Seb states that forces are on each of the elements, but I don't see that in the SDK. I see that force coefficients are on  all the elements. Those are very different things.

I'm trying to understand why CFD breaks the "force element model" (what do YOU call it?). Why does it take so long in rework after enabling that part of the model? That shouldn't happen. Not if you're starting with the equations for mass (continuity), momentum (Navier Stokes), and energy. CFD embodies all those laws of physics. It shouldn't break anything, as simbol described.

 

27 minutes ago, lwt1971 said:

the legacy model was to ease porting of FSX aircraft

Ok, so question: Do devs have to make ANY changes to the aero portion of the FSX models for them to run just fine when Legacy is switched on? Does that mean Legacy = FSX?

39 minutes ago, blingthinger said:

You mean aside from quoting it? mmmmok. This seems to bother you folks a great deal. Why so emotional when the SDK gets literally quoted?
 

Lol projecting much? You can quote the SDK all you want, *that* is not what we're talking about here though are we... we're talking about your non-sensical terms/claims like "force element model" and "Based on the SDK, FSX remains at the core of the force element model.". Whatever videos you're referencing, link directly to timestamps and perhaps we can have a more useful conversation.

What simbol said is that after defining the core modern flight model for a bird and tuning/tweaking it, then enabling CFD on it "de-calibrates" it which requires more tuning again to arrive at the final FM (i.e. multi-stage development). He can explain further if he wants, but to me by "de-calibrate" and "break" he meant that the tuning is effectively "reset" or undone after the CFD pass. Perhaps that's what MS/Asobo designed and intended as part of the CFD development process for the devs. The bottom line is that the final FM is what we get to use/fly so that's all immaterial in the actual flying environment.

As to the work (if any) required to port FSX aircraft when using the legacy flight model, that question is best asked of devs who did that (which I'd guess are far and few between).
 

Edited November 10, 20223 yr by lwt1971

3 minutes ago, lwt1971 said:

He can explain further if he wants

Gotcha. I was pretty sure you don't have a clue on these topics, but thought I'd ask anyways. Fingers crossed that mrueedi will have something to contribute here! He/she seems to know so much. Or at least so much about how I am wrong.

 

4 minutes ago, lwt1971 said:

that question is best asked of devs who did that

See above for the same answer.

12 minutes ago, blingthinger said:

Gotcha. I was pretty sure you don't have a clue on these topics, but thought I'd ask anyways. Fingers crossed that mrueedi will have something to contribute here! He/she seems to know so much. Or at least so much about how I am wrong.

Ya it's pretty clear to all of us here who really has no clue.. but please, keep embarrassing yourself 🙂 .. Just quoting what I said about having simbol explain what he meant by "de-calibrate", i.e. "He can explain further if he wants", to then try and project that as me being clueless is well... lol, ya, very weak, but hey we wouldn't expect anything different from you.
 

Edited November 10, 20223 yr by lwt1971

This is why the cult term gets brought up sometimes 

@simbol,

have you been able to find out why the heck doesn't  a "CFD-enabled Prop" start rotating in a free-running turbine, with it's engine "dead" and blades in feather,  when the aircraft is put into a sideslip ?

Yet, as demonstrated in one of ASOBO's videos, it is sensible to the relative wind enough to "stay still" as an aircraft goes through an aileron roll ?  Strange ?

I have run several tests with the Caravan and B200, and was unable to get the proper behaviour with a feathered prop, and it puzzles me since it's a basic aerodynamic consequence of a blade being an airfoil and a sideslip attitude "offering" it the necessary AoA.

I am willing to test it again in SU11...

I've been able to fly a friends Sting and actually thermal in it for a good while 🙂

The guy, who had never flown a glider, was astonished at "how efficient the climb was" as we entered the thermal 🙂

Of course it's ot a Sinus, but it does a nice work !

Now you turned my interest into your S4 😉

 

Edited November 10, 20223 yr by jcomm

5 hours ago, blingthinger said:

 

Fascinating. Thank you for the designer's perspective! 3rd party devs are the patient magicians of the community.

Can you elaborate on "de-calibrated"? What aspects of the baseline flight model generally need to be adjusted afterwards?

 

What is it about the force element model that necessitates the CFD? It sounds like what you're saying is that the force element model is missing aerodynamic parameters that would otherwise add to the realism of the flight model, correct?

Any parameters you might have achieved initially as per POH such as max speed, lift, etc.you will have to re-tweak. I would be typing all night honestly if you want a full explanation.

What I am saying is that CFD add on top of the FM, it complements it and makes things better. If you wish to have a full discussion on MSFS aerodynamics, parameters, flight physics, etc.you need to talk to Asobo .

S.