Discussion in 'Flight dynamics' started by Brother23, 14 Feb 2007.
Are there special programs?
Hey there, I know that this is an extremely old post, and if you haven't found out already, there are several programs for .AIR files.
Three Main Programs:
Airwrench (free trial 20$ Pay)
Just wanted to start giving back to the community which has helped me in the past.
AFSD, a perfect analysis tool
AirEd / AAM
Jerry Beckwith's Airwrench is a good tool, but totally recalculates the whole air/cfg files for each change. Sometimes it may be more effective to use another editor to change just the parameter of interest. Of course if you go back to AW, then all those changes are undone.
AFSD by Herve Sors is indeed a great utility, invaluable for flight test and adjustment.
If you like free
FS edit comes with the 2004 game pack and it can be dropped into all the sims or you can use it to open files in other locations. Air Ed is great too. but it requires some gusswork as some of the variables are unnamed. FS edit is best here as it has ALMOST all the variables outlined and tabbed. But for the really serious tweaking i do some guessing in AIR ED becuase its more all encompassing and you can cut and paste from other AIRCRAFT (visa vie...if you are building a boxy airplane with loads of power you can meld a CFS2 F4F4 with a F4U/1-A Engine and prop.) RECAP: FS EDIT is for simple AIR ED is for complex.
There's a very good reason why FSX did not include an updated FS Edit program. It has the potential of throughly trashing an .air file at the most inconvenient times.
Anyone who's seriously dedicated to generating 100% compatible .air files can download the FREE Prepar3D SDK and use the asm2air.exe program that MS/ACES wrote to create .air files, but never released to the public...
I'm not a fan of FS Edit, but I think, it can be usefull sometimes. Therefore I tried some time ago to use FSEdit with FSX. But I was not able to run it under FSX or to open air files in other locations. And if I remember my experiments correctliy, FSEdit is not able to open true FSX air files at all, because it needs some sections that are now obsolete.
How do you made it?
Any idea where I can download the asm2air.exe program? I'm getting Prepar3d right now but am having difficulty locating that one. Or is it included? Lockheed Martin's site has a payware option of the SKD and the community version which sounds like is the only one compatible with FSX.
John, asm2air.exe is included in the freely downloadable SDK from Prepar3D, and may be found in this folder:
..\Prepar3D SDK 1.0.2042.0\SimObject Creation Kit\Flight Model SDK
Glancing at the .asm files I notice immediately that for the stability derivatives there are too neglected coefficients:
I have a few quick questions about these.
Firstly, which components did these represent? V stab (CLiv) and fuselage (CL fuselage)? That would make sense to me as these are often neglected.
Finally, are they neglected solely in the .air file? I'm wondering if I can't add the neglected values and have them taken into consideration for the simulation. I realize its being picky about the numbers well after the decimal in the end but I'd be interested in incorporating them if possible.
Oh dear, you're asking questions from a concrete block when it comes to actually "knowing" anything about flight dynamics!
I tend to look at an FDE as a blackbox that has a "goesinta" port and a "comesouta" port. What happens in-between those two ports is a dark mystery to me...
I know nothing about asm2air.exe, but those lift derivatives look mighty close to AirEd Record 1539 in the airfile.
6h=Apparently Lift Slope
The “UNUSED” parameters look to be Cl at alpha=0, and “CL_alpha”, which might be related to the lift slope.
As far as the "UNUSED" declaration goes, you should simply experiment and find out if that's true. Input some large values (one derivative at a time) and check in game for any effect.
The aerodynamic system and particularly the AIR file is the result of a long, slow, undocumented evolution. In the AIR file of previous FS version one can find sometimes oboslete thing and yet not active parameters for future use.
I guess the same is with the FSX and ESP sim engine.
The ESP SKD and the Prepar3D SDK don't covers exactly all sections of the AIR file. E.g. section 1547 (center of moment) is not mentioned at all but still working. Other tokens are still in the AIR file but without function. And some tokens are declared as unused by the SDK but are not zeroed in the original MS AIR files. Resuming: there is still no absolute certrainitiy about the aerodynamic system, as Bill stated.
The first unused line in Section Lift Derivatives was introduced with FS2002 and detected (at that time there was no AIR file SDK at all) as angle of attack at zero lift (AoA @ CL=0). It was a replacement for a similar (and once working!) parameter in section 1101. But many people noted no effects changin the new parameter. Nevertheless, FS9 switched to another way to calculate airfoil incidence, angle of attack, and lift. I think, the line is now really useless.
About the meaning of the second unused line, there was a lot of guessing as introduced but I'm not aware of a definitive and tested decoding of it.
Hi Sergio, it’s been a long time!
Yeah, this ain’t my first time at the rodeo pardner. But see below for my take on those “not active parameters”.
I submit that one’s certainty about the “flight” dynamics of the MSFS series of games is directly proportional to one’s familiarity with it. But I’ll concede that “absolute certainty” is not possible without full disclosure on the subject by Microsoft. Ron Freimuth always maintained that the MSFS developers didn’t know much about it themselves – and I always agreed with that observation. Still do….
As you know, that record (1539) is one that the FS8 FSEdit wrote to the airfile. I’ve never used those, as they always seemed redundant to me, and in one case - tbldb 1545 – very badly done indeed.
The Rec 1101 “Fuselage AoA at Min Induced Drag” still works - I just used it day before yesterday to fine tune induced drag on my project.
Yeah, the defense for that change was “We don’t care what the wing does by itself”.
Problem was, guys would “test” using values that worked in FS2000, then declare a parameter “Inoperative in FS2002”. It took me two years to convince Ron F. that the torque effect wasn’t dead in the airfile!
That’s why I recommended that Mr. Miller raise the values for those “UNUSED” parameters by a large amount (and I’m not talking decimals here), and test in game for effect.
Bottom line on the history of the flight model for MSFS and CFS games as I see it: they peaked with CFS2 as far as high fidelity is concerned, and it started going downhill from there.
Fast forward from CSF2 to the present day with the new “tool” (asm2air.exe), and “ESP”; well, I’ve no idea about “ESP”, I mean, outside of parapsychology, I’m not even sure what that is. I’m guessing it’s related to the infamous “internal memo” that Ron and Jerry were writing about a couple of years before FSX went RTM. Without knowing anything about it, or it’s effectiveness, I’ll offer the ignorant opinion that “asm2air.exe” seems to be the “ESP” analogue of FSEDIT. I’ll stick to AirEd and AAM - even though that may be Luddism in the face of something better for all I know....
Not really. The asm2air.exe program was always used by ACES to compile an original .air file.
It is not an "editor" at all...
Isn't that what MS sold to Lockheed Martin, who are now selling it as Prepar3D? And actually doing some development work on?
Yes. Actually, Microsoft licensed the complete source code to Lockheed-Martin, which they've now released under the tradename Prepar3D.
MS also licensed the compiled run-times of ESP to Flight1 (note: not the source code!) for them to develop and sell pre-packaged "solutions" to customers. I get the impression that Flight1 isn't really happy about the L-M deal though...
I didn’t say it was an editor Bill...but thank you for enlightening me about asm2air.exe’s true identity
In fact, reading over my previous post, I get the distinct impression that I'm completely in the dark about the entire subject of ESP and asm2air.exe. It’s probably apparent to even the most casual observer that I'm an utterly ignorant and backwards fool when it comes to the latest (but probably not the greatest) offerings from MS - and that I intend to stay that way.
As I said earlier, I'll stick to AirED and AAM. Even if they're mere "editors", they work flawlessly, and I know very well how to use them. I've had quite enough of "tools" (FSEdit MakeMdl et al.), operating systems (Vista comes to mind here, Windows 7 ain't much better except in direct comparison with an un-tweaked Vista) and other software produced by Microsoft.
It's absolutely not something I'm interested in.
AFAIK, asm2air.exe was originally from the FS95 days. As it's nothing but a compiler program which uses assembly language files (.asm) for input, what it spits out from the "comesouta" end of the process is what we see as an .air file.
As such, comparing it to FSEdit is apples-and-oranges.
The only reason anyone might possibly find useful would be from the SDK for the asm2air.exe program, and perhaps playing around with it to create totally custom .air files for their own amusement.
For day to day "tweaking" of existing .air files AirED and AAM is sufficent for my needs as well.
FSEdit was (and still is) somewhat buggy, as it was prone to trashing perfectly good .air files. That is the primary reason why ACES didn't include it in FSX's SDK.
Separate names with a comma.