Military AI Works

As it does seem to be causing a lot more consternation than it should I tried a simple amendment to the cfg file.

In the [Reference Speeds] I altered one parm :-
flaps_up_stall_speed = 110

This has the effect of raising the nosewheel earlier to nullify what is being seen on take off.
However, it will also affect rate of climb, rate of descent, and the speed that the flaps extend on approach.

So for those that this issue bugs them try the amended figure. If you find the side effects are not what you would like then I would suggest altering the parm to 120 to see if that balance is more to your liking.

Well, it is not exactly inconspicuous.
I will test the flaps_up_stall_speed change this evening.

Well let me just say that that is not how it looks in FS9. As I said barely noticeable. So that says that the visual nature of it has to do with the P3D flight engine.
If I had noticed it before I would have made my own personal amendments to the fde but I checked and there were none.

I thought so...

From mobile hence short

@Martin
on the next test flight look at the movements of the elevators (in the fs9 version they were animated, but only from the horizontal to up position). You can see that at first are set like the plane is really trying to get the tail off the ground, just like a taildragger. Once the plane is levelled it goes back to almost horizontal position before it moves into the other direction for takeoff. So the plane pretty much does what the elevators are indicating.
Firebird might be right that it's the P3D flight engine behaving differently in that case than FS9 did.

Like I said before I'll give Henk's approach a try and see if there's any change. It's only got to wait until end of next week because I've no way of testing anything right now.

Jean Claude

Martin, I've tried the FS9 model in FSX too and there is no problem. The Mig sits nicely and taxi's and takes-off fine. No one is going to miss the attitude in your screen shots over the past 5 years, never mind Mike Pearson himself. I saw nothing like that.

I was hoping to try the FS9 model in P3D to at least see the model with no gear, just to see how it sits, but it doesn't show at all. Is it possible you could place a copy of the converted aircraft in my FTP folder please? I want to have a look at the datum (CG) position too.

John

Will do, John

From mobile hence short

Martin, could you PM me when you have copied over the files please?

Just out of interest, I managed to get the FS9 model into P3Dv4 (minus the gear of course). The fuselage pitch was just the same as in FS9 and the Mig parked, taxied and departed without a problem using the FS9 FDE.

I also did a second test. I converted the airframe (no animation) to .3ds and imported that into FSX Gmax so that I could export the model in FSX native code. That also worked fine in P3Dv4.

John

In your folder, John

Thanks Martin.

I have the Mig working in P3Dv4 and for the first time I can see the problem, but it's very small and only lasts for a couple of seconds before the aircraft rotates:



What I saw was nothing like the earlier posted screen shot of the problem, so I wondered if frame rate has anything to do with it. I re-tested at 10fps but it was still there. I also checked the datums in the model and they have not moved from the FS9 version.

I think the problem occurs because the main gear compression is so mis-specified, it is not responsive. As the lift takes the weight of the aircraft off the wheels, the Oleos should come down a tad to keep them on the runway a tad longer. I did try substituting the values I calculated, but as I found before, the main wheels protrude upwards through the gear doors. That looks worse than the air gap. There is so little room between the top of the wheel and the gear door, I can't do much.

The far bigger problem with the FDE (in FS9 too) is the take-off roll, which is 40 seconds from brakes off to rotation. I've just watched a clean Mig 29 taking off in a video at RIAT, probably with reduced fuel, but it was airborne in 15 seconds.

Steve's suggestion to change the full_flaps_stall_speed from 120kts to 110 kts, cuts 5 seconds off the take-off time, but it doesn't remove the small air gap before rotation.

Just as a final test, I increased the thrust_scalar from 1.0 to 1.5 to see what it might take to get a realistic take-off. That reduced the take-off time to 25 seconds, so still a fair way to go.

I'm not sure if anyone wants to take the FDE apart, but as Steve and I know, you can spend weeks going round in circles doing that.

The question to ask I think, is how many times a week will users be watching the Mig 29 rotating from a runway in P3D at a position to even notice the air gap?

It doesn't bother me.

John

Just another thought that I've just put to the test. What would happen if I substituted the F-15 FDE (developed my Mike McIntyre and myself) for the Mig FDE? That would retain all the Mig aircraft identities and the first part of the contact points that specify longitudinal and lateral positions of the wheels, but use the rest of the F-15 aircraft,cfg, including the power, lift, gear compression values and the entire F-15 .air file.

With a bit of fine tuning, I got the aircraft to sit right and to take-off without the air gap under the main wheels and in a brakes off to rotation time of 25 seconds. There was just a tiny part of the main wheels projecting through the doors. Of course that was only a test of the take-off, not the entire flight envelope, but it confirmed just what little room there is to jiggle things because the Mig gear modelling and animation is so tight. If it were my model, I would be turning my attention to that.

John

Good news, everyone !
I tested Henk's rotation and the plane takes off very nicely. At the critical point during rotate there isn't much space between the tail fin and ground (I'll have to tell my pilots to be very careful). However it created two unpleasant side effects. At level flight it looks like the plane is still climbing and on approach the angle of attack gets so big that it looks like it's stalling every moment. So I decided to change the rotation into an animation, so I could undo it during flight. After some more test I think I found an acceptable solution.
Thanks again everybody for your input. Heading back to my workbench to update my models.

Jean Claude

What I love is the creativity and the ability to find solutions that let these aircraft shine. You all do very well with Mike‘s legacy.

Thank you, guys!


From mobile hence short

The FDE needs more work. The aircraft is struggling for speed. In my tests, it was taking the same time as a Boeing 747 to take-off.

John

If we are talking about FS9, then I can help, but if we are talking about P3D then I can't test it so I am unable to help.

So now that the 27s have been out for a while just wondering if there was any word on when the 23s might be coming...?

Conversion of the MiG-23 is finished. The package is currently in quality check. As Martin is currently taking a well deserved break it's difficult to say when it will be released.

Jean Claude

Martin is not starting new packages, but will release packages coming out of quality check

Expect QC to be finished this weekend, so if nothing special comes up it shouldn't be too long before release.

Awesome to hear it!