They will if you can speak French 
At least the last time I looked, French/English was a requirement.
Hotfix would have been better option.
I don’t claim any specific expertise but, very roughly, you would programatically apply control inputs and monitor their effect on the flight model. Repeat in different weather scenarios and with different planes.
You’re effectively taking the plane for a test flight but you’re doing it programatically so you have complete control over every variable. Every time the tests are run, they would produce exactly the same results unless you have changed something, consciously or inadvertently.
The number of variables directly under the control of the pilot is not that great, mainly the control surfaces and the throttle. I don’t understand why it would be difficult to build a test suite simulating pilot input. As I say, I’m no expert, but I’m genuinely surprised they don’t have a test suite that would be easy and quick to run and which would immediately flag errors like this one.
Completely unacceptable not to hotfix this.
i doubt that:
- the flight model values can be checked easily, maybe some can be, but not all
- the aerodynamics can be fixed like that, there must be randomness somewhere.
that would also produce a lot of extra work for an issue that happens rarely
It is hard to say. Prepping a release might take time and will only land end of next week.
Original schedule for Sim Update 3 is for the 4th of March. A week later.
Fast releases are great if you can do it. We have a system we develop for a client to be able to push out hotfixes within an hour (and 35 min of that is build/test time). With decent monitoring I can have almost any detected issue fixed in an hour or 2 without disruption. And 99% of the time, the client does not even notice they had an issue.
There is nothing worse than getting emails from complaining clients especially before 8AM. So I make sure I dont
Should BOTH lift_coef_flaps and lift_scalar be divided if both exist?
You could approach it the same way as with level-D simulators by running QTGs? You basically have a response curve based on real aircraft behaviour and then have to show equivalent response from the sim.
Yes, sure. I mean, they must already have complex test suites; they couldn’t have built the sim without them. It’s a mystery why they don’t seem to be making use of them for release testing.
Impossible to find the files for B787
Have you the path and name of file ?
787 is locked I believe
Many thanks for getting back to us with this.
It’s unfortunate but it is what it is. In all of this, I think that we are losing sight of the 99% of people who don’t read these forums and the many people who use the sim casually. It’s all very well for us here on this thread who are into the sim enough to come here and keep tabs on what’s going on. A casual user getting into the sim (for them, probably a game) for the first time right now might head straight for some of the featured landing challenges and, not being able to do them properly because of a bug they’re unaware of, would likely come to the conclusion that the sim is too hard for them and simply abandon it.
Imagine how much their frustration would be exacerbated if they checked for videos on YouTube and couldn’t understand why other people could pull these things off no problem yet they couldn’t?
It’s extremely poor that the sim/game is going to be left in this state for any length of time (let alone several weeks) even though the intentions behind the bug were and ultimately will be good. The one thing I take exception to is the implication that the problem was missed because it wasn’t easy to see.
I’m no software dev, I’m not an aviation expert…but I flew for all of five minutes after updating and noticed it straight away.
You seem to forget the legacy code coming back all the way from the early 2000’s. A large part of the sim engine still runs that.
That said, the modern flight model is brand new and should have proper tests in the day and age.
Same goes for other parts like scenery, where they could have a ‘planet crawler’ looking for anomalies in the terrain vectors. I am pretty sure Bing does not care about spikes, but Asobo should, and with such a feedback loop, it will improve both MSFS and Bing.
No, changing just the former (lift_coef_flaps) would be the easiest.
The “lift_coef_flaps” change was made to address this issue. (And yes, we found it and fixed it before Asobo announced the error.) The change to 'lift_coef_ground_effect_mach_table" was unrelated to the update and was made to improve the effect of ground effect on our specific airplane with our specific flight model.
Think all of the flashy advertising videos we’ve seen with the aerodynamic forces on every square inch of the wings being sampled a gazillion times per second. No doubt a good deal of that is advertising fodder, but I’m sure they must have a serious testing setup. They couldn’t have built the sim without one.
Ahh, I see. Ok. Thank you for the explanation for both. I only change the lift_coef_flaps.
@AwarePlot117729
TBM and C208 community mods already published hotfixes.
I don’t have to learn to fix something I paid someone else to do, with that logic of yours, you should be able to fix everything in your house/car/life yourself, starting with that attitude and the fact you can’t defend your own right to receive a proper service you paid for.