On 10/14/2024 at 11:38 PM, jcomm said:

and the most beautiful gals of the Universe

agreed, south america. 🤣

7 hours ago, Gelrd said:

 

Just chiming in. Essentially it is this right?  Alec is looking at the purple and you are saying the green box is what really matters?

He has no idea what he is talking about, it is an incomprehensive pile of nothing after he got angry that I showed him he was wrong, and even starts to get personal trying to make me feel like I am a fool for doing this. Well, Two aero engineers friends of mine took at look at this post, and they all confirmed his words has nothing to do with anything that has been talked about in here, and has nothing in there remotely close to a sound racionale. And I am the one who will look like a fool... 

The Half Cycle I am measuring is the time between the highest and lowest point of the curve in the same cycle. A Full cycle is would be from the highest point to the next highest point.

So I am measuring how fast the oscillations are in the short period mode, meaning, how quickly the nose reacts when you use elevator. In MSFS2024, it is clearly observed that this happens much faster than in MSFS2020. My timing estimation put almost twice as fast. This could be down to changes in several areas of the Core Sim, if both planes have the same identical FlightModel.cfg files shared, or if not, could be by tweaks in the stability attributes, control power, even CG changes could affect a bit. 

My main point here is that I dont see any improvement so far. I can't say it won't happen, or look inside at what is actually going on internally, but I was looking forward some significant improvement after all the enhancements teased for the MSFS2024 Aerodynamics, and to me it feels just the same, or slighly worse due to the above reasons. 

But you can't unfortunately show any data in here that makes mst people think it's not the greatest evolution since sliced bread, they will instantly go to their confort thinking that a pilot says it feels great, and that is it, must be great then. Luckly for all of us, Real Airplanes manufacturers dont design their planes based on feeling, this only happens when the project is almost ready to ship to the customers. Also, the great addons that everyone praises as realistic also have had very serious process of gathering data, calculations, estimation of values based on known concepts, but this also is completely ignored.

But still I spot some users who like me doesn't take the pill and believe it is all great and sunshine, and have interest to understand why it maybe something doesn't feel right, when everyone says it should be. And go looking for answers. I am happily doing this and will continue to do so for such users that want to base their opinions on data and not on feeling.

 

Edited October 16, 20241 yr by Alec

7 hours ago, Gelrd said:

Alec is looking at the purple and you are saying the green box is what really matters?

No, the settling time of a step response is defined as follows:

First, the measurement starts from control input. It ends, when the signal falls into a rather narrow error band, that describes a recognizeable error. Alec got this completely wrong and still does not accept it, which nullifies his conclusion.

I did take the time, to properly analyse the video and extract the settling times from it. You judge.

MSFS 2020

1.      Pull, 11:29- 16:11 -> 4:42 -> 4700 ms

2.      Push 22:49 -27:39 -> 4:50  -> 4833 ms

3.      Pull 33:22-38:53 -> 5:31 -> 5517 ms

 

MSFS 2024

1.      Push 1:10:02-1:12:52 -> 2:50 -> 2833 ms

2.      Pull 1:16:45-1:18:36 -> 1:51 -> 1850 ms

3.      Push 1:25:08-1:26:58 -> 1:50 -> 1833 ms

I don't need to print the result in large letters that everbody can see that MSFS 2024 is much improved and that what Alec complained about actually is a much improved behavior that is much closer to the real 172.

10 minutes ago, Alec said:

He has no idea what he is talking about, it is an incomprehensive pile of nothing after he got angry that I showed him he was wrong, and even starts to get personal trying to make me feel like I am a fool for doing this. Well, Two aero engineers friends of mine took at look at this post, and they all confirmed his words has nothing to do with anything that has been talked about in here, and has nothing in there remotely close to a sound racionale.

You don't need an aero engineer, to understand the evaluation of a step response. Enter "step response" in google and you will see that the scientifc field which we are talking about is not aerodynamics but math, control theory and system theory. Which is my field. I have absolutely a very good idea about what I am talking. You will not find a single link showing a step response evaluation how you did it.

You entirely missed, that the crucial element in the diagram of the real 172 is not the .57s period of the ringing, but the ~0.8s of the settling time. MSFS 2020 is very far off the correct settling time and MSFS 2024 is much improved in that regard. To be 100% correct, it would need an even shorter settling time.

3 hours ago, jcomm said:

As @Alec says, access to the SDK and being able to record & plot flight data is the way to go.

For the time-wise analysis of the step response the video is good enough though.

we, the GA pilot people, probably didn't test as rigorous as you did and I for one, simply flew the C172 in MSFS 2024 as I would with my own real airplane.  without extreme jerking the rudders and yoke. and that felt very plausible and will be enough for me to re-fly and enjoy my real world routes. I am not going to argue because I don't have any numbers to go by but find your observations interesting. if the behaviour is really wrong, like Alec said, I think Asobo would eventually correct the flight model or perhaps simply adjust the relevant parameters in the corresponding .CFG files. we know they or workingtitle are capable of this, as Matt Nishan has explained how they tweaked the CJ4 to within 2-4 percent of pilot handbook values.

Edited October 16, 20241 yr by turbomax

@jcomm I'm confused. As long as I've known you, you've been the king of feeeel of flight. Now you're liking posts of someone saying feel of flight doesn't matter? What gives?

50 minutes ago, fsiscool said:

You entirely missed, that the crucial element in the diagram of the real 172 is not the .57s period of the ringing, but the ~0.8s of the settling time. MSFS 2020 is very far off the correct settling time and MSFS 2024 is much improved in that regard. To be 100% correct, it would need an even shorter settling time.

"damped frequency is dependent on damping ratio as well as the undamped natural frequency." 

Of course MSFS2024 dampens faster than MSFS2020, the frequency is twice as fast! You are proving nothing here.

At no point did I mention anything about the Short Period Dampening, the frequency of the short period mode is so fast, I didnt even bother to go further and measure in MSFS2024 compared to MSFS2020. It is not important to measure the frequency you say, I am missing the point? Strangely, the U.S. NAVAL TEST PILOT SCHOOL - FLIGHT TEST MANUAL - FIXED WING STABILITY AND CONTROL - Theory and Flight Test Techniques doesnt seem to agree with that affirmation.

 

Hmmm that does remind me of some of the complains about the Flight Model in certain MSFS planes, doesn't it?

1 hour ago, Krakin said:

@jcomm I'm confused. As long as I've known you, you've been the king of feeeel of flight. Now you're liking posts of someone saying feel of flight doesn't matter? What gives?

It's Life ...

I used to say I would never get married... It's the twice I'm in...

I see a lot of numbers and formulas and scientific talk, which I know nothing about.

1 hour ago, turbomax said:

we, the GA pilot people, probably didn't test as rigorous as you did and I for one, simply flew the C172 in MSFS 2024 as I would with my own real airplane.  without extreme jerking the rudders and yoke. and that felt very plausible and will be enough for me to re-fly and enjoy my real world routes.

This is what matters most to me though. You don't yank the yoke as was done in the tests. Flying a small aircraft consists of very subtle control movements. I liked the feel of flight very much, which is more important to me than it being scientifically correct...

20 minutes ago, Rimshot said:

I liked the feel of flight very much, which is more important to me than it being scientifically correct...

lets say Alec tried a scientific approach to test behaviour in extreme situations, that we normally don't encounter. you could also test flying inverted, barrel roles and other aerobatic maneuvers and complain if it doesn't feel right, and you might actually be right. but then again MSFS is not exactly the test bed for Boeing, Airbus or Space  X engineers. considering everything else that MSFS has to offer I can live with that.

Edited October 16, 20241 yr by turbomax

30 minutes ago, Rimshot said:

This is what matters most to me though. You don't yank the yoke as was done in the tests. Flying a small aircraft consists of very subtle control movements. I liked the feel of flight very much, which is more important to me than it being scientifically correct...

You don't have to yank the controls like mad on MSFS... Weather effects an the way they impact the flight dynamics do it for you... 

Look at any GA, Airliner, ... , in MSFS 2020 being hit by even marginal turbulence and / or variable wind... For me that really reduces the amount of pleasure I could get from the sim, given it's surely many "pluses"...

Edited October 16, 20241 yr by jcomm

1 hour ago, jcomm said:

I used to say I would never get married... It's the twice I'm in...

So long as you signed the prenup and have a good lawyer, there shouldn't be any surprises.

I’ve got a 100 hours in the 172 and about 60 in PA28s. I did my ppl in UK Where PA28s are very common trainers and then hour building in sunny South Africa in 172s.

I think there’s two things I can add here. One, no 172 flies the same, each has its quirks and slightly different performance. I flew quite a few different airframes and they all flew just a little differently.

And two, and one that I think makes a huge amount of difference, is the controller you use. I currently use the honeycomb, it’s not bad, but it will not come close to the force feedback Brunner type of yoke. 
 

I didn’t try the tech Alpha as I wanted to wait for the more complete November release. But some of the early feedback on the 172 sounds good to me. I doubt it will be perfect, but I am certain that it will be moving in the right direction.

The sim certainly does look fantastic to me. It’s really Easy to forget how far we’ve come. I remember FS 3 and how cool I thought that was back in 1990!
 

Edited October 16, 20241 yr by antic81

39 minutes ago, Greazer said:

So long as you signed the prenup and have a good lawyer, there shouldn't be any surprises.

Oh, I'm not disappointed 🤣 just as I - after all - am not with MSFS, and won't mind embarking on 2024 the day it get's available 🙂

1 hour ago, antic81 said:

I think there’s two things I can add here. One, no 172 flies the same, each has its quirks and slightly different performance.

This. When I interviewed Seb and he described the new way they do telemetry with sensors on the aircraft to scientifically reproduce all movement, he mentioned clearly that with older aircraft there's going to be much more variance, and that's entirely normal. 

I’ll reserve judgment until I’ve actually tried the new sim , however to me as a “real’ pilot what matters the most is that the aircraft simply feels real.
It doesn’t matter to me so much that it might be mathematically unrealistic, just that it feels like a real aircraft, it’s all about the illusion.  Although it would be nice if things also tied up mathematically to a real aircraft but ultimately I’m not using the sim as an engineering platform.

It can be done, things like the just flight PA38 in XP11 just felt real to me.