Which kind of brings me back to my point of control effectiveness with speed. Not an expert so I have no idea if its implemented or tuneable , but it feels as though the rudder has the same effectiveness on the ground as it does in flight which gives it a light jerky feel, rather than it scaling depending on airspeed, which might be the dampening we are talking about, or “lack of inertia” that is so often a complaint. If I intentionally slam the rudder for 1 second, planes will immediately shoot in that direction rather than a initial slow gradual turn , as though its operating with the effectiveness going at 200knts, which makes the planes feel light. Same with the elevator being very effective at low speeds which produces that light feel.
Designing a hardware or software based dampening filter to remove jitters and spikes would not be rocket science but it is probably a bit beyond a home project and likely simpler to upgrade the controller to using a digital sender (which have no jitter issues) instead of dodgy pots.
What I am about to write applies, in my head at least, until you wrote that last sentence.
The obvious answer is that in the real world when you fly different planes you fly with different controls. Each planes controls were designed, and calibrated, for that aircraft. In the sim when you switch planes you are using the exact same controls, with the same sensitivities. It has to work for all.
But then that should apply to X-Plane as well.
I believe that is the reason why you will get one person say a given plane is twitchy, and another say it’s fine. I think a lot of that also boils down to the quality of your controls, and how you have them set up.
For example, a discussion in another thread about the Cessna 414, I contend it’s fine on pitch, other say it feels twitchy. Last night I spent about an hour doing stalls in the 172 over London. By comparison, the 172 is much twitchier on pitch, but because I’m using a Yoko+ I can deal with that as I have a lot of resolution, and travel on pitch to play with. So far, of the planes I have bought, I haven’t found a single one that, once trimmed out, I can’t make controlled inputs on pitch to enter gentle climbs, or dives with a single finger. That’s nothing to do with my individual skill as as pilot, but to do with the specifications of my yoke.
Someone using a joystick instead of a yoke would disagree.
Yes, I understand what you mean. At the weekend I tried DCS again after a break of about a year. Following a 45 GB update I casually did a free run in the FA 18 and was genuinely shocked at just how right it felt.
I think that you have made a lot of good points here. Let me say that I am not a RL pilot so I could be talking a load of nonsense but most of the aircraft I sim in do not ‘feel’ just right like the F18 did in DCS. Maybe it is because they don’t feel ‘heavy’. A lot of my aircraft certainly seem very ‘pitchy’ too. A factor could also be the turbulence model which I have a feeling could be improved a lot too. I seem to get the same amount of turbulence, for example, in both hot and cold conditions.
My feeling is that the flight issues are down to a mix of a few things.
(1) the flight model is either not quite there yet or developers are not yet experienced enough in taking full advantage of what’s already there.
(2) some of us (including me) are not setting up our controls to the best effect. I set up sensitivity etc by trial and error and rarely get it to my satisfaction. Maybe I should spend a bit more time and effort on set up?
(3) hobanagerik, I think, in a post above wondered perhaps if the ‘quality’ or sensitivity of the controls we use was also a factor. This certainly makes sense. I use, for instance, the T16000m joystick as one of my many controls and I accept that this will probably not be as realistic an experience as, say, a £450 upwards yoke. And yet I still wonder why my cheap joystick still does a good job in DCS.
So, I don’t really know the answer but wonder if it could be a mix of a few things.
I certainly don’t think that the issues are insurmountable and I really do think that we are not that far off. Maybe more tweaking by Asobo and the aircraft devs. and possibly using better controls will get us there in the end.
That’s one of the things I can’t square away easily, but there might be an explanation.
Which is worse? Too twitchy or too slow/sluggish?
Compare two yokes. One with 5" of travel, and a resolution of 256 (8bit). The other with 8" of travel, and a resolution of 4096 (12bits).
Now consider the developer was building their plane, and just happened to have one of those higher specced yokes, and used that to test their flight model.
A user of the higher end yoke is going to find this acceptable, but the user with the lower spec will find it more twitchy.
If the developer had used the lower specced yoke instead, a user of the same hardware may think it feels fine, but a user of the higher specced yoke may find that they have to make more aggressive inputs to get the same response.
Out of the two, I would guess the more twitchy feeling would perhaps feel worse than the sluggish feeling.
What would be really cool is if the flight model could be dynamic, and here I am assuming that the developer had used a real world pilot of that plane to confirm it behaves exactly as it should, and the flight model would adjust itself based on the detected hardware so it felt the same no matter which yoke you used.
But at the end of the day, the physics of it can’t be changed. 5" of travel, with 256 discrete settings is never going to feel as good as 8" of travel with 4096 discrete settings.
I hope my maths are right here.
If you divide 4096 by 8, you get 512. For every inch of travel you have 512 discrete positions.
If you divide 256 by 5 you get 51.2. For every inch of travel you have 51.2 discrete positions.
To move from the centre position on the lower end yoke to either extremity, you have 2.5" of travel.
To move from the centre position on the higher end yoke, you have 4 inches of travel.
The higher end yoke will always feel smoother because of this, and if those figures above are correct, about ten times smoother.
Yes, I understand what you are saying and it does make sense 
When it comes to the resolution, I think that’s kind of getting lost in the details. The most important thing is either matching the travel\range of motion or learning how to compensate for the difference between your hardware and the simulated aircraft’s controls.
I have yet to see anyone ever bring up the extremity deadzone setting, and that’s one of the most important settings for people that use a yoke that doesn’t match the rotation of the yoke of the aircraft you are simulating, such as a CH or Saitek yoke, which are probably the most common yokes since they’ve been around forever. Same applies to sticks. If you’re using a 90 degree yoke or a stick with light GA aircraft like the 172 that has a yoke with 180 degrees of rotation and you’re not using a 50% extremity deadzone (not -50%, but 50%), you’re going to have a more twitchy experience. By default, your inputs will be doubled and using a curve is obviously not the proper solution if you care about getting 1:1 movement. The problem is nobody is out there explaining this to people, so out of the box most people are getting an unecessarily twitchy experience.
Twitchiness is built into the MSFS flight model to simulate the bumpy air you get in summer flights. Just have a look at the aircraft in 3rd person view while sitting on the tarmac with no wind. They twitch. On the other hand, everything in Xplane feels like an airliner: slow and sluggish.
I am excited to see the new XP flight model that they reported on this month. If you believe what he says then great things are coming, but he said it was perfect in XP11 too. I put XP11 back on the hard drive today just to remind myself.
If you have been simming for awhile you probably remember the days when every aircraft had the same flight model but with different skins and sounds. MSFS2020 is not that sim. However, there are plenty of models with “very similar” flight characteristics, but there are some models coming out now that seem to have been built from the ground up. The thing to keep in mind is that the SDK has been evolving steadily, and the level of modelling skill is rising with it.
The nature of the business, getting something to market fast and then getting another one out to keep the money flowing, means there are plenty of half baked models in the sim. I flew one yesterday with the sideslip indicator upside down and non-functional. Try this test: on takeoff, throttle up but don’t touch the stick. See how many models just drive off the end of the runway. 
Still, the future is bright for MSFS2020.
It wasn’t the CAP10 was it?
This is what happened. I was holding full back pressure, and left rudder all the way to the point where the nose was practically on the horizon again.
Microsoft Flight Simulator - 172 Classic Spin behaviour - YouTube.
Yes and it starts to roll opposite the spin direction, very weird. But it doesn’t look as Mickey Mouse as I have seen it before. I will try to find the video of that.
MSFS stimulates 3D columns of rising and falling air based on thermal radiation as part of a virtual 3D atmosphere that results in bumpiness if you fly through them in a light aircraft at the right speed. It’s a lot more complicated than how you’re describing it.
Are you talking about prop wash? I look at my aircraft in the exterior view all the time while it’s sitting on the ramp and I never see it just randomly twitching for no reason…
The inverted indicator on the CAP10 isn’t a bug.
It’s easily demonstrated as well. Fly at midday, and observe the movement of the plane. Now adjust the time to night, and you will see it’s much smoother.
I cannot wait till we get hyrdroplaning on wet runways when you come in too fast 
The whining and gnashing of teeth will be a joy to behold.
Followed by a flurry of YT videos of pilots power sliding on to the ramp.
I wish MSFS would model alpha, beta and inertia better.
Currently when exceeding critical alpha the plane just stops flying. Result of a poorly modeled alpha throughout the range.
Lack of inertia modeling results planes feeling twitchy.
On this topic of inertia/etc that folks bring up, I’d be interested in your case what aircraft you are flying? Are they the Asobo-developed default aircraft? Some add-on birds like the Fenix A320, PMDG 737, Milviz C310 etc model inertia rather well so it’s not a core MSFS FDE issue per se, as outlined in more detail in my original post.
The bumpiness/turbulence discussed above also is more to do with their atmospheric airflow simulation and thermals than the aircraft flight models. They have discussed at length about this in the recent Q&As and the bottom line is it’s realistic in how the bumpiness/turbulence depends on time of day, terrain being flown over, season, etc. They are looking to make some tweaks in SU10 however, and then from the 40th anniversary edition onwards provide a more complete advanced atmospheric airflow simulation (also in order to better support gliders natively). These will include application of their computational fluid dynamics (CFD) tech to atmospheric airflow on a larger scale. Note from the last Q&A where Seb goes into more detail from this timestamp onwards: Twitch
"Currently it is local thermals, that don’t combine with the other air, no vortices, etc. There is research work going on that took the CFD of the airplanes and extended it to 20km around the aircraft, really very very very big CFD running that goes pretty high as well, to simulate airflows in the atmosphere, better updrafts and winds, you get vortices with air going back down. Improvements of airflow arround mountains and terrain"
In the immediate future for SU10, here’s what Seb ended up saying about how thermals work now, improvements coming >= SU10 around thermals and atmospheric airflow, thermals, live weather and wind gusts, etc: Twitch. TLDR for turbulence in SU10: if winds are zero or very low, the thermals will be much less than SU9 so therefore also less turbulence.
Most runways are grooved though, not so likely to hydroplane on those. Don’t think Asobo will create a database of grooved runways.
You can also fly in and out of sunny spots and see the effect it is has the updrafts. That’s the part that a lot of people aren’t seeming to get. MSFS isn’t just lighting those spots up on the ground, it is performing a thermal radiation simulation and using the colors in the satellite imagery to determine how much heat is generated. More heat is generated in that sunny spot than in that shaded spot, and more heat is generated by dark-colored surfaces in that sunny spot than light surfaces, resulting in the air moving up and down in columns depending on not just the intensity of the sunlight, but the scenery. X-Plane doesn’t have this feature, so obviously it will result in behavior that doesn’t match X-Plane. Which is why it’s silly to use X-Plane as the ultimate truth\reference. If X-Plane 11 reflected reality, X-Plane 12 wouldn’t be on the way.
Obviously this feature needs some adjustment and the ability to configure it, but not matching X-Plane is a good thing in this case.
The only flaw I found so far is that it doesn’t seem to respond to snow cover, otherwise it’s pretty good!