Clear skys, 3.6kts of wind at ground level no gusts, and 3.6kts aloft no gusts.
Flying the standard C152, no mods. Straight and level at 2300rpm.
Over completely flat land, 1 mile in from the coast, with the almost nonexistant wind blowing onshore.
At any angle to the wind I regularly get knocked into a 10 degree bank, occasionally a 20 degree bank and regularly knocked up or down 5 degrees of pitch.
The plane will regularly start climbing or sinking between 700fpm and sometimes even 1000fpm.
Commonly the nose will get knocked in the opposite direction to the vertical speed change, e.g. the nose will get knocked down and the pane will start climbing or vice versa.
This all happens half a dozen times a minute.
Over open water just 1 mile away the plane is totally stable and I can literally fly hands off for 20 minutes.
This kind of lurching about in calm conditions over land is imo completely unrealistic. It’s pretty much impossible to even trim the plane effectively and maintain straight and level flight, let alone accurately hold a course or fly a smooth turn. I have done some IRL flying in far more windy conditions and imo the amount of lurching around experienced over land in MSFS is completely over the top.
Something to keep in mind is that when flying IRL you can feel what the plane is doing through your sense of balance and through the seat of your pants, and most of the effort of correcting for gusts and flying straight and level is done in response to that sensation as a learned response that becomes automatic. In the sim you don’t have that feedback so everything has to be done in response to what you see happening on screen, which requires far more concentration and effort and becomes very tiring.
As much as one would want the sim to be as accurate as possible, given the sensory limitations inherent in a computer simulation it might be pragmatic to have the option of a more toned down simulation of turbulence, because the current depiction makes hand flying over land for any length of time simply exhausting.
Some people report this kind of problem, but I’ve never seen it myself. I don’t know why it is that this is inconsistent from player to player.
I fly mostly GA planes (such as the Flight Design CTSL, Cessna 172, String S4 Ultralight, and others) in the Portland, Oregon and Los Angeles, California areas with live weather under VMC. This includes flying in relatively mountainous areas such as up to Big Bear (L35). There is some turbulence in the mountains as expected, but I don’t have the kinds of unexpected climbs or sinks you see over land in general.
I’m on SU12B Beta and have Turbulence set to Realistic. It matches my experience in the few real Student Pilot hours I have, and certainly the experience as a long time commuter pax on small transports.
It’s the thermals that are the problem. If I switch on the 3D thermal visual display I can see what’s happening.
I’m flying in the South East UK around Lydd EGMD. With ground wind set to zero there are only a few thermals which go straight up to about 2000ft after which they disappear, flying in these conditions is smooth. But if I set ground wind speed to 3.6kts then for some bizarre reason the amount of thermal activity goes through the roof. Basically the whole world becomes one big giant thermal which reaches up to about 5000ft, and flying through this is like flying in a gale, the plane is getting bounced all over the place.
In this part of the UK it’s controlled airspace everywhere above 2500ft so all VFR traffic is limited to below that altitude.
I think this is partly my fault actually. I’m using Floyds Epic Clouds and to get light cloud cover I’m using a weather preset with a temperature of 25c which is pretty hot for the UK, and it probably doesn’t have typical pressure and humidity values either. That’s contributing to the excessive thermal activity. I’ve quickly created a custom preset with some more typical values and I’m getting a far smoother flight experience now.
Where is the Turbulence setting though please? I might want to have a play with that too.
I’ve played around a bit more and I still can’t make any sense of it.
If I set standard pressure, temperature and humidity for a cool day in the UK, then changes in pressure and temperature have a discernable but not drastic effect on thermic turbulence.
However the change from 0 winds to a 3 knot wind has a drastic impact on thermic turbulence. At 0 winds there are light thermals which disappear at about 2000ft and the air is smooth, with a 3kt wind layer at ground level the entire world becomes covered with strong thermals which extend up to about 7000ft.
Why on earth would that be? That makes no sense to me. Surely the temperature and cloud cover would have more effect on thermals than a 3kt breeze.
I would suggest that the issue is less about the degree of turbulence ( which I find fairly realistic - hot gusty days in the central California valley are as hard work as you would expect), and more about the unrealistic trim behaviour: in MSFS aircraft do not return to their trimmed attitude as they should when encountering minor bumps and knocks.
Instead, without pilot interventions, the aircraft will lose attitude at an increasing rate. It is this that is really unrealistic as countervailing forces would IRL return the aircraft to a stable attitude. This behaviour seems to be the case for every aircraft and has been since the SIM launched and is now very noticeable when flying in light to moderate turbulence and it what makes it harder work than it ought to be.
That’s a really good point. I hadn’t noticed this before but now you point it out it’s so obvious and it really makes the planes so much more difficult to fly than IRL.
It seems like a pretty basic aspect of a flight model to have got so badly wrong. I thought MSFS was supposed to have a really sophisticated flight model, I remember seeing presentations where they showed off how many hundreds of points on the plane that the airflow was dynamically modelled. It seems to have not turned out so great, the stall and spin characteristics of most of the planes is pretty poor too.
I noticed some time ago that I thought the newly-introduced turbulence was a bit excessive compared to real life. Since then it seemed to have improved.
However I’ve just started flying in VR and the turbulence is again very noticeable, albeit understandable in the areas I’m flying in; just a bit too much.
Hopefully the introduction of adjustable turbulence in SU12 this week will improve matters - that’s right isn’t it, it’s in SU12? I don’t use betas.
I’ve never used VR so this may not help, but if you haven’t already try turning of camera shaking. I use head tracking and the combination of turbulence and camera shake made it even harder to steady the plane.
Slightly related, but a discovery I made over the weekend with TrackIR. I was getting some really bad wobble when I turned my head but kept it still. Solved with the Global Smoothing slider on the TrackIR software (just below speed, above the curved profile panel). Set it to all the way right and problem solved.
Introduces a SLIGHT delay to your movement (very slight) but it’s much much more stable and nicer.
Sadly as excellent as MSFS is in many ways, there are some real core aspects of the flight experience they have not yet got right - trim behaviour, ground handling, stalls and adverse yaw.
Unfortunately as we lack any agreed set of standards or testing regime, most of these issues go unnoticed by the average simmer, particularly as what you experience will depend on your controllers and profiles, what aircraft you fly, your skill levels and your expectations.
All meaning big weaknesses can be accepted as normal by many as there is just too much other stuff going on.
Overall the amount of turbulence fits with my real world flying experience. At my home airport you would not do circuits for example on a hot day in a light plane nor would you go up on a windy day unless the wind is off the ocean.
However, that said, there is something not quite right about the nature of the MSFS turbulence, for example the yawing is not realistic, in real life your whole plane shakes up and down, moves left right, occasionally drops a wing, none of this wobbling around the yaw axis we seem to get.
There seems to be something wrong with the way lateral forces act on the aircraft body in the heart of the flight model:. We see it with excessive and exponential weathervane effect on the ground, we see it with lack of adverse yaw in turns and we see it with the way turblance effects the yaw axis more than pitch and roll.
THIS is the perfect explanation of the problem we have in the sim and why any turbulence or thermal or whatever movement of the air we encounter feels overdone and wrong. The nose constantly comes up or down which makes any manual flight a constant stare at the VSI and why it’s so frustrating to fly with what is meant to be realistic behaviour of the air.
There are some interesting observations/concerns expressed here (Incoming soaring feedback on the SU12 weather update) regarding the thermals. Heat-only thermals tend to overtake everything else. People have been able to “soar” C152s over cities with the engine off.
Treed areas are flagged to disable lift, which destroys ridge lift in many areas. If you fly in an area where there are very broad and distinct lines of trees (Manapouri in South Island New Zealand comes to mind) with a little wind in the middle of the day, the effects are over the top and extend very far up.
I believe SU12 caps the heat only thermals at cloud base but the effects still occur far higher and to a much greater extent than they should.
The yaw thing is perhaps the most annoying but depends greatly on the particular aircraft you are using. The stock B58 had (or still has?) this problem. The tail just wags back and forth even without the presence of effects to pitch and roll. A 3rd party example would be the AH Globe Swift, but I discovered they had set the area of the vertical stabilizer and rudder to 15 square feet each! Reducing these to realistic figures helped the problem significantly. Even so, certain aircraft do it and others don’t, and it doesn’t seem to depend on overall size of the plane.
That’s pretty spot on as well although I have seen some pretty good attempts to do it right, I think it IS possible but probably not as easy as just using what the sim offers at the first glance. In flight the DC6 for example feels extremely good around the vertical axis and its behaviour has been confirmed in many aspects by DC6 rated pilots. Other aircraft like for example the FSW C414 behave extremely realistic as well. It’s a matter of knowledge, proficiency and will to create a good flight model in MSFS, it doesn’t get there automatically. Stalls is the same although many aircraft will simply not offer any hard data regarding how their stalls would turn out. Who would deliberately stall a DC-6 or 747? It’s different for a C172 or course.
Ground Handling is indeed a huge issue in MSFS and it has a lot to do with the wind and how aircraft react to moving air in the sim. If you turn all wind off the planes behave pretty nicely on the ground, almost like they would in reailty from the little comparison I’ve got.
Trim behaviour has always been a problem in the sim since I can remember. We just don’t have the correct hardware. Our sticks and yokes return to the center and the trim system has to cope with that. What MSFS presents however has even less to do with correct trim than FSX and P3D had to offer. Trim (and axes in general) behave very weird in MSFS, too sensitive, not linear and completely unpredictable. Something’s not right if the rudder trim is more effective than the rudder. A lot has to be done on this topic, also with respect to the flight characteristics.
