Jump to content
Oscar_Juliet

Do the aircraft fly properly?

Recommended Posts

I hope that got your attention. I bought IL2-BOS years ago back when it was brand new. I learned to fly flight simulation on IL2 Cliffs of Dover with the TF mods. I moved onto DCS world, P3D, ect. WWII and German aircraft have a special place in my heart. All my WWII experience is on IL2 CLOD in BF109-E series. 

I've spent numerous times trying to get into IL2-BOS just to throw it back into the pile to gather dust. My problem is that I can't seem to get the aircraft to fly decent. I'm hoping someone can help me narrow down why. Maybe it is my expectations from other sims. Or maybe its something I'm doing wrong as a pilot. 

 

I've just spent hours on google, search the forums, and I can't find what I'm looking for. I watch Sheriff's Sim Shack and The Air Combat Tutorial Library a lot. 

Here is my problem:
When I fly the planes it feels like the plane is balanced on a stretched out rubber band. It wobbles side to side and seems to constantly require heavy joystick input to counter act the balance of the plane. In IL2 CLOD I don't seem to have that problem. The planes seem rather stable. When I watch Sheriff's Sim Shack he looks as stable and smooth while flying as any. Just like my experience with CLOD.

I don't know why it feels like the plane is constantly out of balance. There doesn't appear to be a way to trim it out. I also have trim wheels on my stick that can artificially trim an aircraft which doesn't usually have it. Even that doesn't seem to help. I feel like I'm always fighting the aircraft to fly straight and level rather than flying. Everything feels extremely touchy. I'm using CH products with a full hotas and pedals. 

 

Maybe someone will have insight. If a WWII plane is suppose to be this pain in the rear to fly, constantly fighting the stick, then I guess I have to get use to it. Otherwise let me know what I may be doing wrong. 

  • Upvote 2

Share this post


Link to post
Share on other sites

You need to go into the controls and adjust the sensitivity settings. Click on the little axis portrait next to the assigned axis. Although it sounds backwards higher sensitivity setting is more relaxed controls, lower is more linear and twitchy.

Share this post


Link to post
Share on other sites
Posted (edited)

I agree with Redcorn probably a sensitivity issue.

 

Although I don't think this is the problem I've only ever experienced the problems you're stating when I was flying quick missions and one of the updates messed with the turbulence and wind settings. 

 

Probably not the case but worth a try to look lol

 

Hope you figure it out

Edited by TheOldCrow
Spelling error

Share this post


Link to post
Share on other sites
Just now, TheOldCrow said:

I agree with Redcorn probably a sensitivity issue.

 

Although I don't think this is the problem I've only ever experienced the problems you're stating when I was flying quick missions and one of the updates messed with the turbulence and wind settings. 

 

Probably not the case but worth a try to look lol

 

Hope you figure it out

Well I went on Wings of Liberty server and it seems to fly decent. I shot down an A20 and got a bonus kill when a russian plane wrecked into me. It wasn't nearly as bad as the quick mission. I had wind and everything set to off on quick missions and it felt like I was all over the place. If there is a bug that would explain it. 

  • Confused 1

Share this post


Link to post
Share on other sites
1 hour ago, JgonRedcorn said:

You need to go into the controls and adjust the sensitivity settings. Click on the little axis portrait next to the assigned axis. Although it sounds backwards higher sensitivity setting is more relaxed controls, lower is more linear and twitchy.


Yes start here. 
 

The game has come a long way since the original yawfest it was years ago. It seems like the devs have really worked hard to add more “mass” to newer as well (feeling based, not performance based). 
 

Some planes are still a bit too lively at the extremes, but a lot of players state that they enjoy the “feeling of flight” this provides. I’ve decided to just chalk this up to the “artists interpretation” on the part of the devs and to just enjoy the game for what it is. 
 

Regardless of my opinions, hats off to the devs. Their passion is evident. 

  • Like 2

Share this post


Link to post
Share on other sites
1 minute ago, Oscar_Juliet said:

Well I went on Wings of Liberty server and it seems to fly decent. I shot down an A20 and got a bonus kill when a russian plane wrecked into me. It wasn't nearly as bad as the quick mission. I had wind and everything set to off on quick missions and it felt like I was all over the place. If there is a bug that would explain it. 

 

With the difference on the multiplayer servers I would assume it's a bug too. Have you tried starting a career or a campaign to see if it affects it their too?

  • Upvote 1

Share this post


Link to post
Share on other sites

its a mix between hardware, setup and practise I would say.

I personally enjoy the flight models, while I am not confident to judge them from a professional standpoint since I am not a pilot. 
But they feel "good", in the sense that I can understand what the plane is doing. Every plane feels different. Even very similar planes Fw 190 A-3 to Fw 190 A-5. You feel that the center of gravity is different and that you have to trim accordingly. 

  • Upvote 1

Share this post


Link to post
Share on other sites
Posted (edited)
5 hours ago, Oscar_Juliet said:

I hope that got your attention. I bought IL2-BOS years ago back when it was brand new. I learned to fly flight simulation on IL2 Cliffs of Dover with the TF mods. I moved onto DCS world, P3D, ect. WWII and German aircraft have a special place in my heart. All my WWII experience is on IL2 CLOD in BF109-E series. 

I've spent numerous times trying to get into IL2-BOS just to throw it back into the pile to gather dust. My problem is that I can't seem to get the aircraft to fly decent. I'm hoping someone can help me narrow down why. Maybe it is my expectations from other sims. Or maybe its something I'm doing wrong as a pilot. 

 

I've just spent hours on google, search the forums, and I can't find what I'm looking for. I watch Sheriff's Sim Shack and The Air Combat Tutorial Library a lot. 

Here is my problem:
When I fly the planes it feels like the plane is balanced on a stretched out rubber band. It wobbles side to side and seems to constantly require heavy joystick input to counter act the balance of the plane. In IL2 CLOD I don't seem to have that problem. The planes seem rather stable. When I watch Sheriff's Sim Shack he looks as stable and smooth while flying as any. Just like my experience with CLOD.

I don't know why it feels like the plane is constantly out of balance. There doesn't appear to be a way to trim it out. I also have trim wheels on my stick that can artificially trim an aircraft which doesn't usually have it. Even that doesn't seem to help. I feel like I'm always fighting the aircraft to fly straight and level rather than flying. Everything feels extremely touchy. I'm using CH products with a full hotas and pedals. 

 

Maybe someone will have insight. If a WWII plane is suppose to be this pain in the rear to fly, constantly fighting the stick, then I guess I have to get use to it. Otherwise let me know what I may be doing wrong. 

The wobble you are describing in an error in the flight model this game uses. It was much worse before the 2.0 update, where some things were done to try to address it. This is not how planes fly in general. Source: consultation with several real world pilots, including people who have flown ww2 planes. People on this forum who have flown the planes. Watching videos of planes in general flying where the stick movement can be compared to aircraft movement. And the fact that literally no other simulator has felt this way, including DCS which has some of the most sophisticated flight modeling available.

 

The general flight model of this game is certainly not beyond the scope of doubt. For a long time the rudders on most planes did not work and only caused planes to roll. Compression was not implemented for a long time. The behavior of flaps is inconsistent. The wobble was way way worse before 2.0. The developers have constantly improved the model over time and I think they will continue to do this.

 

Despite what people are telling you, stick curve settings are not a solution to this problem. They only help control it under certain circumstances. The issue is that the pitch axis of all planes (although some are worse) is extremely unstable and reacts to stick input rather oddly. Specifically, if you are inducing and upward pitch on a plane in il2 and then you lessen or neutral the stick, the plane will jerk in the direction of decreased pitch input and then rebound back towards the original input before settling.

Edited by Fumes
  • Like 1
  • Thanks 2
  • Upvote 3

Share this post


Link to post
Share on other sites
Posted (edited)

Pretty much what Fumes wrote.

 

As a rw pilot for more than 38 yrs, I also find that, just like in other desktop flightsims I have used, IL-2 suffers from an over done response to perturbations in pitch ( maybe also in yaw, and coupled yaw-roll ) from the trimmed AoA.

 

This affects some aircraft more than other. Probably the ones which came closer to what their rw counterparts feel like are the Spitfires, where they model the neutral pitch stability quite convincingly.

 

In this particular aspect, sometimes I found the MSFS aircraft more realistic in terms of short period oscillations... but then, as we start using the aircraft in that "old" sim many other shortcomings come to rise...

 

DCS also does, IMO, a better job in this particular area, and I don't recall CloD...

 

But, truth is, all summed up, IL-2 is still my preferred flight simulator. It's mi go-to sim in front of a PC when not working 🙂

 

So, yes, there are limitations / quirks, and yes it is getting better from update to update and the series is by far the most rewarding and productive I have ever used in game flight simulation. The development team, as small as I think it is, has been delivering so so so many new features, corrections, enhancements that every penny spent in this sim is more than worth the investment !

 

There aren't perfect flightsims 🙂

Edited by jcomm-il2
  • Upvote 2

Share this post


Link to post
Share on other sites
Posted (edited)

It's indeed a known quirk of the sim.  It used to be much worse, no doubt.

 

As a fellow developer, it strikes me as the effect of excessive moment-of-inertia, coupled with insufficient rotational damping.  Yet this may be due to the subtle little things that even the finest mathematical approximations of flight don't take into full account. That is, there may be nothing wrong with the flight model itself in an academic sense, except that theory and reality don't always meet exactly half-way.

 

 

Slowly but surely the devs are getting to the bottom of it, and we're all confident they won't let go of that bone until it's been chewed all the way through. This team is just as obsessed about aviation as the lot of us. *tips hat*

 

 

Edited by 19//Moach
  • Haha 1

Share this post


Link to post
Share on other sites

I find it different for each plane. When flying the 111 first time when it came it felt like a helium balloon.

I ditched the game for a month and tried the PE 2. quirky yes but totally loved it.

I flown HE 111 very little after that, but The little I have it feel better now. So yes things improve. 

I personally do not like the 109 after flying red so much it just feel too light and nimble. Then I tried the G6 a week ago. That was kind of nice. 

I know many love the plane. But point is I think one have to get used to box airplanes. 

And you might try red planes. In my opinion many of them are more “heavy” stable 

Share this post


Link to post
Share on other sites
Posted (edited)

It's not just a quirk of this sim but as stated above it varies from one game/simulator to another.  Some other flight sims go too far in the opposite direction and the aircraft feel like they're flying on rails.  

 

Almost all of my real-world GA flight time has been in Southern California in the summer when the hot, bumpy air is bouncing you around, so I kind of prefer the unstable feel you get in RoF and in this version of IL2.   I was fortunate enough to get some stick time in a T-6 a few years back and was really struck by the light touch needed on the stick...in normal maneuvers you don't really move the stick so much, it's more like a feeling that you just apply a little bit of finger pressure against it and the aircraft responds (far better than you'd expect for such a big lumbering aircraft like that).  Until we get pressure-sensitive joysticks to emulate that,  it's never gonna be right.

 

All that being said, I also have a full CH Products setup and found some things that help.  For starters:

  • Of course make sure that you've calibrated everything using the windows control panel tool - with the trim wheels on the stick centered at their detente.
  • I'd really recommend that you don't use those trim wheels on the joystick axis for trimming the aircraft in the game.  They're just too course to give the fine adjustment you're looking for to trim for straight and level flight, and are really more for adjusting just the stick for poor calibration, which is ironic as CH stuff usually holds calibration quite well (like, for years).  
  • Instead, map some convenient joystick buttons for trim in the game, as it's coded to give you much finer adjustment. Opinions vary here on using a button or axis.  On my Fighterstick I mapped the pitch trim to the button on the left side of the stick where your thumb goes, forward and back for nose down / nose up.  I mapped aileron trim to the left/right switches on the same button, and then mapped rudder trim with the left CTRL key and the left/right switches on that button.   
  • Speaking of trim I tried using the "throttle" slider on the base of the fighterstick as the adjustment for stabilizer trim on the '109 and didn't like it, I use it for zoom currently, and still looking for a better use for that thing,  I ended up mapping the Bf109 stab trim to the same buttons used for pitch trim (since no plane has both, and it all does the same thing in the end anyway). 
  • Besides setting your sensitivities up as advised above, you may need to go into the Settings / Input Devices menu in the game and set the Joystick noise filter to a higher value, which is meant to get rid of "jitter" on some inputs (like the throttle slider) and smooth things out.  I found a setting of 0.5 was needed to quiet some flutter coming from the throttle slider on the base of the stick (which was REALLY annoying when I tried using that for zoom!)  with no ill effect on any other inputs. 

 

A final thought on trimming for straight and level flight... it's an exercise in patience and alternate thought for new pilots.  Give the aircraft time to cycle and settle in after pitch trim adjustments, and remember pitch controls speed, throttle controls altitude...so if you can't get it to settle in at a constant altitude or attitude, it's usually because you're adjusting the wrong thing.   Small throttle adjustments make for even finer adjustments to climb/sink rate after you get pitch trim dialed in. 

 

G'luck!

 

Edited by Stoopy
  • Like 1

Share this post


Link to post
Share on other sites
Posted (edited)

"The Wobble" in the sim is far better than it used to be, but, it is still there and still wrong.  To me it feels like an under damped automotive suspension.  Hit an undulation and it just keeps bouncing.   The sim seemingly does not take into account the aerodynamic damping caused by airflow over the control surfaces.  It's as if the aircraft are NOT flying through air, but are operating in a vacuum... Oh ... Wait...  that's exactly what is happening.   There is no "airflow" as such, as calculating that is beyond the ability of a desktop PC.  So artificial damping must be introduced by the developers to simulate this.  However, simmers get all bent out of shape and immediately fall back on the "flying on rails" trope, even though that is how real aircraft feel in flight.  THEY DO NOT OSCILLATE ON THEIR AXIS AS ANY OSCILLATION IS DAMPED BY AIRFLOW OVER THE FLIGHT SURFACES.

Edited by BlitzPig_EL
  • Upvote 1

Share this post


Link to post
Share on other sites

It is not  perfect, but I prefer it to what else is on offer. 

 

Sensitivity adjustments can certainly improve things

 

Cheers 

  • Haha 1

Share this post


Link to post
Share on other sites
Posted (edited)

Let's try to keep our blood pressure under control shall we?

 

"Flying on Rails" complaints regarding flight simulation are not necessarily related to expected levels of aerodynamic damping or any "lack of oscillation".  I'm speaking of cases where the flight modeling is so poor that the aircraft will continuously maintain extreme bank and pitch angles as if aerodynamically locked into position in a vice-like grip regardless of airspeed changes due to those attitudes. It's the other, and equally incorrect, extreme from what we have here. Surely you must have encountered that in your past experiences in other sims. 

 

I'm also not sure why you believe airflow modeling can't be done on a modern PC.  It's certainly possible and has been done to various extents, the first example being in Terminal Velocity's "Fly!" Simulator back in the late 90's.  The real issue is, it's tough to do that plus have overhead for other routines to make everyone happy in a combat flight sim like handling AI logic plus ballistics plus weather plus shadings and graphics plus sound plus mission timers and triggers plus scenery object drawing plus multiplayer object and network connection handling plus joystick inputs plus lots of other stuff I'm leaving out. But it's not impossible, just because it is difficult to manage. 

 

 

 

Edited by Stoopy

Share this post


Link to post
Share on other sites
Posted (edited)
7 hours ago, 19//Moach said:

It's indeed a known quirk of the sim.  It used to be much worse, no doubt.

 

As a fellow developer, it strikes me as the effect of excessive moment-of-inertia, coupled with insufficient rotational damping.  Yet this may be due to the subtle little things that even the finest mathematical approximations of flight don't take into full account. That is, there may be nothing wrong with the flight model itself in an academic sense, except that theory and reality don't always meet exactly half-way. 

  

 

Slowly but surely the devs are getting to the bottom of it, and we're all confident they won't let go of that bone until it's been chewed all the way through. This team is just as obsessed about aviation as the lot of us. *tips hat*

 

 

 

I certainly don't know how any of these planes ingame fly IRL because I've never sat in one but the physics feel incredibly close to what I experience when flying gliders. In fact flying in Il-2 made learning to fly the gliders incredibly easy, almost as if I had done it before - it felt intuitive with my prior sim experiences.

 

Adjust you elevator sensitivity and even more your rudder sensitivity, in real life you have proper force feedback, here you need to take these measures to counteract the missing forces (even IRL you can often fly a plane with your little finger).

Edited by 216th_Jordan
  • Haha 1

Share this post


Link to post
Share on other sites

I appreciate all the responses. It is good to know I'm not crazy. I've learned to be way more careful and gentle with movement and inputs than I have with other sims. I've been able to enjoy the flights in the 109 F2.

Share this post


Link to post
Share on other sites
Posted (edited)
22 minutes ago, Oscar_Juliet said:

. It is good to know I'm not crazy.

 

Same here. I experience the same issues and kinda got used to the idea that my joystick didn't dialogue well with this particular sim. 

 

I fiddled around with settings for years and nowadays during most of the flight I can almost forget about the wobble. Its more noticeable when trying to train your guns on a target but still manageable. During landings, though..... damn, how it makes me mad. It seems I can never approach the landing strip without going all rollercoster-like.

Edited by danielprates

Share this post


Link to post
Share on other sites
Posted (edited)

If trimmed and setup correctly pretty much all the planes in the game can be flown on final without joystick input so to speak and
Pitch For Airspeed
Power For Altitude
Works rather well 

Also Xplane Models wind and the like very well using Blade Element Theory 
it simulates something like 10 parts per wing 10 parts per prop. and so on.

"Most other simulators use something called “stability derivatives”(Like IL2:Great Battles) to compute how an airplane flies.
This technique involves simply forcing the nose to return to a centered position along the flight path with a certain acceleration for each degree of offset from straight-ahead flight of the airplane—for every degree of angle of attack the nose is raised, the nose should return to center with a certain acceleration. This is a perfectly nice rule of thumb, but is far too simplistic to use across the flight envelope of the airplane."

 

Stability derivatives will typically say,

Okay, we are flying at Mach 0.8, so we add 5% to our drag due to compressibility,
 

in a situation where blade element theory will say,

Okay, we are flying at Mach 0.8, but the wings are swept at 45 degrees, and the plane is in a 5 degree right side-slip, so the effective sweep on the left wing is only 40 degrees, but the effective sweep on the right wing is 50 degrees, and the plane is rotating at 10 degrees per second to the right, so the advancing wing has an extra 10 knots of speed at the wingtip due to this rotation, but the retreating wingtip has 10 knots less speed due to this rotation, and the roll rate is 30 degrees per second to the right, which increases the angle of attack from nothing at the center of the plane to 2 degrees at the right wingtip and negative 2 degrees at the left wingtip, and the plane is pitching up at 10 degrees per second, which adds 1.5 degrees of angle of attack to the tail and takes away 0.1 degrees angle of attack on the main wing because it is in front of the center of gravity, and the changes in angle of attack cause increase in induced drag on the horizontal stab reduction in induced drag on the forward wing.

Edited by =TBAS=Sschatten14

Share this post


Link to post
Share on other sites
1 hour ago, =TBAS=Sschatten14 said:

If trimmed and setup correctly pretty much all the planes in the game can be flown on final without joystick input so to speak and
Pitch For Airspeed
Power For Altitude
Works rather well 

Also Xplane Models wind and the like very well using Blade Element Theory 
it simulates something like 10 parts per wing 10 parts per prop. and so on.

"Most other simulators use something called “stability derivatives”(Like IL2:Great Battles) to compute how an airplane flies.
This technique involves simply forcing the nose to return to a centered position along the flight path with a certain acceleration for each degree of offset from straight-ahead flight of the airplane—for every degree of angle of attack the nose is raised, the nose should return to center with a certain acceleration. This is a perfectly nice rule of thumb, but is far too simplistic to use across the flight envelope of the airplane."

 

Stability derivatives will typically say,

Okay, we are flying at Mach 0.8, so we add 5% to our drag due to compressibility,
 

in a situation where blade element theory will say,

Okay, we are flying at Mach 0.8, but the wings are swept at 45 degrees, and the plane is in a 5 degree right side-slip, so the effective sweep on the left wing is only 40 degrees, but the effective sweep on the right wing is 50 degrees, and the plane is rotating at 10 degrees per second to the right, so the advancing wing has an extra 10 knots of speed at the wingtip due to this rotation, but the retreating wingtip has 10 knots less speed due to this rotation, and the roll rate is 30 degrees per second to the right, which increases the angle of attack from nothing at the center of the plane to 2 degrees at the right wingtip and negative 2 degrees at the left wingtip, and the plane is pitching up at 10 degrees per second, which adds 1.5 degrees of angle of attack to the tail and takes away 0.1 degrees angle of attack on the main wing because it is in front of the center of gravity, and the changes in angle of attack cause increase in induced drag on the horizontal stab reduction in induced drag on the forward wing.

 

 

Share this post


Link to post
Share on other sites

What X-Plane also has is the ability to adjust 'stability augmentation' as well as control sensitivity:

X-Plane-sensitivity.png

 

I've not used X-Plane much yet, but from an initial impression I'd say that that most aircraft actually need some 'augmentation' at least in the yaw axis, to be comfortably flyable.  So maybe blade-element theory isn't a complete solution (and to be fair, the X-Plane developers don't claim their modelling is perfect, anyway).  

 

Share this post


Link to post
Share on other sites
4 hours ago, danielprates said:

 

 It seems I can never approach the landing strip without going all rollercoster-like.

 

Sounds like you are aiming for the landing strip using elevator input to control your rate of sink or glide path. I set my approach with trim with very little  elevator inputs needed , usually controlling my approach with slight power input changes. The only time I'm dancing on the controls is when I'm about to settle down and during the braking process to keep the tail straight. In turbulent conditions more control authority is required via all axis , but I never aim for the strip like I'm trying to perform a carrier landing (controlled crash).

Share this post


Link to post
Share on other sites
1 hour ago, bzc3lk said:

 

Sounds like you are aiming for the landing strip using elevator input to control your rate of sink or glide path. I set my approach with trim with very little  elevator inputs needed , usually controlling my approach with slight power input changes. The only time I'm dancing on the controls is when I'm about to settle down and during the braking process to keep the tail straight. In turbulent conditions more control authority is required via all axis , but I never aim for the strip like I'm trying to perform a carrier landing (controlled crash).

 

That is how I do it in "by the book" study-level sims like FSX. It is indeed the correct way to do it: throttle slightly to correct rate of sink.  But then again, there you have more tools to do the job right like ILS, DME and the like. I don't know why, in 1940s sims I keep wanting to eyeball everything... bad procedure from my part, I admit.

Share this post


Link to post
Share on other sites
Posted (edited)
26 minutes ago, danielprates said:

 

That is how I do it in "by the book" study-level sims like FSX. It is indeed the correct way to do it: throttle slightly to correct rate of sink.  But then again, there you have more tools to do the job right like ILS, DME and the like. I don't know why, in 1940s sims I keep wanting to eyeball everything... bad procedure from my part, I admit.

 

            I find looking at You Tube articles featuring actual in cockpit flying and taking note of their instruments during take off and landing and replicating the same "numbers" does help me. 

           As an example, the DCS Mig 29 created some issues (porpoising) for me after touch down no matter how I greased it onto the runway, until I studied some footage on the net and copied the real example by holding the nose wheel up after touch down of the main wheels to the letter and my problem was solved.

           I also have a FFB stick that gives me different stick forces according to speed, especially on approach together with muscle memory in relation to these forces may also be a contributing factor for my lack of issues. This is certainly a factor in my decision at the moment, as to whether I drop the hammer and purchase a Virpil set up soon.

      

Edited by bzc3lk

Share this post


Link to post
Share on other sites
Posted (edited)

The performance of the flight dynamics regarding, for instance, static and dynamic pitch stability when using  FFB controllers, is an "open" question for me because I never owned such a stick.

 

I did try ( and didn't like it - but that was long ago with another flight simulator game ) a cheap one, it was NOT the MS FFB I read / heard was very good for it's time.

 

I wonder if the use of a good FFB device in IL2 GB might give me a different feel regarding this "wobbling" effect ?

 

Regarding X-Plane comparisons - I've been using X-Plane in demo since version 2 and as a buyer since version 7 - can't really compare to IL-2 GB in many aspects. By far IL-2 GB does a much better / credible job in terms of "feel of flight" and detailed flight characteristics of it's aircraft than any similar aircraft type I have ever used for X-Plane.

 

The hype around "BET" is a lot bigger than the end result - reason why it's IL-2 I use these days whenever I seat at the desktop PC to pretend I'm a pilot flying a powerful ww2  or awesome ww1 aircraft, even though war and air combat are far from being my "beach"...

 

And, btw, contrarily to what someone suggested in a quoted text above, I do not think IL-2 GB's approach is stab-derivate based - quite on the contrary!

Not saying that parametrically it doesn't use derivates, but I find the approach described in one of the first texts posted by the developers when it was announced as a project gathering 1C and 777 teams, a lot more in the line of the so called X-Plane "BET" approach...

Edited by jcomm-il2
  • Thanks 1

Share this post


Link to post
Share on other sites
Posted (edited)
36 minutes ago, jcomm-il2 said:

The performance of the flight dynamics regarding, for instance, static and dynamic pitch stability when using  FFB controllers, is an "open" question for me because I never owned such a stick.

 

I did try ( and didn't like it - but that was long ago with another flight simulator game ) a cheap one, it was NOT the MS FFB which I read was very good for it's time.

 

I wonder if the use of a good FFB device in IL2 GB might give me a different feel regarding this "wobbling" effect ?

 

Regarding X-Plane comparisons - I've been using X-Plane in demo since version 2 and as a buyer since version 7 - can't really compare to IL-2 GB in many aspects. By far IL-2 GB does a much better / credible job in terms of "feel of flight" and detailed flight characteristics of it's aircraft than any similar aircraft type I have ever used for X-Plane.

 

The hype around "BET" is a lot bigger than the end result - reason why it's IL-2 I use these days whenever I seat at the desktop PC to pretend I'm a pilot flying a powerful ww2  or awesome ww1 aircraft, even though war and air combat are far from being my "beach"...

 

 

Nice to seem someone bring up how much BET doesnt live up to its hype. Doesn't il2 use some form of BET? DCS basically does and il2 seems to show all the same symptoms.

 

I miss the older tabular fm-ing of the early 2000s. It was more predictable. Sure, you generally (although it wasnt impossible) didnt get to experience every little quirk of and aircraft, but at least stuff like top speed, climb rate, and turn rate were generally reliably implemented.Flight modeling before all the new sims came out in 2012-13 had its issues. But Ive never been able to see the rational in sacrificing a predictable flight model where speeds, climbs, turns, dive speeds, flap behavior are all basically correct so that X or Y plane can (theoretically) exhibit its unique adverse yaw characteristics or landing quirks. 

 

All of the newer sims have a big issue with trying to do certain specific minutiae to an extreme level of detail at the expense of everything else. DCS being the most notable culprit. The P-51 flight model in DCS has been in tuning land since its initial release 9 billion years ago. Top speed fluctuating by 15mph at various heights.

 

 

Edited by Fumes
  • Upvote 3

Share this post


Link to post
Share on other sites
Posted (edited)

You shouldn‘t actually need any kind of curves to modulate your stick input. This sim does actually a very decent job of translating stick inputs.

 

The infamous wobble got addressed back in the 2.0 update, and it is much better now, although not eliminated. The real aircraft does not wobble at all, and if provoke it with all your might (it takes all your strength to be rude enough to the aircraft for provoking this), the oscillation is, my impression here, about 0.5 to 1 per second. It is slow.

 

One thing you have to keep in mind is that this sim simulates very fast aircraft (as opposed to a Cessna 172). This means controls become incredibly sensible in the real aircraft and most what is described here as problem is an over controlling of the aircraft. If you fly the real Spitfire, compared to our in game Spitfire, you will experience the exact same problem. I find it surprisingly hard to make the real Spitfires nose exactly where you want it, as your sim muscle is no help. The stick is far away from you (compared to your desk mount) and the response from the stick input is very uneven. A slight touch on the stick, not causing visible deflection, is enough to steer your Spit wherever you want to go. The center play (with no spring force to hold it) of my Saitek X-56 is much larger than stick movements required to fly her in „airliner mode“. Beyond that, you start to need a lot of force. At 500 km/h I had to use force such that I would break any desk mount flight stick. What sets natural limits to your wobbling around the stick in the real world is normally absent in the sim. 

 

My impression is that other flight sims, especially the ones that matured along selling GA planes, are soft on translating stick inputs into control responses. After two decades flying slow planes, it‘s maybe what feels right for fast planes as well.

 

Even without the wobble, you should learn to develop a feeling for the exact stick movements required. Correcting after over controlling will make you going zig zag as well in the real aircraft even without wobble. Going fast will make you only give small sick inputs (you‘d risk ripping off your wings with sudden full deflections). Use smaller inputs in the sim as well.

 

At the speeds these planes are intended for, they fly very much on rails. When you go slow (or very, very fast) things start to change.

 

12 minutes ago, Fumes said:

Personally I miss the older tabular fm-ing of the early 2000s.

Any higher end FM uses both tables and computations. Computations are only convenient within normal flight. Stalling and post stall properties cannot be computed. Airliner sims use huge tables to get systems etc. right. After a certain number of dependencies, you can‘t do it right with a simple forula.

Edited by ZachariasX
  • Thanks 1

Share this post


Link to post
Share on other sites
Posted (edited)
15 minutes ago, ZachariasX said:

 

Any higher end FM uses both tables and computations. Computations are only convenient within normal flight. Stalling and post stall properties cannot be computed. Airliner sims use huge tables to get systems etc. right. After a certain number of dependencies, you can‘t do it right with a simple forula.

Specifically I am referring to the common practice these days or breaking the FM into hundreds or thousands of parts and then hoping the end results is that the Cdo of the plane under X condition turns out to be right. These are all using tables still as its been explained to me. This is the approach DCS uses and its why we have endless tuning issues with their FMs. The older flight models used tables at a more macro level, instead of making the plane 1000 sub models that have to add up right. Thats why in the F-14 dev diary some time ago you could see them talking about how they were trying to tune the turn performance without messing up the rest of the model too much.

 

I dont know really what il2 uses. But whatever it is it appears to have alot of the same problems. Endless tuning. Random unpredictable errors that take time to even figure out what is causing them.

 

To me, what matters in a flight model is getting the airplanes tactical performance right. Top speeds, acceleration, compression, turn performance, flap performance, engine outputs at various altitudes, etc. The more complicated the flight model is made to try to extract out an extreme level of detail and make the plane "feel" correct, the less reliably we seem to get accurate performance in those other areas.

Edited by Fumes
  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites
Posted (edited)

My other sims are Aerowinx PSX and ELITE ( now XTS ).

 

Both PSX and ELITE couldn't be more "table-based" :-), and yet they do their jobs remarkably well, and IMO incomparably better than X-Plane for the same aircraft types they model, in the case of ELITE provided you "vly" them within the "normal IFR envelope" and also completely forget about "ground physics"...

 

Fine tuning a BET-based FDM must be a real headache. And I mean, doing it in DCS or IL-2, since for instance in XP I always starred at the inability to simply "predict" the pitching moments due to flap deployment on my designs using Plane-Maker, and always having to manually add it ( parametrically after all... bummer... ).

 

But honesty, and even though I feel the quirks here & there, there's soemething really special about IL-2 GB that makes me always return to it after my diversions 🙂

Edited by jcomm-il2
  • Upvote 1

Share this post


Link to post
Share on other sites
14 hours ago, =TBAS=Sschatten14 said:


"Most other simulators use something called “stability derivatives”(Like IL2:Great Battles) to compute how an airplane flies.
 

How are you so sure?  I don't remember devs statement about which exactly tech they use.

Share this post


Link to post
Share on other sites
3 hours ago, ZachariasX said:

 

My impression is that other flight sims, especially the ones that matured along selling GA planes, are soft on translating stick inputs into control responses. After two decades flying slow planes, it‘s maybe what feels right for fast planes as well.

 

You have a good point there. Most of us only know the virtual "feel" of the stick, and commercial sims may accustom us in the wrong manner as to what expect from those older warbirds. 

 

In any case, tweaking the pitch sensitivity a few years ago indeed made the game that much enjoyable. Specially so during landings. 

 

Now for something completelly different: I feel I should correct some statements above, lest someone get lost trying to find it. The sensitivity settings are not under "input". There you only find mouse sensitivity and joystick noise filter. The axes' sensitivity settings are actually under key mapping, in the airplane control tab, right beside the key binding gizmo.

Share this post


Link to post
Share on other sites

Real pilot here, also own a warbird, fly it often.

Hated CLOD flight model, totally unrealistic, planes just do not fly like that.

To me IL2 BOS might not be perfect but gives the feeling of flight better than any other sim, Civilian or not (have them all).

Yes there is room for improvement, but is the best out there at the moment IMHO followed by DCS.

  • Like 1
  • Thanks 3
  • Haha 1
  • Upvote 1

Share this post


Link to post
Share on other sites
Posted (edited)
5 hours ago, II./JG77_motoadve said:

Real pilot here, also own a warbird, fly it often.

Hated CLOD flight model, totally unrealistic, planes just do not fly like that.

To me IL2 BOS might not be perfect but gives the feeling of flight better than any other sim, Civilian or not (have them all).

Yes there is room for improvement, but is the best out there at the moment IMHO followed by DCS.


While your experience is a welcome addition, the “real pilot here” is a bit elitist. It’s probably not at all your intention to imply this, but you are 1) not even close to the only “real” pilot with experience in high performance aircraft on this forum / in this thread and 2) undermining and invalidating the posts of others who didn’t start their post by listing their experience. We’ve all had the pleasure of watching your videos and reading past posts, so don’t worry, we’re familiar with your flying resume. 

Edited by Go_Pre
Spelling
  • Sad 1
  • Upvote 1

Share this post


Link to post
Share on other sites
11 hours ago, 307_Tomcat said:

How are you so sure?  I don't remember devs statement about which exactly tech they use.

 

He doesn't know, like a lot of things he writes about here.

  • Haha 1

Share this post


Link to post
Share on other sites

I would say if you want as close to authentic real flight characteristics as possible I would look at X-Plane. Not sure if you get to shoot down anything other than AI though. 

With my limited number of flight hours in a C172 - I think those on the forum that have more experience that I will say the same - If you take BoX as what it is - a game that is trying to simulate flight and projectical behaviour, you won't be disappointed.

  • Upvote 1

Share this post


Link to post
Share on other sites

I am always amazed how some people take a simulator platform as a general reason why (all!) simulated planes have a superior FM, as if it wasn‘t for for detailed knowledge and effort of a programmer to make that particular sim engine give a desired result. FSX, P3D (not so much anymore due to licensing, but initially) as well as XP all have some stock and add-on aircraft that are...  you know... 🙄 While other planes are simply fantastic.

 

Every time I hear about XP‘s airfoils, it feels like back in the 90‘s when I would hear people say things like „A Mac is good for graphics!“ and you try to be happy for them having found some purpose in life.

Share this post


Link to post
Share on other sites
3 hours ago, ZachariasX said:

I am always amazed how some people take a simulator platform as a general reason why (all!) simulated planes have a superior FM, as if it wasn‘t for for detailed knowledge and effort of a programmer to make that particular sim engine give a desired result. FSX, P3D (not so much anymore due to licensing, but initially) as well as XP all have some stock and add-on aircraft that are...  you know... 🙄 While other planes are simply fantastic.

 

Every time I hear about XP‘s airfoils, it feels like back in the 90‘s when I would hear people say things like „A Mac is good for graphics!“ and you try to be happy for them having found some purpose in life.

 

Yup. Any consumer-grade flight-model simulation I've ever heard of requires more than just the input of raw numbers to create a plausible FM. They are tools to assist with hand-crafting, not automata. As for blade element theory, it is clearly useful in analysis, but can't tell you everything. And in terms of the issue raised by the OP, 'wobble', or the degree to which an aircraft will oscillate around an axis after disturbance, it won't on its own tell you much at all, since that clearly depends on the distribution of mass of the aircraft, as well as on aerodynamic forces. And BET calculations don't consider 'blade' mass or rotational inertia, as far as I'm aware. They don't need to, since all they are doing is calculating instantaneous forces. There is clearly more to X-Plane modelling than BET, and accordingly even if BET gave perfect results (which it doesn't, since it is an approximation, and a simplification) other things could be entirely wrong. From my limited experience (only had X-Plane for a couple of weeks) I'd suggest that generally 'other things' probably aren't far off, for the better-modelled aircraft, though there are also some total dogs. 

 

And returning to the 'wobble' issue, one thing that X-Plane can confirm (with a well-tooled model) is that as in real life, 'wobble' depends on more than just the aerodynamic form of the aircraft. X-Plane allows you to move aircraft CG backwards and forwards, and to see how that affects stability. I'd recommend anyone who has X-Plane to try the excellent Ryan Navion freeware aircraft over its CG range (or at least, most of it: you can move it so far aft that it tips backwards on its main wheels, which makes taking off a little weird) With a forward CG, an elevator 'singlet' (the momentary movement of the elevator to its limit, followed by immediate centring ) produces a definite wobble oscillation, whereas with an aft CG, the same singlet results in an immediate change of pitch angle, with next to no oscillation. Generalisations about how real-world aircraft 'wobble' should probably take such things into account too.

  • Like 2
  • Upvote 1

Share this post


Link to post
Share on other sites
1 hour ago, AndyJWest said:

 

Yup. Any consumer-grade flight-model simulation I've ever heard of requires more than just the input of raw numbers to create a plausible FM. They are tools to assist with hand-crafting, not automata. As for blade element theory, it is clearly useful in analysis, but can't tell you everything. And in terms of the issue raised by the OP, 'wobble', or the degree to which an aircraft will oscillate around an axis after disturbance, it won't on its own tell you much at all, since that clearly depends on the distribution of mass of the aircraft, as well as on aerodynamic forces. And BET calculations don't consider 'blade' mass or rotational inertia, as far as I'm aware. They don't need to, since all they are doing is calculating instantaneous forces. There is clearly more to X-Plane modelling than BET, and accordingly even if BET gave perfect results (which it doesn't, since it is an approximation, and a simplification) other things could be entirely wrong. From my limited experience (only had X-Plane for a couple of weeks) I'd suggest that generally 'other things' probably aren't far off, for the better-modelled aircraft, though there are also some total dogs. 

 

And returning to the 'wobble' issue, one thing that X-Plane can confirm (with a well-tooled model) is that as in real life, 'wobble' depends on more than just the aerodynamic form of the aircraft. X-Plane allows you to move aircraft CG backwards and forwards, and to see how that affects stability. I'd recommend anyone who has X-Plane to try the excellent Ryan Navion freeware aircraft over its CG range (or at least, most of it: you can move it so far aft that it tips backwards on its main wheels, which makes taking off a little weird) With a forward CG, an elevator 'singlet' (the momentary movement of the elevator to its limit, followed by immediate centring ) produces a definite wobble oscillation, whereas with an aft CG, the same singlet results in an immediate change of pitch angle, with next to no oscillation. Generalisations about how real-world aircraft 'wobble' should probably take such things into account too.

Yep, that's Static Margin.

I've tried with a few aircraft in IL-2, by varying the fuel load, as well as ammunition, but didn't notice any difference...

That's true about X-Plane, but also about MSFS - on both you can play with your CoG  / CoM vs CoL

Share this post


Link to post
Share on other sites

The RoF Sopwith Camel demonstrated a change of stability with varying fuel load very well - with a full tank (behind the pilot) it was distinctly unstable in pitch, improving somewhat as fuel mass decreased. I haven't tried the FC Camel enough to see if it exhibits the same behaviour.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...