Impressions, part 2

[Sternjaeger]

Ok, so finally I managed to get the new rig and I spent a couple of hours last night over the sim. It's probably because I hadn't flown it since week 1, but it sort of was like the first time, and I still think it's an incredibly solid product.

 

The things I've noticed so far (in no precise order).

 

• yes, the slats on the Bf109 need sorting out, the behaviour and deployment is totally erratic.
• I see what you guys mean with the pitch bouncing, that's definitely weird, but again it would be interesting to know what's the cause (input or FM).
• I haven't managed to see any condensation on wingtips, is it just me?
• there's no pre-stall buffeting on my FF joystick.
• there seems to be a slight delay between input and response (I noticed this particularly with the firing button)
• the big hand counting seconds on the Bf109 clock shouldn't be moving: that was the chronograph hand.
• It would be nice to have a gunsight reticle toggle switch and dimmer adjustment (gunsights were off most of the time, as their brightness was really distracting).
• when diving with the Bf109, the plane tends to pull up when reaching very high speeds.
• I see what some people mean with the tailwheel being a bit too bouncy, but bear in mind that the faster you go the bouncier your tailwheel will be, I think it's almost fine the way it is, the bounciness probably should depend also on the sort of damping/suspension used.
• the rudder-induced spins are very realistic, I think the issues arise when people use joystick rudder instead of pedals. My impression re. the input so far is that it's designed like in the real thing, especially the travel of controls. The truth is that the old IL-2 had more of a "damped" input to compensate for this, whereas this new sim is more "raw", but actually more realistic in its response. It's kind of like learning to fly again I suppose, but I like the progression of rudder response for instance.

 

I'm sure there's something else I forgot about, but I'll try it a bit more later today..

 

So far I'm still very very very very impressed with the solidity of the game, especially the graphics, which didn't show a single glitch on my rig.

Edited January 13, 2014 by Sternjaeger

[OBT-Psycho]

actually on FF buffet before stall, I think it is modeled but it is very weak and maybe too close to the stall. I mean I felt some shake just a blink of an eye before stall, but not sure if it was really that.

 

As for the delay between trigger and actual guns shooting, I think it is understandable, but maybe a bit too long. Seems like a direct portage from WWI sims, just as dispersion of mg17, but still needs to be ironed out.

[6S.Manu]

As for the delay between trigger and actual guns shooting, I think it is understandable, but maybe a bit too long. Seems like a direct portage from WWI sims, just as dispersion of mg17, but still needs to be ironed out.

 

I agree.

[Fifi]

 

 

especially the graphics, which didn't show a single glitch on my rig.

 

Not really a glitch, but do you experienced the latence time for skins to load (switching F2/F1), or even on the Lagg dash board rendering (F2/F1)?

Indeed pretty annoying here.

[OBT-Psycho]

I have this happening, but IIRC it might be graphic engine related as there is the same in ROF. but maybe it is the opportunity to fix this

[kestrel79]

Would love a slight zoom on the gunsight view setup by default instead of having to make your own cameras and such.

[Sternjaeger]

Not really a glitch, but do you experienced the latence time for skins to load (switching F2/F1), or even on the Lagg dash board rendering (F2/F1)?

Indeed pretty annoying here.

 

yes I do, but only on the intro menu, not in game (whereas I've seen some videos of people posting their screen grabs with the same thing happening in game as well). Not a major glitch IMHO, but yeah, it should be addressed.

 

 

Another thing that I've noticed and worries me a bit is the roll rate of the 109 at high speed.

[Wolger]

That is a known bug - texture slow to reload when switching external cams. I see it's a pretty critical one.

[Tektolnes]

Another thing that I've noticed and worries me a bit is the roll rate of the 109 at high speed.

 

What were you expecting to see? I don't have the data for the F-4 but the F-2 data I could find had the following roll rates at 66lbs stickforce:

 

200 kph = 45 deg/sec (0.8rad)

300 kph = 68 deg/sec (1.2rad)

400 kph = 83 deg/sec (1.45rad)

500 kph = 88 deg/sec (1.55rad)

600 kph = 91 deg/sec (1.6rad) - peak value

700 kph = 56 deg/sec (0.98rad)

800 kph = 23 deg/sec (0.4rad)

 

If anything I think the roll rates in BOS are a little high overall. I'd be pretty surprised if the heavy and sluggish Lagg could do a full aileron only roll in 3 seconds (120deg/sec).

[Sternjaeger]

we're quite far from those numbers IMHO, especially with the high speed data, but again it's sort of hard to ascertain with no statistics from the game :-/

[Tektolnes]

I did a few quick tests earlier with a 109 at 50% fuel load and no wing pods, etc.:

 

At 300km/h @ 3k (altitude from the F-2 test data) it took me ~4 seconds (90 deg/sec).

 

At 400km/h @3k it took me ~3 seconds (120 deg/sec) and I couldn't really notice any appreciable difference in the 400-600km/h range at 3k in the few quick tests I did which is fair enough as they should be more or less the same.

 

At 700km/h @ 1.5k it took me ~5 seconds (72 deg/sec) I think. I might try later to climb up to 5k and dive down to 3k just to keep the same test conditions but I'm not expecting much of a different result.

 

All in all both the Lagg and 109 seem to roll too quickly at all speeds.  

[Sternjaeger]

that's interesting man, where did you get that reference data from?

[Rama]

66lbs is 30kN, I'm not sure you can achieve full deflexion with this force, especially with the highest speed.

So it doesn't seems that far off (ingame from given reference). Maybe a bit to optimistic

 

Maybe you should give your references (source) to the dev, in case they don't have them.

[Tektolnes]

that's interesting man, where did you get that reference data from?

 

http://bbs.hitechcreations.com/smf/index.php?topic=160997.0

 

Looks like people have been arguing over this for the past 7 years at least! User there got it from Deutsches Museum. Unfortunately the original docs are gone but Kurfurst has copied in the values a little further down so I can't say for sure how correct the data is. But it does tie in with other stuff I've found on roll rates for WW2 birds

.

There's also the article on the guy who flew 109G-2 Black Six: http://109lair.hobbyvista.com/articles/airframes/black6/bk6_flight.htm:

 

"Roll performance is similar to a Hurricane or elliptical wing tipped Spitfire.  A full stick roll through 360 degrees at 460kph takes 4 to 4.5 seconds without using rudder, and needs a force of around 20 lbf.   One interesting characteristic is that rolls at lower speeds entered at less than 1g, such as a roll-off-the-top or half Cuban, have a markedly lower roll rate to the right than to the left.  Therefore, I always roll left in such manoeuvres."

 

I haven't been able to find anything which would point to the 109 doing 3 second full rolls which, if wrong, would make things off by 30-40% in BOS.

Edited January 13, 2014 by Tektolnes

[VO101_MMaister]

66lbs is 30kN, I'm not sure you can achieve full deflexion with this force, especially with the highest speed.

So it doesn't seems that far off (ingame from given reference). Maybe a bit to optimistic

 

Maybe you should give your references (source) to the dev, in case they don't have them.

Well, with 30kN you could bend the stick. However 66lbs is only 300N what is ca. 30kg. Plenty if you ask me on aileron.

Edited January 13, 2014 by VO101_MMaister

[Tektolnes]

For what it's worth I had a spin in the DCS P51 earlier to try the roll rate there at various speeds. Seems to tie in with the NACA report on roll rates - mediocre at lower speeds while better than most at speeds approaching 400mph. At the upper speed band of 380mph where the Mustang roll rates really comes into it's own (faster than the FW190 at this speed) I could do a full aileron rull with no rudder input in ~4-4.5 seconds which is a good second slower than BOS.

[Rama]

Well, with 30kN you could bend the stick. However 66lbs is only 300N what is ca. 30kg. Plenty if you ask me on aileron.

Yes, sorry for the mistake with the kN

Still, I'm not sure you can reach full aileron deflection at 600 or 700 km/h with a force of 30kg on a Bf109 F4.

[VO101_MMaister]

Yes, sorry for the mistake with the kN

Still, I'm not sure you can reach full aileron deflection at 600 or 700 km/h with a force of 30kg on a Bf109 F4.

We have no exact data about stick forces, but the aileron forces are always significantly lower than the ones on the elevator. As a thumb rule the aircraft designers aim for the 50% of elevator force. 30kg is a lot on ailerons, there are not many pilots who can push the stick sideways with such force. Sounds unrealistic even around the Vd.

 

I don't have the BoS, but if i accept Tektolnes's game test results and compare it with the real life test results the rollrate in the game is much faster already from 3-400 km/h where the stick forces must be really light.

 

MM

[Rama]

Well, I said I'm not sure because British pilots testing a captured Bf109-E reported that above 500 km/h, ailerons were quite stiff, and that they were far to be able to get full authority from them at 600 km/h (around 25% of it with the max strengh they could apply). (in Spitfire: a Test Pilot's Story, from Jeffrey Quill)

Don't know about any similar test report for the Bf109-F4, so I don't know at all if full aileron authority was achievable at high speed. If you have a test report telling it was, I would be interested.

[Finkeren]

It's always dangerous to rely on test results from captured planes, but this seems to have been regarded as fact in more or less all I've read about the Bf 109 as well. On the other hand, there was quite a bit of redesign of the wing from the E to the F models, including a shortening of the ailerons, so maybe there is a difference? 

[Quax]

The difference of the E to the F model was huge in all aspects. 

[Sternjaeger]

Well, I said I'm not sure because British pilots testing a captured Bf109-E reported that above 500 km/h, ailerons were quite stiff, and that they were far to be able to get full authority from them at 600 km/h (around 25% of it with the max strengh they could apply). (in Spitfire: a Test Pilot's Story, from Jeffrey Quill)

 

...and that sort of leads back to the point I was trying to make in the other topic: how do we keep that into account and above all simulate it without people thinking it's a limitation, because frankly it's not, it's a design thing.

[VO101_MMaister]

...and that sort of leads back to the point I was trying to make in the other topic: how do we keep that into account and above all simulate it without people thinking it's a limitation, because frankly it's not, it's a design thing.

 

I think all the abnormalities regarding the FM, one can read on these forums leading to this eternal question of flight simulations. And the only adequate solution would be a proper FFB system with full scale controls.

 

Until then we need to compromise between limited control movements (limited by some sort of human ergonomic factor like in CloD, so joystick and controls wont move 1:1 all the time), or to forget the human factor completely (like on BoS, so far, but then we can get unrealistic aircraft movements, and toward the limits the FM cannot be compared and confirmed by real life test results, because in RL these situation were impossible).

 

I didn`t like the CloD`s system, but now I tend to accept it as a better solution. It is a simulator and as a simulation it should take into consideration the human ergonomic limitations as well, because it was part of the system. Otherwise we will end up with aircrafts flying like they were piloted by superhumans doing manouvers far beyond realism.

Edited January 14, 2014 by VO101_MMaister

[Sternjaeger]

Exactly.

 

I'd rather see everbody limited by the same physiological parameters (which in single career mode could marginally "evolve" with your experience) than having planes flying in an unrealistic manner (which already happens anyway, because majority of simmers never truly experienced prolonged high-G manoeuvres).

Edited January 14, 2014 by Sternjaeger

[Rama]

I didn`t like the CloD`s system, but now I tend to accept it as a better solution. It is a simulator and as a simulation it should take into consideration the human ergonomic limitations as well, because it was part of the system. Otherwise we will end up with aircrafts flying like they were piloted by superhumans doing manouvers far beyond realism.

 

Or even a better solution, 100% "realistic", remove the joystick and all uneeded expensive devices liker rudder, etc.... and use a command interface (keys or even vocal) like "fly to point B", "climb to 5000ft", "attacj nearest ennemy fighter", etc...

No more input problem, no more unrealistic human induced maneuvers, no more user exploit, you can have a complex pilot model never exceeding "physiological limits", you can even add some RPG elements like gaining some extra eyesight capability or arm strengh if you succeed enough mission...... and cherry on the cake, you can chat with your friends while drinking a beer at smoking a cigar at the same time.

 

To go back to "reality", if you want to include some real physiological limits in a "pilot model", then not only you will have to include max limit to arm "strengh", but also the fact that the stick wont move as fast as your joystick, this inevitably leading to the total weird outcome: having the governs moving to the opposite direction as your input from time to time.... that's what I would never want to see.

At least, if you want to include "max stiffness", then it should keep the 1:1 move ratio.

[VO101_MMaister]

Or even a better solution, 100% "realistic", remove the joystick and all uneeded expensive devices liker rudder, etc.... and use a command interface (keys or even vocal) like "fly to point B", "climb to 5000ft", "attacj nearest ennemy fighter", etc...

No more input problem, no more unrealistic human induced maneuvers, no more user exploit, you can have a complex pilot model never exceeding "physiological limits", you can even add some RPG elements like gaining some extra eyesight capability or arm strengh if you succeed enough mission...... and cherry on the cake, you can chat with your friends while drinking a beer at smoking a cigar at the same time.

 

To go back to "reality", if you want to include some real physiological limits in a "pilot model", then not only you will have to include max limit to arm "strengh", but also the fact that the stick wont move as fast as your joystick, this inevitably leading to the total weird outcome: having the governs moving to the opposite direction as your input from time to time.... that's what I would never want to see.

At least, if you want to include "max stiffness", then it should keep the 1:1 move ratio.

Well it is a theoretical question we are discussing here, and there will never be an agreement, because there is no perfect solution. But still it is interesting to discuss it!

After pointing this down, I think you dramatized it a bit too far. We are talking about human factors as barriers here (no 0 or 1 options), between these barriers you operate during real flight as well. How much we take these into consideration is a question of developer decisions.

 

Why do we have black out in games? Why not just pull constant 10G without it? Wouldn`t you find it more realistic since the aircraft would tolerate it?  Why is this human factor is acceptable in games, but not the strength of the pilot? It is part of the flight  and the human barriers! Just like the part of the flight that you can`t pull a stick with 100kg.

 

On the top of that when we are on field of these superhuman manouvers we have no idea how would a real aircraft react. A programmer create some code based on some sort of mathematical scheme, and that is it. Maybe the schema is correct maybe not. Probably not, otherwise factories would not need test flights to figure out how the aircraft acts under different circumstances, they would only use math instead.

 

It is quite easy to find database with avr. human capabilities regarding age and gender. Taking them into the simulation is not a sin, but improving the realism. Like the black out!

 

 

 

 

Edited January 14, 2014 by VO101_MMaister

[Rama]

It is quite easy to find database with avr. human capabilities regarding age and gender. Taking them into the simulation is not a sin, but improving the realism.

No it doesn't.

If your airplane banks to the left while you're pushing the stick to the right, then it's not "improving the realism", it's achieving the very opposite.

And if the ailerons movement isn't related to the stick movement, because you added a delay because of the modelled human strenght, then that something you will see each time you will reverse your joystick move (so very, very often).

I'm sure that with enough cafeine, you can yank a small plastic device (trying to figure a stick...) from full left to full right and reverse at least 5 times per second.... There's absolutly no way you can forbid it to the player. There's no way you can create by software mass and stiffness for this joystick. And I'm sure there are thousands of crazy move you could do that we would even think on right now. There's absolutly no way the developper can insure to find something "realistic" to do with the governs when the player creativity produce such a move.

And if the dev try, if he produce an algorithm that move the govern somewhat "realistic" to some taste, totally different from the joystick move... then what will the inventive player do? He will try to cheat this new algorithm in order to find the best effect for him... and with enough inventivity and enough try, he will succeed, that's a sure thing. You will ALLWAYS find players gaming the game. And now were will we be, in a air combat simulation?.... certainly no, we will be in a game vaguelly related to air combat and not at all to simulation.

There's no way you can call this "improving the realism". leaving the 1:1 scale will allways degrade it.

 

So, max that can be done, if really needed (and I'm still very far to be convinced), is keeping 1:1 scale and adding max stiffness stops. Everything more would be a simulation nightmare.

[VO101_MMaister]

 

So, max that can be done, if really needed (and I'm still very far to be convinced), is keeping 1:1 scale and adding max stiffness stops. Everything more would be a simulation nightmare.

 

Well we are quite agree here. I would be also against to limit the speed of the controls or to change the 1:1 ratio, but the max deflection of the stick could be limited by the max reachable human pushing/pulling force. Just like in CloD by the way.

[DD_bongodriver]

Not entirely sure about much of the last stuff being discussed here, stick length is creeping back in, but proportionality is independent of the size of stick, half deflection on a 30 foot long pole is the same as half deflection on a Xbox thumbstick.....but that's an old story now.

 

my concern here is, where exactly are the figures coming from that are being used to determine this apparent max force? do we actually have the data for all these aircraft as to how much force it takes to reach full deflection at a gazillion kph? fighter control surfaces are relatively small, many of them had aerodynamic balance devices which provided at least some assistance to the pilot using the very dynamic pressure that apparently is accounting for the control stiffness.

[CaptainFlemme]

In some cases it's VERY important to implement human limitations...

 

 E.g., without proper modelling of elevator stiffness at high speed on a early war Bf109, how is it possible to simulate her "dive problem", and the need for the poor pilot to use the trim to get out of a dive?

Therefore, without it, the RAF tactics of diving and then "resourcing" (is that a word? i'm not sure) to break from combat against an Emil is useless. So in that case, keeping the 1:1 scale is greatly degrading the realism... in my view anyway.

 

I hope i'm not saying something stupid, but in my opinion, a "simulation of air combat" is well made if using "real" (for what it's worth) tactics and maneuvers gives results mimicking reality. If not, what's the point? E.g, Counter Strike will not become a "special force simulator", even if the M16 ballistic characteristics, mechanism, sound and smell is reproduced in every tiny details, if it's still possible to shoot from the hip and accurately hit with that M16, or if it's possible to run and shoot with it... so if "human limitations" on using that weapon is not implemented.

 

To return on the topic: I agree for the first impressions on BoS, there is still fine tuning to do on planes or weapon behavior, but for the most part, "so far so good", the foundations seems rock solid :o) !

 

 

do we actually have the data for all these aircraft as to how much force it takes to reach full deflection at a gazillion kph?

 

I'm pretty sure it's possible to do it, as the dimensions and positions of the control surfaces (and compensators/balancing mass) are known, as also the aerodynamic rules creating these forces and the length of the stick of the real plane. It seems to be only a geometric problem... but I'm maybe mistaken.

 

 

PS: sorry to hijack that thread with my faulty english (it's not my native language)

[DD_bongodriver]

Yes, geometry, surface area, leverage, aerodynamic effects, a touch of cable stretch (elasticity), structural warping...........

[Georgio]

While it would be nice to 'simulate' trying to work the controls while pulling 3g, it's surely impractical as most would think their joystick/pedals at fault.

It's a difficult one to resolve but virtually loading the controls at different speeds isn't going to work imo.

[VO101_MMaister]

 

my concern here is, where exactly are the figures coming from that are being used to determine this apparent max force? do we actually have the data for all these aircraft as to how much force it takes to reach full deflection at a gazillion kph? fighter control surfaces are relatively small, many of them had aerodynamic balance devices which provided at least some assistance to the pilot using the very dynamic pressure that apparently is accounting for the control stiffness.

 

Well, that is a good question indeed. However there should be RL tests which give us information about control forces. We can go the other way around as well. From the tests we know what was the max roll or rate of turn at a given airspeed. All we need to do is to use these values as the limit of max. human input force and limit the control`s movement at these values. Even better this way we will follow the RL test data and won`t get unrealistic manouvers.

[Tektolnes]

Well a lot of the aircraft performance data is aircraft limited rather than human input limited e.g. with the roll rate for most of the airspeeds at 500kph or less a person could get the maximum deflection required to roll the aircraft as quick as it could go. The 109 F-2 had a stick force of ~20kg at 500kph which isn't massive. Human input becomes more important when crazy things are attempted but the basic flight models should be straightforward enough to get right without having to worry about human inputs. DCS manages ok without limiting human inputs - the planes physical limitations, hinge forces, etc. keep you honest.

[kestrel79]

I really like the sense of speed you get when flying low. You really ARE moving. After bouncing back n forth between this and Warthunder it feels in slow motion when flying low compared to IL2, which seems more realistic.

[Georgio]

Yep it's the sense of flight that is most impressive in my book. The feeling especially with the LaGG that it's clawing for altitude on take off and that you'd best keep flap on otherwise you're going to drop like a brick.

The will I won't I make it out of that ground attack dive as you pull the stick back to maximum.

The build up of speed in a dive which for me feels totally authentic. I could go on but I'm sure we've all had those '*€#@ me moments' in this great sim.

 

I really like the sense of speed you get when flying low. You really ARE moving. After bouncing back n forth between this and Warthunder it feels in slow motion when flying low compared to IL2, which seems more realistic.