Realism, whats missing?

[MaxGunz]

On realism the fact that IL-2 does not model 3D mass distribution leads to less than real rotational results.

[Treetop64]

Originally Posted by MaxGunz

On realism the fact that IL-2 does not model 3D mass distribution leads to less than real rotational results.

I'd be interested in seeing the MoI numbers, as well as all the other parameters.

At any rate, from the flight modelling point of veiw, I think the very point of "realism" in IL-2 isn't so much about getting the absolute correct results for a particular aircraft, as much as it is about getting the relative performance between all the aircraft in the game within an acceptable bracket.

[MaxGunz]

Thread title is?

[T}{OR]

Originally Posted by K_Freddie

Sorry.. meant with a bit of elevator input + full yaw.

Using a P51 in IL2 and DCS.. I tried to snap roll it using full elevator and rudder with no aerolons. Engine was on idle to minimize torque and propwash.

1) The IL2 replied with immediate wing stall and roll
2) The DCS P51 made an attempt to yaw and nose up.. and refused to do anything. It hung there wobbling, as if in a software loop that was going nowhere. They went off their rockers when I pointed this out....

Having found that great instructional video that I've posted, I set up both aircraft turning with full yaw and slight cross controls, so as to keep it just on the stall zone. I then applied full aerolon which should have induced a stall.

DCS again did nothing and rolled in the direction of application, still refusing to stall when I applied complete cross controls.
IL2 also did not respond fully but did wobble a bit... and was much slower to roll over - but effectively showed that this FM wasn't modeled that much.

Similar results with my (quick) testing. The plane flips over in DCS but does not enter a stall / spin like in IL2. Even in IL2 it is hard to stall it properly. Which one get the job more accurately I do not know. I do know that the landing in P-51 is much closer to what you can see in YT videos than in IL2 where one can put it down soft and easy (ground effect or low speed handling over modeled??).

EDIT: I was too gentle with the controls. Cut back on power suddenly coupled with full left of right rudder and a touch of positive elevator and she snap rolls and stalls immediately. I'll try to do those other maneuvers show in the video you linked.

Which version were you testing this in? IIRC the whole module is still WIP, from sound etc..

Originally Posted by IceFire

Probably the best reason I've ever heard for splurging on some rudder pedals!

Seconded.

[K_Freddie]

Originally Posted by T}{OR

Similar results with my (quick) testing. The plane flips over in DCS but does not enter a stall / spin like in IL2. Even in IL2 it is hard to stall it properly. Which one get the job more accurately I do not know. I do know that the landing in P-51 is much closer to what you can see in YT videos than in IL2 where one can put it down soft and easy (ground effect or low speed handling over modeled??).

EDIT: I was too gentle with the controls. Cut back on power suddenly coupled with full left of right rudder and a touch of positive elevator and she snap rolls and stalls immediately. I'll try to do those other maneuvers show in the video you linked.

Which version were you testing this in? IIRC the whole module is still WIP, from sound etc..

Seconded.

I started with DCS123, and it's nearly the same in DCS124, with the exception that there seems to be more rudder response - I haven't changed my stick settings. Same for IL2 V411 and V412.

DCS is definitely the better feel sim. I was actually very ham-fisted with the DCS stick/rudder inputs.. but to no avail.

Here's an external view of Free flight over Poti (V124) where I show the control surface inputs.. and the P51 refuses to behave
https://www.dropbox.com/s/u9eveurzfk...el_Problem.f4v

All controls settings are Linear and 100% plus no 'fore'play

[K_Freddie]

I can get DCS to so called 'snap roll' if I apply aileron in the direction of roll... but this is not the 'official snap roll', which is the application of elevator and rudder only - as the guy does in that AG video.
He then applies momentary and then opposite aileron, after the roll, to help bring the wing out of stall.

Maybe a full fuselage tank might help, but this is not the idea .

[K_Freddie]

Now if TD can take a sample out of DCS modelling (and extras) and give it the same feel... which I feel would not be too difficult, coz AFAIK, Oleg's FM consisted of a core FM with plugin parameters for aircraft - it could trounce DCS completely. Maybe I can help - please

DCS cannot yet compete with IL2 (forget CLOD) for overall WW2 aircraft.. and I reckon they probably never will.. IL2 has a massive core base and requires a few quirks here and there to bring it into the next/future generation sims.

[T}{OR]

Originally Posted by K_Freddie

I started with DCS123, and it's nearly the same in DCS124, with the exception that there seems to be more rudder response - I haven't changed my stick settings. Same for IL2 V411 and V412.

DCS is definitely the better feel sim. I was actually very ham-fisted with the DCS stick/rudder inputs.. but to no avail.

Here's an external view of Free flight over Poti (V124) where I show the control surface inputs.. and the P51 refuses to behave
https://www.dropbox.com/s/u9eveurzfk...el_Problem.f4v

All controls settings are Linear and 100% plus no 'fore'play

I successfully replicated the same maneuver both in DCS and IL2. Almost exactly the same behavior as in the video you linked. How far this is away from the real thing, I can not comment as I haven't flown one.

As for snap rolling - also nearly identical performance.

TBH, my controls are 100 in IL2, and fine tined with a slight exponential curve (Curvature = 15) in DCS on all 3 main control inputs (ailerons, elevators and rudder).

[K_Freddie]

This is what I get... I'm using buttons for rudders (0 or 100%) at the moment, but it shouldn't have any adverse effects on the idea.

1) First 2 rolls are with elevator, rudder and ailerons into the roll direction = No problem here.
2) Next 2 'rolls' are with crossing the controls = there's a problem here.
3) Last 2 'rolls are with only rudder and elevator (the snap roll) = No go to the right, I was lucky to get it to vaguely roll to the left, but no snap ?

Unless I'm very stupid (shhhh!! ) this is not right. Even the stable Tomahawk flipped over when the controls were 'forced'.

Ignore the prang into the building - I forgot which button the brakes were on

[Gaston]

Originally Posted by Treetop64

I'd be interested in seeing the MoI numbers, as well as all the other parameters.

At any rate, from the flight modelling point of view, I think the very point of "realism" in IL-2 isn't so much about getting the absolute correct results for a particular aircraft, as much as it is about getting the relative performance between all the aircraft in the game within an acceptable bracket.

I think Il-2 is a long, long way from "relative aircraft performance" accuracy, like all simulation games, and should instead just be content to offer some realism on "general aircraft behaviour", which I'm sure it does a fair job of...

If you want a closer approximation of how the different aircraft models actually compared in combat (if any combat pilot account is to be believed -as opposed to test pilot account-, you should substitute aircraft names to get a closer approximation of how they ranked in actual horizontal (and sometimes vertical) combat maneuvers... Unfortunately, this would still leave you with mostly inaccurate vertical performance, which are usually closer as is, so I tried to get them as close as possible taking that into account...:

Spitfire Mk V-: Fly as if a P-47D Razorback

FW-190A-4: Fly as if a Spitfire Mk V

Spitfire L.F. Mk IX: Fly as if a P-47M

FW-190A-5: Fly as if a Spitfire F. Mk IX

Spitfire Mk XIV: Fly as if a P-47M

FW-190A-8: Fly as if Spitfire F. Mk IX, or maybe a Ki-100.

P-47D Razorback: Fly as if an early Spitfire F. M IX

P-47D Bubbletop: Fly as if a Me-109G-2 with Gondolas

Me-109G-6: Fly as if a FW-190D-9

P-51D Mustang: Fly as is maybe...

FW-190D-D9: Fly as if a P-47D Bubbletop

Ki-84 Frank: Fly as if a FW-190D

Ki-100: Fly as if an A6M5, or A6M8 if possible...

The wonderful thing is most Spitfire virtual modelling does seem to overstate wildly the Spitfire's roll rate, so these virtual Spitfires are a fairly good impression of what a FW-190A was actually like in real-life, minus the Spitfire's better climb rate: The Mk V is probably the closest on that account...

And yes the Ki-84 was much faster, but a complete anvil compared to the Ki-100 (and an anvil even compared to the P-47N)...: The Japanese did extensive comparisons with both types, and found one lone Ki-100 could take on 3 Ki-84s and win, then switch pilots and do the same again...

Yes, there is a "kink" in the flight physics somewhere... And we never bothered to find out what it was... You have to remember these specific types of low-wing stressed-skin single-engine aircrafts were truly "active" for under two decades, "football wars" notwithstanding...

Gaston