Acceleration comparisons

[horseback]

Originally Posted by majorfailure

I would bet there has not been a large change in the P-40s FM.
It has all to do with Full throttle height - and the way the power curves get due to it. P-40M very low will have an advantage, then its steeper power curve (higher FTH) crosses the P-40Es, and in medium height it is slower than the P-40E. But the P-40E maxes out at ~3000m, and then the power cirve goes backwards again, and crosses the P-40Ms curve again, so at alts of around 4k the P-40M should be faster again.

No, there definitely were some changes--the Hawk 81A is quite a bit faster in the 4.12 patch at 100m, and so is the P-40M, although I must have tossed the chart I did on it before the new patch came out. Fortunately, I kept my 'original' 4.11.1 version and reloaded the 4.12 version on a different drive from the original DVD and patched up from scratch, so I re-flew the 4.11.1 P-40M and re-recorded the intervals.

Attached is the chart I created just for you; there is very little, if any, change in the P-40E from 4.11.1, but the P-40M is a good deal faster, as one would expect from an upgraded engine and better (full wartime) production quality.

cheers

horseback

[SadoMarxist]

As far as the acceleration comparisons go, there was a nutter on WoP forums a few years ago who claimed that the ability of an aircraft to accelerate can be expressed and evaluated by something called Specific Excess Power (Ps). Furthermore, said person was audacious enough to claim that the information necessary to compare the acceleration performance of one airplane relative to another is contained within the IL- 2 Compare diagrams. I can't find the thread now, but never mind, it was long and boring anyway. However, I was able to find a document which explains Ps:

http://www.aviation.org.uk/docs/flig...-FTM108/c5.pdf

Regarding the trim characteristics of a certain airplane, this will have to do with stability and control. While contributing factors of the control surfaces can be modeled, the peculiarities of inception mechanisms (sticks, yokes, pedals, hydraulic augmentators, balance weights,... ) scarcely can be. Luckily, one can assign trim to axis and then use IL- 2 Joy proggie to adjust the response curve of the trimmer to one's liking .

[horseback]

Regarding the trim characteristics of a certain airplane, this will have to do with stability and control. While contributing factors of the control surfaces can be modeled, the peculiarities of inception mechanisms (sticks, yokes, pedals, hydraulic augmentators, balance weights,... ) scarcely can be. Luckily, one can assign trim to axis and than use IL- 2 Joy proggie to adjust the response curve of the trimmer to one's liking .

While I'm all for maths and technobabble in their proper places, I feel that in a forum like this, they usually do far more to obscure than they do to explain.

As I've pointed out, the trim issue seems to be confined to a relatively small group of aircraft, and I've done extensive experimenting over the eleven years I've been playing this game and its predecessor with trimming methods and controllers. Button trim usually is the more accurate of the two, since it is input in quite tiny increments (about 160 clicks from one extreme to the other), and on most aircraft it is more than adequate, in the sense that you can get to a 'zone' where one more click up or down makes no discernible difference. However, on some aircraft the transition from one click down or one click up always seems to be excessive at any throttle/pp setting; you will either be climbing or losing altitude, in a slight skid or slide or rolling to one side or the other unless you hold your stick absolutely rock steady precisely at angle X/Y. On these same aircraft, I also notice that you need to trim precisely for very small changes in speed and power, or the penalties in performance and speed loss can be severe.

When it consistently happens with aircraft I know were easily trimmed for hands off or at least reasonably stable flight, I have to wonder why, when so many other aircraft described by contemporary pilots act as described in the game (or considerably better). If the programmers can simulate the trimming characteristics of plane A (the P-40 series) from records 70 years old, then how is it that planes B, C and D need to be adjusted at least 150% more often than the acknowledged worst trimming aircraft in the US inventory?

cheers

horseback

[gaunt1]

Originally Posted by SadoMarxist

As far as the acceleration comparisons go, there was a nutter on WoP forums a few years ago who claimed that the ability of an aircraft to accelerate can be expressed and evaluated by something called Specific Excess Power (Ps). Furthermore, said person was audacious enough to claim that the information necessary to compare the acceleration performance of one airplane relative to another is contained within the IL- 2 Compare diagrams. I can't find the thread now, but never mind, it was long and boring anyway. However, I was able to find a document which explains Ps:

http://www.aviation.org.uk/docs/flig...-FTM108/c5.pdf

Im not good at maths at all, so how is that working? Acceleration is proportional to excess power, or exponential?
Im asking this because something is fishy with the acceleration of some aircraft, primarily the La-5FN. Compared to La-5F, it has ~12% more powerful engine, slightly lower weight, and about the same drag. But acceleration is far-far better. So either the La-5F is porked, or the La-5FN is an UFO in this regard. I suspect the latter. Because if we take the factors above into account, I think it should have only about 15-20% better acceleration, for example 270 to 500 should take about 44-48 seconds, not 36.

As far as I know VK-105 Yaks all have very good FM, so LaGG-3 S66 is also fishy, because its slightly better than even the Yak-1b. I doubt that its OK.

Correct me if Im wrong.

[horseback]

Im not good at maths at all, so how is that working? Acceleration is proportional to excess power, or exponential?
Im asking this because something is fishy with the acceleration of some aircraft, primarily the La-5FN. Compared to La-5F, it has ~12% more powerful engine, slightly lower weight, and about the same drag. But acceleration is far-far better. So either the La-5F is porked, or the La-5FN is an UFO in this regard. I suspect the latter. Because if we take the factors above into account, I think it should have only about 15-20% better acceleration, for example 270 to 500 should take about 44-48 seconds, not 36.

As far as I know VK-105 Yaks all have very good FM, so LaGG-3 S66 is also fishy, because its slightly better than even the Yak-1b. I doubt that its OK.

Correct me if Im wrong.

First thing you have to recognize is that my test results have to be expressed as whole seconds; I use the track time shown in the lower right hand corner of the screen, and stop the track at every listed interval (270-350, 350 to 370 and every 10kph after that). While I try to average the interval times of three or most often, four runs, I also note the altitude changes and 'weight' the average accordingly.

Since no one can fly absolutely level acceleration courses in cockpit and in real time (I'm trying to duplicate what the average player could do), no two 'runs' are the same, especially if there is a wide separation between the variometer and the altimeter (the artificial horizon can also be a factor); you can only focus on one thing at a time, and an exceptionally level run will have variations of 'only' 15-20 meters (that's thirty to sixty five feet) over the course of the full run. Many times, I've had changes in altitude of 30 meters or more (up or down in almost even proportion) in an interval lasting less than 4 seconds.

What I have noticed is that some aircraft will literally 'blast' through the early intervals at almost exactly the same times even when there are rather large climbs or losses of altitude (and even if they are considered to be on the 'draggy' side), and some others are affected to varying degrees by varying altitude; I believe that the aircraft that generally have powerful engines tend to do the 'blasting', while the ones with lower power to weight ratios tend to require a very level flight in order to achieve their best times.

Over the total course, aircraft with the weaker engines tend to slow more quickly as drag (which increases exponentially with speed) exerts its influence, even when they are very clean aerodynamically. The Zero series is a good example of this; initially, the engine is able to overcome the inertia of flying at 270kph and pick up speed quickly because the aircraft is so light compared to its fairly large size/wing area, but as drag increases, its acceleration bogs down rather quickly.

Heavier (or compact, in the case of the FW 190A series) but still aerodynamically clean aircraft like the Mustang with a good power to weight ratio will not only 'blast' through the early stages, but the kinetic energy of their greater weight carries them through those sudden little climbs or periods where the trim is out of whack (skids and sideslips) with less of a penalty. At the far ends of a run, I find that staying level and in good trim becomes more critical with the 'heavies'.

If the aircraft is both heavy and underpowered...

As for the 'proportional' aspect of the La-5F vs the La-5FN, my understanding is that the FN had an extra 200 hp at full boost over the F, and that makes it significantly 'quicker' when full boost is applied, a maximum of two minutes in RL, if I remember correctly. On a 3360kg (7400 lb) aircraft, that will be felt. I would expect that as production progressed, there would have been incremental improvements in finish and quality as well, and there was still room for the La-7 with essentially the same powerplant as the La-5FN to 'clean up' and demonstrate even better performance.

As for the LaGG, remember that this was the final version of the aircraft that had been improved quite a bit over the early war versions, and the game assumes it to be in ideal condition, not at the normal RL condition that it apparently came to the Front in. With the same engine & a bit more weight than the Yak-1b, it should be fairly close.

cheers

horseback

[MaxGunz]

Just what planes always button trim with no left over up or down? I've never gotten that except through luck or throttle adjustment.

I have always found the 109's susceptible to nose bob after slowing down even a little.

the peculiarities of inception mechanisms (sticks, yokes, pedals, hydraulic augmentators, balance weights,... ) scarcely can be.

(LOL, that's above average 8th grade level. You lost half the readers.)

Those who don't figure out the implications of the quote above, implications which have been presented over and over for more than 10 years now are still trying to find answers elsewhere. For them it's endless 'persecution' and WTF.

P-51 could be more stable. Move the CoG forward. Then listen to the whines that say stick force is too heavy, I have a book that says so.

[MaxGunz]

Originally Posted by SadoMarxist

As far as the acceleration comparisons go, there was a nutter on WoP forums a few years ago who claimed that the ability of an aircraft to accelerate can be expressed and evaluated by something called Specific Excess Power (Ps). Furthermore, said person was audacious enough to claim that the information necessary to compare the acceleration performance of one airplane relative to another is contained within the IL- 2 Compare diagrams. I can't find the thread now, but never mind, it was long and boring anyway. However, I was able to find a document which explains Ps:

http://www.aviation.org.uk/docs/flig...-FTM108/c5.pdf

Anybody can know that science is just opinion at best and lies straight from the Pits of Hell the rest of the time! Right?

LOL, thanks for the comedy! I couldn't resist adding a line!

IL2Compare is an approximation you can beat at very low speeds, but good for comparison purposes if not all in-game results. One other shortcoming is the single 1000m alt turn curves given. It's possible to look at the climb charts to get an idea of how one plane may gain or lose at some other alt but that's guesswork.

A set of Ps curves for many altitudes for every plane would be a real boon but the hard part is cleanly flying enough stretches to chart even one plane. With NTRKs of flight that conforms (full power always: steady alt straight and level or straight climb at steady speed, check for what else I probably forgot) it should be possible to make charts of some value if not perfect.

IMO steady TAS climb would serve best since it covers multiple altitudes per run.

Last time I was up for it I was asking about using segments of data where for X seconds the necessary variables were steady, and how long X seconds would need to be? That way data might be extracted from tests that are not perfect the whole way. X should not be small is as far as I got.

Anyhow I think that a good set of Ps curves would tell more about the game than we have so far.

[FC99]

Originally Posted by gaunt1

Im not good at maths at all, so how is that working? Acceleration is proportional to excess power, or exponential?
Im asking this because something is fishy with the acceleration of some aircraft, primarily the La-5FN. Compared to La-5F, it has ~12% more powerful engine, slightly lower weight, and about the same drag. But acceleration is far-far better. So either the La-5F is porked, or the La-5FN is an UFO in this regard. I suspect the latter. Because if we take the factors above into account, I think it should have only about 15-20% better acceleration, for example 270 to 500 should take about 44-48 seconds, not 36.

Ps=(VT-VD)/W [m/s]
VT(La5FN) > VT(La5F)  -  more powerful engine -> more thrust
DT(La5FN) < DT(La5F)  -  cleaner airframe -> less drag
W(La5FN) < W(La5F)    -  La5FN is lighter 


Ps=dEh/dt
Eh=h+V^2/2g [m]   - Energy height

From that you can calculate acceleration.

All of the above combine = big difference in performance
When you know specific excess power, you can calculate climb and acceleration performance. Or you can go backward and measure acceleration(Like horseback is doing) and calculate Ps and rate of climb from it.

[horseback]

Originally Posted by MaxGunz

Just what planes always button trim with no left over up or down? I've never gotten that except through luck or throttle adjustment.

I have always found the 109's susceptible to nose bob after slowing down even a little.

(LOL, that's above average 8th grade level. You lost half the readers.)

Those who don't figure out the implications of the quote above, implications which have been presented over and over for more than 10 years now are still trying to find answers elsewhere. For them it's endless 'persecution' and WTF.

P-51 could be more stable. Move the CoG forward. Then listen to the whines that say stick force is too heavy, I have a book that says so.

The key word in my description is excessive. Most aircraft have an easily determined 'sweet spot' where just holding your hand on the stick will keep the aircraft level and steady (not banking to one side or the other). If you take your hand off the stick, the change will be gradual over two or three seconds, maybe five to ten degrees of roll (and it will usually be random--left one time, right the next). "Nose bob" is one thing and most aircraft have it to a greater or lesser degree. Usually the greater the nose bob, the more the aircraft is out of trim; if you are way out of trim, the bob becomes a yo-yo. If you need to constantly re-trim for the slightest change in state (nose up, nose down, bank, 5-10 KPH gain or loss of speed, 5% more or less throttle or prop pitch), you're screwed.

With the Spitfire (any Spit/Seafire) in this game, there is a tendency to roll right at all speeds; it has no aileron trim and the tendency and amount of pressure to the left is the same throughout its speed range. Take your hand off the stick and it will roll 15 - 20 degrees in about two seconds. The P-47 has the same tendency, except if you add one click of left aileron trim, it has the tendency to roll left at the same strength: about 15-20 degrees in a couple of seconds. I call that 'excessive'.

I also wonder where it comes from, because in over fifty years of reading and asking former and current pilots of these types (say "Spitfire" to Bob Hoover, and you would get twenty minutes contrasting and comparing it to the Mustang, P-39 and P-40; unfortunately, when I asked him the cassette recorder hadn't been invented yet) about every little thing, and no one ever said that there was always this little bit of pressure to roll right. Very short 'throw' on the elevators vs a full arc on the ailerons, yes. Tendency to quickly overheat on the ground, yes. No incipient roll to the right was ever mentioned in print or verbally.

With the Mustang (and to a greater degree, the Hellcat, Corsair and P-47 in that order), you cannot achieve the state of consistent trim I described above; there is always that bit of pressure against the stick and one click up means that you are fighting a climb and one click down means that you're in a shallow dive (and either case means that the rudder is going to need a tiny bit of pressure one way or the other and it will inevitably be too much and the 'ball' will shoot across the T&B indicator). You can do it with the Airacobras and the P-40s, but not the later, more sophisticated designs, which were all described as equal or (much) better in this regard.

The in-game P-38 (like the Mustang and the others) needs constant elevator trim adjustment every 10-15 kph of of speed up or down, contrary to every description I've read or heard of the real thing. As I've pointed out, WWII era pilots all carped constantly about the fact that the P-40 series all needed to be trimmed for speed changes of as little as 10 miles an hour (16 kph). That is about four miles an hour more than you need to trim the in-game P-51 for.

Excessive.

You like to go on about game controller joysticks vs the extended 'real things', but how is it that the 109 and the so many other aircraft modeled in the game are capable of precise flight with easy trimmability but this one group of aircraft all described as equal or superior (by both sides) in this respect cannot be modeled to be equally capable of that precision without an inhuman awareness of trim state (that is not accurately depicted by the instruments' display) and an ability to input micro units of trim at precisely the right time?

It is not just a matter of one aircraft's center of gravity or people not having their joysticks adjusted properly. There is a basic error at work here. It could be mathematical or it could be a personal prejudice. Something is stacking the deck against these aircraft.

cheers

horseback

[MaxGunz]

I "trim" the roll in the Spits with rudder and zero slip at the same time. The cause is propwash, the fix is rudder.

As to why the toy/short sticks matter is because every little bit of motion is magnified on that short-short lever without the balance of forces a real stick has. We have centering springs always pull their way. This is a major disconnect from reality.

IRL it's not a big problem to hold the stick against slight force. You know the moment that changes on your fingers. But try trimming for hands off and change your speed 10 mph. It won't be the same as IL2, the real stick will move a tiny bit. That's feedback we don't get, part of the disconnect.

Another problem is there is no weight distribution in IL2 models. It's all calculated from center. The heavier planes may suffer because of that. I don't know what to say about power in IL2 FM but that it has to line up with weight somewhere in there and IIRC there was some problem with differential power in turning with the P-38 that got down to a modeling shortcut -- perhaps thrust gets averaged and applied on center, FatCat can probably tell.

I'd rather not focus on the small cracks in the model except to find ways around them. I get the feeling that you would get so much out of a full length stick setup that you'd rename your firstborn or something. Think of the difference between a twisty stick and a good set of pedals then take that into 2 more axes.