Moderators: wdolson, Don Bowen, mogami
We do not have the option of different maneuverability at different altitude for the same plane. We do not have the option of selecting tactics either. We need some reasonable compromises - averages if you will - or we just have nothing at all.
ORIGINAL: herwin
ORIGINAL: RevRick
But maneuverability is not just a turn rate or radius,
The point is that maximum sustained turn rate is at a slow speed. Note that a turn is accelerated movement. Power = mav, so you need low m or v to get high a. A plane flying that slow may be hard to shoot down, but was also a sitting duck against diving attacks. I studied this in some detail in bats versus bugs. Bugs have quick reaction times and high turn rates, but lack the power to get out of the way if being tail-chased in a dive.
REPLY: Not so. In Vietnam - early days - the enemy preferred the MiG-17 - for its superiority in the "horizontal" - turn rate. Our first solution was to use the A-1 Skyraider - which both had guns (most our jets didn't) and the ability to turn better. We could not use diving tactics with most our planes - but the F-8s could - but it was not thought to be effective.
it's also roll rate, dive speed and a host of other things.
Climb rate is almost purely power loading, since altitude is potential energy. Top speed is power loading versus drag, here speed is kinetic energy and you want a high wing loading, since a low wing loading generates a lot of drag. Roll rate is rotational inertia--how compact the airframe is and whether the engine is in-line or rotary. Dive speed is drag and, to some extent, power loading, since it involves the conversion of altitude to speed. I designed a fighter game about 20 years ago, and the primary factor that the player had to manage was energy.
REPLY: The first clause is exactly right - and while it should have been obvious to me - it was not. It is why Gary's algorithm is not that bad. ROC is giving us power loading - and power loading has a lot to do with maneuverability.
Just taking climb rate and top speed into account does not work. I just read some USAAC reports about the P-39D which were copies of the RAF reports that showed the P-39 more maneuverable at low altitude than the Spitfire. But, at higher altitudes, it lost power so much that it was easily surpassed by the Spitfires speed. Which leads to the propensity for the game to have AI bombers attack at 10K feet. At that altitude, the P-39D ought to chew up bombing raids and give the A6Ms a tussle as well. Maybe we ought to have Deck/Low/Medium/High altitude Maneuverabilty ratings and attack settings. Maneuverability ratings based more on turn rate, role rate, and critical altitude as well as speed and climb rates.
Deck (torpedo/strafing/skip bombing) 100-1000 feet
Low (ground/fighter bomber/dive bomber attacks) 1000-5000
Medium 6000-16000
High 17000 and up.
or some such. Maybe CAP could not intercept more than two levels lower or one level higher, which would lead most people to set CAP medium, which was where most of it was anyway
ORIGINAL: ChezDaJez
We do not have the option of different maneuverability at different altitude for the same plane. We do not have the option of selecting tactics either. We need some reasonable compromises - averages if you will - or we just have nothing at all.
Incorrect. The P-39 is penalized maneuver points the higher it flies in WitP.
Chez
ORIGINAL: Sardaukar
One point to clarify with "maneuverability" (as others have already done) is for example combats that Spitfire Vs had with Fw-190As. While former was a lot better turning fighter, Fw-190 had way better roll-rate. For similar period aircraft, that and other characteristics made Fw-190 superior in combat versus Spitfire V..which pilots called "clipped, cropped and clapped"..[:D]. Those disadvantage were adressed in Spitfire IX.
You may have missed this - there are two penalized slots - but we fixed this. Neither P-39 nor P-400 are penalized by code in RHS. We put planes in those slots that don't go high enough to get the penalty! We rate planes PURELY on their staticstics - and in our case that INCLUDES altitude penalties - but it is the SAME penalty for all planes - not a special one.
ORIGINAL: el cid again
Unless you can show how to use something, I am unable to use it.
The P-47 for example may have a tremendous roll rate but is this counterbalanced by the size of the target profile presented...
ORIGINAL: Nemo121
Well it is germane since in discussing how to model something most authorities would agree that one must first decide on an ideal high-fidelity representation and then, unfortunately, pare back from there as computing power, research resources and source material availability dictate.
REPLY: OK - in principle I agree. However, in practice I want to use some part of the EXISTING data set to generate the maneuverability value. Anything - absolutely anything else - requires a massive amount of manpower in research. It is I admit worse for RHS - since we have more plane types - and so daunting. Worse, since I do plane research, and have a background in doing this professionally, I realize that to do this with respect to a sort of data NOT included in standard materials, is wholly impractical: we will be able to get this data for a number of more famous planes, but certainly not for rare ones and those that never became operational - and likely in between lots of cases where a plane simply does not have that information anywhere we can find it. I MIGHT be able to derive it - but in that case we are likely to end up right where we started: we are letting ROC and speed be the sources we derive from - and the equation 20% ROC and 80% speed seems to match remarkably well my models (where 42 data types are used). IF the data were both easy to get and apply - and is certainly better - I would love to have it. I am skeptical we can get all of it. I am skeptical we can calculate it in the other cases in a way everyone agrees is fair. And it looks a lot like the work involved would be massive (measured in man-months) for a marginal return (less than 5% improvement at a seat of the pants guess). This latter because checking a few cases of the crude model we use against sophisticated models shows amazing similarities in the ratio of values for a composite maneuverability rating. We have something that is a lot better than you think. Check it out.
So, is roll rate important? Absolutely, it was a key component in why the DR I and Sopwith Camel were such effective dogfighters and why the FW-190, when flown by an expert pilot, could out-perform many Allied types. The reason being that roll rate did play one of the main parts in determining how quickly the plane could TRANSITION between specific tactics ( left turn, 180 degree barrel roll, split S etc) and, as such, allowed expert pilots to transition through OODA loops more quickly than their opponents thus gaining a decisive temporal advantage in the OODA loop.
REPLY: I do not disagree in principle. I am not sure it matters. We do not have roll rate data for a single type in the present data set. I know of no single source to get roll rate data to a uniform standard. In fact I know of no combination of sources that is likely to produce this data for even a majority of planes in the set. I suspect there is no way to get it for every last type: I will give you 100:1 odds we must estimate it for at least some planes and 10:1 odds we have to estimate it for a major fraction (quite likely a majority) of planes of interest. Please address this in a way that says: I have scholarly data on virtually all these planes that needs no research whatever - and I will jump on your bandwagon.
QUALIFIER: roll rate matters not a whit in many tactical situations - so we need to come up with a way to figure what portion of the time it matters - how to weight the data? I am skeptical we can create a consensus on the relative import of speed, ROC, roll rate, turn rate, dive rate, etc - but I am willing to join and adopt any such consensus - and to participate in a process to find it - which this might be the start of.
E.g. if your opponent is relying on you to take the same time as him to enact the counter to his manoeuvre but you can actually transition into that manoeuvre 1 - 2 seconds quicker than he has allowed for then you've just gained those 1 to 2 seconds to put yourself in a better position at the end of the manoeuvre either to take a shot or to just startle your opponent and force him into a manoeuvre which you can transition into even more quickly ( thus continually shaping his reactions and forcing each of them to end up with him in a poorer and poorer position until such time as you get into a killing position).
Even in that slowest moving of intellectual pursuits one can see the OODA loop in action so anything which gives an advantage in the OODA loop ( as roll rate does) will give advantage in air combat.
Obviously though if it is not possible to either research or calculate roll rate then it cannot be used. OTOH my understanding of it is that it can be calculated by taking into account engine torque, wing loading and control surface area as what you are essentially measuring is how quickly torque and air hitting the control surfaces can "flip" an object weighing x kilos etc etc through y degrees. Of course researching the necessary calculations etc and figuring out how to combine them into such a broad term as "manoeuverability" is probably impossible. It is a pity that fighters which give an OODA advantage ( such as many of the Japanese light fighters) don't have this represented... although, to be fair, some unexpected US heavy fighters would also benefit.
ORIGINAL: ChezDaJez
You may have missed this - there are two penalized slots - but we fixed this. Neither P-39 nor P-400 are penalized by code in RHS. We put planes in those slots that don't go high enough to get the penalty! We rate planes PURELY on their staticstics - and in our case that INCLUDES altitude penalties - but it is the SAME penalty for all planes - not a special one.
You complain that the game engine won't allow you to model aircraft performance at various altitudes and then you remove what little performance penalty was incorporated. In effect you are saying that the P-39/400 will now perform as well at 20000 feet as it does at 5000 feet. Go figure.
REPLY: MANY planes in the game had performance issues at various altitudes. We discussed this and a consensus formed about how to address it. It was agreed that it was wrong to penalize these two particular planes in this way.
We do model the altitude issue for them - in a different way. See the discussion of "operating altitude." In effect these planes won't be able to reach as high as other fighters do - and in fact not even as great a fraction of their service ceiling as fighters with turbosupercharged engines or jet engines. It is a simple solution - but one we can apply to all planes directly proportional to their data and in a uniform way. I don't know why you think it is right to penalize a P-39 but not a Jake, for example? We came up with a system that covers all types of planes - and gives relative disadvantages and advantages on a uniform basis - and then just made the special code irrelevant. In general this is much better design practice.
Just like the nerfing of the A6M2's range so that it can no longer fly one of its most important (and historical) missions of the war, that of escorting Bettys from Rabaul to Guadalcanal.
REPLY: Excuse me? What is this about? We did NOT "nerf" (that is a word?) the range of the A6M2. It is EXACTLY the same as it was in CHS. Further - and note this carefully - I did this work under CHS supervision - and when asked to increase the range of the A6M2 - I found the difference was just 3 minutes of flight time each way - or 9 minutes total for a ferry range flight. I recommended to my supervisor we allow it - and was soundly opposed - on the basis that if we could "nerf" numbers in favor of one plane we would have to do it for many planes. The resultant decision - which RHS has simply honored after it came to exist - was that we would NOT "nerf" data - even if it was very close.
You are confusing a map issue with a range issue. The map is distorted. The range data is right. All planes suffer or benefit according to the specific points the map has a specific error in - and these errors are as great as 50% at some points - and often 20%. You can always look at some "historical mission" and find some case where the map is not right - and then get upset with "the plane cannot fly its historical mission." But don't think it is right to modify the range data - you would end up with planes flying too far most of the time. Why give a Zero a range advantage and not every other plane? By the same fraction of course.
Frankly I hope to fix this - and am working on it too - by fixing the map - going over to a spherical projection system. I want a real fix - not a pretend one. It is not right to make the range of one plane more - and it is also not right to make the average range for all planes more. It would be better to run all planes at 5 or 10 % less than max - because no sane person flies to 100% of range - and those that do lose the plane sometimes. .
Another reason to not use RHS. No thanks.
Chez
ORIGINAL: treespider
ORIGINAL: el cid again
Unless you can show how to use something, I am unable to use it.
This is not directed at El cid but to the audience in general prompted by Cids above statement
Two points:
1. Do any of you even know what Manueverability is supposed to represent in the game? A layman may look at the word Maneuverability and think that it is supposed to represent turn rate, roll rate and a variety of other rates. However what if it also represents things like target size, profile, or any number of other intangible items??? The P-47 for example may have a tremendous roll rate but is this counterbalanced by the size of the target profile presented...which may or may not be factored into maneuverability. So now once you figure out what Maneuverability is supposed to represent on to point # 2
2. How does the game even utilize it? As Cid stated above you cannot use it unless you know how to use it.
ORIGINAL: ChezDaJez
The P-47 for example may have a tremendous roll rate but is this counterbalanced by the size of the target profile presented...
The Corsair was the largest single engine fighter in the war and its maneuver rating is excellent so I doybt that size is a part of it.
Chez
ORIGINAL: ChezDaJez
I do not know how GG defined what maneuver is but it is not maximum speed nor is it ROC related. If speed and ROC were the primary factors in maneuverability then the X-15, Mig-25, SR-71 and others would all be extremely maneuverable. That is not the case.
Maneuver is also not a straight line. It is, simply put, the ability to change direction. Speed certainly influences maneuver but it is not a component of it except when a maneuver may require a minimum amount of speed such as when performing a loop.
For the sake of simplicity I will confine my comments to single-engine fighters.
Roll rate, initial turn rate, sustained turn rate, acceleration, zoom climb and initial dive acceleration are all part of what I believe constitutes "maneuverability." These factors are heavily influenced by the aircraft design and include factors such as weight, drag coefficient, horsepower, # of propeller blades, length, wingspan, and wing area along with many others. Wing loading and wing aspect ratios can be computed if necessary. And all of this data is readily available in books and some is on the web.
Again an aircraft’s maximum speed and sustained rate of climb are not generally maneuverability factors but aircraft speed at the beginning of the maneuver does influence the ability to complete or prevent a maneuver.
Roll rate is the ability of the aircraft to revolve around its longitudinal axis. As the ability to roll is not a true maneuver, roll rates are not very important in and of themselves. The importance of roll rate however becomes clear as the aircraft needs to roll in order to begin most maneuvers except a zoom climb.
Roll values for a 360 degree roll (in seconds) for selected aircraft at 200 knots and 10K feet are as follows:
A6M2: 5.9 seconds
A6M3: 4.3 seconds
Ki-43-II: 5.7 seconds
F4F-4: 5.5 seconds
F6F: 5.4 seconds
F4U-1D: 3.7 seconds
P-40E: 4.6 seconds
As you can see, the Corsair is the king in rolling ability but the A6M3 and the P-40E aren’t far behind far behind. However as speed increases, the time required for Japanese fighters to complete a roll climbs tremendously. For example, an A6M2 at 300 knots requires 14.8 seconds to roll whereas a Corsair requires 2.5 seconds! An A6M2 at 330 mph requires 21.6 seconds!
The initial turn rate is the maximum turn rate that can be temporarily achieved at the during a turn. It will result in stalling if held to long. It is not a sustainable turn rate. Many pilots refer to this as the ability to pull your own aircraft's nose to a target.
I do not have initial turn rate values for any aircraft but I do have 180 degree turn times. In other words, what is the fastest an aircraft can complete a 180 degree turn at 250mph (The time needed to roll 90 degrees for the maneuver is factored in):
A6M2: 5.1 seconds
A6M3: 5.1 seconds
Ki-43-II: 5.2 seconds
F4F-4: 5.8 seconds
F6F: 6.8 seconds
F4U-1D: 7.3 seconds
P-40E: 6.5 seconds
This gives you an idea of what the initial turn rate might be like. As you can see, the Japanese aircraft can turn very quickly. The F4F-4 is the only US aircraft that comes close though the P-40 also could pull lead for a short time. The Hellcat and Corsair were both prone to stalling when pulled too tight too fast. You could divide the time into 180 degrees to compute a # degrees per second rate but I’m not sure it would accurately reflect their true ability.
Sustained turn rate is the ability to maintain the smallest possible turn radius without stalling. The Zero and Oscar excelled at this due to their excellent wing loading values.
The following sustained turn times apply (250mph @ 10K feet) and depict the time required for an aircraft to complete two 360 degree circles:
A6M2: 25.7 seconds
A6M3: 27 seconds
Ki-43-II: 25.4 seconds
F4F-4: 33.6 seconds
F6F: 36 seconds
F4U-1D: 49 seconds
P-40E: 25 seconds
The following data represents the sustained turning radii (in feet) for various aircraft:
A6M2: 339 feet
A6M3: 403 feet
Ki-43-II: 338 feet
F4F-4: 519 feet
F6F: 646 feet
F4U-1D: 700 feet
P-40E: 597 feet
As good as the Corsair was, the Japanese fighters would be on its tail within 2 circles if it attempted a turning fight. There was a reason pilots said “never turn with a Japanese fighter.”
Acceleration is the ability to increase speed and its rate is altitude dependent. It technically is not a component of maneuver but can be an important factor in combat nonetheless.
The following values represent the time required for an aircraft to accelerate from 150mph to 250mph:
A6M2: 55.7 seconds
A6M3: 49.6 seconds
Ki-43-II: 32.7 seconds
F4F-4: 73.1 seconds
F6F: 34.1 seconds
F4U-1D: 33.2 seconds
P-40E: 55.4 seconds
The Oscars could out accelerate the Corsair and Hellcat at low speeds. The Zero could also but the advantage was fleeting once those 2000 hp engines with their 4-bladed props got going. The Wildcat couldn’t out accelerate a seagull.
A zoom climb is simply the ability to gain as much altitude in as short a time before stalling and is typically measured in feet per minute.
A6M2: 4400 fpm
A6M3: 4900 fpm
Ki-43-II: 5100 fpm
F4F-4: 4300 fpm
F6F: 5100 fpm
F4U-1D: 5300 fpm
P-40E: 4900 fpm
I don’t have much faith in these particular values as I think they are too low, especially for the A6M2 and the F4U-1D. Plus they don’t indicate how much altitude is gained before stall occurs.
Like I said, the data is out there if you want to take the time and effort to produce an accurate product. Using maximum speed and sustained rate of climb is not going to give accurate maneuver values. I obtained this data from a variety of sources over several years. Unfortunately some of my notes failed to attribute some of the sources but the vast majority were found in Francillon’s works and quite a bit was obtained from US flight training manuals for various aircraft. Sufficient data does exist to build accurate aircraft performance models for any one who is willing to take the time to research and develop it.
Chez
Finally - I think you may be confusing reality with WITP. Maneuverability is a name given to a field - and it may or may not be a very good name for that field. But we are stuck with a very small number of fields to use with each plane. It is quite clear that speed is a major determinant in maneuverabilty as used by the model - and we cannot expect the model to work if we were to remove it entirely. It must remain a major determinant - even if we reduce it (as, in fact, I did, for RHS).
Two thoughts right off the bat, 1) Francillion does not mention Roll rate, initial turn rate, sustained turning radii, aircraft acceleration from 150mph to 250mph, or Sustained turn rate. So what reference material do you advocate to be the standard? 2) What formula are you advocationg to use with WitP "manuever" rating? With the silence of the WitP gods, one must surmise that they do not care to divulge any information. El Cid has offered a formula and an explanation, given the fact the "manuever" is somewhat of a misnomer. We might be arguing with over-analysis over a value that we do not know of with definite conviction. I am sure El Cid would consider any suggestions that would additionally improve the game play. But I do not agree your point if no other formula is offered so a comparison can be made.
ORIGINAL: ChezDaJez
Two thoughts right off the bat, 1) Francillion does not mention Roll rate, initial turn rate, sustained turning radii, aircraft acceleration from 150mph to 250mph, or Sustained turn rate. So what reference material do you advocate to be the standard? 2) What formula are you advocationg to use with WitP "manuever" rating? With the silence of the WitP gods, one must surmise that they do not care to divulge any information. El Cid has offered a formula and an explanation, given the fact the "manuever" is somewhat of a misnomer. We might be arguing with over-analysis over a value that we do not know of with definite conviction. I am sure El Cid would consider any suggestions that would additionally improve the game play. But I do not agree your point if no other formula is offered so a comparison can be made.
You are correct. Francillon doesn't give exact numbers but it soes provide a wealth of other information that can be used to determine wing loading factors, aspect ratios, and weight to H/P ratios. As I said most of my research was conducted a few years ago and I didn't log all of my sources. Francillon was one that was very helpful. Some of the Osprey books gives some data. Some came from captured aircraft flight testing. Some comes from sources like Bauer and Mikesh. Mikesh is particularly helpful. Some of it even came from magazine articles such as in Air Classics and I used that data when I had no other choice. There is no single source where you will get all of this information.
The data that I presented above is a compilation from literally dozens of sources that I have put into a spreadsheet. Much of it came from the public library, some came from books I purchased. Some of it was interpretations that said XXX aircraft could climb at 3000 fpm and the YYY aircraft was 300 fpm slower. So I took that to mean YYY aircraft could climb at 2700 fpm. A good deal also came from written interviews from fighter pilots. They were particularly helpful when I had data that contradicted other data. Some of it even came from various flight simulators and the forums that supported them. Quite a bit came from warbird forum discussions similar to the discussions we have on this forum.
Unfortunately, the data I have is not complete and I am missing chunks that adversely affect the quality of the data. Some of it I've had to extrapolate such as for the A6M3 roll rate. I had 3 or 4 different sources that listed different rates. So I averaged them. But the data is certainly a good starting point for determining COMPARATIVE maneuver ratings. It really doesn't matter whether the Corsair's roll rate was 2.5 seconds at 300 mph and that the A6M2's roll rate was 14.8 seconds so long as it is recognized that the Corsair rolled about 6 times faster at that speed.
As far as how to develop a formula for maneuverability, I don't know. I'm not a mathmatician. I do know that using maximum speed and ROC of climb values will produce a highly flawed model. Otherwise the Me-163 Komet would be the most maneuverable fighter of the war. I happen to believe roll rate is more important than turn rate. I also believe that zoom ability is more important than maximum dive speed. Dive ability has one brutal limitation... its called the ground, you can only dive so far and if your at less than 5000 feet diving ability is greatly restricted, especially in a P-40 that needed nearly that much altitude just to pull out of any dive over 350mph. Maneuverability is a very complex attribute given that it changes based upon airpeed and altitude. A P-39 is fast and maneuverable at 5000 feet. At 20000 feet it is a sitting duck. The Zero is a very fine aircraft below 250mph. Above 275, it sucks. But surely there is some way to meld the numbers into a single value that yields a reasonable representation of the aircraft being modeled.
Once that is straightened out, I have another wrench to throw into the works. And that is tactics. Team tactics are far more important than any aircraft's individual physical maneuvering ability. The Wildcat proves that. Taken one on one, the A6M2 should tear up the Wildcat. Taken 2 on 2 and the story begins to change. 4 on 4 and the Wildcat becomes the favorite. The problem with incorporating tactics into maneuverability is that it is time based. The allies used poor tactics early in the war but pioneers such as Thach helped changed that so that by 1943 everyone knew what not to do against a Zero.
BTW, the data I presented in my post above should not be taken as gospel on the whole. Most of it is correct but some of it was computed and some of it was averaged from multiple sources. Those numbers were the best that I could come up with and still reasonably defend. The main point in showing those numbers is that they accurately reflect the relative differences between aircraft more so than absolute differences. I developed them for my own personal use.
Chez
