Echo38

That's what causes it. It's a very old bug; a forum search should show up previous reports. The only workaround is to refrain from using dynamic weather. : /

I should point out that even a force-feedback joystick would not solve this problem, unless it had a hyper-strong mechanism greatly exceeding that of commercially-available FF gaming joysticks. With the forces that we're talking about at high speeds, you'd break the FF mechanism in the shoddy gaming sticks out there. Or the mechanism would stop providing force, if there was a safety of this nature. Either way, the stick would, when entering high speeds in the sim, fairly quickly stop providing force, effectively turning into a non-FF joystick whenever the high speeds came into play.

I don't doubt this at all! But, if I'm interpreting the posts aright, we have a case of two real P-51 pilots disagreeing about one of the finer points of the hydraulic system, or something like that. Not surprising, given how often pilot input diverges, especially with something like this that cannot easily be tested. Perhaps going directly to the people whose job it is to maintain that hydraulic system would be better than talking to the pilots.

Speaking of mechanics, perhaps its time we stopped asking the pilots and started asking the folks who maintain the bird. Surely someone up top is able to contact the the Fighter Collection's mechanics about this question?

This is all quite confusing to me, but the one thing that everyone seems to agree on is that deploying the P-51D's flaps at high speed (and/or leaving them deployed and overspeeding) would damage something, but it doesn't in the sim. Is this assessment accurate? Whether or not the flaps should retract or freeze in place upon overspeeding is a different, though related, question: Retract under force, but unable to deploy again due to damage? Retract under force, but able to deploy again despite damage? Freeze in place, unable to deploy or retract due to damage? Freeze in place, but able to retract if the lever is returned, despite damage? USARStarkey: why is Mr. Fahey so hostile toward DCS? While I enthusiastically welcome any criticism of a real warbird pilot, in the spirit of making the sim closer to the reality, I must say that the attitude displayed in the lines you (appear to have) quoted indicates vehement contempt, and I'm perplexed as to what makes him have such a low opinion of the sim as a whole. We all know that DCS isn't perfect, and is missing certain important elements such as engine idle behavior. However, the attitude that I got from your quotes isn't that merely that Mr. Fahey recognizes errors in the sim, but rather that he doesn't see an value in it at all as anything other than a mere game. And that's puzzling, because the sim is highly accurate in many areas and pushes boundaries that no other P.C. flight sim has approached (and has gained the respect of many real warbird pilots). What's the story? Why does he appear to be so emotionally opposed to anyone regarding DCS as an excellent P-51D P.C. simulator with some flaws?

There's nothing unrealistic about having a P-38L dogfighting an A6M. However, I'll concede the point about the Pacific; this is really the wrong thread for me to be arguing in favor of a P-38L for fighting the Me 109K ; )

Incidentally, I'd like to point out that I have no problem with the idea of matching P-38G versus A6M. My concern lies with the P-38's ability to fight its most challenging (and commonly-encountered) opponent, the Me 109. And since we have an Me 109K, the P-38G would be a poor match for that.

There comes a point where one aircraft has too many advantages over the other in critical areas. If it's something like P-38 versus Zero, the P-38 is indeed superior in general, but the superiority isn't absolute; the P-38 can't win a knife-fight (i.e. a dogfight that occurs entirely within gun range) with the Zero, all else equal. But when you get a situation where one aircraft is faster, climbs better, and turns better, then the other aircraft is screwed in anything approaching an equal situation. An extreme example would be F4F versus A6M5 Zero. The latter does virtually everything better in the dogfight, and so a team of F4F's versus a team of A6M5s will generally get stomped if the number of pilots is the same and the skill pool is roughly equal. This is shitty balance, and there's no need for it. One can have a well-balanced planeset without reducing the fidelity of the aircraft simulation, simply by choosing the most appropriately-balanced model whenever possible. In the case of the P-38L, it's also the most well-documented, so I see no reason to choose a lesser model and ensure that the P-38 is an underdog in yet another virtual depiction.

So, if the P-38 is going to be (as you claim) inferior to the Me 109K, then it makes sense to use the best P-38 model rather than a lesser one, which would exacerbate the mismatch. Since the Me 109 was the most ubiquitous opponent of the P-38, it doesn't make sense to force a mismatch by taking a less-than-the-best model P-38 to face the best model Me 109, especially if the best P-38 is already going to generally be at a slight disadvantage in dogfights against the 109. The P-38L is unquestionably the best match for an Me 109, out of all of the P-38 models, and the P-38L is also a suitable match for dogfighting against the A6M Zero (the Zero being clearly better at horizontal maneuvering, but the P-38 having the ability to extend & gain energy before closing again).

The P-38L was the definitive variant of the P-38, the one with most of the aircraft's initial problems remedied or at least minimized. The L is also the one that's most likely to be a decent match for the Me 109K and other aircraft we're to end up getting. Since DCS aircraft so far have been modelled to factory spec', I expect that our hypothetical DCS P-38L will have a War Emergency Power setting of 1600 hp. at 60" & 3000 RPM (which allowed a top speed of 420 MPH), rather than the commonly used settings of 1700+ hp. at 64" or 66" & 3200 RPM (which allowed a top speed of ~440 MPH). This means that our P-38 would be less competent versus her contemporary opponents in the sim than she generally was in reality. The P-38 was a heavy fighter; despite the excellent lifting wing & the Fowler flaps, she also relied largely on power-to-weight ratio to achieve her good slow-speed turning ability (did you know that the P-38 turned considerably better than the P-51, below 250 MPH, where the P-38 could safely use the Fowlers?). Having two engines, each offering approximately as much as the single engine of her standard opponent, the Me 109, gave the P-38 sustained turning abilities similar to those of the Me 109 (which weighed about half as much as the P-38). Remove a large chunk of the horsepower which the P-38 historically had, and she loses a critical component of that triangle (the latter being high-lift wing, Fowler flaps, and high power-to-weight ratio). With the lower horsepower that we're sure to get (judging by our P-51D), we're going to have a hard enough time without also having an inferior model selected. For this reason, the best model of P-38 should be selected to ensure the most equal match against her primary opponents; that model is the L. (It also makes sense as the other aircraft so far modelled have been very late-war models.)

Our monitors are too small for any realistic flight simulator. Our real-life FoV is roughly 170 degrees, including peripheral. Assuming you've got a "normal"-size monitor, and sit with your head two or three feet away from it (the normal situation for such a monitor and P.C. gaming), then your monitor is only occupying about 20 to 30 degrees of that 170-degree FoV. Therefore, if the zoom level of the sim is such that an object (for example, an aircraft) is displayed on your monitor at the same apparent size to your eye, as would match it if it were sitting outside your window in real life, then your FoV in the sim is, mathematically, only ~1/7 of your real-life FoV, and you'd be flying nearly blind. Try it; zoom in until your virtual hand on the stick in the virtual cockpit looks about the same size as your real-life hand on your desktop joystick, and try flying around that way without changing the zoom level. So, in order to cram the ~170 degrees of FoV that we have in real life, into the ~25 degrees that we have between our eye and the corners of our monitor, a simulator has to make the total image size smaller than reality. You can't have real-life image size and real-life FoV at the same time, when your eye-to-monitor angle is smaller than your real-life FoV. It's physically impossible. One must understand at least some of the real-life physics of cameras & lenses (zoom & FoV, in particular), in order to understand why any ultra-realistic flight simulator on the P.C. (i.e. with normal-size monitors and user-to-monitor distance) will be harder to see things in than real life. Games like IL-2 and War Thunder, which are less pure of simulators, make dots unrealistically large relative to things around them, in order to make them easier to spot (older games like IL-2 also had blander terrain, against which dots are easier to spot due to their being less background clutter); the objective is to use an unrealistic method (dot scaling) to "make up for" an unrealistic handicap (small monitors), but two wrongs don't exactly equal a right. Which doesn't mean that I believe that DCS's method is necessarily the best way to do it, by the way ... but in terms of simulation purity, it's the most accurate depiction, and our little monitors are to blame for the not-lifelike results. You can't blame the sim for modelling things accurately, when the actual problem is a result of our inadequate hardware. I do think that it would be good if we had an option for scaled-up dots, to keep both camps happy. In the end, any of the methods are going to have significant drawbacks, and the end result can never be entirely realistic. From the point of view of a gamer, it's better to have aircraft-dots larger than they should be (compared to the tree-dots around them), like in War Thunder, but from the point of view of a purist simmer, it's less inaccurate to have the aircraft rendered at their true size within the game world, like in DCS even if that means that they are (like everything else in the game world) too small on the monitor, when the FoV is set to higher than the eye-to-monitor-edges angle. The former makes distant aircraft unrealistically easy to see under some circumstances, and the latter makes things unrealistically difficult to see under some circumstances. Such is the trouble of trying to put a large picture onto a small monitor!

They look larger from default-FoV external view than from the default-FoV cockpit view because the default FoV is different between external and internal. Neither is "wrong;" there is no single true-to-life zoom level which the simulator could enforce for greater accuracy, because the real-life image size depends on your monitor size and the distance from your (real) eye to your monitor. Hence variable zoom/FoV in any P.C. flight simulator worth its salt. Last time I did the calculation, with my old monitor and chair setup, ~90% zoomed-in resulted in the real-life apparent image size. However, because of the small size of my monitor (relative to my real-life FoV), this meant that the FoV in the sim had to be ridiculously low, since being 90% zoomed-in is being 9/10s of the way to being at the minimum FoV. Again, not the sim's fault, but rather the fault of our small monitors. Moreover, ~90% zoomed-in would not be the real life apparent image size if I moved my chair further from (or closer to) my monitor.

I want a P-38L more than anything, but I must admit I'd be rather disappointed if Yo-Yo himself weren't working on it.

What kind of joystick & pedals are you using?

I agree with Fastfreddie. As someone who once topped "leaderboards" in several different flight sim/games (years ago ... I'm rusty as heck now), I suggest using the method I used. First, a basic level of competence is needed before starting "advanced" training (because you won't learn much if you never know what hit you). This is accomplished by practicing offline until you are comfortable with takeoffs, landings, and shooting down an AI fighter. Once you're at that point, the next step is to venture online and don't do what the majority of aerial combat simmers do, which is to look for easy targets (already engaged with someone else, already damaged, and/or at a much inferior beginning energy state, etc.) to backstab. This doesn't help learning much. One learns far more from difficult situations than from easy ones. The key is to, once you have the basics down, always take the most challenging fights you can find* (if you're winning more often than not, then you aren't properly challenging yourself). I myself favored duelling higher-skilled opponents, because this format allows the maximum rate of learning. Spend as little time flying around looking for combat as possible, and as much time actually in the dogfight as you can. An air-start duel mission and a ground-start mission with close (<5 miles apart) airfields and a no-"vulching" rule is ideal (each of the two is a bit better than the other at teaching different skillsets). Examine your replays afterward whenever possible, especially if you aren't sure what you did wrong (and/or what your opponent did well). Watch them from various perspectives (first-person, third-person your airplane, third-person enemy airplane) to maximize your understanding of what happened in the fight. Do these things with dedication and patience, and almost anyone can become a leaderboard-topping virtual top gun ace pilot. It may take a few years, as in my case, but I've known someone who managed to get there in little more than six months. The more hours you fly a day, the better your rate of learning; I consider two hours a day about the minimum for becoming & remaining competitive, although you can probably make do with an hour a day if you aren't interested in becoming "the best." Two hours a day is better than four hours every other day, because the skills you learn are so perishable that even short breaks result in a noticeable decay (especially at the higher levels of skill). As an example, when I was at my peak, I would virtually never accidentally stall & spin in a dogfight, but if I took a few days off, upon returning I would suffer from accidental stalls for my first ~hour or so back in. In the end, how well you do -- how good you become at virtual dogfighting -- is largely determined by how much you care -- how much you value this activity & how intent you are on becoming proficient. There are methods & tips that can help you accelerate your learning, but (as you surely already know from getting this far) there's nothing that can replace hard work and pushing yourself as hard as you can. (Although having the proper simming equipment is also required; the best craftsman can't do much with broken tools.) * This doesn't mean always picking fights where you're outnumbered and/or have a grossly inferior airplane (etc.) than your opponent's, by the way. A co-E 1-vs.-3 isn't what I call "a challenge," because it's an unwinnable situation (assuming competent enemies in good fighters), from which little is to be learned other than the basic, obvious fact that one man can't beat three without some kind of massive advantage. A proper challenge is one where the conditions are as near to equal as possible except for skill, in which your opponent's is greater than yours. This is the most effective way to learn.

Yes; I've already stated that I consider "two 30mm shells on average" to be a reasonable figure, and that's a very short burst. Sometimes you'll get lucky with one, and sometimes you'll get unlucky and need three or more. Not every detonation on the wing is going to hit a spar or fuel tank. You need to look not only at impact location, but also angle. As I indicated earlier, these were perfect conditions attempting to determine the maximum damage these rounds could do to various parts of the airplane. The angle and location of impact were carefully chosen and, from a stationary shooter to a stationary target, there was little difficulty in hitting the exact place. (By the way, the Spitfire weighed quite a bit less than a P-51, and wasn't nearly as sturdy & rugged.) In the air, however, there are many factors (even if the pilot's skill is not one of them) which can cause the damage to be reduced. Slipstream & wingtip vortexes don't merely make aiming difficult; they can also subtly affect the trajectory of the rounds themselves (I don't expect it to be much, but it's there). The target aircraft may not be straight & level and presenting you with a zero-deflection shot. As a result of these and others, the shell can "graze" and, although this may still result in a detonation of the shell, much of the explosive energy can thus be directed elsewhere rather than into the target. It can also punch through a thin part of the airplane (e.g. the trailing end of a wing) and expend most of its explosive energy after exiting, doing little more than making a three-centimeter hole. Furthermore, even on direct hits, it has been well-established that sizable chunks can be taken out of the wings of a solidly-built fighter (e.g. any of the main U.S. fighters) without impeding their ability to fly under normal conditions. They're no longer combat-worthy, but they aren't shot down, either. Aside from examples of cannon and flak hits, fighters sometimes collided with trees and telephone poles and came out on top, sometimes bringing ~foot-thick pieces of wood home, lodged in the wing. Fighter aircraft generally aren't delicate flowers whose wings readily depart them at one hard smack, even from a cannon, and I've found it very unusual to see a wing coming off of an aircraft in gun camera footage. I've seen it a few times (mostly Japanese lightweights), but very few. This is true. It's been many years since I've had the heart to browse gun camera films (the older I get, the more the reality of my simulated hobby bothers me), and I have no reason to devote hours of my time (these "little" posts have eaten enough!) working to try to prove this point to anyone. I trust Yo-Yo to do his own homework, because, so far, whenever he has devoted his attention to an issue, his level of care has been exemplary. (I believe that the reason for the current Mk. 108 damage problem is that he has been spread too thin, so to speak, to give it his full attention.) My objective here was to provide a pointer in the right direction for any who are open to doing research themselves, because the evidence is out there. I dislike seeing the myth that the Mk. 108 was a regular one-hit-kill weapon (especially regarding wings, which were more resistant to damage than flight sim/games generally give them credit for) spread around without any contrary suggestion offered, hence my little pointers.

While some aircraft were indeed taken down by a single 30mm shell, this was not something that happened constantly. As I said in my previous post, that's what happened under "perfect" conditions. Often, however, the target aircraft survived multiple hits. Even fighters could withstand several, with some luck. If our virtual fighters are regularly eating five or six with no problem, then I agree that there's likely a flaw in the simulation, but the opposite extreme (that of ~1.1 hits required on average to make a kill) would be just as much as a flaw, because that isn't how it went in reality. Even that video of the Spitfire ... three hits to the rear fuselage, and the fuel tank caught on fire, but it doesn't appear to have blasted the tail off, or anything similarly dramatic. If one browses through gun camera footage, it doesn't take long to see a fair number of cases where aircraft survive multiple hits from these cannons. On the other hand, it's quite rare to find cases where a single hit takes down an aircraft. As far as I can tell from the variety of sources I've examined, the notion, commonly propagated by flight simmers, that the Mk. 108 cannon should usually kill any fighter with one hit, is a myth. Two hits average is a more accurate estimate.

Oh, I'm afraid you're going to have a hard time getting those!

Yeah, that crater wasn't made by an aircraft exploding. Someone seems to have dropped some seriously heavy ordnance on you.

The Mediterranean is not only the least well-represented major theater in flight sims & games, it's also the most interesting to me, because of the Mediterranean Sea's having been the stage (so to speak) for the ancient history of Western civilization. It has some beautiful geography & climate, too.

Folks, the P-51D does have the option of a pilot model in first-person, you know.

The most interesting part of this to me is that real-life Great War replica operators have sought out people within the DCS community, regarding developing an aircraft for the sim. As far as I know, this is a first in the history of flight sims, that a real-life warbird pilot unrelated to the development of flight sims would be impressed enough with a P.C. flight sim to actively express interest in aiding its development. Great War airplanes aren't my dish of tea, but I think it would be great for the flight simming community if one of the experienced third-party DCS developers were to do a maximum-fidelity module for some of the renowned First World War fighters. Rise of Flight generally does a very good job of portraying the physics of flying and so on, but it doesn't model full systems management, so for those who are looking for the closest thing they can get to flying the real airplane, there's something missing. Therefore, there's still room for a "DCS standard" (i.e. max-fidelity, full engine/systems management) portrayal of Great War aircraft, and saying "no need for that, we have RoF" overlooks something important. I myself am not much interested in flying Great War, because Second World War and Korean War aircraft are where my heart is. However, there are plenty of people who would buy a max-fidelity Great War fighter module. And, happily, there appears to be an increasing number of third-party developers who are willing and able to do work of this sort.

It's important to not base one's {impressions of how these airplanes ought to handle} upon shoddy sim/games of the past. {Warbirds, IL-2, Aces High, War Thunder, and the like} did a rather poor job of accurately portraying these aircraft. I suspect that {the assumption that the FW 190 ought to roll faster and turn worse} comes from having played those outdated games, in which the characteristics of the FW 190 were surely exaggerated or otherwise misrepresented (as was proven to be the case with other aircraft in those games).

I can't comment on the accuracy of the effects in the sim, but do note that the oft-posted picture of the British fire test was an example of the "perfect hit," which is the exception and not the norm. Perusing gun camera footage indicates that one-hit kills were extremely rare. In fact, out of all of the hours of gun camera footage that I've seen, the only one-hit kills I've seen were either due to the detonation of a cannon's ammunition magazine, or the strike was on a Japanese light fighter such as the Oscar, which had relatively weak construction. Note that the horizontal stabilizer of the B-17 in the picture wasn't blown clean off, as some suggest should happen to the average fighter's wing upon getting hit with a single round. The average fighter's wing was not flimsy even compared to a B-17's horizontal stabilizer. The P-51, for example, was on the sturdy side, structurally, and I wouldn't expect its wing to fall off easier than a B-17's stabilizer (although, in a hard turn of the sort a fighter makes and a bomber doesn't, the G forces can, of course, have a catastrophic effect on the damaged wing, which a bomber wouldn't normally have a problem with, due to lower G-forces).

Nice work.