Squiffy

Well the F-106 test is a good example of that already. A couple points to keep in mind: 1) To me, "dispersion" IS the median, or already includes it as a matter of fact. Let the sim engine find the median, according to dispersion. [if needed an 85-90% circle would be just great compared to how it works now. Lots of room between 80 and 95%] 2) Mounted to the airframe or "in flight" ... don't get lost in Einstein's Special Theory of Relativity here. The gun system should function independent of the flight dynamics and fire relative to the airframe. Unless you are seriously modelling g loading and such for jamming, nothing more is needed. We already have barrel warp and that has problems too. The airframe will pitch and roll and "rope a dope" plenty all on it's own. That is already modeled and rendered via the aircraft. Errors and inaccuracies seem over modeled at times and double stamped for some reason. And of course, some weapons are just fine on other aircraft and units. Please don't bring up the "third party argument." If you allow it in game, including the inconsistencies, you endorsed it. To recap, don't get hung up on "medians", we are already dealing with "dispersion." Don't get hung up on "in flight", the sim already models that. It could be we are dealing with unintended "double calculated rendering" in the math for some aircraft, weapons, and shells.

Oops, I goofed. I copied from the stock file, pasted over by the Stalinist update process ;) Here's the one you want. And also note the blue boxes with the equation down below. This table says the the MiG-15 has a coefficient of lift of 1.1 at Mach = 0.0, apparently, when cymax < Cy. I bet we can tweak cymax to be = cya or 0.067 at Mach 0.0. This may kill the Sputnik effect. I am also looking into modelling the imploding fuel tanks issue from under pressure after half emptied.

Cool. Thank you very much NineLine. Please remember we have eye witness testimony of A2G strafing from a Hornet pilot. It may just be that there is a calibration problem or execution problem between the files and the code or the files and the engine, that sort of thing. Even though the math is not working out perfectly, tweaking it for game play, in-game realism and FUN will sell copies.

Check this thread for MiG-15 SFM edits https://forums.eagle.ru/showpost.php?p=3954635&postcount=22

Sure! Actually, the g limit may be the easiest way to adjust the dogfight challenge any further than the "average" setting in game. I adjusted the low speed stuff because of the AI energy fighting style. It may be even more accurate and supportable to edit the higher speed stuff because that's where the Sabre excelled. The fat wing on the MiG is really something when you see it up close in real life. Also the wing fences. Those things added drag too so it was not an elegant solution. It's like looking at a Hurricane wing versus a Spitfire. Trying to push that through the air becomes a problem at high speed.

Also check orange boxes for pitch agility and drag brakes. I think I edited these too.

rick are you posting these? Sorry to be dense but where can I find them?

OMG! Please ED, you can do it! No one is perfect and everyone makes mistakes. If this gets fixed on the Viper et al., DCS will rock the summer! If not, I will buy another tank of gas for my Ferrari instead ;)

Do you think this thing will by fixed on the Viper? Why or why not?

Was busy at work before I finally clicked post above... Now, I will try to be diplomatic but this kind of thing really makes one question the possibility of bias. I got hammered like many noobs for griping about things before I fully understood them when I first joined. But then I also learned just how badly things had been fouled up by so-called fixes that were not thought out. This rhetorical deflection seems like a defense mechanism and an attempt to distract with smoke and mirrors from simple failings that would otherwise be so easy to fix. "Why?" is the natural question that follows. The empirical evidence and proof with language and photos in the YouTube video shows the code data does not work correctly and does not achieve the "8 mil" spoken of by General Dynamics staff. I recently found a bug in the MiG-19 g-suit detail. It got fixed fairly quickly. It was an obvious violation of the standard set forth in the very same entry in the lua. What is that other than bias? And as far as gunnery errors go... Give us the chance to screw up wind and aim and over correction fair and square. We don't need these human errors programmed into every weapon! That's what the sim is for!

I know! I mean come on... Somebody screwed up. 0.60 = 60%, 0.06 = 6%. I think the word "mil" got warped in translation. Even so, whatever was coded in the file is not working correctly. Even the 3 barrel gun for helicopters got corrected (with the same info!). The 1000 inch bore sight panel in the video shows good grouping. The 1000 foot target showed good and similar "conical" grouping within the 14 foot square target panel, and 8 foot circle bulls-eye. Remember, and this is true for the M2,M3 machine guns, an 8 foot "dispersion" at 1000 feet is fantastic! 1000 feet, that is 0.18 or so miles. 14 feet is less than the wingspan of a MiG-15. And how many western aircraft does this effect?

Try these to start. It's a little aggressive but you will only see slight changes in game because of the uber space dynamics of the MiG's energy and climb. In the .lua find and change to: Mach_max = 0.92, -- Max speed in Mach (for AI) Ny_min = -1, -- Min G (for AI) Ny_max = 5.0, -- Max G (for AI) Ny_max_e = 5.0, -- ?? Max G (for AI) AOA_take_off = 0.17, -- AoA in take off radians (for AI) bank_angle_max = 80, -- Max bank angle (for AI) ... g_suit = 0.02, ... And the most important... especially the lower speed/mach number entries for Ai's energy fighting style. see screenie. Make changes to items in the yellow box. This is low speed pitch and roll effectiveness which I wanted to reduce. I made my edits below Mach 0.5 which should be about 300 knots or less. the Ai likes to Sputnik around stall so this will make them regret it a bit more. Try this and also edit your M2 shells in the weapons/ shell.lua to have somewhere between 875 to 899 m/s for better ballistics and longer range. Cheers! "No guts, no glory!" ;)

Here is my F-86 uber M3 barrel warp or lack thereof. There is a huge amount of room for adjustment here but nothing has been done. { name = "HeatEffectExt" , shot_heat = 1.012, barrel_k = 0.045 * 1.0, body_k = 0.025 * 1.0 }, shot heat = heat added by each round? barrel_k = barrel deflection * some multiple? body_k = bullet or weapon deflection * some multiple? An explanation of this will help a lot.

MiG-15 uses on the 23mm: { name = "HeatEffectExt" , shot_heat = 7.823, barrel_k = 0.462 * 2.7, body_k = 0.462 * 14.3 }, and on the 37mm { name = "HeatEffectExt" , shot_heat = 7.823, barrel_k = 0.462 * 2.7, body_k = 0.462 * 14.3 }, or exactly the same... as the M3 on the Sabre? What? you mean we have cannon accuracy and barrel deflection on a 50 cal barrel? How do you explain the exact same equation across 3 different weapons, 2 of which are canons and NOT high muzzle velocity, rifled machine guns? The barrel circular error option was much better and way more flexible. Or should I say was used more flexibly. You can see the changes out to 4 decimal places or more, when the cursed barrel warp equation was ham-fisted across 3 dissimilar weapons. At least the Doras different canons got full consideration in the barrel warp effort. I think it ended there. If we hear why the equations work the way they do we may learn more but up to now it looks like an oversight. :(

Do you see what I see? The Dora's MG 131 machine gun has the exact same barrel dynamics as an F-86 Sabre, How is this possible? Spitfire uses on the 303: barrel_circular_error = 0.0009, FW-190 A8 (AI) uses on 131: {name = "HeatEffectExt", shot_heat = 4.103, barrel_k = 0.462 * 3.7, body_k = 0.462 * 14.3}, Which is fine if it's an MG 131.

Dora uses: Guns = { -- LH MG 151/20 MG_151_20({ barrel_circular_error = 0.0007, plus: {name = "HeatEffectExt", shot_heat = 20.9, barrel_k = 0.462 * 16.6, body_k = 0.462 * 35.4}, AND MG 131! {name = "HeatEffectExt", shot_heat = 4.103, barrel_k = 0.462 * 3.7, body_k = 0.462 * 14.3}, Sabre: {name = "HeatEffectExt", shot_heat = 4.103, barrel_k = 0.462 * 3.7, body_k = 0.462 * 14.3},

OK P-51 uses something different: -- Guns Guns = { -- Left wing outboard M2 BrowningM2({ muzzle_pos = {-0.045, -0.410, -2.427}, effect_arg_number = 436, mixes = { {2, 2, 2, 2, 2, 2, 2, 4}, {2}, {4} }, count = 270, barrel_circular_error = 0.0024,

Possible but the fact that the exact line of code is copied into the MiG file suggests something entirely different, and along the lines of a project management failure as opposed to dogged, hard fought, super attention to detail historical research. I will check other .luas and see if I can find more.

But we paid for SFM. It's wrong, and not too difficult to correct. The MiG's climbing prowess is historical, but also not that difficult to adjust for reality's sake. Some members describe it as "flying in space" where its energy is way too high, as if gravity was not in the equation, or drag and whatnot. The SFM matrix is not hard to fix. In fact, the MiG's thick wing which made it great at climbing, made it suck at rolling and going fast. These facts are NOT optimized in the sim which again comes down to decisions and priorities. They can be fixed. I will post some of my SFM edits later. Also g limit and max bank angle can also be edited for ai. When people say "stop comparing to ai and fly online" I say bull. I paid for ai and not fixing it is a cop out. If needed, I will fix it and you can pay me for the pleasure ;)

Woah, I just saw this now. That's funny, either way the exact similarity of the two different weapons having the same barrel dynamics shows a lack of effort or a simple oversight. I think it is really hard to imagine that a canon with low muzzle velocity but big punch would have the exact same barrel dynamics of machine gun with high muzzle velocity and smaller punch. Maybe it's based on metallurgy and some kind of average? But the weapons are different and barrels must be too. And of course the rate of fire fix was not a fix but an exponential worsening because it wasn't thought out. I agree because this very aspect of the Sabre was one of it's saving graces, that kept it competitive in the new jet environment. Now we may also see that the Sabre barrel dynamics were based on something earlier still and THAT may be the culprit. I have seen the lines elsewhere, I think we are on to something.

Has anyone posted updates yet? The errors with the shape and blue color in the stock wing and stars really bugs me ;) Maybe some did appear that way but the darker blue and more balanced proportions are definitely better.

I have played with the SFMs. They are nice matrices with various data points set according to mach number. Roll rate and pitch rate were the first I tried at low speed entries for the scissor maneuver I was talking about. It can be done and can be blended across the matrix. Max g in turns, max angle of bank appear elsewhere but can be used to limit performance. It's working I am tweaking my beta install after the update. the damage effects are pretty cool with the smoke and shrapnel from good hits. Also, the Sabre must have higher M2 (M3) 50 cal shell velocity closer to 899 m/s. Extend life of round out to 10 units and maximum range to about 2000 m. Looks much more realistic and you can get hits on migs in higher speed distance shots with more g as it should be. And then when you close in like in a stall turn maneuver, you can shred them with accurate fire in less than 300 ft. Like you should with a 50 cal. The barrel heat dynamics are an exact copy of the Mig 15 cannon barrel. How is that possible? I have edited that down but they are different guns with different shells and the iron and tin probably came from different mines ;)

Thanks guys. With the complexity of these modules it's understandable. But another thing I noticed was in the barrel dynamics lines of both the 15 and the Sabre. They are exact copies and it's not clear which barrel was created first, canon or MG? Is this really intended or did some things slip by? I bet this kind of thing happens in module development but the consequences can be severe and make a big difference with these tightly matched ships. 19 v Phantom, and even 21 v Phantom will be another match up to watch (maybe). Good to know you are onto things. Thanks!

It appears the entire barrel line in the F-86 is copied from the MiG-15. Or in other words, the barrel dynamics of the M2, M3, 50 cal are based on the canons of the MiG. That may be the source of the difficulty. I backed them off way down by at least 75% and the groupings improved a lot. I can post my lines later.

Good move. I looked at the SFM for all three. I think we can tweak the roll rates and some of the AOA figures in there but I need to unpack them a bit more. I can ask my bro who's an aerospace engineer for NASA/Lockheed Martin. (Orbital Systems) We built an electric RC trainer together. Rebuilt the firewall after my classic control reversal maiden crash. Then he built a KitFox with his buddy from LM, we test flew it together (2nd flight). 4 cylinder turbo. Now he is building a KitWolf(?), a 4 seater, with his wife. The wing was in the basement a year ago with clickos in place :D