Range effects and DP boxes
Contribute Now
Goal amount for this year: 8000 USD, Received: 4,425.00 USD (55%)
Welcome Back, Unregistered     Today's Date:       Current Time: 

Please Support the SOH "2023 Charter Membership Drive"

All online donations at this time are through PayPal

Snailmail donations can still be sent, see this thread


Alternate Donations

Please log in before making your donation.




Accepted Payment Types

Page 1 of 2 12 LastLast
Results 1 to 25 of 33

Thread: Range effects and DP boxes

  1. #1

    Range effects and DP boxes

    I've set up a "mission" using a P38, with the guns angled to shoot straight ahead at ground targets while sat/parked on the runway, and placed a series of buildings at increasing distances. The results so far seem to show conclusively that there are built in range effects causing decreasing hitting power as the bullets travel further. Changing the bullet flight time (?) which is usually 2, appears to affect the distance that the bullets become ineffective and how quickly they lose hitting power. Some of the tutorials available indicate that the bullets retain their damage points over distance but this seems to be not the case. Similarly I have been experimenting with damage boxes where aircraft typically have a separate "pilot" box, or fuel tank box, that is often completely enclosed in the fuselage box. Some of the tutorials have indicated that bullets will travel through one box and into the next, but again this seems not to be the case. Placing a damage box inside or behind another box does provide protection. I decided to put some serious effort into this as I was getting frustrated by some otherwise excellent visual aircraft models that were almost impossible to shoot down, caused by a combination of dp boxes within other boxes, and "systems" being damaged that CFS2 did not recognise. Some explosion effects such as "BOMB" seem to penetrate through DP boxes but the effects are variable, sometimes objects nearby survive and sometimes identical objects further away get destroyed. Other weapons such as rockets appear to be unable to penetrate through one DP box into the next.

  2. #2
    Very interesting. Please keep reporting your results.

  3. #3
    Hmm interesting. One good tactic to bring down bombers is to stay behind their firing range or the range they start shooting which could be a good distance depending on the aircraft. Some mods shorten the range a lot meaning you can nearly fly into them before they shoot. Itís harder to shoot a longer distanceís as in real life itself but the bullets will take them down if you hit Ďem. Dp boxes work by hit points meaning they can absorb some damage without being completely destroyed. They generally donít travel through into other boxes like they do in real life in most cases. Itís still a good setup though and where you get hit at on the plane should effect the way it flies. Anyway itís a great game and itís getting better as it ageís especially in the mod department not so much in the Windows department where it seems to being taken in most of the flak. Some of these planes are supposed to be hard to take down. Many a German could tell you not to stop firing at the Jug or P47 as those who thought the smoked one realized it was still flying. If you have a aircraft you believe is to strong itís easy to dial them into a more realistic range or preferred range by editing the Dp with the age old software must people drop out of college trying to learn (Notepad).

  4. #4
    To help with this, I've altered the stock P-38 to fire fairly level and also managed to change the panel view so that this is also looking ahead level, with the gunsights roughly correct for 600m, and zoomed in x4, which has helped a lot!. If I set the cannon to fire at 769 m/s, with a target 625m away, the bullets start to fall short with the flight time set at 0.77, but 100% of bullets are hitting the target with the flight time higher than 0.77s. I've played with the weight of the bullets and it doesn't seem to affect the trajectory at all. Neither does changing from mg to cannon. The velocity though does affect the trajectory. Initially I thought that the trajectory was being affected by bullet weight etc, but using the x4 zoom shows that the aircraft sits slightly differently each time on the runway. Target is an oil tank with 1500 damage points. At default flight time of 2, it takes 56 rounds of 30dp to destroy the tank. With the bullet flight time set at 1.5s, it takes 71 rounds, and with the flight time set at 0.90s it took about 250 rounds to destroy the tank. At the 0.90s setting, a second tank directly behind the first, so approx. 10m further away, took another 300 rounds without exploding. The point of impact/trajectory seemed to be identical for each flight time setting. With flight time of 0.75s I was getting 28% of bullets hitting/reaching the target. 66% at 0.76 and 98 - 100% at 0.77. 100% hits at 0.78s but not seeming to cause any noticeable damage. In addition it seems that the ammo weight is placed at the [GUNSTATION] location, and that the muzzle flash/firing point should be placed approx. the barrel length further forward in the [guns.x]. If the bullet weight is not affecting the trajectory or hitting power, then it seems sensible to use the weight of the rounds rather than of just the bullet, especially on aircraft where this is going to affect the inertia and cg.

    This is what I used for the stock P-38 guns in order to get them all hitting a level target at 600m

    gunstation.0=0,24,1,0.08,839,2,0.01,750,2,40,1d1*1 4,0.086,0.298,3.477,5.91965,0,0,0,0,0,0,1.6
    gunstation.1=0,24,1,0.08,839,2,0.01,750,2,40,1d1*1 4,-0.2,0.179,3.420,5.91965,0,0,0,0,0,0,1.6
    gunstation.2=0,24,1,0.08,839,2,0.01,750,2,40,1d1*1 4,0.2,0.181,3.486,5.91965,0,0,0,0,0,0,1.6
    gunstation.3=0,24,1,0.08,839,2,0.01,750,2,40,1d1*1 4,-0.086,0.299,3.571,5.91965,0,0,0,0,0,0,1.6
    gunstation.4=1,31,2,0.11,769,0.9,0.05,500,4,40,1d1 *30,0,0.094,3.639,5.74945,0,0,0,0,0,0,4.4
    gunstation.5=2,-1,4,0.00,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
    gunstation.6=3,21,8,0.00,0,0,0,0,0,0,0,0,0,0,0,0,0 ,0,0,0,0,0

    and added this to the panel cfg

    [VIEWS]
    VIEW_FORWARD_DIR=6.600, 0.000, 0.000
    VIEW_FORWARD_ZOOM=4.000

  5. #5
    To check out the enclosed dp boxes, I took the stock Val, enlarged the pilot box but keeping it completely within the fuselage box, and then increased the damage points of every system except the pilot, to several thousand each, changed the pilot effect from DEATH to BOMB and then tried shooting at it. If any bullets at all where getting through to the pilot dp box then the pilot would die fairly quickly but no bullets gets through. If something else explodes nearby though it can kill the pilot.

  6. #6
    Quote Originally Posted by RWILLS View Post
    To check out the enclosed dp boxes, I took the stock Val, enlarged the pilot box but keeping it completely within the fuselage box, and then increased the damage points of every system except the pilot, to several thousand each, changed the pilot effect from DEATH to BOMB and then tried shooting at it. If any bullets at all where getting through to the pilot dp box then the pilot would die fairly quickly but no bullets gets through. If something else explodes nearby though it can kill the pilot.
    The bullets don't travel through one end of a plane into another. When the bullet hits it takes away a certain amount of hitpoints from the system it hits and that's it it wasn't designed to break a front leg bone tear through the heart then exit the other side like a deer rifle. Effects does play a roll after a bullet is done so that in a round about way makes certain hits more damaging then others because that fire will keep burning and it's damaging the system it is in and explosions have a big impact too. It's not a perfect setup to real life but we can make it real without getting up in the real air and start slinging projectiles at each other then it would no longer be a game because of the try again button would be eliminated.

  7. #7
    Results so far;
    At longer range >800m the tracer seems to be higher than the normal bullets. This may be deliberate as tracer rounds are often faster through the air (less drag than a normal bullet.) Reducing the tracer % seems to help get more bullets on target at long range, depending on the size and shape of the target. At ranges of more than 850m, it seems effectively impossible to get 100% hits even on a large (Oil Storage Tank) target. Shooting low seems to get more hits than aiming for the centre. I think that possibly, to achieve a spread of shot, the software possibly gradually turns bullets into "flyers" as the range increases. Effectively like having bullets with too much powder in the case, but never having any with too little. Some anomalies in the terrain mesh (Port Moresby!!) also may have affected initial results, so I switched to using Wake Is. I've found the B-25 (TR_B-25C-1_HllsFr) ideal for this as it needs hardly any adjusting to shoot straight and level along the ground, and means I can use the same aircraft with the same settings in the air later. The oil tank target I'm using is AF_OilTankB1, which has a dp box 14m x 14m x 10m high, as a target. When the bullets hit the target, the flash is always in the same place, which seems to be the nominal aim/trajectory point, and with the oil tank at 900m range, I get most hits consistently by aiming near to the bottom of the tank hardy any bullets seem to drop short which I've verified by using a lower building in front of the tank. Increasing the flight time from 2 to 4, generally seems to improve the hit % as well as increasing the power of each bullet. The damage points in the GUNSTATION seem to be about right at typical combat range, with the bullets seeming to be more powerful at closer range (than defined in the GUNSTATION) and less powerful at longer range.

  8. #8

  9. #9
    Seems I have a bit of information to digest. Appreciate the effort and keep up the good work.

    Cheers
    Aussie

  10. #10
    some conclusions;

    The "flight time" seems to use the defined muzzle velocity but it defines much more than the time of flight. It defines the range at which the bullets become ineffective, with lighter bullets losing power more rapidly.
    The weight of the bullet for calculating the decrease of power is taken from the defined hit points and NOT from the defined ammo weight. Therefore the ammo weight should be the complete weight of each round, so typically 4.8 ish for a 0.50 round.

    Increasing the speed of the bullet makes it more effective at longer range.

    The defined hit points in the gunstation approximate to the hitting power at around 500 - 550m. The hitting power at shorter range is much greater.

    At a range of 1600m, I obtained 99.9% accuracy irrespective of what combination of mg/cannon/trigger I used. at shorter range the trigger 2 is clearly more accurate. The main difference seems to be that the trigger1 is less accurate with longer sustained fire (as if the barrels are overheating/distorting?).

    My earlier conclusions regarding tracer may be not the case and there seemed to be some issues initially with with terrain mesh.

    The damage box (for a ground object?) seems to be treated as if any bullets near the side and top are treated as possibly having reduced effectiveness. When shooting very near the edge, I think some bullets enter the side at a shallow angle giving more damage per hit than a burst of fire that is close to the edge of the box but completely inside. Possibly the base of a ground object box is treated differently than the base of an flying object.

    The speed of the round seems to be added to the aircraft speed, so range may be much more when attacking a ground object at high speed. I've checked this out by firing a really slow bullet with tracer and watching it hovering around in front of the aircraft before falling away. There are gravity effects on the trajectory but not and aero drag. - bullets may well accelerate when fired downhill, and may lose power more quickly when fired uphill?

    The damage points seem to treat individual bullets as integer values and seem to ignore decimal places. So a 6.5 bullet is no more effective than a 6. (I may be wrong but that is what I am getting so far).

    The rate of loss of hitting power is quite variable at times. In "sniper" mode with the trigger2 and a low rate of fire to give single shots, the accuracy and hitting power are very consistent. At longer range, the trigger1 shows multiple impact flashes at the target - indicating a spread of shot.

    An additional summary of results attached that is clearer/simplified.


    The B25 I used did not sit properly when spawned and after I adjusted the static pitch and static cg height, I had to slightly adjust the gunstation upangle. For the very long range shots when my shots were initially falling short, I could get more range by switching the engine on and as the nose lifts slightly this allowed longer shots, which helped a lot in estimating what adjustments to make as the range increased. The "target I used was cfsfueltrailer but made the damage box larger to ensure successful hits, 14m wide x 12m tall. The aircraft I ended up using was the TR_B-25C-1_HllsFr, with an additional central, forward firing gunstation that I could use for single shots.


    Other factors that I haven't properly checked out so far;

    The type of aircraft may effect things (fighter, bomber etc?)
    In the air, flying and moving fast will obviously change things but is going to be tricky to set up and get results.
    AI guns may behave completely differently!!!
    Coriolis effect is probably being ignored?
    A heavy cannon on a light aircraft may be less accurate?
    AI flak/AA guns need looking into. I can probably set up a test for this reasonably easily and try to count the number of hits from a slow firing AA gun? Would need a simple aircraft with a single hit box and single system.
    CFS2 clearly has a lot of random elements built in, so absolutely anything could affect the guns!
    Attached Thumbnails Attached Thumbnails gunstation range effects.pdf  

  11. #11
    If you want to be able to repeat the test setup - ?
    "Mission" file below, for the longer range test, at Wake but a lot further back along the runway, (take off position -1150); The shorter range tests are most easily done by redefining the take off position rather than constantly repositioning the target!! I think that at -1150 the aircraft is just off the flatten area as it needs more adjustment to the guns than I expected, but it seems to sit consistently after spawning so maybe the trajectory of the bullets towards the end of their flight is a bit different to what I expected.
    The other vehicle is to help indicate the range!

    I made some small adjustments to the B25 panel, cfg, dp files which I will post/detail separately.


    cfsfueltrailer mods to .dp file (box size and damage points)

    [BOXES]
    box.0=%box_name.0%,-1.60,-0.01,-7.00,1.60,11.99,7.00


    [SYSTEMS]
    system.0=%system_name.0%,500,-1


    [mission]
    version=2.0
    title=%title_string%
    summary=
    objective=
    intelligence=
    mission_type=1
    allegiance=3
    total_goals=1
    player_aircraft=3892
    player_aircraft_name=%player_aircraft_name_string%
    airfield=%airfield_string%
    date=1/6/1943
    time=12:00


    [unit.0]
    id=9100
    type_id=3892
    name=%name_string.0%
    is_player_aircraft=1
    allegiance=3
    skill=1
    payload=0


    [formation.0]
    id=6000
    type=air
    waypoint_path_id=5000
    directive=1
    players_formation=1
    points=2
    unit_id_position.0=9100,0,0.0,0.0,0.0


    [waypoint_path.0]
    id=5000
    waypoint.0=2,1,N19* 17.04',E166* 39.07',+0,,,,0,,8170,0,1
    waypoint.1=4096,0,N19* 17.25',E166* 39.24',+735,,,,224,,0,0,1


    [unit.1]
    id=1007
    type_id=4132
    name=%name_string.1%
    heading=125
    allegiance=3
    skill=1
    points=0


    [formation.1]
    id=6001
    type=ground
    waypoint_path_id=5002
    directive=0
    unit_id_position.0=1007,0,0.0,0.0,0.0


    [waypoint_path.1]
    id=5002
    waypoint.0=1,1,N19* 17.25',E166* 39.21',+0,,,,0,,0
    waypoint.1=1,1,N19* 17.23',E166* 39.24',+0,,,,0,,0


    [unit.2]
    id=9099
    type_id=4313
    position_lat=N19 17.2420
    position_lon=E166 39.2242
    position_alt=+0
    position_pbh=0.0,0.0,315.0


    [unit_type.0]
    type_id=3892
    unit_family=1
    name=TR_B-25C-1_HllsFr


    [unit_type.1]
    type_id=4132
    unit_family=3
    name=amamb


    [unit_type.2]
    type_id=4313
    unit_family=4
    name=cfsfueltrailer


    [runway.0]
    id=8170
    heading=34
    units_across=2
    base_lat=N19* 17.04'
    base_lon=E166* 39.07'
    base_alt=+0
    takeoff_start_pos=-1150,0
    takeoff_stop_pos=1600,200
    landing_start_pos=5000,350
    landing_touchdown_pos=550,0
    landing_stop_pos=-250,0


    [strings]
    "title_string"=wake range test 1600m fuel truck
    "player_aircraft_name_string"=TR_B-25C-1 Hell's Fire 500th BS
    "airfield_string"=Wake4
    "name_string.0"=Argo
    "name_string.1"=Bronco

  12. #12
    TR_B-25C-1_HllsFr

    minor tweaks

    Panel.cfg file I added this

    [VIEWS]
    view_forward_WINDOWS=MAIN_PANEL
    VIEW_FORWARD_DIR=0.75, 0.0, 0.0
    VIEW_FORWARD_ZOOM=4.000


    aircraft.cfg file I adjusted with these values (in red) for the wheels contact points and the cg height/pitch to get the aircraft to sit/spawn nicely instead of jumping around, and also to be more level for what I needed;

    point.0=1, 15.84, 0.00, -8.05, 1400, 0, 1.500, 45.0, 0.4600, 2.5, 0.70, 6.0, 7.0, 0
    point.1=1, -1.82, -9.23, -8.10, 2000, 1, 2.500, 0.0, 0.4765, 2.5, 0.86, 7.5, 8.5, 2
    point.2=1, -1.82, 9.23, -8.10, 2000, 2, 2.500, 0.0, 0.4765, 2.5, 0.86, 7.0, 8.0, 3




    static_cg_height = 7.754
    static_pitch = 0.338

  13. #13
    Initial tests with someone firing back at me!

    I used a small biplane (A1N1)and reduced it to a single .dp box and single system with 2000 damage points. I left this at the take-off position so was a static target, and lifted the tail to horizontal in the static position.

    For the AI gunner I took a tank (Matilda) and gave it a gun with 25 hit points.

    With mg and either trigger it was hopelessly inaccurate (hit me about 5 times in 250 shots from moderate range). With cannon it was very accurate at anything under 1000m range.

    Interestingly it always stops firing after achieving 50% damage. I have changed my damage points (500, 1000, 2000) and the hitting power (100, 50, 25) and still it goes up to 50% target damage and stops firing.
    At very close range where it was actually able to hit the target, the mg setting also stopped firing after achieving 50% damage.


    at close to 1000m it took 68 hits to give me 50% damage.
    at close to 200m range it took 68 hits to give 50% damage
    this was repeatable and very consistent , varying sometime by no more than +/- 1 hit.
    This works out to a damage caused of about 15 points per hit even though the cannon was set at 25 points.




    At 1400m range I let it fire at me for a long time and it didn't achieve a single hit.

    It also completely ignored any attempt by me to slow down the rate of fire below about 1 shot every 1.5 seconds.

    This was a vehicle so may not represent what an AI aircraft does, but it's a start!!

  14. #14
    I did 6 runs against 1tank and each time it damaged me to 50% and stopped firing.
    Added a 2nd tank and they damaged me to 75% and stopped firing.
    Added a 3rd tank and they killed me 100%.

    Went back to 2 tanks and they killed me 100%!
    Went back to a single tank and it killed me 100%, and now does this every time!!!

  15. #15
    First interesting discovery is that the tank's rate of fire is controlled partly by the "flight time", so I have it firing once every 10 seconds now by putting 10.0 in the "firing interval", and 10.0 in the "flight time".
    With the flight time set at 2, this was also the maximum firing interval I was able to achieve. I need to try that with some aircraft to see if the same applies.

    gunstation.0=1,-2,2,10.0,839,10.0,.......etc

    The AI tank is working consistently now, so I should have some decent/reliable results soon.

  16. #16
    CFS2 AI Gunner test Date 27/01/2023
    AI gunner
    Location. Opposite ends of Wake Is. runway
    Vehicle, Initially moving a few m before stopping.
    Possible fired a couple of rounds before stopping
    Pointing reasonable straight at target so no issues with limits on slew or elevation of gun.
    Matiltda 2 tank
    Modified gunstation
    gunstation.0=1,-2,2,4.00,839,4,0.08,1400,0,100,1d1*25,0.256,1.712,-0.064,0,0,0,-90,90,15,-1,4.6
    range 1100m on level ground
    Target (Player)
    JP_A1N1
    Modified to have a single hit box.
    box.0=%box_name.0%,-4.50,-0.70,1.035,4.50,1.30,-4.56
    frontal target area 9m wide x 5.2m high
    Actual Point of aim in relation to the hit box is unknown at this stage!
    Single System
    system.0=%system_name.0%,2000,3
    Gun hit points per round defined in gunstation = 25
    Hits on target to destroy target 100% damage = 69
    Damage points of target box = 2000
    Actual average damage per hit = 28.99
    Total Rounds Fired (4 seconds per round for 30 minutes) = 450
    Hit Rate Fraction = 0.15
    Most Consectutive Hits = 3

  17. #17
    I've repeated this at close range = 200m.
    100% hit rate
    Exactly the same number of hits to completely destroy the target so no change in hitting power with range for this particular AI gun.

    The unexpected thing though is that the firing rate is much faster at 200m (approx. 1 every second) than at 1100m (one every 4 seconds as defined in the gunstation), and seems to be possibly influenced by the round hitting the target and therefore no longer needing to be tracked? If I can get it to miss a few times and see if this results in a longer pause before the next shot then this may be confirmed. I did get the impression, that in the 1100m range test, the shot following a hit was slightly quicker, such as when it achieved 3 consecutive hits a couple of times.

  18. #18
    I've got the A1N1 in the air behind a Devastator, modified to have a single rear gun, slow firing so I can count the hits, and with a hitting power defined at 25.

    I had to slow the Devastator's minimum speed in the dp file so I could keep up with it, but even having done that, and with only a single "system" defined (Fuselage), it's incredibly difficult to hold position long enough to take 100% damage, and at the same time count the number of hits, all while trying not to get too close as this causes the Devastator to take avoiding action.

    Also discovered that I need to display "Health better than xx%" rather than "less than xx%" for it to notify me promptly. If I ask for "less than xx%" to be displayed it doesn't display anything until 100% damage is reached.so it needs a combination of true/not true in the events setup for each step in the % damage to be displayed.

    When being shot at on the ground, by the AI tank it was very easy to count the number of hits, which matched the audible hits. In the air though, there is a lot of "Dr Seuss" clunking and thunking that seem to be near misses and may or not be causing damage as they do not register in the text display as hits.

    The original twin rear gunner seemed to be able to shoot me down with approx 28 hits that I counted, but I think each "hit" was a double hit, and it was difficult to count.

    With a single rear gunner the "better than 95% health" display cuts out after 5 confirmed (in the text display) hits. 5% health = 100 points so 5 hits is in the right area for a 25 hit point gun. It seems that several other clunks/thunks that seemed like hits, were not actually causing any damage.

    I'm going to modify the Devastator aircraft category to see if that helps it fly straight, and also change the "system" in the A1N1 to something like the "radio" to see if this keeps me flying fast enough to keep in range of the Devastator's rear gun as the damage gets close to 100%.

  19. #19
    Changed the Devastator to category 2, and now it flies straight and doesn't take evasive action.

    Added a small 0.5m x 0.5m x 0.5m box in the centre of the main "single" box I made for the A1N1 and then placed the Engine, Pilot, Rudder, Port Elevator and Starboard Elevator systems entirely inside that box. Now I have a player aircraft that can sustain 100% damage to the main box (2000 hit points) and remain flyable and under control. I was expecting to have to put more systems than this inside the small box, but have tested it and it works fine with just these systems protected. Previously, without defining these systems, CFS2 was clearly degrading the performance of the engine etc as the damage to the main hit box increased.

    The AI gunner definitely seems to cause some damage with near misses that are not registered as hits so it's going to be difficult to work out the exact hitting power of the AI rear gunner at different ranges etc, but a larger main hit box on the A1N1 might be enough to get 100% full hits.

  20. #20
    Another interesting discovery is that the co-ordinates for the hit boxes must be in a specific order, in order to register hits/damage. - This would make a box impervious to damage as the aircraft can be used without any issues.

    This works, and will take hits;
    -4.50,-0.70,1.035,4.50,1.30,-4.56

    But this doesn't take any hits even though the box is the same size and shape;
    4.50,-0.70,1.035,-4.50,1.30,-4.56


    It looks like the lowest number needs to be in the first set of co-ordinates as is the case for the x and y values.
    The "z" values don't seem to be the correct way around based on the logic for why the "x" values might have to be the way they are. - I'll try turning the plane sideways to expose that face to fire and see what happens

  21. #21
    The aiming point for the AI is clearly the 0,0,0 position of the target model.
    If the damage box is just 0.05m above the 0,0,0 point then the AI cannon will miss the box every time at a range of 200m, (static AI gun and static target).

    The machine gun AI will hit a box 2m wide by 0.45m tall, placed 0.05m above the 0,0,0 point fairly regularly due to it's built in scatter of shot, while the cannon misses this target every time from 200m as it is too accurate aiming for the 0,0,0 point.

    This would make something like a Twin Mustang quite difficult for the AI to hit in a stern chase, where both fuselages are well away from the 0,0,0 point.

  22. #22
    Keep up the good work. The world needs more Smart people. I passed the exam myself but I cheated so glad yíall are here figuring this stuff out.

  23. #23
    quote

    Another interesting discovery is that the co-ordinates for the hit boxes must be in a specific order, in order to register hits/damage. - This would make a box impervious to damage as the aircraft can be used without any issues.

    This works, and will take hits;
    -4.50,-0.70,1.035,4.50,1.30,-4.56

    But this doesn't take any hits even though the box is the same size and shape;
    4.50,-0.70,1.035,-4.50,1.30,-4.56

    I,m wondering if the attacking direction - rear to front was the reasoning the 1% dp rebuilds , ??

  24. #24
    At 190m range, the "veteran" AI cannon was able to get 100% of it's shots into a 0.050m x 0.050m target box, centred at the targets 0,0,0 position, and 95% of the shots into a 0.030 x 0.030m box.

    In contrast the veteran mg gun was unable to get any hits inside a 0.50 x 0.50m box.
    The ace mg got 14% of it's shots into the 0.50 x 0.50m box.

    A 4 x 4m box was hit 62% of the time by both the veteran and ace mg.

    A 2 x 2m box was hit 18% of the time by the veteran and 44% by the ace mg.


    With a target box 8m wide x 5m high, again centred at the target's 0,0,0, both the veteran and ace mg achieved 80% hits.

    Moving the target box around shows that the mg is more complex than the cannon, and I think even more so for multiple mg. There are possible implications for how best to harmonize mgs on fighter aircraft.

    For interest I set up a slow firing AI "mortar" firing at 50 m/s at 45 degrees elevation, with a few degrees of freedom and this does open fire on me at around the 190 -210m range tested.
    gunstation.1=0,-2,2,2.00,54,20,0.01,200,0,100,1d1*250,0.000,1.912,-0.064,-45.0,0,0,-90,90,5,-5,4.6

    One thing I'm finding is that some of the buildings I try out to give a reliable and simple target, don't have damage boxes that particularly match the visual model. CD_Barracks_Large_Bergen for example had a damage box that was quite a lot smaller than the visual model. The original .dp has a box that is about 14m long, and the visual model is 80m long! Similar with some aircraft, where you fire away at what you see, only to find that the damage box is somewhere else!

  25. #25
    I made a slow firing , slow speed cannon (added it to the AF_IJA_75mmType41_MtGun, which didn't actually have a gun!);

    gunstation.0=1,-2,2,2.00,68,6.00,0.01,300,0,100,1d1*25,0.000,1.000 ,0.700,-10.0,0,0,-90,90,10,-10,4.6

    This slow muzzle velocity has revealed that there are in fact some anomalies, possibly specific the the ends of runways, and maybe only some runways. Whatever it is, is enough to confuse the AI gunner into thinking that the target is further away than it really is. I think for most situations this is not going to be noticeable though as the faster bullets will have a flat enough trajectory to still hit the target regularly. It doesn't seem to affect the damage caused from hits. AI guns firing from +/-90 (each side) towards the player aircraft seem to be ok. Maybe it's linked to the sometimes rapid burst of speed an AI aircraft has as it's about to land? When I added an AI wingman behind me on the runway, the AI was less affected.

    I also found that the AI gunner seems to prefer to have it's gun pointing in the middle of any ability to elevate up/down. So if a gun can fire from 0 to 90 degrees elevation, the AI gunner prefers to have 45 degrees elevation as it's starting point, and have +/-45 freedom. If you start off centre and have unequal up/down elevation freedom in the gunstation then the AI gunner is not so good.

    I also tried a slow speed (100 m/s) flak gun and was able to observe the way that the AI gunner predicts where you are going to be when the flak arrives. You can fly along at almost the same speed as the flak and watch what it does. By limiting the slew, I was able to approach a flak gun from behind and see it fire ahead to where I would be before I reached it.

Members who have read this thread: 63

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •