Welcome to Keen Software House Forums! Log in or Sign up to interact with the KSH community.
  1. You are currently browsing our forum as a guest. Create your own forum account to access all forum functionality.

Torpedos, which is the new best design?

Discussion in 'Community Creations' started by mecra, Dec 9, 2016.

Thread Status:
This last post in this thread was made more than 31 days old.
  1. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    Yes but it means you need a radio on the missile, with the range only being 5km and the enemy ship will be able to see the radio. and the longer range means takes longer for the ship camera to charge. where as if you launch the missile and it has it's own scanning lidar looking ahead it can do final lock on and terminal guidance. Almost needed for long range from 2km to 50km (or more) engagements.
     
  2. Uranius Crack Trainee Engineer

    Messages:
    47
    But you can place a lot more cameras on your launching ship than on your missile, and you would be able to manually track the target
     
  3. Whiplash141 Junior Engineer

    Messages:
    958
    I used to use a bug to do this, but now, I am using the new Inter-Grid Communication :)

    Edit: Derp, I thought you asked "how" not "if" :p
    But yes, for semi-active, you need to broadcast information to the missile.
    --- Automerge ---
    I think you misunderstand how broadcasting works. You don't need to broadcast more than a meter on the missile to be able to receive messages. The firing ship; however, needs to broadcast info, but the range can be up to 50 km for a large grid shooter. I kind of like this mechanic: to fire a salvo, you have to expose yourself for a short time just like how in ship warfare firing up sonar gives away your position. It is definitely more fair than missiles being fired from a direction that you can't pinpoint.
    [​IMG]
    (Picture courtesy of rexxar.)

    Also... homing missiles are too easy to dodge :( I miss GFD man
     
  4. Uranius Crack Trainee Engineer

    Messages:
    47
    Wouldn't missiles be harder to dodge if you keep raycasting from your launching ship? Since you can use more cameras and are able to manually keep track of your target
     
  5. Whiplash141 Junior Engineer

    Messages:
    958
    Yes! That is why I prefer Semi-Active systems to homing. I have written a code for that, but right now I'm trying to sort my target leading algorithm.
     
  6. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    Doh... oh this is going to be interesting[​IMG]
     
    • Like Like x 1
  7. SF-1 Raptor Junior Engineer

    Messages:
    740
    :stare::eek:
    I really wish I understood how you did this.
     
  8. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    Ok.. I like.

    22km Lidar Fire Control integration test with launch bay and gps/lidar homing missile

    [​IMG]
    [​IMG]
    --- Automerge ---
    Got a hit at 42km this time on stationary and on a 5 m/s object ( not changing course).
     
  9. Wellstat Apprentice Engineer

    Messages:
    212
    If the cross grid comms is completely removed, then the missile's Semi Active mode will be exactly the same as GPS mode. Since the missile only sees coordinates sent from the launching ship and do not need to know how that coord is retrieved.
     
  10. Sinbad Senior Engineer

    Messages:
    2,788
    @Whiplash141, based on the pic you provided above, you could step the range of the antenna on the launching ship in sync with the maximum speed of the torpedo. That way the range at which the antenna is visible is only ever a hundred meters or so farther away than the torpedo.
    Targets wouldnt be able to see the launch vessels guidance antenna until the torpedo was a few hundred meters away. Far too late for it to be a warning to them.
     
    • Like Like x 4
  11. Wellstat Apprentice Engineer

    Messages:
    212
    I am also making changes to my Lidar Homing Script to incorporate this, but that will be on the missile end. Definitely will require another script on the launching ship to send the lock on info.
     
  12. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    What format you think you will want them as?
    I have been using the GPS format since that is a standard. or would you rather have 2 vector3d with current position and current velocity of the target?

    I have adjusted the Lidar Mapper and the Lidar Fire Control to match watch your missile script wants.

    I could even send along another missile with it's own scanning array to try to give local guidance to near by missiles. Or send out a missile/drone with laser comms to firing ship to do the fire control from closer range.
     
  13. Whiplash141 Junior Engineer

    Messages:
    958
    Yeah, I already have an antenna scaling feature built into the missiles, but I'll look into trying to do this on the antenna's end. Might be some issues when working with large and small grid antennae interacting because of the massive broadcast range difference :/
     
  14. Sinbad Senior Engineer

    Messages:
    2,788
    The torpedo can have a range of 1m and still receive fine, you just wont get any telemetry back from it. The fastest the torpedo can move away from the launcher is 100m/s, so for a stationary launcher, the antenna just has to extend its range 100m every second (plus a hundred meters or so range as an initial launch range) and the torpedo should always be in range of the launcher.
    This all gets fiddly with a moving launch platform. Remembering the launch co ordinates and time, and the initial launch vector, you can estimate how far the torpedo has moved away from the launch coordinates, and in what direction, that can give an estimated position of the torpedo. The same calculation can be made for the launching ship, but with the benefit of already knowing its current position. Then the distance between the current position of the launcher to the estimated position of the torpedo can be calculated. Add a fudge factor based on how much the target vessel has moved since launch (and thus how much the torpedo has manouvered away from the initial launch vector) and you should be able to scale the guidance antenna range well enough to remain undected.
    Or you could just make the guidance range expand at the maximum combined speed of 200m/s, assuming torpedo and launcher both depart the launch coordinates at max speed in opposite directions as a worst case scenario. But that could potentially leave your launcher detected for the last 25km of torpedo travel, or a little over 4 minutes.
     
  15. Whiplash141 Junior Engineer

    Messages:
    958
    I'm aware. Issue is without a broadcasting antenna or beacon that you can see, the missiles position is not properly synced to client in MP :/

    In vanilla. This would make the code problematic for use in modded worlds. I don't like needing users to open up the code and change crap... that's how stuff gets broken lol

    Sweet lord, that is a boatload of assumptions and guestimations :D

    There is also the issues of:
    1. What if I fire multiple missiles
    2. When do I reset the antenna range
    3. What if the missile gets out of range
    4. Shooting ship movement will break many assumptions
    5. Missiles don't always fly linearly, course correction isn't being accounted for
    etc...
     
  16. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    Yep.

    The FFG i'm designing to use the Lidar Fire Control and Wellstat's missile code. I can toss 4 light missiles (1.5m x 6.5m x 26m ) every 12-20seconds at ranges of 50km.. well you can try farther but unless they are broadcasting a radio or beacon, you need to have a lidar mapper give you back a gps of enemy position before you can even lidar scan that spot.
     
  17. Sinbad Senior Engineer

    Messages:
    2,788
    Haha, yes lots of assumptions. But thats what the fudge factor is good for. Estimated torpedo position would form a pool of error, or rather a sphere of uncertainty. As long as the guidance antenna range covers all of that area, it will probably cover the torpedo.
    Multiple missiles, simulteneous launch should be ok if you use a larger range on each torpedo's antenna, hopefully the group will stay in range of each other and will form their own mini relay network. As long as one of them can see the guidance antenna, it should be able to relay to the rest of the salvo. For serial launches, track the first launched, the others should all be closer. Then again, the power of this new intergrid capability could also lead to having multiple launch controllers, each managing its own salvo to different targets.
    When to reset range? Well, you know how far away the target is, you know how much its moved since you launched (and so how much extra time the torpedo will take to get there) so you know the minimum engagement time, as well as a fair idea of how much time torpedo course corrections will add to that. Add another 10% and assume the engagement is over. In any case, even if you are wrong by 10 seconds or so, the torpedo would likely be close enough for autonomous terminal guidance.
    If the missile gets out of range, there are a few things it could do. First would be checking if it can aquire the target autonomously in terminal guidance mode. If not, maybe have it loiter for a few seconds in an attempt to allow the guidance antenna range to catch up. If part of the message sent to the torpedo is the launchers position and motion vector, then the torpedo should be able to steer to reestablish guidance based on the last known position and course of the launcher. Depends on how smart you want your torpedo to be. In any case, sending this info is a good idea for establishing an engagement zone. Consider loosing your antenna with a torpedo capable of autonomous terminal guidance in range. Without the knowledge of where its launcher is likely to be, it may engage its launcher.
    As far as the motion of the target and launcher effecting where the torpedo is, if you use a recursive vector sum of both the targets movement and the launchers movement (each update you calculate the motion vector for that interval, then sum it with the sum of all previous intervals) then you can have the total target movement since launch, with a direction. Same for the launcher. Add the target info to the straight running estimated torpedo position (launch coordinades and initial launch vector, extended by 100m/s along that vector for each second since launch) and assuming the torpedo has been correcting course you should be within a hundred meters or so of its actual position. Work out how far that is from the launcher, and add 10%. That range should be good enough most of the time.

    And thats all we need right? You only need to get lucky one, the target has to be lucky every single time you launch.
     
  18. Whiplash141 Junior Engineer

    Messages:
    958
    Nope. Not with TOW guided missiles lol.
     
  19. jonnytaco Apprentice Engineer

    Messages:
    228
    "lets go that way!" is pretty much what they know with your TOW OGS right? I forget when you changed it but they also make their way back to the targeting vector before continuing forward too I think.
     
    • Agree Agree x 1
  20. Whiplash141 Junior Engineer

    Messages:
    958
    Lol yep! All they know is the missile's origin and the direction that they should go. And yeah, they will hug that target vector as closely as they can. I also added a way to toggle the missile's antenna "on" and "off" in flight to hide it from the enemy as well as allowing you to toggle spiraling on/off :)
     
  21. jonnytaco Apprentice Engineer

    Messages:
    228
    You know i'm going to use that hell out of that spiral off function :p.
     
  22. Mattk50 Trainee Engineer

    Messages:
    69
    I think im missing some underlying mechanic here, those seem to be rods of heavy armor with gyros on the end with a merge block in the middle for some reason, but how are they supposed to separate? Is there a way to get merge blocks to attach to each rod or must you have an explosive hidden somewhere inside?
     
  23. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    I tend to like cheap quick to make long range light missile.
    --- Automerge ---
    Or fire off axis and then send a target update so the missiles come in 10-45 degrees off from the firing ship.
     
  24. tankmayvin Senior Engineer

    Messages:
    2,863
    The merge blocks have two functions:

    1) They let you change the payload on the fly. Never really used this in practice but it's useful when designing.

    2) It is the mount point for the rods. Deep inside that thing there is a warhead mounted on the front of the MB you can see and there are sensors inside the rod assembly that fire the warhead when it detects the target. The warhead pops the MB (which is 9x1 and pops easily), and the rods fly free.

    The gyros are there because they massively increase damage, or did when the design was originally implemented. They are inerted (turned off)

    Whips script lets you spin the missile so you can get dispersion of the rods for higher volumetric damage, but you'd need to fuse much earlier than 50m to get max damage. I need to put in a raycast fusing system so you can burst at a set distance from the target.
     
    • Friendly Friendly x 1
  25. iN5URG3NT Senior Engineer

    Messages:
    1,132
    Right, this thread has a lot of experienced PMW builders who have no doubt fired hundreds of missiles in testing. I got a question for you fellas:

    What are the advantages and disadvantages of the PB Triggered Warhead Cluster Missile versus the Solid Rod Missile and what features/blocks are most important to both?

    For this question you can assume launching and guiding are not an issue and neither are enemy turrets. Propulsion will be Hydrogen and the missile will be small grid. I'm mostly interested to hear about in game results. Like heavy armour vs blast door, sensor placement, tip design etc.
     
  26. tankmayvin Senior Engineer

    Messages:
    2,863
    I've never gotten a pb triggered warhead stack to work reliably in MP, even vs a perfectly static target. KE missiles work 100% of the time and ore bombs are somewhere in between - their damage heavily depends on the particular target grid.

    Nukes are amazing in principle, but they just aren't reliable enough for me to use. To get reliable detonation the standoff has to be set too high and the resulting damage just isn't great.

    The main issue with KE is that it needs to be very heavy to avoid splatting or deflecting.
     
  27. Uranius Crack Trainee Engineer

    Messages:
    47
    small ship warheads don't scale very well and they also run the risk of prematurely exploding (though I think this will change in the near future). A solid rod is easier to make and will often do more damage than a simple warhead, dependig on the weight of your projectile.
    If you want to make a small missile that still does some damage, then warheads are a better option in my experience.
     
  28. tankmayvin Senior Engineer

    Messages:
    2,863
    They deal utterly amazing damaged when stacked if you can get them to go off properly.
     
    • Agree Agree x 1
  29. jonnytaco Apprentice Engineer

    Messages:
    228
    Warhead missiles tend to cause the most amount of damage for the mass of the missile. As far as maximizing your blast effect, laying out your warheads in a fatter rather than longer pattern greatly improves efficiency. If you're using sensors to trigger the warheads I'd suggest using 1 at the front and 1 at the back with overlapping fields to provide a level of redundancy if it impacts before detonation. Ideally I'd recommend using raycast to trip the warheads because they update 6 times faster. A sensor at 100 m/s will cover 10 meters before updates, a camera with a 2m raycast will move 1.6m between updates. This greatly increases reliability and gives much greater control of where you want the warheads to go off.

    As for downsides using warheads. If you are firing them from an internal bay you need to be very careful with your clearances. Last thing you want is for the thing to go off in your hull. He missiles are also much more vulnerable overall so engineering ways to mitigate this can be more complicated. HE missiles also tend to not penetrate as deeply although this is arguable after changes to collisions. You also need allot of warheads which costs allot of magnesium.

    KE missiles are cheap and resilient. I honestly have not really found a reliable penetrator design since the changes so my knowledge here is rather limited :/. I do know that ore bombs can be some of the most effective and cheap payloads however they often have very high mass and currently inflict almost no dmg to ships with locked rotors.
     
    • Agree Agree x 1
  30. Elfi Wolfe Apprentice Engineer

    Messages:
    498
    Or fire off axis and then send a target update so the missiles come in 10-45 degrees off fr
    I use dual sensor trigger warheads on a light missile. In best case fires warhead opens a hole and second warhead is somewhere inside. Worst case is a single explosion on surface.
     
Thread Status:
This last post in this thread was made more than 31 days old.