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. Cadde Trainee Engineer

    Messages:
    99
    Thank you for a constructive argument rather than "LOL I AM GOOD YOU ARE BAD" that i have been having with @tankmayvin so far.
    And i totally agree with you, all those points are things one would have to overcome. And the PB isn't... Or rather, the interfaces/functions we are given... are far from optimal for such a task. It can be done but when one spends a few days just working out the kinks out of an PB autopilot as well as barely understanding how to work with 3D vectors, matrices and quaternions it becomes an exercise in keeping ones frustrations at bay.

    But it's slowly getting there, between actually playing the game of course. Once the basics are set up, the implementation is just a matter of giving the right instructions. Making something orbit in an ellipse while also aiming at a target is a cakewalk at that point. Making it move erratic is just a matter of changing the attack angle of that ellipse.
    Or, if the drone detects that it's no longer useful, it can always go for a ram and self destruct.

    The one most important thing about a drone here is that it costs basically nothing, is small and extremely nimble. It should out accelerate anything beyond a reactor with all thrusters on it.
    Hmm... Come to think of it... I wonder how quickly a reactor with 4 small ion thrusters accelerates? Gotta test that some time.

    Oh and by the way, gyros become super effective when the grid mass is low or there's enough gyros. One can keep all thrusters pointing in the same direction and still maintain perfect movement control. The only downside is you lose sight of the target while changing your velocity vectors. But imagine a drone that is nothing but an accelerator with a gun on it. It moves in, fires a few shots, angles itself 90 degrees to one side and blasts off, re-targets and fires. Rinse and repeat.
    0-100 m/s in what? 3 seconds? I dunno, i haven't tested it yet but i shall!

    EDIT: Oh and since you mentioned "one script fits all" and the supposed server limit on PB's. It's of course entirely possible to control all drones with a single PB currently because you can keep references. A server restart would be painful but the whole point of drones is they are disposable. Once you launch them you don't expect to see them return again. They are in your control for as long as they have fuel, life or the server pops.
     
    Last edited: Dec 30, 2016
  2. tankmayvin Senior Engineer

    Messages:
    2,863
    I've used it in solo survival only, where it works very well against the sort of piloting the current AI scripts are capable of.

    Do you have a script for a drone that can orbit around a ship at 750 m doing evasion? That would be extremely useful.

    I would definitely use an aim assisted grav cannon to compliment missiles for pot-shotting pirates etc during base defense.
     
  3. GrindyGears Senior Engineer

    Messages:
    1,787
    @Cadde I think the point tank was making was simple: Unless you actually have a prototype built, or some form of tangible design, there's very little value in your claim that its "superior" in theory it sounds great, but there are a number of technical issue surrounding self replicating machines, the first that comes to mind is resources, those pesky little things that welders need in order to build a ship, unless you program it smart enough to go grab resources when it decides it needs a friend, your infinite replication will grind to a halt. Another issue (that may have been fixed idk) is blueprints when built are sterile, they don't contain another blueprint, i've looked into this before, for a tunneling machine, but that kinda brought me to a halt.

    If i told you in theory I have a wheel based miner (small grid) that could dig down close to 100m from the surface and use only about 50kW of power and has over 100 rotors and over 100 pistons, would you believe me? I mean, I don't actually at the moment.... Right now I have a prototype (mechanisms only) that can go around 80m deep, some longer segments could easily fix that. What's preventing me from finishing/going full size, is the currently broken merge blocks that don't allow me to have enough pistons to deal with the physical load it must endure.





    Basically if you can't back up your claim, don't try and state its superiority until proven so.

    Edit: Removed spoiler tag to try and fix images....
     
    Last edited: Dec 30, 2016
    • Like Like x 2
  4. tankmayvin Senior Engineer

    Messages:
    2,863
    For the record I think it's important to note that I have never once claimed to be better than anyone, I was balking at the notion that someone can claim they have something that will beat absolutely everyone else, when they then freely admit the entire superior weapon project is actually just vapour, doesn't exist yet.

    Imgur links are broken :(

    I think you can have nested BPs now, but it's been know to cause all sorts of various problems so lots of people have moved away from them.

    A bigger problem is that timers and the like don't properly inherit the blocks referenced by the parent of the blueprint, and so printed grids almost always seen some sort of manual tweaking before being operational.

    There are also extensive issues with welding up projections from the same grid as the projection and these come and go but can totally break projects. I'm fond of self welding and I constantly have ships just crap out on me and refuse to repair battle damage and the like.
     
  5. Wellstat Apprentice Engineer

    Messages:
    212
    The evasion script uses GFD, so to get it working i will need to change it to use data from mothership instead (ship size and coordinates). The basic principle is constant changing acceleration by varying the side thrusters randomly (your ship appears to constantly curve).

    Probably posting this in 1 to 2 weeks.
     
    • Like Like x 1
  6. Renfield37 Trainee Engineer

    Messages:
    97
    Sinbad, I use a barrel-less design. The first photo shows what the front end of my stone thrower attachment looks like. The second photo shows the size of each boulder in realistic inventory settings. Next, if you look at the small asteroid I used to anchor my heavy armor demolition wall to, both of the boulders generated are about that size, maybe a bit smaller. The last two photos show the damage it does in vanilla. It only does this clean slice through effect on voxel anchored stations. For ships and non-voxel anchored station, it breaks through about 4 or so layers I think. I haven't really done much testing with it lately. I know if the ship is moving away or has good angled armor it will bounce off it though.

    sheesh imgur album linking is all sorts of messed up, so i'm just gonna post direct links




     
    Last edited: Dec 30, 2016
    • Informative Informative x 1
  7. Sinbad Senior Engineer

    Messages:
    2,788
    colour me confused. ive got this setup, i adopted what i could see in your screen shot, because my opposing welder/grinder wasnt working:

    [​IMG]
    weld it, load it with 5000 stone in one stack (only stripping turrets for now), grind it... and nothing. i get no Nugget of Station Negation.
    my accelerator works great, i just cant drop this boulder.
     
    Last edited: Dec 30, 2016
  8. Renfield37 Trainee Engineer

    Messages:
    97
    oh nvm, those are stabilizers. you have a rack of grav gens on the back, right? If so, I haven't thought about doing that with my build. Would help out the accuracy a lot. Thanks for the idea.

    crap my post got deleted, stupid auto merge thing.

    Anyway, what I said was you need to make sure your grinder is filled up with filler so that it only has room to eat the components from the cargo container. I permanently keep 41,095kg of stone in my grinders. I also have the conveyer pathways separated. One direct line from the grinder to the welder and another from the stone storage to the small cargo. I also have sorters in each line that filter stone. The grinder-to-welder path blacklists stone and the stone storage-to-smallcargo whitelists stone. Hope that helps.
     
    Last edited: Dec 30, 2016
    • Like Like x 1
  9. Sinbad Senior Engineer

    Messages:
    2,788
    ill PM you my barrel setup, its fiddlyer than it looks...
    --- Automerge ---
    that makes sense. unfortunately it wont work in my creative lab. ill need to rub some survival on the floor after i modify the breech. thanks for the tips! ill be disarming pirate bases in no time. and once i can dismantle one or two and work out the lines for their reactors i might be able to gut a base in a couple of shots.

    ps, sorry for derailing the torpedo thread with talk about fancy boulder chuckers. ill stop now. yay torpedos, torpedos rock!

    on the torpedo note, has anyone ever made a torpedo built around a rocket launcher? loaded with just enough missiles to launch continuously for the last 800m, then set up to do so. it could work either as a decoy launcher, or soften up the target before the main payload hits.
    fiddliness of script aside: advantages? disadvantages? thoughts?
     
    • Like Like x 1
  10. Wellstat Apprentice Engineer

    Messages:
    212
    I should also start revamping my Sogeki Onsa gravity cannon soon. Releasing a camera raycast version of it.
     
  11. Ronin1973 Master Engineer

    Messages:
    4,890

    I'll +1 that. I test my stuff against targets... preferably other ships on the workshop that are fully armed and shooting back. I put moving targets on a remote and have them travel around 10km back and forth.

    You can claim to have an awesome weapon design by hitting a static cube of heavy armor. But not many people have a big cube of heavy armor as their base. PMW behave differently against ships versus static grids. Your awesome missile might just be a dud when it gets bitch-slapped with 20 rockets before connecting with the target.

    1. Survivability to impact/detonation.
    2. Damage to target.
    3. Cost of PMW

    Those are the factors in order of priority for me. If it doesn't reach the target, what's the point? How much damage can it inflict? Your first hit might be a surprise or the opponent might under estimate the strength of your attack. But after that, he/she is now deciding how to respond. And finally, the cost. Weapons are expensive. Weapons that destroy themselves are especially expensive. While a consideration, the cost is at the bottom. The cost of the weapon is known. The cost of losing an encounter is unknown. You could lose your ship, your whole base, your entire faction. So worrying about cost is exists. But only in terms relative to your assets. On a vanilla game, mining vast quantities of resources is pretty easy... especially in procedural asteroids. If you've added a few scripts like Talyden's Inventory Management, you need only feed ore into the base and the rest can be automated as far as levels of production.
     
  12. tankmayvin Senior Engineer

    Messages:
    2,863
    His idea is actually to make self replicating drones that will swarm to GPS coords of his choosing and destroy anything there before it has any time to respond using hundreds (400+ turrets). According to him swarms can't be beat and we're all just chumps that will sit there being shot at.

    His drone turrets somehow will beat everyone else's turret systems and do so much damage that they won't have time to jump away.

    My only cost criteria is that on average, the missile strikes must do more damage in components to the enemy than they cost me to build. Which is usually not hard to meet because of the cost of large-large thrusters, reactors and the like.
     
  13. Ronin1973 Master Engineer

    Messages:
    4,890
    A large drone swarm system... hahaha... post it to Youtube in action... (still laughing). The logistics ALONE would be a non-starter. You'd be hard pressed to build any decent drone from a static system. You would have to weld using pistons and/or rotors. How many drones.... 40? Good luck on building 40 of them in a day... on demand? How do these drones swarm a target without colliding into each other? Even with collision avoidance you'd have a traffic jam of drones trying to work out who should go where. Not to mention how much LAG 40 ships will cause to anyone near them. What do you do with the drones after the battle? That's a lot of grinding. How do you recall them? These are all problems someone who ACTUALLY who has actually BUILT and TESTED such a system would immediately know the answer to... since these logistics would give a solid answer to the VIABILITY of such a system (which in all likeliness isn't viable). You're just being trolled by someone who is pretty ignorant of the game and is passive aggressive. You do have a bit of a temper on the forums, Tank. I've seen it. I don't disagree with your sentiment, though. I just tend to filter what I say. :)

    I think costing more than the damage inflicted is arbitrary. If the ship/station is a major threat then whatever you spend to dispatch the enemy can easily be made up over a short amount of time. The loss of materials isn't at the point of detonation but when you actually build them. If you're defeated, you're probably going to lose those materials as a spoil of war anyways. If you prevail, you take his stuff.

    Unless he has a mighty swarm of self-replicating drones that can just overwhelm anything they come up against... just cuz....
     
    • Friendly Friendly x 1
  14. Cadde Trainee Engineer

    Messages:
    99
    Or drones... You know, it's not a far cry from making a drone orbit something or move to centimeter levels of precision to have them field a welder of their own to make new drones.

    Again, drones building drones is not a problem. And resources are plentiful in this game. If you are starved for resources you are doing it wrong.

    By programming them not to of course. And for everything else that isn't a drone, they can easily raycast in the direction of travel to find obstacles and find a pocket large enough to fit through.

    You have a source on this? That it's the ships that cause lag and not your shitty connection?

    Which part of "disposable" was unclear to you? I wouldn't need to spend much time at all to initiate the self replication program or even the resource gathering program. All i need is to tell drones the coordinates of a pocket of RESOURCE X and they would go there and roomba up the whole resource until there's not a drop of it left.
    Your lack of engineering mindset boggles me to be honest.

    Or maybe you deniers who clearly haven't thought of the ACTUAL limitations in such a system are the real trolls? Even if it's because of your ignorance.
    It's possible, it's just a matter of doing it IF you have the skills to do it. How many lines of code have you written in your life?

    If profit was the name of the game, i wouldn't be using drones. I would infiltrate and rape his base.
    The whole point of a drone swarm is to ruin someone's day. And with a drone swarm, the last thing i would need is to take someone else's resources.
     
  15. Ronin1973 Master Engineer

    Messages:
    4,890

    Dude, you're a troll... SHOW me... talk is dirt cheap. You KNOW how to do all of this? You have a working copy of world/ship that DOES this?

    You haven't a clue about the logistics involved. You haven't done it. You are purely conjecture. You are cobbling together things you have SEEN but haven't DONE.

    Again... SHOW ME what YOU have done and you can sit at the table.
     
  16. Timberwolf Apprentice Engineer

    Messages:
    246
    Whatever happened to good old chemically propelled projectiles?
     
  17. 3eepoint Junior Engineer

    Messages:
    926
    KEEN....

    ...ok srsl, just the standard joke, its more like that projectiles lack a mean of dependable hitting a target on long range. You can fire a PMW from 50km and it will hit if correctly guided. Vanilla turrets have a range of only 800m and do not preaim, so a moving target is not verry likely to be hit, thats another reason why pmw's are that effective, they are small and fast and inflict massive damege. A nightmare for a targeting system that can only aim AT a target and not WHERE IT WILL BE when the projectile arrives. Even if there are scripts to compensate the preaim, there are only modded guns that have more reach or rotorturrets, wich are more like Nukes in MP and also need a guidiance script and can get really clumsy.
    For things like spine mounted mass accelerators, there is currently nothing that hits the spot. When you propel rocks/ingots, you have the speedlimit and have to aim right the moment you shoot, also, thrash rmoval is a thing.... For Modded weapons, there are one or two modular aproaches, one which is broken and one that more qualifys as a shotgun on Horse stereoids or something, but not a scaleable massdriver....

    ...if the damege model in SE for collisions is realistic is a whole other thing ...
     
  18. Timberwolf Apprentice Engineer

    Messages:
    246
    At this point any long range guns that get added run the risk of being instantly obsolete unless they're better than PMWs, at which point, yay powercreep.
     
  19. tankmayvin Senior Engineer

    Messages:
    2,863
    That's the gats and the rockets.

    SE isn't a "build custom weapons from blocks game" - grav guns and PMWs are really just a happy coincidence/meta-game component that arose out of what the engine can do.

    If you want a game that was designed from the ground up as a "buid weapons from blocks" as the core mechanics, you really, really, really need to play From the Depths.
    --- Automerge ---
    I'm trying to find the Youtube video, but someone actually build a drone system that held formation and slowly made its way to a target or screened the mothership. Simspeed when combat was initiated was as bad as you would expect.

    I mean large numbers of stock turrets lag the crap out of the game without the drone overhead.
     
  20. Wellstat Apprentice Engineer

    Messages:
    212
    I believe you are referring to Rdavs fleet command. Below is one of the videos at 14:45 mark, showing live footage of the drones in action:

     
  21. Robotnik V Apprentice Engineer

    Messages:
    382
    We're getting slightly off topic here, drones aren't torpedoes. If you want to discuss drones, I recommend you make a new thread. I'm talking to everyone, not one specific person.
     
  22. tankmayvin Senior Engineer

    Messages:
    2,863
    Yeah, that's the script! But I'm trying to find a specific video where someone created a big fleet of those buggers and it turned into an epic slideshow.
    --- Automerge ---
    Everything is a torpedo when it runs out of ammunition.....or when your cockpit gets blown up and your fighter plows into the thing you were shooting at.
     
  23. Wellstat Apprentice Engineer

    Messages:
    212
    New script idea.... Divine Wind Script
     
  24. Cadde Trainee Engineer

    Messages:
    99
    At least this i can agree with.
     
  25. tankmayvin Senior Engineer

    Messages:
    2,863
    Suicide already seems to be built in to the keen drones. At the very least you have to dodge their debris, but definitely can't argue with a Kamikaze button
     
  26. mhalpern Senior Engineer

    Messages:
    2,119
    Even if you could get it to work, grinding and welding takes time, and you don't want to leave hundreds of self replicating machines on the field, you might not have a grey goo scenario, but that's because you'll crash the server long before that, welders have a limited range, so the swarm may be more about nuescence than a threat, as you wouldn't be able to armor or arm them much, that's the best case

    As for lag, moving objects have always caused lag physically impossible for them not to, in any game engine
    --- Automerge ---
    Resources may be plentiful, but refining and assembling takes time, and the true problem is getting the right resources for it to self sustain, in an automated fashion, also complex multigrid with legitimate in game tools isn't available and self replicating multigrid would be a networking nightmare in a script.
    --- Automerge ---
    In a PMW you have to consider that your typical enemy, at least for their large structures, is going to build in redundancy so simply punching a hole won't do you much good, you also want to consider cost, how long and how many resources does it take to make for however effective it is, if it makes it hard to supply my fleet with appropriate munitions it's too expensive, if I go through more platinum in a typical engagement than it would take to build an FAC, then it's too expensive, you get the picture
     
  27. FakeName Trainee Engineer

    Messages:
    34
    I like to keep torpedos as simple as possible. I have build some easy, survival and resource friendly dumbfire torpedos, anti gravity shield torpedos (launched by a gravity gun), TV-rockets (anti-gravity shield) and some mines and bombs which impact upon detonation or, in combination with a hull penetrator, which will fly into that hole and explode inside the ship.

    Usually my torpedos are not larger than a 5x1x1 or 7x3x3 slots and mostly consist out of blast door blocks (steel plates) to make the mresource friendly. They can penetarte up to 4 layers of heavy armor and in some rare (I believe it is a bug) scenarios even up to 7-8 layers.
    All vanilla, ofcourse and do not require any scripts or programable blocks so they should work on ANY server. If anyone is interested just get in touch with me. :)
     
    • Agree Agree x 2
  28. mhalpern Senior Engineer

    Messages:
    2,119
    One theory that I have had that with camera ray cast is only one step from being practical is a centripetal airburst hydrogen torpedo, what you do is put armor piercing fragments around the torpedo connected to the main body by non deforming blocks, you don't want the fragments to touch each other, and once it is close enough to the target, it will spin on its alron axis and then detonate a warhead, said warhead will destroy the non deforming blocks, releasing the flak granting a nice area effect of where you may find holes, the raycast helps with guidance and proximity triggers, but the one step is that we can't transfer hydrogen from one tank to another yet, only from oxygen generators to tanks, unless they changed that, it wouldn't need to be completely filled, and the large size of the hydrogen tanks is an advantage as you can work in plenty of connection points for the flak, the bigger problem I see is building them with a static, piston-less welding set up, as is preferred for torpedoes and other ship board manufacturing
     
  29. Ronin1973 Master Engineer

    Messages:
    4,890
    I built a weld system that functions without rotors or pistons. Due to the limited range of welders, this dictates the shape of my PMW. I also design the missiles to fit through an airtight hangar door array. The array opening is four blocks wide. The weapon can be four blocks wide, but the outer blocks can't be full height blocks else the missile won't pass through. The base of the airtight hangar doors are just a hair wider than a full block.

    If you can do it without pistons or rotors, then youo never have the issues in planetary gravity or having a PMW explode in the heart of its mothership....full of warheads sitting next to five other PMWs full of warheads. If you're playing in survival, you're more apt to blow yourself up than anyone else if you're using pistons and rotors.

    This is stuff you learn by DOING and testing designs under in-game survival settings...on a live server; rather than just predicting success based on conjecture.
     
    • Like Like x 1
  30. mhalpern Senior Engineer

    Messages:
    2,119
    For me it's a matter of time, and having other stuff to do, and being limited to a laptop, but I try to keep up to date enough to understand the challenges keeping the system mechanically and logically simple and basing much on things that I have seen work improves the chances that it will in fact work, but that is why I specify it as a theory, because the difference between theory and practice is that in theory everything should work.
     
Thread Status:
This last post in this thread was made more than 31 days old.