|
Post by bdcarrillo on Nov 9, 2016 20:04:36 GMT
I still haven't achieved satisfactory/combat effective results.
Without a 10% mass fraction warhead I can achieve through-the-same-hole accuracy. Stick a payload on and it goes to rubbish on terminal. Tumbles, overcompensation, straight up misses on static targets.
Still having issues with mid-course corrections not occurring. I expect to see a nudge or two but nothing happens.
I'll put up a pic or two tonight of my test missile, if I can dial it in a bit better.
I also noticed that automatically fired missiles come out of the ship nose first (perpendicular) whereas commanded launches kick out parallel to the ship. Kinda seemed odd to me.
My other observation, so far, is that these guidance strategies aren't going to be a one size fits all; they'll likely need to be customized per missile.
|
|
|
Post by ross128 on Nov 10, 2016 1:39:44 GMT
The big puzzle now: what combination of missile characteristics and homing pattern can get a missile through a 1000km engagement range? One thing I've noticed is self-igniting propellants (like F/H) have a tendency toward being unable to turn their motors off, while propellants that require an ignition source tend to leak a bit of fuel before shutting off, unless under laser fire in which case it'll fail to shut off (laser acting as ignition source?). I'm guessing that the acceleration overwhelms whatever shutoff valve that may be present, so the missile can only cut the engine by cutting the ignition source. Only when the acceleration has stopped can the shutoff valve be closed (hence the delay even when successful). Fuels that require ignition may also find themselves unable to shut off if the engine is running too hot (the engine needs to cool off fast enough that it doesn't auto-ignite the fuel with its own heat). That'll be a major blow to fluorine-based propellants: without the ability to cut the engine, those will be limited to short-range only. O/H is capable of engine shutoff (tested in-game), so it's a strong contender for long-range. So the question is, what kind of engine characteristics, how much dV, and what homing pattern do we need to prevent a 1kw/1000km infrared laser from granting immunity to missiles? Edit: What seems to be working well for me so far is around 6g of acceleration, 6k-8k dV, and a 20% dV APN boost phase, unguided coast (to make sure the engines cut), and PN terminal phase with 2.0 damping. At 6gs the fuel cutoff valve doesn't have any problems actually cutting the fuel flow, which means at that low acceleration I can use F/H without getting runaway rockets. That setup can get all the way through a 1000km engagement range and still hit, though it requires a lot of patience and if the laser can actually do damage that far out, quite a lot of missiles will get fried on the way. KKVs can get quite hilarious when you give them that much room to accelerate though.
|
|
|
Post by dwwolf on Nov 10, 2016 11:19:33 GMT
I wonder what kind of gimble angle people use on their missile engines. I'm starting to think its more important than people realise.
|
|
|
Post by dragonkid11 on Nov 10, 2016 12:33:05 GMT
I go for 5 degree gimble angle but it seems even that is not stable enough.
|
|
|
Post by apophys on Nov 10, 2016 12:51:33 GMT
Maybe slower gimbal rotation?
|
|
|
Post by redparadize on Nov 10, 2016 14:12:15 GMT
Try lowering trust. too much G or too quick turn time seem to be a problem.
|
|
|
Post by bdcarrillo on Nov 10, 2016 14:38:01 GMT
Try lowering trust. too much G or too quick turn time seem to be a problem. Unless you have a very anemic engine or very massive missile, your thrust to weight ratio (G) will always skyrocket on terminal. Either situation would likely result in a drastically slower missile that would get swatted like a fly on approach. Fast, durable, cheap... Pick two kinda situation.
|
|
aiyel
Junior Member
Posts: 83
|
Post by aiyel on Nov 10, 2016 15:14:38 GMT
One key thing I discovered was that the homing success depends heavily on the missile cg and engine gimbal. I had many missiles tumble around at mid or terminal phases until fine tuning the missile itself. It seems like the remote control constantly overcompensates, regardless of settings. That's where I had to physically limit the missiles ability to respond. I had a tough time figuring out the dampening... Wildly different ratios didn't seem to make obvious differences. I did some reading on the homing techniques, which helped me figure out the different strategies. An in-game tooltip blurb would be good. IIRC most guided weapons tend towards this - the control surfaces (gimbals for us) are either neutral, or all out in one direction or the other, meaning the bomb isn't really "on target" except on average over time (and hopefully when it detonates/impacts, since being on target on average over time means it won't e far enough off target then to miss). It's not so much most weapons. Guided missiles, for example, tend to heavily favor proportional guidance to conserve their momentum. The only systems I know of that do this are laser guided bombs and certain categories of laser guided missile (those being laser beam-riding rather than spot seeking) and it's mostly a cost-and-weight saving measure with older tech.
|
|
|
Post by cuddlefish on Nov 12, 2016 3:47:15 GMT
I still haven't achieved satisfactory/combat effective results. Without a 10% mass fraction warhead I can achieve through-the-same-hole accuracy. Stick a payload on and it goes to rubbish on terminal. Tumbles, overcompensation, straight up misses on static targets. Still having issues with mid-course corrections not occurring. I expect to see a nudge or two but nothing happens. I'll put up a pic or two tonight of my test missile, if I can dial it in a bit better. I also noticed that automatically fired missiles come out of the ship nose first (perpendicular) whereas commanded launches kick out parallel to the ship. Kinda seemed odd to me. My other observation, so far, is that these guidance strategies aren't going to be a one size fits all; they'll likely need to be customized per missile. This seems like it might be a center of mass issue, if adding a payload is causing serious performance shifts. Try adding some ballast elsewhere and see how that changes things.
|
|
|
Post by bdcarrillo on Nov 12, 2016 22:00:04 GMT
I'm still tinkering with the design and know that a CG shift is likely causing the weird tumble on terminal. I've had decent luck with half degree gimballing at minimum speed.
I did find a potential bug... Commanded missile launches with a 100% fuel burn on boost/mid flew just fine even past emptying their tanks, but automated launches constantly disabled themselved when they ran out of delta-v. This ocurred in the same combat.
Just for reference: Pursuit- straight at target Deviated pursuit- angle slightly ahead of target (hard coded angle unk) Proportional Navigation- adjusts based on difference in vectors to lead target Augmented Proportional Nav-PN with additional factors relating to correcting for target maneuvering
|
|
|
Post by tessfield on Nov 13, 2016 20:14:28 GMT
Ugh, has anyone an idea how to avoid missiles exploding just because they got close to a drone? I have a flak missile which has a hard range of 50m and activation range of 500m and while it performs just fine against a single cap ship, the second drones come into play, they start blowing up prematurely.
Launcher is set to Target Ships, Flak Modules are set to Target Ships.
I guess drones aren't considered shots...? Any ideas how to avoid this?
EDIT: Lowering detonation range fixed this, as expected, though not what I wanted to do.
On a related note, has anyone managed to create a non-nuclear-explosion-based-missile for destroying drones and/or missiles?
On a related note, maybe a bug, it looks like flak is expanding in 1D rather than 2D, which makes anti-something-tiny flak missiles really inneffective. Can anyone confirm or is it just me? I'm using a group of 7 1800x5g flak, then another group of 3 on top of that (for a total of 90kg flak and 10kg explosive).
|
|
|
Post by jageriv on Nov 14, 2016 3:44:43 GMT
Hm, what problem exactly is this fixing? Are people getting more accuracy out of their missiles? Is this somehow easier on the processor?
While this does seem neat, I'm not sure why this was needed now.
|
|
|
Post by someusername6 on Nov 14, 2016 4:00:53 GMT
Hm, what problem exactly is this fixing? Are people getting more accuracy out of their missiles? Is this somehow easier on the processor? While this does seem neat, I'm not sure why this was needed now. The whole idea of module and ship design seems to be to be able to co.pare various designs and find ones we like best for a variety of scenarios. Turns out that missile guidance logic has a measurable effect on missile performance and it is easier to expose, rather than always follow a hard coded behavior. It is one more thing we can optimize, and while I don't know how to do it yet so I am a fan of the option.
|
|
|
Post by jageriv on Nov 14, 2016 7:03:40 GMT
Oh, certainly, it looks like it might be something cool and nice to have, assuming it does make a meaningful difference in game (haven't really had the chance to experiment and see). I was just wondering why this had priority over some other things, is all, and what immediate problem was getting fixed with this.
|
|
|
Post by dragonkid11 on Nov 14, 2016 8:48:04 GMT
jageriv, have you played the game before patch 1.0.7?
Because before the patch, missiles are utterly crap in actually homing onto the target. The only way to hit anything with missiles back then is with sheer luck or sheer firepower.
|
|