|
Post by thorneel on Aug 1, 2017 22:11:07 GMT
What happens if you drop the fire rate (so you don't have to stop and resume fire manually) and play the opposite side, so the laser boat tries to dodge?
|
|
|
Post by jtyotjotjipaefvj on Aug 1, 2017 22:36:38 GMT
Can you even get the AI to do fast intercepts reliably? Also, I'm not sure if you can get it to ignore range that far away. I don't really feel like learning how to mod AI behaviours just to test that. Besides, I don't think dodging will do much since the laser boat needs to be head-on towards the target to fire and has a 60s turnabout time, so it can't really dodge while also firing.
As for the manual staggering, I was thinking it would be more effective at hitting the laser boat if it did try to dodge, but that didn't turn out to be necessary. Also, you could up the ammo count a lot without affecting the ship's performance in any relevant way.
|
|
|
Post by apophys on Aug 2, 2017 7:13:21 GMT
after some tweaking, I was able to create a drone that can destroy the laser boat without dying itself. The secret lies in its two meter thick diamond armor plate that covers only the nose of the ship. It provides a thick enough obstacle for the laser that the drone can empty its ammo bin without losing too many cannons, and can survive until the laser ship is torn into bits by the cannon fire. [...] hopefully finally proving that lasers are not the only good weapon in the game, as some people seem to think. This is excellent; designs along this line might actually overturn the meta. princesskibble and jasonvance will want to see this. I agree that shooting through armor is acceptable due to the IRL ability to poke small holes to shoot though. Orbital mechanics will make the hole not in line with the laser beam, so the gun is in fact protected by such a setup. Do note that the drone design as presented makes use of a partial armor bug: your 2 meters of frontal diamond is almost not counted for cost and mass. A 4.69 m radius, 2m thick diamond rad shield is 10.5 Mc. Also, you would probably get better armor cost-effectiveness with selenium; that would be 1.25 Mc for the same size. Still, it does a successful kill of a competitive laserstar at 1 Mm without dying, and that's a very big deal. It's the first kinetic design I've seen to demonstrate doing so. I will need to try to beat this with lasers now. Can you provide the *.txt file here? (P.S. Your current statement of lasers being not the only good weapon is a lot more reasonable, and supported, than your prior statement that lasers are proven to not be the best weapon due to a fight between two hugely overpriced beasts.) Besides, I don't think dodging will do much since the laser boat needs to be head-on towards the target to fire and has a 60s turnabout time, so it can't really dodge while also firing. It cannot move at all while firing, because the MPDs would take 100% of the reactor output to use, and it has no electricity to spare while weapons are active. In-combat dodging hasn't been necessary before, so I elected not to include such a feature in the ship. A broadside design with NTRs would be able to dodge (I might have to make one of those if I can't produce a cheaper laser that would brute-force through 2m of frontal selenium). Alternately, side-mounted NTRs on a nose-forward ship would do the same, but AI in CoaDE does not handle those properly.
|
|
|
Post by jtyotjotjipaefvj on Aug 2, 2017 7:59:00 GMT
I will need to try to beat this with lasers now. Can you provide the *.txt file here? Here you go: New Spacecraft 7.txt (7.64 KB) A lot of the modules have generic names so you might run into conflicts though. Do note that the drone design as presented makes use of a partial armor bug: your 2 meters of frontal diamond is almost not counted for cost and mass. A 4.69 m radius, 2m thick diamond rad shield is 10.5 Mc. Also, you would probably get better armor cost-effectiveness with selenium; that would be 1.25 Mc for the same size. I'm not sure if it would be durable enough without using the armor exploit. I guess I could try to make a version that can carry the radiation shield dv-wise while staying cheaper. I'm also fairly sure that NTRs would make the laser star pretty much immune to kinetics. You wouldn't need much dv to survive one battle, and you can get acceleration in the several g range quite cheaply. With a target that agile, you'd need an extremely fast missile with high-velocity railguns at the very least to have a hope of hitting.
|
|
|
Post by treptoplax on Aug 2, 2017 13:02:36 GMT
I've had some success with the micro-railgun-drones I just posted in the Share Your Designs thread; they will often kill Deep Fryers and occasionally survive with as little as one 3Kc drone per 100Mw laser. There may be something funky going on with targeting or damage model for extremely small frontal area... I can repost that here with a carrier if people are interested. Going to have to try to hybridize it with this other thing now Ultimately I suspect the lasers will be forced to a broadside MPD/resistojet model, which will be much harder.
|
|
|
Post by Durandal on Aug 2, 2017 13:17:46 GMT
I've had some success with the micro-railgun-drones I just posted in the Share Your Designs thread; they will often kill Deep Fryers and occasionally survive with as little as one 3Kc drone per 100Mw laser. There may be something funky going on with targeting or damage model for extremely small frontal area... I can repost that here with a carrier if people are interested. Going to have to try to hybridize it with this other thing now Ultimately I suspect the lasers will be forced to a broadside MPD/resistojet model, which will be much harder. I know I would like to see it. Are you using a boost vehicle or do the drones themselves burn towards the target? I would also love to see broadsides make a comeback. Ive always thought they had merit.
|
|
|
Post by treptoplax on Aug 2, 2017 15:33:58 GMT
I've had some success with the micro-railgun-drones I just posted in the Share Your Designs thread; they will often kill Deep Fryers and occasionally survive with as little as one 3Kc drone per 100Mw laser. There may be something funky going on with targeting or damage model for extremely small frontal area... I can repost that here with a carrier if people are interested. Going to have to try to hybridize it with this other thing now Ultimately I suspect the lasers will be forced to a broadside MPD/resistojet model, which will be much harder. I know I would like to see it. Are you using a boost vehicle or do the drones themselves burn towards the target? I would also love to see broadsides make a comeback. Ive always thought they had merit. Surprisingly, neither (although I do think ultimately a boost vehicle would be useful and I have one I'm testing) - high speed intercept isn't required (or really even desired unless it's *really* high speed - 10+Km/s). That's the funny thing about these drones; they don't even really want to cross the death zone. They're trying to outrange the lasers (!) by being too small to hit at long range. Occasionally one even survives with the gun functional - this is frankly a little weird, and I suspect what's happening is that the distributed targeting assigns lasers randomly (not evenly!) across all drones, even the ones whose guns have already been destroyed, and a few can slip through the cracks entirely because hard-killing the drone is much tougher than just taking out the gun (which has a big block of nitrile rubber behind it).
|
|
|
Post by treptoplax on Aug 2, 2017 16:02:25 GMT
I will need to try to beat this with lasers now. Can you provide the *.txt file here? Here you go: A lot of the modules have generic names so you might run into conflicts though. [snipped] There's a design-name-appending tool linked from this thread: dedup/renamer you may find useful for exports.
|
|
|
Post by jtyotjotjipaefvj on Aug 2, 2017 16:55:59 GMT
I managed to improve my previous design by quite a bit. I changed the conventional cannon into a 10 km/s 4gram 1MW railgun, meaning the target gets less time to dodge. I also slimmed the design down by switching to methane fuel and making the drone longer. This cuts down on head-on surface area, which means the armor cap should be lighter too if I switch to a design that doesn't use the partial armor exploit. Another improvement is the increased acceleration, which means that I no longer need to do a high-velocity intercept at the start of combat. Since the drones reach 12 km/s in 18 seconds, they can start stationary and still reach full speed before they've suffered any significant damage. The launcher can field 20 of the drones for ~2Mc, and the drones are durable enough that I didn't lose a single gun when destroying one Deep Fryer. I'm pretty sure 20 drones could kill 4 deep fryers simultaneously without major losses, since the 4 railguns would each target one laser boat. Here's an export of the design: Laser removal lamp.txt (6.89 KB) Picture links of the designs if you don't want to try it for yourself: dronelauncherThere's a design-name-appending tool linked from this thread: dedup/renamer you may find useful for exports. I'm on Windows so I don't feel like trying to make that work. Maybe I'll make a windows version at some point, that seems like a useful tool to have.
|
|
|
Post by treptoplax on Aug 2, 2017 17:03:46 GMT
I managed to improve my previous design by quite a bit. I changed the conventional cannon into a 10 km/s 4gram 1MW railgun, meaning the target gets less time to dodge. I also slimmed the design down by switching to methane fuel and making the drone longer. This cuts down on head-on surface area, which means the armor cap should be lighter too if I switch to a design that doesn't use the partial armor exploit. Another improvement is the increased acceleration, which means that I no longer need to do a high-velocity intercept at the start of combat. Since the drones reach 12 km/s in 18 seconds, they can start stationary and still reach full speed before they've suffered any significant damage. The launcher can field 20 of the drones for ~2Mc, and the drones are durable enough that I didn't lose a single gun when destroying one Deep Fryer. I'm pretty sure 20 drones could kill 4 deep fryers simultaneously without major losses, since the 4 railguns would each target one laser boat. Here's an export of the design: Picture links of the designs if you don't want to try it for yourself: dronelauncherThere's a design-name-appending tool linked from this thread: dedup/renamer you may find useful for exports. I'm on Windows so I don't feel like trying to make that work. Maybe I'll make a windows version at some point, that seems like a useful tool to have. You can use it directly from web using the provided link! Requires cutting-and-pasting the whole design, which is a bit annoying but doable. It's a bit of a kludge but a lot better than doing it by hand.
|
|
|
Post by jtyotjotjipaefvj on Aug 2, 2017 21:49:53 GMT
I did a quick test with a 1.7m thick diamond radiation shield, and it looks like the drones are effective even in a completely legit loadout. The shield brought the combined price up to 11 Mc though, so I'll need to do some further testing to see how cheap I can make them. Here's an image showing one surviving drone. The turrets aren't even protected by the shield, but they still survived fine with their own diamond armor.
|
|
|
Post by jtyotjotjipaefvj on Aug 2, 2017 22:58:36 GMT
After some more tweaking I managed to beat the laser star with just ~2Mc worth of drones, including the carrier. I only lost 3 drones out of 15 used. You can see the test run here: I switched to a Boron Nitride radiation shield for the front armor plate, as it seems to offer quite good laser protection and doesn't cost or weigh too much. I wouldn't be surprised if there are better materials for that, but this seems to work well enough. A 31 cm thick plate can keep the drones alive until the projectiles hit home, and the drones cost "only" 114 kc a pop. I also placed the rail guns in front of the armor cap to make the design completely exploit-free, at least as far as I can tell. They should be able to handle armored laser boats well enough too, although the railgun might need some tweaking to make it more accurate in that case. Mobile targets might be a little more problematic, though that should also decrease the laser boat's wattage per credit, allowing more drones to be used against it. Design .txt here, now with tagged modules to avoid conflicts: Laser removal lamp export.txt (7.54 KB) Here's a better picture of the drone, in case you just want to see the layout in a better light than in the video:
|
|
|
Post by Durandal on Aug 3, 2017 1:22:04 GMT
I did a quick test with a 1.7m thick diamond radiation shield, and it looks like the drones are effective even in a completely legit loadout. The shield brought the combined price up to 11 Mc though, so I'll need to do some further testing to see how cheap I can make them. Here's an image showing one surviving drone. The turrets aren't even protected by the shield, but they still survived fine with their own diamond armor. I would say this one would be exploit free, and it should still add protection to the guns instead of mounting them on the nose. A shot showing a closeup of the barrels level with the hull would help, but it looks like the guns are mounted so that they fire clear of the forward armor. Thought about adding a large spacer so it's got an armored nosecone to defect sandblasters? A rad shield bulkhead would still protect against getting hulled by lasers.
|
|
|
Post by treptoplax on Aug 3, 2017 3:19:39 GMT
Ok, here's a glass-cannon-killer I'm pretty happy with. As noted, the basic idea is pretty simple; a railgun drone so small it can't be accurately targeted at long range. Here's the drone: And here's a carrier: mincarrierX.txt (15.29 KB) The carrier has some bonus missiles and little point-defense guns; it's a stripped-down version of my tournament entry I had lying around. Operating on AI with Launcher or Aggressive/Reckless settings I haven't been able to defend against it with a single Deep Fryer (although a little extra combat acceleration would probably go a long way). It's probably possible to do a bit better with gun/armor/engine, although they're all moderately optimized. (Note: if you're using this to test against laser ships, I suggest disabling/removing the missile launcher. It's mostly there to intercept incoming drones, and is only capable of saturation attacks on relatively puny laser arrays).
|
|
|
Post by treptoplax on Aug 3, 2017 3:29:27 GMT
I did a quick test with a 1.7m thick diamond radiation shield, and it looks like the drones are effective even in a completely legit loadout. The shield brought the combined price up to 11 Mc though, so I'll need to do some further testing to see how cheap I can make them. Here's an image showing one surviving drone. The turrets aren't even protected by the shield, but they still survived fine with their own diamond armor. [snip] Have you seen this thread of anti-laser-armor testing?: David's Fantastic Spreadsheet of Laser TestsSome weird things are really effective against lasers if that's all you care about (nitrile rubber, PTFE, aramid fiber, aerogels).
|
|