|
Post by blothorn on Sept 30, 2016 22:15:14 GMT
Automatic re-engagement would be nice--usually once I fire missiles, I want them to keep going until they engage or run out of DV forcing the enemy to dodge; I am never going to say "oops, that interception is too expensive, going to let them carry on out of orbit and hope they find an interesting target in solar orbit."
That said, after an enemy adjustment you are usually not too far off interception; if your missiles have even a modest reserve of DV you can often just wait until you are close to change course (by which point dodging is going to be too expensive, or even require more acceleration than a capital ship has). Readjusting course every 10m is going to a lot of trouble to eke out a tiny advantage, which it sounds as if you do not need. You can also direct your fleet to a near-interception and then launch missiles from a few hundred km. Or you could exploit AI limitations and put your missiles on an almost-but-not-quite interception course at the start and then burn for the intercept at the last minute; I have never seen the AI dodge away from a non-interception course.
Also, I think this goes without saying but just in case, if you are playing DV exhaustion games with missiles you should never have more than one volley in flight at a time.
|
|
|
Post by blothorn on Sept 30, 2016 20:41:41 GMT
Interesting---are you dispensing with the detonator to achieve that projectile weight? Or was my understanding that you need a control module for proximity detonation to work incorrect?
(And do those results against laser frigates hold if you turn their laser range up to that of the frag guns? Weapon range limits are really beginning to irk me---I can think of no earthly reason you would want to hold fire until you close to your "effective" range, when it is extremely likely that you will be dead before that happens.)
As far as resistance to lasers go, a handful of us have concluded that you want armor with very low thermal conductivity--I forget whether turrets have access to silica aerogel, but otherwise (or if you want something with a modicum of effectiveness against kinetics) look at the ceramics. However, with a coilgun I suspect your primary risk is the coil itself, which has far fewer options. (Why do we not have the option to choose the coil armor material?)
|
|
|
Post by blothorn on Sept 30, 2016 19:35:35 GMT
A note on thermal conductivity. Having a high or a low conductivity is better against lasers in various cases. If the laser is slow to burn through your hull (i.e. low irradiance caused by low focus, long range, or low power), having a high conductivity is beneficial. This is because the armor can radiate the heat away to nearby armor tiles and keep from ablating effectively. Now consider a laser with a high irradiance. If it has a high enough irradiance, it will ablate your armor whether you want it to or not, so a low conductivity is better, because it keeps the heat from penetrating very far through the material. In other words, it prevents the heat from spreading, so the laser, no matter how high power, can only boil away surface layers at a time. Yes---this is a very powerful laser relative to the size of the targets, which definitely biases things in favor of the low-conductivity armor. However, my thinking is that you generally want to minimax defenses: a strength does not compensate for a vulnerability. I think that in most situations, I would trade increased damage from underpowered lasers (that are not going to kill me anytime soon, regardless) for a bit more time against a more powerful laser. Missiles, as noted, only have to survive seconds anyway. For a capship, being able to take weaker fire without taking permanent damage sounds a lot more attractive, but CDE combat is fast and brutal. If the armor buys me a minute or two of life, I shall not complain much about its state afterward. Interestingly, I tried somewhat the conductive+non-conductive composite; putting basalt outside the carbon was decisively superior than the other way around. My theory is that while the carbon was intact, it was spreading the heat across the surface of the basalt. Since the carbon has a higher melting point than the basalt, this increases the surface area of basalt being heated, allowing much more efficient transfer of heat to the basalt. To get conductive-over-nonconductive armor to work, I think you want the melting point of the nonconductive layer to be well above that of the conductive layer, in the hope that in the face of a powerful beam the conductive layer will ablate away before it has much chance to damage the nonconductive layer. Unfortunately, you really want a high melting point in conductive armor, and I cannot think of a good nonconductive armor with a high melting point: the only materials with conductivity below 20 W/m^2 and melting points above 3000K are relatively dense metals with fairly unattractive profiles otherwise. This does mean that you probably do not need to worry about putting a low-melting-point, high-conductivity ballistic armor over a medium-melting-point nonconductive thermal armor (e.g. spider silk over basalt), but do not expect the outer armor to do much other than get out of the way against a laser. Nonconductive over conductive is silly from a thermal perspective---you only take advantage of the strength of conductive armor against weak lasers after they have burned through the nonconductive armor, and then the nonconductive outer layer will prevent the conductive layer from radiating heat outward. Nonetheless, I suspect it is the way to go if you want to combine a weak nonconductive thermal armor with a high-melting-point conductive material that is stronger against kinetics. (Although then you need to worry about the nonconductive material impairing the anti-kinetic ability of the interior layer; I suspect that in some cases it would help stabilize projectiles that would shatter or ricochet, similarly to the soft AP caps on old naval projectiles.)
|
|
|
Post by blothorn on Sept 30, 2016 18:19:40 GMT
I have---I prefer nukes because with present fuse options, near-misses with flak are nearly useless, whereas a near-miss with a 106kt nuke will shred anything exposed (notably engines). And while nukes are bad at dealing internal damage they strip off Whipple shields far better than flak, so I suspect you can get a great synergy with a nuke launcher to strip poorly-armored exposed components and shields and a high-velocity sabot gun to finish the enemy off with internal damage.
|
|
|
Post by blothorn on Sept 30, 2016 10:17:46 GMT
My bid for "lightest 10kt nuke": And for when you need a nuke you can use to cheat at bowling: (Although I think that at this point making smaller nukes is a bit silly; the detonator and armor begin to outweigh the warhead.)
|
|
|
Post by blothorn on Sept 30, 2016 9:22:28 GMT
Several more tests: 7.9mm spider silk averaged 5 seconds, 7.2mm aramid fiber 208, 1cm UHMPWE 48, 6.9cm silica aerogel 238. The aramid fiber and silica gel both got all missiles past the target, roughly equalling the performance of basalt fiber. Distinguishing among those three will take a more difficult test. (Note that times in this test are a roughly log scale of effectiveness: spider silk-armored missiles died in a second apiece to a 350MW/m^2 beam at 30km; the missiles that passed the target survived a 20GW/m^2 beam at less than 3km). The terrible performance of spider silk seems to disprove the theory that being fibrous is advantageous. Note, however, that spider silk has a fairly good thermal conductivity while aramid fiber has one of the lowest of a solid material (comparable to basalt fiber) and silica aerogel the absolute lowest. Similarly, UHMPWE performs on a level with carbon materials with much higher melting points and has ~10x lower thermal conductivity than graphite/amorphous carbon. This seems a fairly decisive argument that low thermal conductivity is a huge advantage. What to look at next? There are a bunch of ceramics with low conductivity, but all have poor tensile strengths and I would be concerned about their not pulling their weight against kinetics (although that is where I would look for projectile armor, as they are much less likely to see kinetic impacts, and using the expensive fibres seems wasteful). Several fibers combine superb tensile strengths and low conductivities, but all are ruinously expensive and cannot be mounted on turrets. It would be nice to have comparable data for kinetics; the "Raw Steel" made it sound as if the ultimate tensile strength was about the only thing that matters, but if the high shear moduli of the ceramics make them respectable armor (which I think they should be) they bear more investigation. I also definitely want to test capship scales (centimeter thicknesses and larger surface areas). It would also be interesting to play things out at longer ranges; I suspect few things other than missiles will want to come within 30km of a 100MW laser, and it may be that a lot of the mid-range materials are perfectly adequate at lower intensities (where conductance and re-radiation can offset a much greater portion of incoming heat). Here is the laser. Intensity at the starting range (just beyond 30km) is around 350MW/m^2. I welcome any attempts at replication with other colors.
|
|
|
Post by blothorn on Sept 30, 2016 6:59:34 GMT
Nifty. That may make nuke coilguns properly practical---one shooting mine is absolutely devastating, but not at all compact. I do wonder how they get 20000 kg/m^3 of fusion fuel in there...
|
|
|
Post by blothorn on Sept 30, 2016 5:43:24 GMT
People want pocket nukes?
|
|
|
Post by blothorn on Sept 30, 2016 3:46:52 GMT
See the attached crash log. Interestingly, the first time this happened the game did not save user designs and I lost most of the original components. I made a similar (but slightly different) engine, reactor, and radiator, rebuilt the ship, made a new launcher---and it crashed again. The only thing in common between the two was the general configuration and the laser, and I got another crash after dropping the laser.
The design is the "Solar Flare" drone, found on the "Light Carrier" warship.
|
|
|
Post by blothorn on Sept 30, 2016 2:33:56 GMT
Also, thermal insulation for cryogenic propellants is probably a very different problem in vacuum.
I think the big reason to use something other than UHMWPE in-game is cost---a random tank was 250t/398kc in UHMWPE and 251t/337kc in boron. Exotic materials can probably make a much bigger performance/price impact in more critical applications. I am also not certain whether you need to worry about your armor heating up and melting your tanks with radiant heat (probably a particular concern for missiles, if it does happen).
Edit: but yes, completely ignoring chemical reactivities is annoying. Having to worry about flourine reacting with the tank walls (and possibly your ship if the tank ruptured) would make flourine/methane propulsion much better balanced.
|
|
|
Post by blothorn on Sept 29, 2016 23:32:26 GMT
You need to go to the ship designer and make a payload containing it. Be sure to add some good thermal armor; I have no idea what the stock nuke has, but the reinforced c-c a lot of stock designs use is simply terrible.
|
|
|
Post by blothorn on Sept 29, 2016 22:44:15 GMT
That is mildly terrifying. How effective is it in practice?
|
|
|
Post by blothorn on Sept 29, 2016 20:33:48 GMT
Back of the envelope, firing a 12.7kg projectile at 1km/s every 2.5s requires 5.08MW, so you might want to start by upping power.
|
|
|
Post by blothorn on Sept 29, 2016 20:28:56 GMT
I agree that AI vs AI is suboptimal---but I think it much better than human vs AI. The former involves working around the AI's limitations, the latter exploiting them. A lot of strategies I have seen only really work because the AI is stupid (decoying missiles with low-value targets) or does not exploit the engine fully (in particular, not ignoring range).
I would like to see a scriptable AI--it would be a fair bit of work, but would allow crowdsourcing the (far harder, I think) problem of actually developing good AI.
|
|
|
Post by blothorn on Sept 29, 2016 18:56:49 GMT
I am very excited about the possibility of tournaments, but I agree that human-vs-AI is probably unworkable. Specifically: * Autocalculated kinetic ranges are much lower than effective, and AFAIK the AI never uses "ignore weapon range". * It is easy to get AI missiles to engage decoys where a player would probably just scatter--and if not, the AI has weaknesses in handling missile thrust to get hits on maneuvering targets. Put those huge advantages together with more subtle optimizations, and I suspect that most competent players would go 3:0 against everything.
What would work, and would probably take relatively little developer time, is allowing AI-vs-AI battles. That is how almost every tournament is run in the other design game I play (From the Depths), and has worked quite well. (Incidentally, FTD did get workable multiplayer when it was still a one-man dev team, although I think MP has always had enough severe bugs to just teeter on the edge of usability.)
Edit: I also think there are some rough edges in the physics calculations (particularly for reactors) that need to be hammered out.
|
|