|
Post by jtyotjotjipaefvj on Sept 25, 2017 16:43:07 GMT
I managed to beat three gunships with this highly technical design: it technically has armor and technically it has a weapon in addition to the missile launchers. Design blueprints for ship, launcher and missile: My original plan was to do a high velocity intercept to avoid missiles and quickly spit out the full missile storage and then use the high acceleration to escape but that turned out to be unnecessary. Since the micromissiles can take down flak missiles easily enough, I can do a slow approach and hover at 150 km, slowly launching missiles when needed. With better rationing and some luck, the missile storage might be able to take down a fourth gunship too, but I can't be bothered to optimize so much. Below is a combat video in two parts since I accidentally stopped recording after the missiles were done and I can't be bothered to glue the clips together. Proof of PD capabilities: Proof of Gunship-destroying capabilities: Weirdly enough the missiles seem to have better accuracy against flak missiles. You'd think the gunship would be an easier target to hit but you'd be wrong! As for points, I think I earned 3 kc * 1.03 km/s * 858 Mc = 2.65 * 10^12 c²m/s points. For secondary score I got 3.01 g * ~10 kg * (2 or 6 depending on how you count) = 590 kgm/s² = 590 kN. Or 1.77 MN if each launcher counts as a separate system. Your point system is pretty silly btw.
|
|
|
Post by jtyotjotjipaefvj on Sept 24, 2017 12:54:29 GMT
I made a missile that has twice the payload, 600 m/s more dv and more armor as well as a slimmer diameter than the current winning design, while costing a bit over half as much. I'd be surprised if anyone was able to beat this with a NTR design, since Fl-H rockets are so cheap by comparison. With Fl-H the engine costs nothing and the fuel is cheaper than methane as well. It has 1 cm of graphite aerogel with an added 0.5 mm PTFE cap in the front for extra laser protection. It should survive lasers quite well as long as it faces the laser, since it's so slim and the PTFE cap should survive a few glancing hits. For the engine I used a standard AE F-H engine modified a bit to make it slimmer. relevant stats below: Mass: 22.4 kg Cost: 33.6 c dv: 10.6 km/s acceleration: 13.6 g -> 116 g burn time: 33.1s turnabout time: 0.898s diameter: 14.5 cm cross section: 406000 mm² payload: 1 kg TNT-Lead flak bomb Edit: here's a video as proof that the missiles actually work and can kill things dead.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 22:34:08 GMT
That's a pretty nice design. KKVs are fun.I might suggest upping your fuel flow rate so you get even shorter burn times at higher thrust, no point in having any fuel left when it hits. As for bringing down the price on my missiles, I've been tweaking my nuke designs. Sadly, I doubt I can get much more boom for my buck, my 12 kt nukes mass 52 kg and cost 2.16 kc, and that's probably the best I can get. Similarly, my 10 kg flak bombs use Nitrocellulose for the charge and oddly, platinum for the shrapnel, as it's the best balance between cheap and dense that I can find. They're maybe 150c apiece. Oh, another thing: Is there any reason to use Aramid Fiber instead of Boron Filament? It just seems way more expensive for very little difference. Just switching the partial Aramid layer in my design for Boron filament cut the price by a round a kilocred. You can make nukes both lighter and cheaper than yours. See here for reference: As for aramid vs boron fibers, I haven't noticed much difference between them either. I always use either boron filament or spider silk as my spall liners.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 22:05:41 GMT
So your comment about quantity over quality got me thinking, resulting in this design: It's a dirt-cheap KKV with 10 km/s dv, 340c per missile and 13.5 kg dry mass. It's enough to punch through the armor of a gunship in one hit, and you can fire as many of them as you like without running out of money. Which is what the above ship does. It has 24 launchers each firing 6 missiles per second, resulting in a good rate of 144 missiles launched per second. It carries 3000 missiles but still costs a bit under 9 Mc. Below is a video showing how it works against a gunship. Even though 90% of the missiles launched missed completely, there's still enough hits to take out the gunship in one 2-second salvo. Plus the cloud of missiles coming out of the launchers looks pretty great.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 21:04:55 GMT
Fair point with the size there. Here's a different version with 3 fixed-mount methane NTRs and 5.4 km/s. It's a bit bigger than yours but slightly cheaper and lighter. You can't really go cheaper than this, the propulsion parts only cost 500c, with most of the cost coming from the nukes. You could shave off around 70 kg by switching to a hydrogen deuteride fuel but then you end up with doubled fuel volume. So really it depends on what kind of tradeoff you want between dv, cost and fuel volume. Also, remember that shrapnel energy increases with the square of missile velocity, so this one would have around tripled fragment energy compared to yours. The added speed also decreases the time enemy PD has to kill missiles before they hit, which means more missiles make it to the target. At 20 km/s, missiles are pretty much impossible to stop completely, since even if you take out the controller or warhead, the missile has enough dry mass to punch straight through pretty much any armor you can have. They also arrive quickly from hundreds of kilometers away so you can launch them from a lot further away in combat. I rarely use missiles that go under 10 km/s, but that's mostly just personal preference, I don't know if it's actually optimal.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 20:00:08 GMT
Hydrogen Deuteride is also a good propellant for heavy payloads. I was able to double your dv while adding only 10kg of mass and not quite doubling the price per missile, so I'd consider that an overall improvement. The only downside is the increased size but if you're going twice as fast it might not be an issue.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 17:10:38 GMT
Maximum cost, mass and burn time do nothing atm? It would be great if something like this could be integrated into the game, together with graphs. They should work fine. They only break if no configuration fulfills the requirements, in that case they just show some engine that happened to be the first in a list. The way I implemented constraints isn't very robust, so it can't deal with the case that no engine is valid.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 16:57:24 GMT
I added in support for tank mass and improved the suggestion display a bit. Below is a bunch of screenshots with the spreadsheet and ingame versions of the suggested missiles. The tank mass is still not quite correct but it looks to be pretty close. I'll see if I can find the error in my calculations. It also doesn't take the tank wall costs into account currently, which should be trivial to implement. Screenshots:
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 14:38:24 GMT
fgdfgfthgr's propellant data sheet gave me an idea for a ship calculator that can run through many different engine configurations for a given payload and dv budget. You can force additional constraints for maximum ship cost, mass and burn time as well. It's still fairly rough and limited in the engine data department but you should be able to customize it easily on your own. This should allow you to design new ships faster, for example if you want a light missile with high acceleration and 10 km/s dv, you can quickly see what the performance would be with different thrusters without having to manually try out all combinations. If you know your payload mass, eg. the mass of armor, control modules and weapons, just plug them into the calculator and see which engine gives the performance you want. Here's the link to the newest calculator version: drive.google.com/open?id=1VKWwRyyJELmRPPUvF2YEJRr7kD2gSxd2iBvYnRzLrboTests for two different cases in spoilers below to verify that the thing works correctly. First a test for a 50-man craft with 20 km/s dv, so that both MPDs and NTRs show up in the recommendations. calculator output: ship in-game: Next, a more extreme 100 km/s test to get the GW MPDs to show up. calculator output: in-game:
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 13:47:57 GMT
I made a quick mockup of such a calculator: It uses the three different types of NTR in AE modules, with thrust ranging from 10 kN to 10 MN. It's fairly easy to add in new engines and propellants but not quite zero effort. I might add in some way to input a cost function or choose what type of ship you want to make and have the sheet give you a suggestion based on that. Not sure how exactly I'd do it though.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 13:15:57 GMT
It's under BLUR and BLUR_CROSS in the PostProcess.frag. Modifying those seems to do the job. Any pointers on what exactly I should change, please? Flailing around there have only resulted in the game not launching so far... You have to make a valid GLSL shader. If the shader compiler produces any errors, the game fails to launch completely with no hits at what caused those errors. You could look up some shader tutorials for openGL but it's a terrible way to do it if you don't already know how to write GLSL.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 11:31:45 GMT
It would be useful to have total mass required, mass ratio and fuel volume for the different dv brackets. Another useful feature might be a calculator where you input required payload mass and it gives this table for the given payload so you can choose your propellant based on whether you need a small, cheap or light missile for example. I want to have those functions, too. But how? I guess you'd need to record engine mass and cost by hand. After that you could compute dv and fuel mass from the required mass ratio for a given EV, then compute fuel volume and cost and add them to the engine's properties. It would be even better if there was some way to compute mass and cost for the engine automatically but I guess that's not easily doable. Recording the engine data by hand would be a hassle, but it could be made easier if you chose just a few propellants known to be good and design a number of engines with different thrust levels. Or just used Apophys's module collection.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 10:45:46 GMT
1. I use boron nitrate, so this sheet is boron nitrate, this table was not designed to give the best performance to all people but myself. 2. There is no unit in my xkm/s cost part. It's just relative value. 3. Neutrons did not consider because this will lead me into hard work. 4. You can give me a better equation or some enhance on readability, if not, I can't improve this table anymore. It would be useful to have total mass required, mass ratio and fuel volume for the different dv brackets. Another useful feature might be a calculator where you input required payload mass and it gives this table for the given payload so you can choose your propellant based on whether you need a small, cheap or light missile for example.
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 10:28:26 GMT
Even after logging in it says I don't have rights to access this. Either way, not maximising EV sounds like a terrible idea if you want to choose a propellant. EV and optimized TWR are really the only things that matter. Maybe propellant density and price too, but they're usually secondary. 1. You should be able to access now. 2. The EV didn't maximise mean I didn't maximise it to the highest temperature an NTR can have, but the highest temperature of Boron Nitrate, I use Boron Nitrate to make the rocket. You should use at least titanium diboride, maybe even tantalum hafnium carbide as the control rod. This information isn't really super useful if it doesn't show the best performance of each propellant. I'm not a 100% but I don't think reactor temperature affects thrust or TWR linearly, so the order of propellants might well change if you optimize the NTR. Another note is that 30 RPM is rarely the optimal choice for pump RPM. You should minimize the engine mass for the given mass flow rate and maximize the exit velocity to get the maximum performance for each fuel. I'm also not sure if you take into account how propellants affect reactor performance. The heavier propellants tend to make reactors terribly inefficient since they absorb neutrons between fuel rods. I think. For example, here's a hydrogen deuteride rocket that is cheaper than any of your 5 km/s dv rockets but gets 11.9 km/s dv and a fairly ridiculous acceleration too. According to your table I'd need 16 kc to get even 9 km/s which is clearly untrue. I'm not sure how other propellants would relate to this but you can see how your data could be very misleading. Designs in spoilers:
|
|
|
Post by jtyotjotjipaefvj on Sept 23, 2017 9:22:53 GMT
I've noticed that setting bloom to 1 does the job quite well too.
|
|