Jod

I was just about to create a topic about this. I have a 114 ton craft with 12 LT-2 landing struts. When I tried launching it with the struts starting off deployed - the LaunchPad blew up It seems fine when I launch it with the landing struts retracted though. I mean come on, it's not really even that heavy! Good thing I test all the expensive craft in sandbox before launching them in career mode, or I would be sunk. These retrievable turbojet engines and intakes cost a fortune! After playing around with it a bit more - I've devised an Ultimate LaunchPad Demolition Tool. It completely obliterates the LaunchPad and gleefully continues bouncing above the rubble, unscathed by the carnage.

I've ran into a very pesky memory problem on 32bit winXP. Which is odd since I have 4GB of RAM installed, the game usually runs perfectly on high settings and uses less than 1GB of RAM(no mods installed). After every couple of launches when I return to the space center it crashes with an OutOfMemoryException: Out of memory. The crash logs are spammed with "can't allocate memory to textures" errors: DynamicHeapAllocator out of memory - Could not get memory for large allocationCould not allocate memory: System out of memory! Trying to allocate: 5592408B with 32 alignment. MemoryLabel: Texture Lots of them. And textures really glitch out eventually, especially in the mission control interface. So far I got different stack traces in every crash, here's two of them: ========== OUTPUTING STACK TRACE ================== (0x007960C8) (KSP): (filename not available): ??$Transfer@V?$StreamedBinaryRead@$00@@@AnimationEvent@@QAEXAAV?$StreamedBinaryRead@$00@@@Z + 0x512a8 (0x0079778D) (KSP): (filename not available): ??$Transfer@V?$StreamedBinaryRead@$00@@@AnimationEvent@@QAEXAAV?$StreamedBinaryRead@$00@@@Z + 0x5296d (0x0058225B) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x94d6b (0x00581ED2) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x949e2 (0x0058268B) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x9519b (0x005EA0FD) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0xfcc0d (0x7C80B713) (kernel32): (filename not available): GetModuleFileNameA + 0x1b4 ========== END OF STACKTRACE =========== ========== OUTPUTING STACK TRACE ================== (0x10118F0A) (mono): (filename not available): (function-name not available) + 0x0 (0x10106B71) (mono): (filename not available): mono_jit_info_get_code_start + 0xaea5 (0x10106BC5) (mono): (filename not available): mono_jit_info_get_code_start + 0xaef9 (0x10106EAC) (mono): (filename not available): mono_unity_liveness_has_parent_class + 0xb1 (0x101070C6) (mono): (filename not available): mono_unity_liveness_has_parent_class + 0x2cb (0x10107217) (mono): (filename not available): mono_unity_liveness_has_parent_class + 0x41c (0x1010724E) (mono): (filename not available): mono_unity_liveness_has_parent_class + 0x453 (0x101074D0) (mono): (filename not available): mono_unity_liveness_calculation_from_root + 0x1e (0x0057D0FB) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x8fc0b (0x0057D1B6) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x8fcc6 (0x0058F9C7) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0xa24d7 (0x00581399) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x93ea9 (0x00581854) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x94364 (0x005829D9) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x954e9 (0x00582B84) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x95694 (0x005F3E4B) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x10695b (0x005F566B) (KSP): (filename not available): ??$Transfer@VSafeBinaryRead@@@Behaviour@@QAEXAAVSafeBinaryRead@@@Z + 0x10817b (0x0085E948) (KSP): (filename not available): ??$Transfer@V?$StreamedBinaryRead@$00@@@AnimationEvent@@QAEXAAV?$StreamedBinaryRead@$00@@@Z + 0x119b28 (0x00893830) (KSP): (filename not available): ??$Transfer@V?$StreamedBinaryRead@$00@@@AnimationEvent@@QAEXAAV?$StreamedBinaryRead@$00@@@Z + 0x14ea10 (0x7C817067) (kernel32): (filename not available): RegisterWaitForInputIdle + 0x49 ========== END OF STACKTRACE =========== Here is the most recent output log: http://m.uploadedit.com/b041/1412975681617.txt After some thought - it could be my video card. GT520 only has 1GB video RAM... I will try to reduce texture quality and see if it helps. Edit: Yep, after turning off some of the landscape settings and lowering texture quality the crashes stopped happening.

That's... a bit too glitchy for me. I was thinking along the lines of a seat for 1 kerbal, lots of infiniglide flaps since we don't have propellers yet, and some ions for when I'm halfway out of the gravity well.

Wow I can't believe this thread is still kicking. So essentially it can be done with exploits, huh... I guess I'll be trying to make a semi-infiniglider with ion engines then.

Fixed the first post. I meant electricity from PB-NUKs and/or solar panels. Will try to make a wider and longer rover. The one I have is great for landing and taking off without a skycrane but not for driving around at high speeds. Upd: Made one that does 15 m/s relatively stable. Can possibly make an even faster one but meh. Flagged 2 arches, got over 30km land distance.

Aerospikes must always be ahead of CoM or the Drag resistance will screw you over at high speeds, literally. The rest of the engines - behind.

Battery power as in - electricity. I am currently trying to get the land nav ribbon for the moon, but it takes ridiculously long. At 10 m/s 30km is almost an hour long trip, so I was trying to get at least 20 m/s which usually ended up in the rover toppling and parts exploding due to terrain not being level.

Has anyone done this? 10 m/s as I can tell is a safe speed at which even if you crash your pilot in a seat will most likely survive. Is there a way to build a rover that can travel at 20 m/s or higher on electricity alone? And not crash and kill anyone riding it? I'm guessing seats are out of the question since kerbals tend to fall out of them a lot.

-your aircraft goes out of control at high speeds? -changing CoL and CoT positions does nothing? -you've tried everything and nothing works? My boy, you have Center of Drag issues. Move the air intakes as far back as possible. Also - if you use aerospikes they need to be in front of CoM. It solves this problem for me anyway. Every time my craft goes nuts at high altitudes I find that it's because I placed aerospikes in the back or air intakes in the front. No exceptions so far.

Ok... 30 000m then. What is the altitude from which you can see munar arches btw?

So... I'm planning to send hive ships full of probes to Eve. If I understood correctly - I should assemble my fleet at just above Mun orbit(12 000 000m)?

I am placing my Mun station at 13 000 m right now. Maybe I too should reconsider... I was initially thinking of 150 000 m but suddenly realized I will be sending landers and probes back and forth and it would take a while.

I don't know what the FTL engine will be, but it will probably not work with liquid fuel and not be efficient within kerbol SOI to avoid huge balancing issues. Hence - making fuel depots now is shooting blind. Depending on the stats of FTL drive there are several plausible locations(including Eeloo orbit in case FTL has ungodly weight that I have to drag with me across kerbol system), but I would probably go for Duna orbit. It is at the same inclination as Kerbin, allows aerobraking to save a bit of liquid fuel, has low escape velocity and can easily be resupplied. Even if FTL works only outside kerbol SOI, it is easy to calculate how much you need to slow down before entering the system to end up in Duna orbit. I don't think using a gravity slingshot effect to slow down will be useful, but if it will then Kerbin orbit is the way to go, maybe even Eve. Jool is a gamble, currently it has too low gravity for a gas giant but if it gets increased(or there will be a larger gas giant in another system) then it will be the planet to go to slingshot some speed down and then aerobrake the rest. It will heavily rely on resources too. If mining will be implemented then outer rim planets like Eeloo and Jool are the more likely targets. The decision whether to make the station in Kerbol orbit or near the planets should be based on craft fuel reserves, remember that you need to transport fuel to the refueling station too so making a fuel depot that can easily refuel ships but can't be easily refueled itself is still inefficient.

Sounds like a shader that does not account for surface normal direction.

Ok, so essentially - I will need to split my air intake and jet engine action groups. Close 3/4 the intakes at lower altitudes, then open them and turn off 3/4 of the jet engines at higher altitudes... Will give it a try.

How do you exactly save that much? Because I get flame-outs at 27 000m altitude no matter what I do: decreasing engine throttle, adding MOAR intakes, increasing number of turbojets up to 20... I am working on a 200 ton SSTO too btw, able to carry a payload of 40 tons. Gets to orbit burning two orange fuel tanks almost to 0. Consists of only 400 parts though, anything more than that starts to lag badly. I am impatient though, if your method is waiting for 10+ minutes while the weak-ass jets accelerate the damn thing then I would prefer assembling more efficient interplanetary craft in orbit.

I use SPH SSTOs. One for payloads up to 20 tons: And one for payloads up to 40: Cheap. Reliable. Reusable. A bitch to construct. And a hell to work out the kinks. Have not used VAB ever yet.

I finally managed to SSTO some orange tanks to space Dispatched one tank near Eve too. Half empty though. +Sent a probe to Eve surface.

Try struts, the physics engine don't count them as mass I think. At least that's what the wiki says.

You people do realize that a bit faster and you will leave Kerbin SOI with this. First aero-hop to Duna/Eve and back, anyone? *Ok, not a bit, about twice as fast.

Nah, only when I'm desperate enough to use it instead of the conventional methods. So maybe later.

I was going to try something like this for my Eve lander in case my current configuration failed. I am planning to use a tug that catches landers sub-orbit and pulls them to stable orbits anyway, so I might consider this in the future even if it is a little bit exploitative.

So... You think this can fly on Eve as well? As in - landing, picking up kerbals and transporting them to orbit?

To actually land on the runway you will need a good symmetrical rudder that does not contribute to roll when changing heading. Also - you need to kill engines earlier, 100m/s is hardly a healthy speed for landing. I'm not a good pilot either so I just add a lot of wings so I can glide at 40 m/s just above ground almost indefinitely. I use only the SPH for all my designs and it is HARD to balance all 4 force centers with only 1 plane of symmetry. So do expect twitchiness unless the craft is fine-tuned empirically. Getting a reusable spaceplane took me a week of hard work. Check out my tutorial on balancing, it might help: http://forum.kerbalspaceprogram.com/showthread.php/33681-Jods-list-of-aerodynamics-issues

Ugh. I've been calculating delta v for weeks now, still trying to get a small compact Eve lifter. Thrust is basically the force that is applied to CoT. To get Thrust-to-Weight-Ratio you simply divide your thrust by 9,81*mass. So for example 1 aerospike can lift no more than 18 tons on Kerbin vertically. It is a good idea to have your TWR above 1 even if you use wings because current drag model gives absurd amounts of aerodynamic friction. I usually prefer TWR at about 1.05 for rocket engines or 1.4 if i use jet engines. Remember that as you consume fuel your weight will be reduced and TWR will increase. Delta-v is technically an advanced analogy to mileage. The more delta v the further your craft can get. It depends on your fuel amount and on your Isp. I am currently working on a small craft with 1 aerospike and 2 LV-Ns that has 2/3 of its mass as fuel, in total it has around 7000 delta-v according to formula: delta-v = Isp*9,81*LN(full mass/dry mass) Where dry mass is mass without fuel. The problem is actually getting a craft that has huge amounts of fuel relative to its total mass and efficient engines that can lift it all at the same time. Because efficient engines weigh a lot you have to combine them with other engines in a way that would give you a lot of TWR at take-off and an efficient burn later. This gets really hard, especially getting fuel ratio up when your Isp and TWR are capped.