Fraktal

I actually did some experimentation with the HG-5 on relay sats a couple of months ago. Three relays of two HG-5s each, located in geosynchronous orbit above the Mun and spaced out 120° relative of each other, are just barely strong enough to pick up a single Communotron sitting on the far side of the Mun, bounce the signal between each other and make contact with Kerbin at DSN level 2. Signal strength for the Communotron will be very weak (~5%) but will never lose connection completely. You might have some interruptions in non-equatorial valleys and polar latitudes, though. Taking this relay design, adding on two more HG-5s and launching this quadruple-antenna version into a 300 km orbit around Kerbin will result in the Kerbin sat being able to briefly pick up signals from the munar relay constellation during closest approach. This quadruple-antenna version will also be more than strong enough to reach Kerbin from Minmus at DSN level 2. So if you want a relay within Kerbin's SOI, quadruple HG-5s per sat are a good choice unless you want direct Mun-to-Minmus communication. On the other hand, I was recently considering whether it would be more effective to use dual antenna relays around the Mun at below-geosynchronous altitude plus a single quadruple-antenna relay sitting on the Mun's surface facing Kerbin. I mean, it's going to be in the shade for several days per orbit, but since relays don't need power to transmit...

This thread is going to be useful for you.

You don't waste. One transmission at 100% signal strength sends back all the science you can send back in one transmission. If the signal strength is lower than that, you obviously transmit less, but the missing portion is not lost in the sense that you can never ever get it by fetching a second (or third, or fourth, etc.) copy later on. If you go into the R&D facility at the KSC and open the archive, the bars next to each experiment/biome/situation combination will tell you proportionally how much science from there you've sent back to Kerbin out of the total you can send back (which, in addition to the number next to the bar that says the exact value you've sent back so far, lets you gauge how much is left to collect). As long as the bar isn't 100% full, you can still get the remainder at any time (but the fuller it is, the less science you get per repeated experiment, so you eventually get to the point where it's simply not worth it to send another mission there).

I split my station into modules that do one single thing (crew module, lab module, fuel module, power module, antenna module, etc. with any necessary size adaptors where necessary) and bring those up either one by one, or 2-3 at a time, depending on how big the module is (crew, lab and fuel modules plus six-way station cores for more docking ports are obviously too big to lift more than one at a time, but I did pull off three at a time before). All of the modules have a docking port on both ends so that the worker pod can grab it by either end to slot it into place by the other end. This also means that once I don't need a module, I can send up an empty tug, have the worker pod unplug the module and slot it into the tug, which then proceeds to haul the module back down to the ground for recovery (provided it survives the landing, that is). The worker pod has the following components: 1 single-seat command pod, although you can go fully unmanned if you'd like 1 docking port at the top node of the command pod 1 OKTO core for SAS (so that Engineers can handle it, freeing up the Pilots for other missions) 1 of the smallest battery, mostly for emergency 2 OX STAT solar panels on opposite sides of the worker pod to keep your reaction wheels going 2 lights pointed in the same direction as the docking port in case the module hauling lasts long enough for the station to be eclipsed and timewarping until the eclipse is over would mean drifting away from the station 2 roundified monoprop tanks, this plus the command pod's monoprop tank is generously enough fuel for fetching-hauling-docking multiple modules all over the station And of course, don't forget monoprop storage on the station itself so that the worker pod can refuel at least a few times before you need more fuel from the ground. 1 extra reaction wheel to make it easier to turn while hauling the payload 2 single-thruster RCS blocks per dorsal, ventral, port, starboard, 4 for fore and aft, all set to translation only and balanced for zero torque while not carrying a payload (I use RCS Build Aid) Don't waste RCS on rotation, use the reaction wheels for it. RCS rotate only if the payload is very heavy (like a topped off Rockomax/Kerbodyne fuel tank) You can add on structural parts for bling, if you like, I just keep it barebones myself to extend the fuel endurance via weight savings The tug itself is unmanned and consists of a lower stage as regular finned booster, an upper stage as a probe core with batteries, plus fuel tanks and engines radially mounted on the end of outward-pointing girders (to provide enough clearance for the engine exhaust to not torch the payload), plus a downwards-facing fairing base as upper stage, with the payload sandwiched between the two stages. When the lower stage with its fins is expended, the upper stage pulls the payload behind it rather than pushing it, making it aerodynamically stable due to the payload's drag being at the rear and much less prone to flexing if you try to steer during flight. Once the payload has been delivered, the tug can use its remaining fuel to deorbit itself and parachute down for a soft landing, leaving no debris in orbit.

And that's why you launch aerodynamically unwieldy and too-large-to-fit-inside-fairing payloads in a pull configuration, rather than a push configuration. Unrealistic? Definitely. Ugly? Maybe. But if the payload's drag auto-stabilizes the rocket, who am I to complain? And that's why I don't equip my tugs with RCS at all. I just tow them next to the station, zero out relative speed, and let the RCS-only worker pod permanently stationed at the space station come over and move the payload to where it needs to be. At least it gives my engineers something to do other than twiddling their thumbs. That's what makes KSP great. There are multiple approaches to pretty much everything and nothing forces you to play how the game expects you to.

Static suction so that said 4 Rapiers don't flame out on the runway.

Give us a DLC focused on allowing ships to operate independently of Kerbin. Reactors fueled with Fissionable. First model available somewhere around tech tree tier 7, significantly before RTGs, but there's no way to obtain Fissionable outside Kerbin at this point. I'm thinking Rockomax or Kerbodyne diameter. The existing drills can mine Raw Fissionable, which is a separate resource from Ore and cannot be processed with the existing Convert-O-Trons. Instead, Raw Fissionable needs its own machinery to process into Fissionable. Additionally, the tech tree node that gives the RTG would also unlock a breeder reactor. This one isn't as powerful or efficient as the first reactor, but in addition to consuming Fissionable, it also slowly refines Raw Fissionable into Fissionable as part of its normal operation. The conversion rate is slower than the dedicated processing machinery, but faster than the breeder reactor's own Fissionable usage, making it quasi self-sustainable similar to using a Fuel Cell Array to power a Convert-O-Tron in LF/OX mode. Xenon extraction. Should be a separate resource from Ore. Maybe an option to scoop it from the air, which obviously requires a celestial body with an atmosphere, making it harder to transport. If LF/OX rockets are meant to be hydrolox, possibly a way to gather LF from solar wind with equipment that can be installed on space stations or deep space vessels. It's only LF, though, without OX, so you still need to bring OX from elsewhere. It also doesn't work in atmosphere. Station-specific stuff, yeeeeesssss. Gimme. One idea I thought of in this regard is a fuel scoop for passively generating LF over time using EC. Not restricted to stations, of course, but slow enough that a station meant to be left unattended for long periods would be the most convenient place to put it and heavy enough to make lugging it around on a ship inefficient. Optional features: Make its LF production scale with distance from Kerbol (ie. it works better the stronger the solar wind is). Make it stop working if Kerbol is occluded by another celestial body (going with simple occlusion since it would be too complex to simulate planetary magnetospheres). And for those who want to lug the thing around rather than leaving it on a station, allow it to gather LF super-fast by dipping into Jool's atmosphere. Life support yes, but make it optional and introduce a difficulty option for scaling how fast the supplies get depleted. All crewed parts can recycle air by expending EC, at a rate about equivalent to a single kerbal's air consumption; this recycling function can be toggled in the part's context menu and is off by default. If you bring more than one crew member per pod, you need an extra part to break even: Early-game, you have a simple, lightweight CO2 filter that can only recycle a finite amount of air before shutting down for the rest of the mission. Mid-game, you get a CO2 scrubber whose filters can be replaced during the mission. Extra filters are carried in a separate part and replacement can be carried out by a kerbal on EVA. Optionally, engineer kerbals would get some extra capacity out of each filter if they're the ones doing the switch. Late-game, you get full air recyclers with the same EC-to-air conversion as crewed parts, making you able to supply an arbitrarily large crew for as long as you have power. Jet engine air intakes automatically refill the vessel's air supply for as long as the vessel is in Kerbin's or Laythe's atmosphere. ISRU can also produce air from Ore at a slow rate. Hey, if it can already cook up oxidizer... Speaking of which, oxidizer can be converted into air in case of emergency. Handy if the emergency is you not having enough fuel left to leave, but a rescue mission is on the way so you're not using the fuel you have left anyway. Whether food and water would be separated or not, doesn't matter. But if they are, explicitly allow water (but not food!) to be obtained via ISRU. A part that contains Spare Parts (Junk, maybe?) for repairing components that can be broken without destroying them outright, like solar panels and retractable antennas. If on-site manufacturing of these parts from Ore would be allowed, make it require its own foundry/fabricator part in addition to/instead of a Convert-O-Tron.

Cargo bay, service bay or FAR.

Windows 10 does it even on desktop. I looked it up.

It's probably Windows or the graphics driver thinking the game doesn't require a high enough graphics performance to warrant using the dedicated card, so it fires up the integrated card instead in the name of conserving power. It's a really annoying feature both Nvidia and AMD have been pushing in recent years. Go into the Catalyst Control Center (quick googling says your dedicated card is AMD) and check the power plan settings.

As far as I'm aware, KSP does not model airflow obstruction on radially attached parts.

...you know, that's not actually a bad idea. In fact, make it a difficulty option to allow each celestial body to rotate/orbit in retrograde, independently of each other.

This has been the case since at least 1.5. I only ever had landing legs, externally-mounted science equipment and maybe fins burn off, everything else always remained in one piece.

I personally pitch the main wing about 1-2°, then pitch the smaller tail/canards in the opposite direction to fine-tune until the CoL and CoM almost overlap and the plane wants to very slightly pitch up while flying at normal cruising speed without SAS. I don't know how much extra drag this causes compared to trimming the elevators, but I don't think it's that much since the elevators are almost completely flat against the wing 99% of the time due to the plane naturally wanting to stay in the air without having to be forced.

It didn't, as far as I'm aware. Fixed radiators cool whatever they're directly attached to plus whatever's directly attached to that.

Which service module is it?

Yep, obstruction is the no.1 reason why a relay wouldn't work if it's in range. I recently had such a problem myself; turns out 500 km above Kerbin is not high enough for three relays to all see each other with KSP's default celestial body occlusion multiplier. Have you, perhaps, considered a secondary relay orbiting Ike to maintain connectivity in the event of an occlusion? I mean, the chance of both sats being eclipsed simultaneously is small if their respective orbital periods are not close multiples of each other.

One more scenario where this is useful: emergency parachutes for aircraft which you only intend to use via abort group.

I have a lot of pre-made designs, but all too often I find mission requirements falling squarely between two designs in such a way that one is too weak but the next is too powerful. My designs usually use 2-4 SRB or non-gimbaled boosters around a gimballed core. Sometimes I use a non-gimballed core with a gimballed upper stage, in addition to the boosters.

It is if removing most of the science from Minmus bottlenecks the player to the point where you'll have to reach Duna with something built out of 200+ 1.25m parts lagging the game to oblivion because you can't afford any Rockomax parts without launching unmanned rovers at every single celestial body outside the Kerbin SOI. That is, if you even have the antenna range to reach there because you can't afford the RA-2 either. I'm not saying it's impossible (I know better than to say that to an experienced KSP player). I'm merely saying that if the default difficulty is too easy for you, the solution isn't to raise the difficulty for everyone. That way lies madness. Some think being able to reach the top of the tech tree within the Kerbin SOI is cheap, so what? KSP isn't a competitive game and you're not forced to play as such, so let people play however they want.

I'm playing with 10%. Right now, I completely harvested Kerbin, the Mun, Minmus and near-Kerbin Kerbol orbit for all science I've got unlocked (crew/EVA reports, temperature, pressure, goo, material exposure and surface sample) in each biome and each situation, plus transmitted temperature and pressure readings from all biomes on Ike and Gilly, plus some orbital data of the same of Eve and Duna. This is the current state of my tech tree, 2 years and 338 days after starting this particular save: A grand total of ONE of these latest nodes have been unlocked with a laboratory. All of the rest have been painstakingly collected via grinding. If all but two of Minmus' biomes were to be eliminated, I'd be lacking another three nodes. So yes, Minmus in its current form is REQUIRED for the stock tech tree at minimum possible science income in order to get the gear for a manned Duna flight. Nukes and ions? I'd be happy to have them before my first manned mission to Eeloo if I'm intent on visiting everything in order of proximity to Kerbin without excessive reliance on laboratories (which is another thing the "science system is too easy" crowd keeps complaining about year by year). Thus is my opinion.

You do realize that would make a 10% science game ultra-difficult, right? As in, you'd barely be able to unlock the Mk1 landing can before running out of science in Kerbin's SOI. Don't complain about the game being too easy if you're playing on easy mode all the time, mate. If you feel it's too easy to get up the tech tree, that science multiplier option is there for a reason. Yes, you're probably going to ask what masochistic individual would play that way (I do), but like it or not, it's still a stock option. And if your proposed change breaks the stock game, you're not approaching the problem the right way.

It's a bug that appeared in 1.7 and still hasn't been fixed. If the fairing is between the root part (or is the root part) and the interstage node, the node won't detach when the fairing opens. Makes it impossible to launch multiple satellites in one go without having to waste payload mass on decouplers to detach the sats from what should be thin air.

You still need to use decouplers between the interstage node and the payload, though. Few versions ago there was an update that keeps interstage nodes attached even after the fairing is popped open.

What about four RA-2 satellites mounted on the end of girders in an X shape for a visually pleasing ground-based relay?