Jump to content

What are the most important things you've learned about playing KSP to pass on?


Recommended Posts

A good advice if you are on a rover: switch into docking mode when controlling a rover, meaning "push the little planet thingy down the little rocket thingy" as you can see in this image. Also read the entry on the wiki, enligthening: http://wiki.kerbalspaceprogram.com/wiki/Rover

Docking_lin.png

Now about mechjeb:

One thing I need to say, KSP navball certainly isn't stupid. There are a few things that could be improved on it but in general it's an ultimate navigating device. It's fine by me if you didn't spend enough effort to learn how it works and how to use it using available sources (forums/tutorials/youtube/wiki) but you shouldn't blame the navball for it.

This exactly.

MechJeb and any autopilot mod takes away the learning experience, and that´s my baseline point. If you are learning the ropes of the game you NEED to be able to read all the standard instruments on screen, including the navball.

I´m not bashing anyone here, if you like mods and do things the easy way then do it, is your way to play the game and that´s fine and dandy. But if you really want to get involved and understand the game mechanics then ditch mods and challenge yourself. You should be able to dock two spaceships using only the navball before installing mechjeb.

Edited by Wooks
Link to comment
Share on other sites

I'm really struggling to hold my tongue here. To put it bluntly, if you can't find east on the navball then you're in no position to be "passing on important information".

Sure, the captain of a ship might not need to know how every single part of the ship operates, but he should at least know how to read a compass.

Link to comment
Share on other sites

A good advice if you are on a rover: switch into docking mode when controlling a rover, meaning "push the little planet thingy down the little rocket thingy" as you can see in this image.

I will add this:

1) Disable steering for the "rear" wheels on your rovers to allow better high-speed handling, as you will not turn as sharply and be less likely to flip over. The downside of course is a larger turning radius.

2) Keep SAS and your reaction wheels on when driving, and use docking mode/remap the keys so you are not pitching forward when trying to drive forward. This will help keep you rover stable even at high speeds. The downside is it can make turning difficult since the reaction torque will resist the wheel's attempt to turn the rover.

3) To combat the downside of (2), use both turn and roll controls to "lean in" to a turn. Roll (Q and E) still works in docking mode, and by rolling into the turn you are putting more weight on the inside wheels, giving better grip and helping you turn. You are also helping to counter any centrifugal force on your craft from the turn itself making it less likely you'll flip.

4) If you have a lot of torque, periodically turn SAS off for a moment to make sure all your rover's wheels are making contact. I've been able to do some unintended wheelies...

5) When trying to get up steep slopes, consider going back to staging mode. Attempting to go forward will now also pitch your rover forward, which will help push the front wheels down and improve traction. Provided, of course, you don't have so much torque that you do a forward flip!

6) Consider putting a fuel tank and thrusters pointing upwards to provide down force for extra traction.

Here's my Science Rover on Mun crawling up a 45 degree slope with relatively little effort thanks to the upward-pointing traction thrusters:

screenshot348.jpg

7) Consider putting a fuel tank and thrusters pointing downwards to help you jump over peaks and soften landings if you accidentally (or deliberately?) drive off a cliff.

8) Use structural trusses as bumpers to help guard delicate parts from crashes... as long as those bumpers don't obstruct the function of the parts they protect.

=Smidge=

Link to comment
Share on other sites

I feel required to point out that the captain of a ship does not nessessarily know how to do everything required to make that ship operate. He may not be able to pilot the ship, or repair the engines/trim the sails by himself. The captain's job is to oversee everyone whose job it is to do all of that. He's required to know enough to get his job done, but you can't just take the captain of the ship and put him on any given task and expect it will work out.

Furthermore, there are distinctions between the commander of a vessel (traditionally a captain, but also possibly called a skipper or, in unusual circumstances, a coxswain might hold the position,) and the master. (Also the owner as well, but in this case the owner is irrelevant.) The master of a ship makes the big decisions - the master is the one who says "We're sailing this boat to Shanghai," or "we're sailing for Portland." The captain is the one who organizes it all, and the navigator and pilot are the ones who set the course and apply the throttle.

I'm really, really kind of ****e at setting a course on my own, and even worse at applying the throttle except under the most forgiving of conditions, and by that I mean Minmus. That doesn't mean I shouldn't be allowed to play KSP until I "get over it." That's what MechJeb does - it's a handy dandy pilot and navigator I can give orders to and take over from when and if I have the skills to do it... Or when and if I realize I'm trying to dock a lander to a mothership and I haven't unlocked the auto-dock feature yet. (That was hairy, and sketchy as all hell 'cause the lander wasn't anything resembling RCS-balanced, but kinda fun, too. I never would have been able to do it without MJ to make a hohman transfer to put me within 200m of the mothership, though.)

I know what MJ is doing, but with that stupid navball interface - and the fact that sometimes you're coming up on a parabolic apoapsis far, far too fast for me to get a maneuver planned using the damn thing and orient the craft appropriately - I'd be lucky to get anything done. Not to mention the fact that the navball isn't very helpful, I certainly can't find a heading on it. I'm lucky if I can orientate to the horizon using it, forget about "finding east" on it, and most of what I know about which icons on the damn thing mean what is because MJ actually spells out in english what it's tracking when it orients to an icon.

You only need a ladder when you're going somewhere with gravy heavy enough you can't just RCS up to the hatch.

Also, some experience from me: Kerbs tend to explode when they touch land and are exceeding a velocity threshold. It doesn't actually matter what rate they were approaching the ground, just that they're going too fast in any direction. So you can be skimming the ground very quickly and only clip it with a Kerb's toe, and he'll explode as if he'd rode a solid rocket booster straight into the dirt.

East is 90 degrees, South 180, West 270, North 0.

Link to comment
Share on other sites

Docking is easy, as long as you have RCS symmetry. You put RCS on the center of mass or equidistant from it, and you can translate without a problem.

Don't forget those solar panels! Especially for probes: never go anywhere without either backup batteries or a standard panel.

RCS farther out from the COM gives you better leverage: RCS thruster on a girder will give you more control than on the hull.

Throttle back on your jets when you have an SSTO at high altitudes to milk a little bit more time out of the jets. Keep doing this until you can no longer keep your speed accelerating.

Pitch to the 90 degree mark when launching for an extra Delta V boost.

Link to comment
Share on other sites

Mechjeb for me was pretty much the tutorial system. It's not as good as a human at some stuff such as docking (uses loads of RCS) and atmospheric flight when you have FAR installed.

I do still have mechjeb and install it on all my craft and use it for the boring stuff ( keeping aligned during a 5 min burn for example).

I don't feel Mechjeb has hindered my understanding the game or prevented my from slowly accumulating the skills to fly and land all my designs. Furthermore I think it should be recommended to new plays as a method learn how orbital manoeuvres are done

For me KSP is 90% a game of designing stuff, flying things is just to prove my designs meet there performance objectives, this if an autopilot can do I'm happy to let it. (However, it is stratifying knowing you can do it on your own)

But as a lot of people have already said the number 1 piece of advice is:

Learn how to read the Navball!!!

Link to comment
Share on other sites

Pilot finds challenge in piloting the... thing he has put together

Engineer can find challenge in making a craft even he can handle.

Of course, most of us have something from both sides, and that's a great thing about KSP - you can compensate for one of these abilities with the other. Then you get something you can use - and with that you can learn what you lack on both sides, and perfect the both! And then you'll be able to create and pilot something that seemed impossible before.

P.S. I'm actually a terrible pilot. Have played Orbiter a bit - never got the Delta Glider to properly land on a runway. Aircrafts in KSP? Not too good either... At least I learned to pilot landers.

Link to comment
Share on other sites

I guess I'll add my two cents.

Remap the rover controls to arrow keys and remove the bindings for moving the camera with buttons. This will help a lot with rovers.

Use procedural mods, like procedural wings, fairings, stretchySRBs etc. You will have a lot more options like this.

Have a way to calculate your Dv. Be it mechjeb, Kerbal Engineer or a paper and a pencil, it's the best way to avoid making overpowered rockets.

Try to keep your space clean. Ditch empty stages so that they'll fall back into the atmosphere, collide with a planet/moon or if you really have to leave it in orbit, make sure you can deorbit it, either by putting a probe core on it and leaving some fuel, or by bringing another rocket to do it.

When you make a probe, try to place the solar panels so that at least one would face the sun no matter which direction you're pointed.

Link to comment
Share on other sites

My best lessons to pass on:

1. Test, test, test! Whatever you want your craft to do in space should be tested on Kerbin as much as possible. Make sure Kerbals can easily get back into your lander, make sure your staging is right, make sure the landing legs extend far enough to clear the engine, etc. Test each stage independently. Nothing worse than investing hours in an interplanetary mission to find that one of your later stages is somehow broken.

2. Don't forget electricity. I've been playing since before electricity was implemented, so I often forget to include it, especially on the core stage of my booster that I want to deorbit once it delivers its payload. A couple of solar panels and a battery weigh almost nothing but will save you the frustration of a craft with dV left and no way to use it.

3. The navball. Learn it, love it! No part of the interface is more important. The visuals of your ship are pretty but they tell you almost nothing about your craft's orientation. When doing maneuvers manually your eyes should be glued to it.

4. Generally, less is more. Big craft are cool and all, but they make the game laggy and are more vulnerable to the Kraken. Learn to minimize your part count and mass, especially in the upper stages.

5. Occasionally, more is less. Landing on a high gravity world? It's generally more efficient to leave your transfer stage in orbit and dock with it after ascent than it is to bring all that mass down and have to boost it back up.

6. Have fun and don't worry about other people. Use mods and cheats if they are fun for you. Whatever goals you set for your space program only have to satisfy you.

***

As for the whole MechJeb thing, here's my take:

I think it's a valuable learning tool. Watch what it does when you give it an order, examine the maneuver nodes it sets up, and think about how it is making things happen. Not everyone learns the same way, so saying that everyone should become a good pilot through trial and error before using MechJeb does some people a disservice.

If craft design and mission planning are more enjoyable for you than piloting, then use MechJeb and don't look back. It is frustrating to have a capable craft fail a mission due to pilot error. It also allows consistent, repeatable flights for testing designs or repetitive chores (I use it for launching/rendezvous/docking of tankers to fill interplanetary craft that I launched empty).

That said, there is sometimes great satisfaction in manual piloting. I don't think I've had a more satisfying moment is KSP than my first successful Mun landing (back when the Mun was the only destination). Docking manually is satisfying, too. Personally, I make landings on each body manually until they become repetitive, than I let MechJeb handle it.

Basically, I think MechJeb is a great enhancement to the game. It is useful for beginners and experts alike, for reducing tedium, frustration and learning time.

Link to comment
Share on other sites

Some tips for lander testing:

0) Before sending it to another planet, test your lander around Kerbin. It's better to be sure than to find that it can't do anything at the place. Also it's good to know what you can and what you can't at this TWR.

1) Ultra-light (Gilly, Bop, Pol) - just test on Minmus. It may be much of an overkill for some of them delta-v wise (you might start from suborbital trajectory in this case and test some hops around the landing site), but still you have to have at least this level TWR for comfortable piloting. Ensure that you have means to secure your craft on the surface - it's too easy to tip over in light gravity.

2) Mun-grade (Eeloo, Dres, Ike) - test it on the Mun! Dres and Ike are actually easier targets, but it's better to have some extra. Again, you may use suborbital tests with starting at already reduced speed and at least getting the same value when starting back.

3) Double-Mun-Grade (Moho, Vall, Duna) - try double Mun landing at half throttle (a bit overkill, but still) or direct landing from Kerbin-Mun transfer trajectory and going directly back to Kerbin. For Duna landing with aerobraking try starting at couple km above the Mun with minimal velocity, and also run some chute tests in Kerbin atmosphere.

4) Laythe - land it on Kerbin and get back to orbit (or at least 2 km/s). Ensure it's capable of landing on a slope and in water (there're almost no good even spots)

5) Tylo - try simulating "landing" from 100 km Kerbin orbit to 70 km altitude and returning back to 100 km orbit (and better ensure it can get to orbit - you'll need that extra delta-v for the actual landing). Also note the fuel levels during the "landing" and try dropping the craft with these fuel levels on Kerbin's surface to test touchdown sequence. (P.S. if it's SSTO rocket, it should be enough, just ensure it is landable)

6) Eve - I don't know. Landing with chutes can be tested on Kerbin, but the ascent... I'm not sure there is appropriate simulation for that atmosphere. Maybe try to put pitch at 45 degrees right from the pad? If it reaches escape velocity - it should probably be good enough.

Edited by Alchemist
Link to comment
Share on other sites

1) Do things when you're ready to do them, because you want to do them, and because you need the challenge. Don't do them because you see other people doing it, and you feel you must do it too.

2) Try not to get discouraged if your design looks like a toaster next to someone's gleaming, gold plated beacon to the science gods.

3) Never tell someone how to "fix" their design because it isn't how you would do it. When they ask why it's broken, give them ideas how to fix the design, not how to do it your way.

4) It's a game, enjoy it as such and let others enjoy it how they would like. If they aren't enjoying it, ask them why and give ideas of what to do, not how to play it your way.

5) Kerbals have a heck of a lot of EVA propellant, and can do some amazing suborbital moon hops themselves to collect some of that science-y goodness if you can pick a good landing site.

6) Go launch an orbiter just for the sake of playing around with the orbit. Don't plan on doing any science, landing on the Mun, or whatever. Just go fly and watch Kerbin rotate underneath you. (i.e. Don't forget to look out the window once in a while.)

Link to comment
Share on other sites

1. If mods are your thing, do them, if not don't.

2. Regardless of your choice for #1, remember, you set your own goals, hopefully you set all of them with fun as mission priority number one.

3. The navball is your best friend, it removes concerns about which way the camera is pointing as you try to maneuver your craft.

4.When in doubt about delta-v add more boosters OR reduce weight.

5. Struts and symmetry do not always a stable craft make. If your rocket is spinning, a lot, and you've loaded your craft with SAS, RCS, and wings/fins, it's likely a strut and/or symmetry problem.

6. Scot Manley's videos are excellent for inspirations, however, they should not be the expectations you set for yourself.

7. Stock KSP aerodynamics means that if there's a drag measurement on the part your using, then it will exert drag regardless of location on your craft.

8. When leaving a body, there is such a thing as wasted speed, you don't need to hit mach 7 to make orbit.

9. Remember to have some power sources on your vehicles, I can't tell you how many probes I've lost and Kerbals I've stranded because I forgot batteries and/or solar panels

10. It's a sandbox game, there are millions of ways to do something, for you the only important way is the one that you want to do.

Link to comment
Share on other sites

0) Take notes. Remember *what* you did, *why* you did it, and why it worked/dint work/went boom. Refer to these notes! They will remove a lot of frustration.

1) Less is more. (except fun... moar boostahs is always more fun ;) )

2) Don't glue something across your capsule hatch! My Jeb got to Mun, and...... stared at a door welded-shut by a thermometer overlapping the edge!

3) Electricity. You need electricity to open stowed panels. Probes need electricity to do *anything*. Don't assume because your engine has a generator you are ok.

4) Use enough parachutes. No good going to the Mun, and landing your capsule at 2m/s faster than its tolerance (boom)

5) Tall and skinny is for fashion models. Squat and ugly is for steering straight, especially during launches.

6) If you find yourself adding more struts, and more, and more.... refer to rule 1. You are doing it wrong. Beginner rockets are NOT mean to mass 2000++ tons, just to go to Mun.

7) Use Mechjeb. See how it does things. *then switch it off*. It ruins your fun!

8) Structural girders make for excellent landing legs! A bit heavy, but you can land with XXXm/s of horizontal speed, if careful.

9) If doing science, don't goto Mun. Don't even bother with Kerbin much. Go to MinMus! Easy science, low grabbity, less delta-v than Mun.

Orbital stuff - longrange

10a) If you want to go faster, the best place to do it is at the bottom of a gravity well.

10b) If you want to change direction, the best place to do it is as far from a gravity well as you can get. And as early as possible. No, earlier even than that!

Meaning: Do your interplanetary speedup burn as low and as fast as practical. Oberth will reward you.

But if you need to change your orbital inclination, do it as far out as available.

And if you are going to adjust your direction with a mid-course correction, do it sooner. a 100m/s fix just entering Mun's SOI could have been a 2m/s fix just after leavin Kerbin.

10c) If you need to stop, most fuel-efficient is *right* when you need it. Brake for landing as late as safe.

10d) If you need to go, the best time is right at the start. No use carrying all that fuel on the whole trip, if it can just as well be spent early.

Small Maneuvers in orbit.

*Relax. It makes sense, but you might need to unscrew your brain first.

If in orbit, and you want to go down, thrust against your movement. This will slow you down, which will speed you up even faster once you get down.

If you need to catch up to something in your orbit but a bit ahead of you, *slow down*. that will drop you down, drop your altitude, which speeds you up, and allows you to gain on it.

If you need to catch up with something at your altitude, but off to the side... have a break. It will come to you by itself, as its orbit intersects yours.

Wierd things to remember: (where KSP ignored fizziks, or invents its own)

*Kerbalnaut eva pack has looooots of juice. That refills free when ending your eva.(in .23 at least). Jeb can jump on Mimus, orbit, and land again using just his backpack.

*Drag is silly. If it is on your rocket, it drags. Placement doesnt matter. Angle doesnt matter. a FULL fuel tank falls just as fast as an empty one. in KSP, drag depends on MASS, not surface facing the wind or any namby-pamby realworld concepts like that.

*Planetary gravity ONLY affects you while in its Sphere of Influence. Step 1 mm outside of that, and it might as well not exist. Only ONE gravity source affect your ship at any one time.

Sorry for rambling... Seems i want to say a lot more than I expected!

Link to comment
Share on other sites

Hope to not duplicate (but the first post give the direction with "radialize" ;) ): vertical stack decoupler (the circular ones) are very weak and when the weight over them is greater than what was below (typically boosters), the boosters goes up just before running out of fuel and make a big firework (and some Kerbals think it's the Jeb day festival, again !).

Using KER help a lot as you may achieve better result with less weight and less money (for when it will count ;) ).

As a crazy moronic gigantic rocket made with big red fueltanks + Mansail engines will not get as far as a smart and small/"lightweight" rocket with a good Delta-V.

For the other stuffes learned : orbital manoeuver mostly. I'm still lame at landing anything, even VTOL flying is not my cup of tea.

And generally speaking, I've learned a lot about rocket science :). Far more than I have imagine first.

Link to comment
Share on other sites

3. Learn to dock, and get good at it. It opens up a whole new world... well worlds actually. Why struggle getting a massive ship into orbit when you can split it into two parts and have them docked in less than half an hour?

"

how do I dock when I cant even rendezvous?

Link to comment
Share on other sites

how do I dock when I cant even rendezvous?

Well, if you want to learn properly how to perform the maneuvers on your own, learn the controls and how to read the instruments, you can start watching tutorials like this one, Scott Manley has a handful of tutorials for beginners that will get you doing rendevouz and docking in a snap without ever need to install any autopilot mod.

I highly reccomend subscribing to Scott youtube channel, you will learn tons of stuff in an easy curve instead of bashing your head against the screen trying to understand what MechJeb is doing without guidance.

Edited by Wooks
Link to comment
Share on other sites

Knowing what MechJeb is doing isn't terribly helpful if doing it requires the kind of precision that you just can't get without an autopilot or being Scott Manley. Knowing the theory behind a Hohmann Transfer and intercepting a vehicle in its orbit and being able to do it on the fly are quite different things.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...