Jump to content

How to learn the game (as opposite to just follow others)?


Recommended Posts

Hi all,

I'm a very fresh player in the Kerbal Universe and at the moment I'm taking my babysteps. However, I wonder if there is a good way to learn the game. Now I know there are tons of tutorials and I could just build the rockets they build there. But somehow I feel that this is not the best way to learn. What I fail to see (atm) is WHY certain rockets work, while others don't. For example, with my first few tries to reach the orbit, I often have the engines to overheat. My guess is that I have a too heavy rocket. But is there a way to "see" (or calculate) it beforehand? How much "power" do I need to reach the orbit and can I go from there?

So basically, what I'm looking for is basically tutorials about the tech, not about how to reach Mun as quickly as possible. I don't mind taking minor steps one after the other, if I have the feeling of learning and making progress.

Does that somehow make sense?

Link to comment
Share on other sites

If you don't want to copy others then... just go and experiment! Try different configurations and combinations to see what works, what doesn't work and what needs some more tweaking.

As for engines overheating: if these are liquid engines and they are critically overheating, just throttle down a bit. If you put a bunch of SRBs together and they blow up - don't put them so close to each other. Also winglets may help with heat dissipation.

Link to comment
Share on other sites

The big thing would be to pick up the (simpler than you'd expect) basics of rocketry and orbital mechanics. Nyrath's atomic rockets site gives a good overview. For specific pieces of math, I would point to the rocket equation and later Kepler's laws. For just getting to orbit, the important thing is horizontal speed; you only need enough vertical to get out of the atmosphere.

The overheat bar is merely a warning. If the engines don't explode, they're still fine.

Link to comment
Share on other sites

  Ellegon said:
Does that somehow make sense?

Yes!

-You should only worry about overheating if it blows up your engine. I have not seen any overheating explosions in recent updates, but I have heard the RAPIER is prone to overheating.

-Asparagus staging will help you quite a lot in the early part of this steep learning curve, but some see it as cheating/unrealistic. (I find it ugly)

-VAB construction works with a "tree," in other words every part is connected to one adjacent part, all leading back ultimately to a root part. This means no rings, nets, etc. (It is possible, but requires docking ports and a tolerance for very low framerate)

-"Moar boosters" is not always a good idea, struts are your friend.

That's about all to tell without a full orbital tutorial. Have fun!

EDIT: The "edge" of the atmosphere can be measured by where you can timewarp at 5x.

Edited by DisarmingBaton5
Link to comment
Share on other sites

There are ways to get a lot of information about your rocket before you launch, especially if you use mods like MechJeb, but a trial and error approach will make you a much more competent designer than relying on basic sums about whether your rocket is going to work. There are different approaches to reaching orbit depending on staging, rocket design, fuel types, power settings, rocket weight, etc. The easiest launches generally just mean pressing 'go' and waiting till you're high enough to turn into orbit, but easy does not equal efficient, in fuel or time. Sometimes a rocket that's on the very edge of flying itself apart is actually exactly what you want because you're trying to do something outrageous, like launch an enormous space station intact. As long as your rocket stays just this side of flying apart, you're ok, but only experience will warn you if your rocket is a basketcase.

More importantly, if you build 5 rockets, 3 fail, one just succeeds, and one is a comfortable success, you're already learning what you need to know about rocket design. The failures weren't a waste of time - they taught you how not to do it.

Edit: Everything I need in terms of technical help I get from the wiki.

Edited by The_Rocketeer
Link to comment
Share on other sites

I learned by reading up on physics on wikipedia. Looked up a random term I wanted to know - in this case, delta-v - and just followed links and cross-references from there. Browsed easily 30+ pages, all of which had interesting tidbits that helped me along.

Link to comment
Share on other sites

My suggestion would be first play with the career mode up until you have unlocked 2-8 tech nodes, this should give you a feel for the parts and what they do.

Then play sandbox for a while and just try strapping huge engines and stuff together.

I took ages to discover that engines only explode when the heat bar is totally full.

Multi-couplers like quad couplers and tri-couplers are likely to make engines over heat because they seems to pump the heat into the other engines. Don't use them if you have heat problems.

For orbit you want to go up till the air thins (you will never get orbit speed otherwise) at about 10k. Then go 45 degrees till your AP is at about 60k then go fully sideways until your AP is at 70k. Then wait until you are near the AP and burn again to make your orbit round.

Also don't try going too fast in low atmosphere, going above 163m/s below 5k is wasting fuel trying to force air out the way.

In the end just enjoy making crazy contraptions in sandbox that's how I learned. There is no money so computing power is the only restriction.

Link to comment
Share on other sites

In order for a rocket to 'work' to the extent of getting into orbit it needs to go far enough, fast enough, in the right direction.

In reverse order -

1. The right direction is how you fly your ascent path and reading about things like 'gravity turn' and trying some variations is far better than going straight up then turning left/right.

2. Going fast enough means adding engines until your TWR (Thrust to Weight (mass in Kerbin's gravity) Ratio) is higher than 1. Below 1 gravity wins. At exactly 1 your rocket can 'hover' but not accelerate against gravity. Most people recommend a launch TWR of 1.6 - 1.7 because that lets the rocket accelerate through the high-gravity, thick atmosphere reasonably quickly. In the game as it is you'll have to do the book-keeping of your mass and your engines' thrust to work this out. Download Kerbal Engineer Redux (KER) from the spaceport even if it's just to show you this!

3. The third requirement is going far enough (deltaV) which means adding fuel. As you add fuel your mass increases so your TWR drops so you might need to add more engines, which need more fuel, etc. etc. .... deltaV is measured in m/s of acceleration and is something else you'll either have to keep recalculating or use KER for. You need 4,500+m/s deltaV to get into orbit from Kerbin's surface (apparently; I need rather more ^^).

So; if you can work out which way to go, have engines that are powerful enough to go that way and enough fuel to get there your rockets will work.

[This is the order I learnt in and it's only once I had all three that I started to think I was designing rockets, rather than just making it up as I went along.]

Edited by Pecan
Link to comment
Share on other sites

I use a 3/4 rule for staging. Each stage is three times bigger than the one on top of it. So if you had one small fuel tank and a small engine, the next stage would have 3 small tanks and a 3x more powerful engine. This will get you anywhere you need to go as long as you keep adding stages.

Link to comment
Share on other sites

Hmmm.... I think I see what you mean. You don't want a total copy-me instruction set, but you'd like to know some concepts to consider so you're not flying purely blind. I'll list a few to help out. And I agree that Werner Von Kerman and Scott Manley are good resources on YouTube. These really helped me get over the hump using maneuver nodes early on and later rendezvous and docking (it felt impossible).

Anyway... some thoughts on learn-as-you-go:

1) Learn your gauges... they tell you a lot and most are ignored. Example: The atmosphere thickness gauge at the top let's you know the air resistance. You don't have to have someone tell you the numbers... you can see this yourself realtime! Add to it homework reading from NASA... speed and air resistance in low atmosphere, gravity-turn, when/why to throttle up (and why they don't just have it "floored" the whole time), etc. Then use the book smarts w/ the indicators in-game to create your own concepts. And remember, Kerbin's size, gravity, atmosphere thickness are different than Earth, so copy NASA in concept only... their exact math won't help.

2) Heat happens three ways that I can tell... direct contact with a hot part (can be a problem), radiant (not a big deal but not ignored), and by exhaust (almost always a problem). So, space your engines out and they won't overheat, test to see if they start to overheat but don't actually blow up. See if you can use seperators so hot parts don't touch or a heat-sink... something to absorb/spread the heat (a big metal part takes more heat from an engine than a non-metal part). Put a decoupler on one stage and the engine start on the next so you can space the timing and avoid explosions...

3) ALWAYS LOOK AT THE WEIGHT OF PARTS. I can't stress this enough as it's not always obvious visually for given size/materials. It's easy to just keep piling on parts and you'll end up overbuilding badly. This compounds problems when trying to reach a new personal goal (farther planet, heavier lander, etc). Again, you don't have to calculate it, but if you always think, "could I do without this or use a lesser part?", you'll be on a good path.

4) Use probes when trying something new. These can be SUPER light, often handle easier and require less resources than manned capsules... good to learn concepts w/ less complication/weight. So, while you're learning, think of not killing Kerbals for the sake of skill advancement :-)

5) Build w/ modular parts. Save "tops" of rockets as something like "One Man Orbiter" or "Unmanned Science Lander" and then label it's contents in the description, then a full launch version when you have booster, orbital, transfer stages and maybe add a "FULL" suffix so you know that's a launch ready version. While building though, save the (usually bigger/complex) booster stage for this as something like "1m LKO booster" or "3m Heavy Asparagus Booster " and save that to the sub assemblies. You can then mix/match chunks you know work well and speed up your build-to-test time versus buld-from-scratch every time.

6) Don't just add more struts on a problematic rocket. It's an easy "fix" but actually does add to the weight, air resistance AND the part count. Sometimes you can build a more orderly rocket and reduce complexity. That said, struts are vital... just think of the physics of push-pull (they seem to only work that way... not torsion or flex... just tension/compression) and use them sparingly when you can.

7) Print up the controls/keybindings for the game and keep it handy. Example: You probably don't use RCS translation controls yet. And if you get to where you need them and have no idea they exist, they won't do much good!

Link to comment
Share on other sites

Kerbal Engineer Redux or Mechjeb (both are mods) will give you technical readouts of your ship. Weight, Thrust to weight, Delta V (learn what Delta V is!) for each stage.

Once you know how much Delta V your craft has and how much DV it takes to get to orbit it becomes a case of just slapping stuff together and getting it to orbit and then the system! It quickly becomes second nature

Edited by vetrox
Link to comment
Share on other sites

Also a good way of learning is reverse engineering: not just copy someone else's design, but try to analyze each element. Then try adjusting each element and see what that does. After getting some experience you may try recreating some real world rocket designs - you may get some good ideas, but just copying real rocket to KSP is not the optimal way, because some parameters may strongly differ... So try recreating and then optimizing a real life design.

Link to comment
Share on other sites

I took the basics other taught, learned a lot about orbital mechanics and went from there. I even went so far to design my own Excel spreadsheet with all the formulas and stuff so I could build my rocket outside of the game and get all the calculations (ISP, delta-V, TWR, etc). The bonus effect of this was I learned a lot about excel as well!

I was just like you. I struggled to get to the Mun because I refused to follow what everyone else said...the general consensus to getting to the Mun, take off, wait till the Mun is a few degree above the horizon and burn...but that does not really help does it? Sure it gets you to the Mun, but how can you use that information to go other places, because you never really learned what was happening behind the scenes to make that happen.

  frizzank said:
I use a 3/4 rule for staging. Each stage is three times bigger than the one on top of it. So if you had one small fuel tank and a small engine, the next stage would have 3 small tanks and a 3x more powerful engine. This will get you anywhere you need to go as long as you keep adding stages.

Although a great tip, this is not really helping the OP. Why are you using this method? What function of rocket building does this serve? This is what the OP was talking about. Not just blindly following others tips without really knowing why he is following them. "Because it just works" is not a valid argument for some people who want to be able to apply the information to their own rockets and know WHY it works. :)

  vetrox said:
Kerbal Engineer Redux or Mechjeb (both are mods) will give you technical readouts of your ship. Weight, Thrust to weight, Delta V (learn what Delta V is!) for each stage.

Once you know how much Delta V your craft has and how much DV it takes to get to orbit it becomes a case of just slapping stuff together and getting it to orbit and then the system! It quickly becomes second nature

This a million times. Many think using some addons is "cheating", however, unless you are already a rocket scientists, you will be unable to get all this information and still have fun. Remember, ALL this information is figured out months ahead of time before any real life rocket is ever built. Why should we build rockets blind in KSP?

Now whether you want to use the auto-pilot functions to fly the rocket without any user input, that is up to you. Some people use it only after they mastered certain tedious and boring techniques because after you have gotten 100 rockets into stable LKO, it just gets repetitive. Others use it for harder maneuvers, to learn what is going on so they can try and emulate it. I tend to use it right now just for assistance, when I take off I set it to keep a 90,90,90 orientation off the pad, and manually do the gravity turn and all orbital burns to make a stable orbit. If my rocket is not designed well, then even using the A.S.S assistance in mechjeb, it will not fly straight. Just again, it takes away the tediousness of the boring parts of take off, especially when I am learning to dock and will do dozen of identical take offs as i fail my docking attempts.

And again, the raw information you get from Mechjeb while building your rocket, and while flying, make the game more dynamic to me. It is less guessing at this point and more science. You can see the effect changing fuel and engines and parts has on performance. And if you are in career mode (I HIGHLY suggest this, you learn to do more with less real fast) then you don't vne unlock the more powerful functions of Mechjeb until later in the tech tree, after you have already learned to do a lot of flying hands on.

Edited by jedensuscg
Link to comment
Share on other sites

The single most important thing if you want to understand rockets is to learn what the rocket equation means in practice.

In a nutshell:

the more fuel you have, the bigger a change to the speed of your rocket you can make BUT it is subject to severe diminishing returns

And:

Each type of engine has an efficiency rating, its specific impulse.

the bigger the specific impulse of an engine, the bigger a change to the speed of your rocket you can make

-----------------------------

So - you have a rocket, and it can make a change of speed (delta v) of 2000ms-1, say.

You put some more fuel on it. Now it can make a change of speed of 3000ms-1.

You put a lot more fuel on it. Now it can make a change of speed of... 3500ms-1.

You put vast quantities of fuel on it. Now it can make a change of speed of... 3700ms-1. Severe diminishing returns. And it handles like a pig because there's so much fuel.

-----------------------------

You have a rocket. It has an engine with a specific impulse of 300. It can do 2000ms-1.

You change the engine on the rocket for one with a specific impulse of 800. Now it can do 5330ms-1.

You add another engine of the same type (specific impulse 800) to the rocket. It still does 5330ms-1. Actually a little less, because it's carrying another engine.

You add ten more engines of the same type. It still does 5330ms-1... actually a lot less, because ten nuclear engines are very heavy. :)

-----------------------------

When you're designing rockets to get off a planet, there's two other things to consider. You need to beat gravity (so you need enough thrust to be accelerating upwards) and you need to beat air resistance.

The first is easy, if your rocket moves upwards faster than an arthritic snail then you're doing ok.

The second is also easy; you can more or less ignore air resistance for now so long as your rocket is not glowing red hot with flames all around. If it does that you're going too fast.

A third thing: you need to turn the rocket, when you're high enough, so that you're going sideways around the planet - so that you're in orbit - and not just boosting straight upwards.

Usually people fly up to around 10km and then start slowly turning the rocket sideways, until by the time it leaves the atmosphere it is pointing towards the horizon.

Link to comment
Share on other sites

Because the game is still early in development, there can be no manual that wouldn't be obsolete before it was completed. Also, it would take a thick book to cover every aspect of this complex game anyway. Most of us learned by just trying things and seeing how they work. When you get stuck, asking a specific question is much easier for others to answer than a general, "How do I KSP?" :D And don't be afraid to ask those questions, as many people on this forum actually enjoy helping other players.

Link to comment
Share on other sites

Thank you all for the great tips and links. That gives me a lot to read into and to try out. So far I like the "trial and error" way and I don't see fails as a bad thing. However, I'm sure there are good ways to improve myself. I also like the idea of reverse engineering. I also try to see patterns in tutorials. For example: I watch/read 3 or 4 tutorials for the first orbit flight and try to see what they do in similar ways or differently.

And yes, I will try to not post questions I could have researched with 5 minutes googling... I promise :)

Thanks again to all who took the time to answer.

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

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...