What is your creative process like when you have an idea in KSP?

[Little Katie]

I see some really really cool space crafts in this forum and it always blows my mind.

I was wondering when you guys have ideas in your head, what is it like between the idea and the finished craft? Like do you know in your mind what parts will make your idea happen? Does it keep you up at night imagining how you will fit things together? Do you ever draw or look at a picture so you have something to work with? Or write it down in words instead of pictures? Or something else entirely? And I was also wondering how long it took before you really knew all the parts well enough to just put something together (something that actually works), or if that is an ability that kind of comes natural.

[Godot]

That´s actually relatively easy for me ....

basically it works like this:

[Greenfire32]

With the exception of space stations, I usually just load up the VAB/ SPH and go nuts. When I do space stations, I usually build the entire thing as a "Concept" craft and then build the individual pieces as needed.

This allows me to see how "cool" the station will be before I go through the trouble of launching and assembling everything...and then de-orbiting the station because I hate it.

[Agost]

Well, I sometimes think about KSP while not playing ( normal problem solving, like getting everything inside a fairing or so ), but nothing sleep-breaking.

[Slam_Jones]

I'm not a stranger to writing out/drawing plans I have for future ships (though my art skills are certainly lacking).

If I haven't played KSP in a while, I'll usually end up thinking about old ships I've made, and what I can do to improve on them. Lately I've been doing SSTO's and Tourist Space Stations (pretty stations that only funtion to complete contracts and... look pretty).

A lot of times, though, I'll just load up the VAB and start throwing parts together. Sometimes I come up with something cool But usually not.

[Cirocco]

Step 1: see something somewhere that gives me a flash of inspiration.

step 2: forget about said flash of inspiration

step 3: get bored at work

step 4: remember flash of inspiration

step 5: start drawing crude pen designs in one of my notebooks

step 6: get home, fire up KSP, try to build said design, notice several critical design errors/limitations, close out game in frustration

step 7: repeat step 3 through 6 until design is no longer crippled

step 8: perfect design until it accounts for ridiculous self-imposed limitations and demands

step 9: either launch design, take pictures, post up on the forums and feel awesome, or completely forget about design and have it be erased upon next update/game restart.

I have 2 works in progress that have just gotten into step 8 (one space station and one interstellar mothership). I wonder if I'll ever finish them before 1.0 comes out

[Taki117]

Generally speaking I design for the goal. I have a particular mission to accomplish, and I build a craft to accomplish that mission.

[SpaceExplorer]

When I have an idea, I usually just go to the VAB/SPH and start building. Although when I am making an interplanetary craft, base, or anything that needs to be assembled in orbit, I usually build it as one big station in the VAB then break it into components to be launched up separately.

[Warhawk]

Everything I do is part of a larger system, which means various creations have tasks that all work together to get somewhere. Individual design is secondary to the program as a whole. Naturally compatibility between ships could be a problem, but thanks to the wonderful world of standardized docking points and the Kerbal Attachment System mod, I can avoid the worst of real life's problems in that regard.

The system is governed by three tenets: The Mission, the Means, and the Future. Mission is straightforward: the short to medium range goal, be it landing on the Mun or orbiting Duna. The Means is determined by what parts I have available; I refuse to speed-tech in career so I have identifiable eras and limitations in technology are always foremost on my mind. And lastly I say the Future because I don't want to send up half a million credits of space junk once the job is over.

From that point on, each ship/station/device has a given role and is produced and launched in the order that is needed. I have an unwritten checklist for each kind of vehicle, drawing directly on real-life principles such as redundant systems, emergency abort modes, generous life support requirements, radiation shielding, and so on. Various mods make these things pertinent, like life support and random system failures, but even when there aren't mods I do it for the sake of the experience.

But the intention of designing for the future is important, which is why I give all of my designs more than they ought to have for the Mission they're meant for. A fuel station will have extra docking ports for more tankage. A space tug will have mounting points for different docking port sizes and claws to carry different payloads. Landers will be rated for the most demanding planet in a system, such that they can be used on any other moon or planet nearby, assuming complete recovery. Some consider this wasteful, but I have the last laugh when my 20 year old station is feeding argon gas to nuclear-electric spacecraft en route to Eeloo, at a fraction of the cost of their expendable ships.

As for specific ships themselves, I already have a good idea of what each part is capable of. It helps to stay as stock as possible, since this is experience gleaned from repeated use. Likewise, I have "families" of designs rather than custom launchers/spacecraft/landers for each and every sortie, further improving both reliability (incremental improvements) and flight crew experience (my own). Starting with an objective within a mission, and knowing what each part is capable of, I build vessels all in one go, and then test them as separate parts: A rocket or station designed part by part will not show you the overall mass distribution and/or performance until it's assembled, whereas building the whole thing in one shot and then fine tuning it in chunks avoids that problem. Kerbal Construction Time gives you a "Simulation" mode that costs money but doesn't actually blow things up and get astronauts killed. I use it to test each segment in sequence.

Other than that, there aren't any hard and fast rules I follow. The order that I build and test segments of ships, which are segments of systems, is arbitrary until things really get moving in space. From then on time is a factor due to crew and part endurance.

[Starwhip]

I've played around with the stock parts for so long that I can envision stuff just clicking together, kinda like in the Lego Movie with the holographic parts. (With the part numbers, which, by the way, was ABSOLUTELY GENIUS! I swear I was the only one in the theater who realized what they meant.)

And then I kinda go

Usually it blows up the moment physics kicks in because I forgot to strut something or my staging's off.

[Lhathron the Elf]

I often don't really even play KSP until I have something worthwhile to do. Sometimes this is caused by mods demanding something be done. Or sometimes I have designs that I've made that are just super sucky but that I really like, so I try to make them better. Sometimes I try to make them better using mod parts and things like that.

For example, I have RemoteTech and so recently I finally got around to setting up my communications network, before realizing that it sucks and that the probes themselves are worse. So now I'm reworking my satellites and I've just installed a GPS mod so that I can also put little GPS tracking devices on my probes to tell my little Kerbals where the hell they are.

Yeah, that's pretty much how my process is. It's either 1) Get a mod and do random stuff with it, or 2) Realize my designs suck and start messing with them. But in designing things, I usually just look in the VAB and put everything that I need to be on something onto it and then try to make it look pretty. So no real process goes on there.

[EladDv]

ummm...i see someone made something get inspiration and then try to make it better/smaller/more powerful

[Herbin]

My creations usually go" the form follows the function " route. Sometimes not pretty

[Vaporo]

My design process:

1. I want to go to the Mun today using NovaPunch parts.

2. Huh... my lander looks kind of dinky and boring on top of that big rocket. Why don't I add an some extra fuel?

3. *adds ten extra fuel tanks*

4. Huh... my launch stage looks small and weak to carry my lander. I should probably add more engines.

5. *adds ten more asparagus stages*

6. If game is not crashing, go to step 2.

7. Gee, that took a long time to build. I think I'll browse the forums for a bit before launching.

8. See Eve lander.

9. Wow! That guy's Eve lander is smaller than my Mun lander! He must be really good at this game.

[Flixxbeatz]

When an idea pops, typically it goes like this:

1 - Build the craft on either VAB or SPH. Even the minor details and aesthetics are laid out. This will typically take me more than 30 minutes.

2 - Test it.

3 - If it fails (flies erratically, goes back to kerbin, explodes for no reason, veering off runway, etc), revert to launch. If success, skip to Step 6.

4 - Think of appropriate modifications and apply it.

5 - Repeat Step 2.

6 - Launch it to orbit, if it really meant to go to orbit and beyond.

[Starhawk]

I was wondering when you guys have ideas in your head, what is it like between the idea and the finished craft? Like do you know in your mind what parts will make your idea happen?

Now that I've built a lot of different ships, I sometimes know which parts I want immediately. Other times I have sort of a vague idea, and put stuff together and dismantle it in the VAB until I start to zero in on a design.

Does it keep you up at night imagining how you will fit things together?

When I was doing my Eve Rocks Challenge entry I definitely lay awake thinking about how to make the thing work.

Do you ever draw or look at a picture so you have something to work with? Or write it down in words instead of pictures? Or something else entirely?

I use the VAB or the SPH as my sketchpad. After a lot of hours spent building, the process became more automatic and it's now easier for me to 'find my way' to the design I want.

And I was also wondering how long it took before you really knew all the parts well enough to just put something together (something that actually works), or if that is an ability that kind of comes natural.

It took me many, many hours of playtime before I felt familiar with most of the parts. It helps that I played career while I was learning, so I was exposed to the parts a few at a time and had time to read the descriptions and try to understand how they fit into the scheme. I also spent a lot of time reading the Wiki and later, the forums, to understand design better. It's certainly been a steep learning curve, but it sure has been fun.

I think these are interesting questions.

Happy landings!

[Xavven]

I start by building the payload in the VAB and checking its mass, then I switch to an Excel spreadsheet to calculate the engines and fuel I'll need for the delta-v required since I don't use KER.

The spreadsheet doesn't tell all, though. Sometimes it looks great on paper but the parts end up fitting poorly or you get the wrong overall shape, for example a tall and spindly lander instead of a stout, stable one. It helps me to use a combination of math and human judgment.

When coming up with wildly new mission concepts, I spend a good amount of time visualizing the function of each vessel/launcher/lander/transfer/etc. and how they all fit together in broad terms. Once I have an overall mission mode that works, I can apply it to almost any celestial body, so I don't have to do that level of mental work for every single mission-- just the delta-v calculations as stated above. By standardizing in this way, all that really changes for any given type of mission is amount of fuel and the size of the engines.

[Aragosnat]

Depends on the fun level. If building for myself. I finish contracts before starting as I build costly things. But, for contracts. I build what is needed and then some as I try to make sure the craft can do multiple ones to be extra efficent. Maybe. But, it is mostly try and fail or succeed. Mostly fail unless using old working sort of design.

[Little Katie]

Now that I've built a lot of different ships, I sometimes know which parts I want immediately. Other times I have sort of a vague idea, and put stuff together and dismantle it in the VAB until I start to zero in on a design.

When I was doing my Eve Rocks Challenge entry I definitely lay awake thinking about how to make the thing work.

I use the VAB or the SPH as my sketchpad. After a lot of hours spent building, the process became more automatic and it's now easier for me to 'find my way' to the design I want.

It took me many, many hours of playtime before I felt familiar with most of the parts. It helps that I played career while I was learning, so I was exposed to the parts a few at a time and had time to read the descriptions and try to understand how they fit into the scheme. I also spent a lot of time reading the Wiki and later, the forums, to understand design better. It's certainly been a steep learning curve, but it sure has been fun.

I think these are interesting questions.

Happy landings!

Oh thank goodness, I am not alone then!

[nothingSpecial]

1. Have an idea.

2. Make a basic outline (in head): main steps, main limitations, overall design, etc etc.

2. Run KSP first time in several days/weeks.

3. Start to put things in VAB one onto other.

4. Look at payoad, check delta-V, try to check for resources/antennas/RCS etc etc.

5. Design rocket.

6. Test it, look for major flaws, revert, fix, rinse, repeat.

7. Let it lay for some time, usually until this or next evening.

8. Open, launch. Look at major-but-not-visible flaws (my first manned interplanetary mission had excess delta-V, therefore excess parts, therefore was too laggy; I did only one Mun assist instead of two Mun and two Ike assists, and I still had the half of total fuel untouched; I also had a major headache managing different engine blocks manually.)

9. Learn to avoid those flaws next times.

10. Try to make next mission right after that one.

11. Lose interest.

12. Try to remember ideas for it for the next time.

[Newt]

Sketches on paper, in blue fountain pen ink. In the rain (sometimes).

This was how I have built many of my craft. I sketch them out, think about what I need, count astronauts and delta v and mass.

Then I go build them, test fly parts, and go back to the paper to sketch out improvements and tweaks, and repeat until I get where I want to be.

That or I just sort of build something and see how far I can go with it, usually with slight changes along the way. When I was new to the game, I decided to build a commsat. When I got to orbit, I noticed I still had a stage full of fuel, so I flew to Mun (first Mun orbit ever for me). then I still had fuel left, so I shot the stage at the Mun, and had an impactor. Then I went to the VAB, and tweaked it so I could land the whole stage intact. I still had a ton of fuel, so I stuck a second probe on top of that, (ion engine), landed one part of the vehicle on Mun, and flew the ion engined probe to Laythe aerocapture/orbit. That was a pretty fun succession, which was essentially undirected. But I also have fun just playing with ideas in my head for hours, trying to do exactly what I need on the first go.

[RatchetinSpace]

Craft come to me two ways:

If the class I'm in is boring then I'll usually pull out my text book and scribble down a design for a craft. I tend to draw out the mission first and all the manouvers I'll have to complete to get there and then design a mission profile from there on.

If I want to build a utilitarian craft like an SSTO or launch vehicle I'll usually just jump into ksp and let the parts and struts fly.

I then test over and over and over again, every system and subsystem and backup subsystem must be tested to make sure it will work, and I use MechJeb for my delta V calculations. Once I'm sure it will work I launch it and start the mission. Knowing me though there will probably still be something wrong that I over looked.

[AbacusWizard]

I generally start with a specific ability (or list of abilities) in mind, and consider what parts would be needed to accomplish that ability. Sometimes I sketch out a general design on paper (especially for large orbital stations and surface outposts); other times I just grab those parts in the VAB and start mashing them together. Once I've assembled the necessary parts, I start considering how it's going to get to where it's going.

[Camaron]

Process? You mean like organized thought or something?

I should try some of that.

[Leoworm]

1. Decide if it´s a mission craft, or a multi-purpose craft.

2. Just randomly start sticking parts together and get pissed that I don´t have enough budget.

3. Make the rocket less awesome and more budget friendly (Also, making it budget friendly makes it a lot safer)

4. Try it.

5. Watch the pretty Explosions.

6. Quit the game for 10 minutes.

7. Go back to Step 2 until it doesn´t blow up anymore.

8. Be happy and go back to Step 1.