This game is frustratingly amazing and I love it

[iNVERTEDpLAZMA]

Hello all.

This game is great. It tugs at that little kid part of me that wants to go out and fly into space and be an astronaut. I keep trying to design rockets, bases, and space stations and I keep failing miserably. I don't say that because i'm asking for help or because I'm stuck. I'm just taking a break from blowing up my space stations and venting a little. I'm sure you all know this frustration.

I'm playing with mechjeb. I've got the rendezvous pilot engaged and I watch it drive my rocket carrying 6 nearly full rockomax jump-64 fuel tanks right into my refueling station and suddenly the dark side of the moon is no longer dark. This thing is huge and it doesn't turn around very quickly at all so if it gets going too quickly towards the station it's game over. I've got to line them up like 1km away and then crawl in at like .5 to 1m/s because there is no turning around to correct and I've tried some mono-propellants but the mass of this just makes it complicated as all heck. 
I've spent the last 4 hours trying to get fuel up to the space station around Kerbal so i can move the fuel transporter to the moon to resupply my first command center and pick up the science they're holding onto.

 

I would absolutely love to see this as a multiplayer game. How nice would it be to communicate with your friend that's driving the OTHER ship and they can keep it on target while approach to dock.
I've got ship 1 set to track the target then I switch to ship 2 to set the same but since i've left ship 1 the command seems to drop so she just starts floating off track.

 

Time for bed, i'll try tomorrow.

[Streetwind]

Yeah, we've all been there

Don't worry, you'll get better! I recommend not relying on Mechjeb too much, by the way. It's supposed to be a tool, not a crutch - you should learn to rendezvous and dock without it. That will give you a better feel for how your craft handles. Control problems on final approach are 100% solvable through better spacecraft design!

(Also, starting to learn with smaller craft generally works better.)

[MinimalMinmus]

Docking became MUUUUUUUUCH more  easy after I stopped using pilots/SAS under level 3.

Another tip: it took me a while to understand how, for optimal RCS use, the monoprop thrusters must be as close as possible from the center of mass.

Edited February 7, 2017 by MinimalMinmus

[CodaThePortalmaster]

There is DarkMultiPlayer, which adds multiplayer support for ksp, although two different vehicles controlled by two people that close to each other tend to bug out.

[Lunar Sea]

2 hours ago, MinimalMinmus said:

Docking became MUUUUUUUUCH more  easy after I stopped using pilots/SAS under level 3.

Another tip: it took me a while to understand how, for optimal RCS use, the monoprop thrusters must be as close as possible from the center of mass.

All you really need is the prograde/retrograde vector which every level 1 pilot has. Not that hard to find and point to the target marker oneself if need be. 

And RCS should be close to COM in the x-y planes but not necessarily in the z, if you are using them for attitude control they will be more efficient if you have some radial distance from the center. 

[Rocket In My Pocket]

If monoprop thrusters aren't big or powerful enough for your uses just place an actual engine on the front of the vessel facing forwards. (Something small like a terrier maybe.) Switch between the main and braking thrusters via an action group when you need to slow down.

Also, as others have suggested; ditch Mechjeb. You won't ever learn to do it well if you are always relying on a computer to do it for you, you may even end up finding that you can do a better job via hand especially with large or unusual crafts.

Finally, far be it from me to question how you play as you sound like you are having fun and that's what matters....but that whole operation sounds quite excessive for just getting science from the Mun.

[Lunar Sea]

24 minutes ago, Rocket In My Pocket said:

ditch Mechjeb

This a billion times. Just do it. 

[N_Danger]

 Mech jeb, which I unashamedly used for 200+ hours, it is not perfect in all respects. 

Docking is one of those 'not perfect' things.

All of mech jeb's autopilots are wonky at times and waste a crap ton of fuel at all times. Using the information readout functions in mech jeb is highly recommended instead.

And another vote for learn to do dock(and rendezvous) manually, if you play science or career  It really opens up the game early on in the tech tree if you can.

 

 

[regex]

Welcome to the club.

[SpacedCowboy]

What a refreshing rant! No negativity here, nothing but prograde.

[Ohm is Futile]

14 hours ago, Lunar Sea said:

16 hours ago, MinimalMinmus said:

Docking became MUUUUUUUUCH more  easy after I stopped using pilots/SAS under level 3.

Another tip: it took me a while to understand how, for optimal RCS use, the monoprop thrusters must be as close as possible from the center of mass.

All you really need is the prograde/retrograde vector which every level 1 pilot has. Not that hard to find and point to the target marker oneself if need be. 

And RCS should be close to COM in the x-y planes but not necessarily in the z, if you are using them for attitude control they will be more efficient if you have some radial distance from the center. 

Actually, I would argue all you really need is attitude hold, period. Prograde/retrograde hold are useful, mind you, but I don't see those being more useful than good old attitude hold. In fact, I could imagine it spinning your ship when you really don't need it to, like if you bump your target just a little off target and the ports don't stick for some reason. That could result in SAS swinging your ship around and hitting your target like a batter striking a home-run.

As for RCS ports being close to CoM, that's absolutely, totally false. What they need to be is balanced according to CoM, but distance hardly matters. You are right about the attitude control, though I wouldn't really use RCS for that.

As for the OP, happy to see your attitude (haha) towards the game. I am slightly puzzled as to why you'd need 6 nearly full orange tanks anywhere else than on an ascent stage, but it's simply too much of a kerbal thing to do for me to judge.

[ruairiau]

I've been getting back into Kerbal after some time away. I've never managed to get further than the Mun but I'm determined this time to complete the career mode. I've been watching a lot of Scott Manley and Quill18 on YouTube which has really helped me understand the game, how to build better ships and how to fly better as well. I'd suggest taking a break now and then and watching their videos. Quill18 has a new series of the career mode for version 1.2 which I think is awesome.

[Ohm is Futile]

3 minutes ago, ruairiau said:

I've been getting back into Kerbal after some time away. I've never managed to get further than the Mun but I'm determined this time to complete the career mode. I've been watching a lot of Scott Manley and Quill18 on YouTube which has really helped me understand the game, how to build better ships and how to fly better as well. I'd suggest taking a break now and then and watching their videos. Quill18 has a new series of the career mode for version 1.2 which I think is awesome.

As much as I like Quill, I find he makes strange assumptions and statements for a game like KSP. I'll admit I haven't watched his latest series, but the previous ones made me feel like prying the mouse off his hands and stealing his headset. I much prefer Scott Manley, Marcus House or Mark Thrimm when it comes to KSP.

I have to ask, what is it you intend to do to "complete" career mode?

[ruairiau]

2 minutes ago, Ohm is Futile said:

I have to ask, what is it you intend to do to "complete" career mode?

The science tech tree

[Birdco_Space]

All you really need for docking, if you are a fan of mods, is NavyFish's Docking Port alignment indicator.  Once you learn the interface, it makes docking so much more efficient, and easily accomplishable with just SAS heading hold.  Most of my docking maneuvers are accomplished with about 5m/s of RCS delta v once you finalize the intercept and match velocities with your main engines.

 

For RDVU, the most important thing to remember that adding thrust with your heading offset from your retrograde velocity marker pushes your retrograde velocity vector away from your heading on the nav ball, and if your retrograde velocity is sitting ontop of your anti target marker, and you are relatively close to the target, you should have a close RDVU.  How I work this is to set up an initial RDVU with a maneuver node either a half or a full orbit ahead ot the expected intercept.  warp to a position about 3 minutes before RDVU, then gradually (and occasionally) retrograde burn, off setting slightly to line up retrograde vector with anti target marker.  Keeps your engines pointed in the general direction of target (allowing for emergency stops) and gets you there relatively quickly.   

[iNVERTEDpLAZMA]

 

9 hours ago, Ohm is Futile said:

As for the OP, happy to see your attitude (haha) towards the game. I am slightly puzzled as to why you'd need 6 nearly full orange tanks anywhere else than on an ascent stage, but it's simply too much of a kerbal thing to do for me to judge.

Thanks everyone for the advice and feedback. I will ditch Mechjeb eventually. It has helped me get a feel for the mechanics of the game. I did try a large number of hours solo and after about 10 hours of unsuccessful launches (I hadn't figured out SAS was a thing yet), I started researching mods. For now, though, it lets me focus on other parts of the game. For example, my need for 6 large orange tanks!

I'm designing and building a fueling station. the idea is that I can launch into orbit with a bare minimum of fuel, empty tanks, and heavy rocketry. Once in orbit I'll fill up with fuel then launch outwards from there. I was attempting to do this docking with only a primary engine. No thrusters or monopropellant.

Also, I will take a look further through the forums and sticky posts later today. Can't Kerbal 'too' much at work, but does anyone know if there is a glossary posted here somewhere?

[Nathair]

On 2017-02-07 at 9:45 AM, Lunar Sea said:

And RCS should be close to COM in the x-y planes but not necessarily in the z

I'm curious about the reasoning here. Why should your torque be applied closer to the CoM?

 

14 hours ago, Ohm is Futile said:

As for RCS ports being close to CoM, that's absolutely, totally false. What they need to be is balanced according to CoM, but distance hardly matters.

I hate to contradict you but in the application of torque distance matters a lot, in simple cases it directly multiplies your force.

[Lunar Sea]

2 minutes ago, Nathair said:

I'm curious about the reasoning here. Why should your torque be applied closer to the CoM?

 

I hate to contradict you but in the application of torque distance matters a lot, in simple cases it directly multiplies your force.

I'm talking about RCS for translation, having it close to the CoM reduces attitude changes - but like Ohm pointed out it just has to be balanced around the CoM. However since I try to use as few RCS engines as possible I don't have them spread out all over the craft, just near the CoM. 

[Sharpy]

In short: few RCS, near CoM if for translation only. Twice as many, spread as far as possible, balanced - for translation+rotation.

[Ohm is Futile]

34 minutes ago, Nathair said:

I hate to contradict you but in the application of torque distance matters a lot, in simple cases it directly multiplies your force.

Torque, yes, thrust no. If you only use RCS to translate, then torque doesn't matter. In the case of thrust, all you want is the CoM to be in line with the center of thrust and to do this you care about mass vs. thrust balance, not distance with the CoM.

 

2 minutes ago, Sharpy said:

In short: few RCS, near CoM if for translation only.

No. That's exactly not it, you could stick your RCS blocks as far away as you can, so long as they are balanced your ship will translate without inducing torque. If you want to use RCS for attitude + translation, then yes, putting it as far away from the CoM as possible is better, it will create more torque.

10 hours ago, ruairiau said:

The science tech tree

...but then you don't really need to go further than Kerbin. 

[Sharpy]

16 minutes ago, Ohm is Futile said:

No. That's exactly not it, you could stick your RCS blocks as far away as you can, so long as they are balanced

Except to balance them you need twice as many.

Place six unidirectional thrusters: one directly starboard, one directly port, one aft, one bow, one dorsal, one ventral from the CoM and you have the craft both balanced and all directions for translation covered. Place any of them far from CoM and you need a pair.

And if you can't place them in given direction and don't want to put a pair, but e.g. balance the unwanted rotation with reaction wheel while translating, then placing them near CoM but not in line with it will still be more beneficial since it will inflict less torque.

Say, you are allowed only to place three 4-directional thruster blocks and one reaction wheel on your craft (a straightforward, long, round rocket). And you need ALL degrees of freedom. What's the optimal placement?

Edited February 8, 2017 by Sharpy

[Ohm is Futile]

3 minutes ago, Sharpy said:

Except to balance them you need twice as many.

Yes, you do.

3 minutes ago, Sharpy said:

Place six unidirectional thrusters: one directly starboard, one directly port, one aft, one bow, one dorsal, one ventral from the CoM and you have the craft both balanced and all directions for translation covered. Place any of them far from CoM and you need a pair.

In fact, you can do it very simply: 4-way symmetry of the 4-way RCS blocks on-top of the CoM, not simply near it. You could also do 2 4-way RCS blocks and stick 2 more linears on top of them, but you're not going to have equal amounts of thrust in every direction, although the same could be said for the previous method since you're going to have twice the forward and backward facing RCS ports. Either way uses 4 parts and negligible amounts of mass (unless you're using a tiny probe, I guess).

6 minutes ago, Sharpy said:

And if you can't place them in given direction and don't want to put a pair, but e.g. balance the unwanted rotation with reaction wheel while translating, then placing them near CoM but not in line with it will still be more beneficial since it will inflict less torque.

True. Although I find it endlessly annoying when translating induces torque, it's hard to avoid with ships assembled in space and fluctuating fuel stores. I've found that the best way to fight it is to actually turn off RCS and SAS and then tap the SAS again and re-align your attitude after each important maneuver. Of course, planning ahead and minimizing torque in the first place is a good idea, and I suppose this is part of the reason why I dislike using RCS for torque so much.

11 minutes ago, Sharpy said:

Say, you are allowed only to place three 4-directional thruster blocks and one reaction wheel on your craft And you need ALL degrees of freedom. What's the optimal placement?

I would probably skip one degree of freedom, put two symmetrical 4-way RCS blocks on top of the CoM and rotate the ship around itself whenever I need to translate the other way. Not too sure what would happen with 3-way symmetry here.

[Sharpy]

9 minutes ago, Ohm is Futile said:

. I've found that the best way to fight it is to actually turn off RCS and SAS and then tap the SAS again and re-align your attitude after each important maneuver.

You know, that when you press caps lock (switch precise mode on), RCS thrust is automatically balanced if only possible?

[Ohm is Futile]

22 minutes ago, Sharpy said:

You know, that when you press caps lock (switch precise mode on), RCS thrust is automatically balanced if only possible?

Define "if only possible". But oh my... if that works, you've made my... month? Learn something new everyday. I mean, I knew about precision mode, but I didn't think it would help with unbalanced RCS. Would that imply having more than a single set of RCS ports and then the thrust being automatically adjusted across them to account for unbalanced CoM/CoT? That would be simply too good to be true.

[Sharpy]

1 hour ago, Ohm is Futile said:

Define "if only possible".

Less than 90% imbalance. Say, your RCS operating at 10% power causes less undesired torque than your RW can exert at 100%. Or one thruster from a pair that is supposed to be balanced but isn't, is less than 10x the distance from CoM as the other.

But if you stick a single thruster sideways on the nose cone of a rocket without a reaction wheel, there's no magic power that can prevent it from exerting torque. Or if you surround the nose with 10 vernors and place only one monoprop thruster on the tail, the game won't throttle the vernors THAT much far down.

Of course this works only with SAS active. Without SAS, respective thrusters just fire at 10% their thrust, no balancing. And it wastes some more RCS fuel than strictly necessary - but then in precise mode you're operating at 10% of normal fuel flow.