KSP Challenge: Impact the Mun!

[dvader]

16 hours ago, mystifeid said:

12min 28sec

That is impressive! The screenshots (especially from the bigger Mun Crash 4) are really nice. But, how do you even build such a huge thing? I can't "zoom out" enough in the VAB to even get close to such a size.

I tried to use smarts instead of brute force this weekend and calculated the optimal acceleration for an idealised continuously staging rocket given a certain weight. The good news is that a perfect Vector based rocket of 5Mkg and a constant acceleration of 9g could have about 29 kdV and get a time of 9.2 minutes. The bad news is that I can't build a perfect rocket... There were lots of self-induced vibrations, explosions, reloads and debugging and redesigns before I got an impact at 13:30 which is much worse than my previous non-smarts-based rocket... But, the graphs look nice!

 

I take no responsibility for the correctness of my calculations...

[mystifeid]

5 hours ago, dvader said:

But, how do you even build such a huge thing? I can't "zoom out" enough in the VAB to even get close to such a size.

You need to move the whole thing. Left clicking on the root makes the craft stick to your pointer until released with another click. Shift - left clicking on any part does the same thing. The VAB will not allow you to move the root outside the building but you can move it right back into a corner which should easily allow you to add, for example, four or five more mammoth stages. After that you could try re-rooting the craft and moving it further although this might affect the symmetry but I've never done this because by the time you've got that far it's taking a long, long, long time to test each stage. (Last time I stopped well short of the limit). Also if you don't go right to the limit you can still add launch stabilizers (lots of them) - without these your craft will usually quickly break up on the launchpad but since time is likely moving at one hundredth of it's normal pace you can still launch perfectly given the right timing. It's interesting that the space bar seems to preempt everything else.

Don't try to move the root of your craft into the corner of the VAB all in one go - do it bit by bit. Likewise when you are finished adding stages - if you do not have lots of launch stabilizers and the craft is at an awkward height - it will help to move the root back to the center of the VAB.

This information will be like a curse. If you thought this challenge was a time sink before, things are about to get ten times worse. The biggest of my craft had 72 mammoths in each of the three outside stages and adding the last stage was a nightmare. Man, did it take a long time to test it. Mercifully, the first time I did, it flew straight and true.

This is also memory crash territory (for me) so I save often with different sequentially numbered names. The biggest craft also required me to quit the game and reload it in between finishing building and launching.

And at this point, you need to be sure about each stage. While you can change the fuel load of the two outer stages fairly easily, changing stages inside of these can be problematic and will often (for me) lead to a memory crash.

5 hours ago, dvader said:

The bad news is that I can't build a perfect rocket... There were lots of self-induced vibrations, explosions, reloads and debugging and redesigns

I know what you mean. The good news is that I've learnt a lot about building things (quickly). The thing that does my head in is the symmetry. It's impossible to maximize space with rockets and keep even spacing. What looks harmless in an inner stage often turns into a deal breaker ten stages out.

5 hours ago, dvader said:

But, the graphs look nice!

Very, very nice! I'd like to see a 9 minute 29km/s craft! I hope you get there.

Anyway I find it great fun getting rockets to 10km/s. The good ones I'll keep building until they get to 15km/s. Not so much fun. The really exceptional ones I take further (which is painful), otherwise I start again. The process has left me feeling drained so I think I have to go back to driving my rover around Eeloo for a while. Good luck!

Edited July 10, 2018 by mystifeid

[dvader]

22 hours ago, mystifeid said:

Shift - left clicking on any part does the same thing.

Anyway I find it great fun getting rockets to 10km/s. The good ones I'll keep building until they get to 15km/s. Not so much fun.

Thanks for the tip! I also got a bit tired after last weekends repeated failures, reloads and sub-par performance. But, there will be a next time.

I liked this challenge because it wasn't just about dVs but about a good combination of dV and TWR. I've built high dV rockets before, some tiny high TWR rockets (for the high G challenges) but not high TWR and high dV at the same time. I still believe there is some "great" design waiting to be found but the problem was that the clever/complicated deisgns didn't scale to the dVs needed due to shaking, collisions and strut-adding-hell. I believe a successfull design must be simple so it can effectively be built at a large scale.

Have fun driving around Eeloo. I'm off to continue my NCD Caveman attempt (well, that and vacations).

[wizzlebippi]

At 15:02, I think I'm close to the max of what an xbox can accomplish.  Didn't bother to look at the total part count, but those vectors sound like an evil horn.  

[tonimark]

still not made it:

lost for 2 minutes

[mystifeid]

New time - 12mins 8secs - no mods, stock install. Attained 23.7km/s at 8:15.

And look ma, no Vectors. First time. But what the rocket did have was 312 Mammoths, 96 Mainsails, 61 Skiffs and 1 Spark. All up only 3407 parts and 33.5 kilotons. Finally realized that gimballing engines, alternators and probe core charge make batteries and reaction wheels totally unnecessary. And the craft ran only on SAS hold all the way to the Mun - with no course corrections.

Before having to move the root to add the last six stages this rocket was already hitting the Mun in 12:50. Which is why I persevered. Funny that another 264 Mammoths shaved just 42 seconds off this time.

Imgur album - https://imgur.com/a/BMhECja

Nice fourth stage separation

Edited July 19, 2018 by mystifeid

[mystifeid]

@tonimark  - here is a recipe for success. To begin, start a new sandbox game. Even better, copy your KSP folder to another location and remove all the mods because you sure don't need them for this challenge and then you'll have a stock install for future stock challenges besides your regular modded install. If you decide to do this, again, start a new sandbox game. You will need to do this because your launch time is going to be 1:37:30 (that's 1hr 37min 30sec) on day 1. If you launch at this time all you will need to do is hit the spacebar a couple of times to stage your rocket and the example below will happily plow into the Mun in 19 minutes and 33 seconds after attaining 11.47km/s. It has three stages and just 22 parts which are listed below from top to bottom. Start with number 2 - the probe core. Out of habit I always autostrut every part in the (central) stack to root which stops a lot of wobble.

Also, before launching, turn on SAS hold (the "t" key) and leave it on for the duration and remember that no other input is required other than that required to stage the rocket.

After you have seen how easy it can be, I'm hoping that you can do much better. (Anyways, I'd be disappointed if you pop up a screenshot here with a time around 19:33)

1. Small nose cone (Aerodynamics)
2. Probodyne Okto2 Probe Core (Pods)
3. 3 x Oscar B Fuel Tank (Fuel Tanks)
4. Spark Engine (Engines)
5. TD-06 Decoupler (Coupling)
6. FL-A10 Adapter (Structural)
7. FL-A151L Fuel Tank Adapter (Fuel Tanks)
8. FL-TX900 Fuel Tank (Fuel Tanks)
9. Skiff Engine (Engines)
10. TD-18 Decoupler (Coupling)
11. FL-A215 Fuel Tank Adapter (Fuel Tanks)
12. Kerbodyne ADTP-2-3 (Structural)
13. Kerbodyne S3-S4 Adapter Tank (Fuel Tanks)
14. Kerbodyne S4-512 Fuel Tank (Fuel Tanks)
15. Kerbodyne Engine Cluster Adapter Tank (Fuel Tanks)
16. 5 x Mainsail Engine (Engines)

Edited July 19, 2018 by mystifeid

[tonimark]

oh k i will try this later

[cratercracker]

@Darth Badie! Here is a little patch, as an alternative to the official ones.

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

60

80

100

 

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

Decided to abandon the part with text because it looked a bit unfitting.

 

[Darth Badie]

1 hour ago, cratercracker said:

@Darth Badie! Here is a little patch, as an alternative to the official ones.

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

60

80

100

 

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

Decided to abandon the part with text because it looked a bit unfitting.

 

Love it! Amazing as always! Thanks @cratercracker

[Enceos]

On 7/11/2018 at 5:10 PM, wizzlebippi said:

At 15:02, I think I'm close to the max of what an xbox can accomplish.  Didn't bother to look at the total part count, but those vectors sound like an evil horn.  

 

Now that you mentioned it, I always hear a sound of horn instead of rocket engines... My brain was rewired! o.O

[JoE Smash]

On 7/9/2018 at 6:02 PM, dvader said:

That is impressive! The screenshots (especially from the bigger Mun Crash 4) are really nice. But, how do you even build such a huge thing? I can't "zoom out" enough in the VAB to even get close to such a size.

I tried to use smarts instead of brute force this weekend and calculated the optimal acceleration for an idealised continuously staging rocket given a certain weight. The good news is that a perfect Vector based rocket of 5Mkg and a constant acceleration of 9g could have about 29 kdV and get a time of 9.2 minutes. The bad news is that I can't build a perfect rocket... There were lots of self-induced vibrations, explosions, reloads and debugging and redesigns before I got an impact at 13:30 which is much worse than my previous non-smarts-based rocket... But, the graphs look nice!

 

I take no responsibility for the correctness of my calculations...

There is a mod called "Hangar Extender." It adds a little button on the bottom of the VAB/SPH screen (looks like an arrow I think) that when you press it, it takes you outside the respective hangar. It allows you to basically build as big as you computer can handle without crashing....I imagine....

It says version 1.4.1 but it has been working fine with 1.4.3 and probably 1.4.4 as well. I got it off CKAN...

 

[mystifeid]

14 hours ago, JoE Smash said:

There is a mod called "Hangar Extender."

That might be useful if I wasn't trying this stock and if the VAB didn't eat the 15 gig of RAM I can give it so easily. But I find it's not just the size of the craft that matters, it's what you do and how you do it that counts.

New time - under 12 minutes!! - 11min 55sec. The craft is similar to the last one but with less fuel in the inner rings and it's called "Strike Me Down Now Lord".

The "Strike Me Down Now Lord" lifting off.

Imgur album - https://imgur.com/a/nKeTcwe

Edited July 22, 2018 by mystifeid

[swjr-swis]

2 hours ago, mystifeid said:

The "Strike Me Down Now Lord" lifting off. 

Not so much lifting off as pushing Kerbin out of the way.

Somewhere at the other side of the kerbal universe, an alien scientist picks up their first evidence of a gravitational wave, and theorizes that two massive black holes must've collided....

[Darth Badie]

Thanks everyone for participating!  You will receive your badges in the next hours  

I hope you had fun