Jump to content

Speedrun v1.2


Cunjo Carl

Recommended Posts

       Real-time All-flags Speedrun v1.2         (WIP)
 

** Sorry in advance if it's not displaying properly. Try zooming out by pressing Ctrl - until it looks right!
 

           FLAGS.jpg      

One year ago I staked blood sweat and tears on a wager that you could plant flags on everything in a single sitting of KSP career mode.  Though I had to claw my way to the finish line (curse you Laythe) I proved it was possible in only 8 hours.

Now we have version 1.2 with its wonderful conveniences. Flow priority! Rigid Joints! Loadmetas!! It's time to relight that torch and get back in the fighting ring. I hear a sub-6hour calling.

Join me on my epic quest to reinvent the run for v1.2 and hammer down that time. And I do mean epic quest, I'm expecting nothing less than skeletons and harpies.
 

Rules:
• Plant a flag on everything and return
• No serious clipping
• 255 part max
• No glitches :o (update 5/5/17)
 

Broad overview:
Part 1: The Shallows            Gathering resources in the Kerbin system
Part 2: Misc.                         Planting flags on all the easy planets
Part 3. Eve                           The unfair last boss of KSP
Part 4. Jool                           Jool 5 challenge on the clock
 

Comments and suggestions are heartily welcomed!

 

I'll be sharing this journey using the power of gifs! Go ahead, click the spoiler box to see a show of the flags from the previous run plus gifs, pure and distilled, from the World Wide Web of 1999.

Spoiler

 

twirlystar.gif

hamster3.gifhamster3.gifhamster3.gif

twirlystar.gif

ezgif_com_optimize_7.gif

twirlystar.gif

hamster3.gifhamster3.gifhamster3.gif

twirlystar.gif

                                                                        The many flags of the original speedrun!

 

                                                                                          I did warn you :)



Part 1: The Shallows

What's going on:
We have a lot to accomplish while still on the shallow side of the shoal! Most pressing is we need about 2M credits and 2K science. This is easy enough if we gobble up story contracts and the delicious Minmus science, but how to do it quick? Ideally we'll be able collect all the goodies we'll need within the first hour, and I think that's still in reach!
Saddly, the immolative staging that gave the v1.0 run its incredible early speed has been removed. To make matters worse, the invaluable materials bay is now locked behind a couple layers of science. Finally, the previous megaload of 5 Minmus biomes has been spread out in an unfortunately sensible way. To our advantage is rigid joints, a higher available part count and the experiment storage unit. Not really a fair trade, but it's a long speedrun, we can spend a few minutes unfurling our sails.

 

 Goal: First let's get access to the terrier and science bays, while drumming up enough cash to upgrade the launch pad.

 

The Fancy First Flier: A Charming Dead-end

We could make our first stride towards this goal by using the first launch to its fullest abilities. Using some charming tricks, we can lift 4 mystery goo containers up to space, and land far enough away from KSC to break every record all in one shot! How can we do this without any decouplers for staging? It turns out the truss struts make excellent decouplers when rocket fire is applied to them. By sticking a truss strut between stages, the bottom stage will blow off when the top stage fires. Handy! The problem is they're very draggy, so we can only really use this trick for the first set of stages. Above that, we'll rely on the lower stages to eventually just blow away on their own.

We have enough mass for 11 'Flea' SRBs, which we need to divvy into stages quite carfeully. Too few rockets and we lose to gravity, too many and we lose to drag! We'll use a big cluster to burst up to 300m/s, then a few thrust-limitted rockets to steadily maintain this speed up to 8km, and finally, and finally a series of single staged rockets to push faster and faster all the way up to space! Along the way, we'll gobble up science. Nomnomnom.

Once we get to space though, we'll need to come back, and a pod with an empty SRB attached will just nose-dive and hit the ground at 500m/s- way too fast for the little parachute to do any good. We can use the basic fins as wings to save us though! They'll decelerate our reentry, and guide our decent. They work so well that the final stage can actually fly like a glider on its way back down. It's fun! I suppose we shouldn't dally though.

Only trouble is, if we put fins on our pod we'll need fins at several key places throughout the rocket to keep it stable on the way up. This setup is far from optimized, but it does the job consistently enough.

The setup nets us 170k funds and 61 science in only 5 minutes!

                                                                                                                                                                                                         Kerbin_Bar.jpg

 

ezgif_com_optimize_2.gif          

The Fancy First Flier


Pictures:
1. "Thanks, I've got it."
2. (VAB) Carefully balanced thrust
3. Burst off the launch pad!
4. Toodle out of the soup
5. Single rockets push us to space!
6. Woh, space is high
7. Basic fins for decelerating
8. Basic fins for flying
9. Enjoying gliding
10. Empty SRB provides 'crumple zone'
11. Success!
12. +61 Science, +170k funds

 

.... So...

Why am I not jumping for joy? Well, 5 minutes is actually a long time for 60 science. We'll be needing it by the truckload soon! I think this ship is a little dead-end, and that's ok. There'll be lots of hits and misses while striving for that final goal. It's fun to learn from and enjoy all the little detours along the way. Science marches on!

  These are the first steps of a very long journey- I'm looking forward to it!

 

 

Edited by Cunjo Carl
Link to comment
Share on other sites

53 minutes ago, eloquentJane said:

This post is doing strange things to the forum layout.

Strange, yet very pleasing to the eyes! I like it.

And good luck with the challenge. It sounds like a difficult one.

By the way, do you consider Moho as an easy visit? 

Link to comment
Share on other sites

14 hours ago, xendelaar said:

Strange, yet very pleasing to the eyes! I like it.

And good luck with the challenge. It sounds like a difficult one.

By the way, do you consider Moho as an easy visit? 

 I suppose I always do things a little strange, it keeps life interesting! It is indeed a tricky challenge, but very rewarding to work towards and exciting to actually run. I hope it makes for an interesting read.
Yep, here we count Moho as an easy one, along with Eeloo, Duna, Ike and Dres. Just for the record, it definitely wasn't easy the first time I went there!

 

 

13 hours ago, eloquentJane said:

So am I the only one seeing it show up like this?

I can't even use the horizontal scroll bar to see what's going on, it cuts off half of the post.

No, you're definitely not the only one it'll show up like that for, but I am concerned it won't let you scroll! I'll take it the horizontal scroll bar in your screenshot doesn't work?

So a bit of why it looks like that, since Imgur stopped working I've had my heart set on presenting something a news letter format. I've always liked the look, and I feel like it works well. Unfortunately in this forum you need to size tables in terms of pixels, not em or (getting fancy) frame %, which makes them hang off the edge on many resolutions. I knew this in advance, and just figured scrolling could suffice despite the funny look. Given it doesn't in your case, could I ask you to try one other thing?

Simply pressing Ctrl - (zoom out) a couple times should make it display correctly, though on most browsers you'll need to zoom back to normal afterwards. If even that doesn't work, please let me know, and I'll just edit it back to a normal format. In any case, thanks for giving it a look and letting me know the troubles you had!

 

 

Edited by Cunjo Carl
wording
Link to comment
Share on other sites

6 minutes ago, Cunjo Carl said:

No, you're definitely not the only one it'll show up like that for, but I am concerned it won't let you scroll! I'll take it the horizontal scroll bar in your screenshot doesn't work?

It does work, but it doesn't seem to scroll far enough to show the whole post.

 

6 minutes ago, Cunjo Carl said:

Simply pressing Ctrl - (zoom out) a couple times should make it display correctly

Thanks, I'll keep that in mind.

Link to comment
Share on other sites

This week, we make our first bid at the early game science rush.
5.5min - 180 science - 255k funds

 

1
Humble Hopper

cap.png
ezgif_2_86b9675e39.gif

     Concept:
     • Hop off the launch pad and gobble up science

     Lessons Learned:
     • We can get the thermometer and barometer from a
            single launch

     Results:
     • 1 min - 27 science - 35k funds 
 

 

2
Mesosphere Meanderer

cap.png
ezgif_2_47e8485fbb.gif

     Concept:
     • Take a leap to space to reap its higher science

     Lessons Learned:
     • Though fun and effective, this mission took longer than I
              expected.
     • Will probably replace it with a pair of shorter missions
     • Using rockets to cushion landings has been
              wonderfully consistent in v.1.2.

     Results:
     •  3.5 min - 126 science - 220k funds

3    
Bay 'n Booster     

Cap.png
ezgif_2_7b5ed6b5ff.gif

     Concept:
     • Use the newly earned materials bays for easy science

     Lessons Learned:
     • Easy and effective, this mission can probably be scaled up
             for greater science.

     Results:
     • 45 sec - 26 science - 0 funds


 

 

Though not optimized, this is definitely a solid enough start to move forward! We're about 5 minutes in and already have access to the Thumper, the Terrier and autostruts along with plenty of money in the bank to upgrade facilities and make a launch. Next stop is orbit, Mun and Minmus for some more cash, assuming we're ready and the storyline missions haven't changed too much.

KSP Storyline Mission Mechanics: Then and Now
 

              

      

       v1.0.5
o  Collect science data
o  Launch our first vessel
o  Escape the atmosphere
-> Get to Orbit
   o  Dock in orbit *D
   o  Return to Kerbin from orbit *:P
   o  EVA in Kerbin orbit *A
   -> Flyby the Mun
        o  Gather Science from Mun
        o  EVA in Orbit of Mun *A
        -> Orbit Mun
             -> Flyby Minmus
                o  Gather Science from Minmus
                o  EVA in Orbit of Mun *A
                -> Orbit Minmus
                   o  Flyby Sun
                   o  Explore Ike
                   -> Explore Duna
                        ... Explore missions go on ...

 

 

 


 
   Key
-> ("Critical Path") All missions in the indent beneath this become available when accomplished
o  ("Optional") Will add itself to the list of available missions

*A  = only if you upgraded your astronaut complex
*D  = only if docking ports have been researched
*:P = (particularly gets in the way!)

You will get a single mission at random from the list available.

Reloading the moment the mission is chosen will reroll the dice,
so it was common to save and reload in several specific places.


If you accomplish one of the goals without a mission, the mission will not appear. I call it "poisoning the mission".

If a mission is offered but you miss it, it's gone... FOREVER. Which was a huge problem in casual play.

Additionally, non-story "Plant a flag" missions open when you orbit a given place.

Alright! Time to retest the system. Are you ready? I'm ready.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

 

                      

 ** I WASN'T READY!! It's so complicated :confused: 

        v1.2.2
o  Collect science data
o  Launch our first vessel
o  Escape the atmosphere
-> Get to Orbit (Clears optional intro missions)
   -  EVA in Kerbin orbit *A
   -  Rendezvous in orbit *P
   -> Return to Kerbin from orbit
      -  EVA Near Mun *A (required for land on Mun)
      o  Gather Science from Mun (required for land on Mun)
      o  Return from flyby of the Mun
      -> Flyby the Mun
          -  Dock on Kerbin *D
          -  Rendezvous in Orbit of Mun *P
          -  Transfer Kerbal in Orbit of Mun *A *P
          o  Return from Mun Orbit
          -> Orbit the Mun
               o  Plant flag on Mun *A
               o  Walk on Mun *A
               o  Return from being landed on Mun
               -> Land on Mun
                   -> Return from Mun surface (Clears optional Mun missions)
                      o  Collect Science from Minmus
                      o  Return from Minmus flyby ??probably
                      -> Flyby Minmus
                          -  EVA near Minmus *A
                          o  Return to Kerbin from orbit of Minmus
                          -> Orbit Minmus
                              - Dock two vessels in Orbit of Minmus *D
                              - Rendezvous in Minmus Orbit of Minmus
                                 ... likely land/flag missions ...
   Key
-> ("Critical Path") All missions in the indent beneath this become available when accomplished. These often appear with optional missions tacked on.
-  ("Hogs") When it becomes available will typically supersede critical path missions. If many hogs are available, game will choose randomly from any level. They often appear with other missions tacked on from the same level, including (rarely) critical path missions.
o  ("Optional") Will be cleared by specific higher level missions. Some can place themselves in higher level lineups. Rare to see alone with other optional missions.

*A     = Only if you upgraded your astronaut complex
*P     = Waits to show up until you've made progress in missions (Minmus)   or have any two facilities upgrades excluding the astronaut complex. Some combinations inbetween also seem to flag them.
*D     = Only if docking ports have been researched
:P     = Everything is getting in everything's way!


You will get a mission with 1-3 objectives chosen randomly from a given level.

Reloading the moment the mission is chosen will reroll the dice.

The more items are in a single mission, the more each item is worth.

If you accomplish one of the goals without a mission, the mission will not appear. I call it "poisoning the mission".

Additionally, non-story "Plant a flag" missions open when you orbit a given place.

 

resized_solobar.png

And those are the mechanics we're working with/against in the early game. The lists were gathered experimentally so some minor matters may be amiss, but they should be mostly on. By far the most important change is the "Land on the Mun" mission that was added to the critical path for v1.2, which will force a huge departure from the v1.0.5 run. The Mun is a much harder place to land than Minmus, so we'll probably tweak the first missions to push for even more science and funds early on at the cost of a few minutes.

Upcoming Objectives:
1. Get an astronaut complex before orbiting Kerbin. Do *not* upgrade R&D or Tracking facilities.
2. EVA in Kerbin orbit and return using the mission.
3. Go to Mun following missions until flag planted. Do not return.
4. Go to Minmus following mission until in orbit. Finally upgrade R&D and Tracking facilities
5. Ignore missions from here, we'll coast on money from World's Firsts. Do Mun/Minmus science.

              Next week: Charting a course to the Mun.

 

But for now, we can boggle at the enormously profitable, erroneously easy missions that are available if only you know where to look.

image.png

 

Edited by Cunjo Carl
Still still figuring out the new story mission setup....
Link to comment
Share on other sites

We'll be making most of our money through the lucrative but commmmplicated story missions, but we'll also earn some quick petty cash from World's Firsts as well! Here's the breakdown:

World's First Funds

Right at Home Goodies
  • Speed records (9.6k each)  25, 80, 250, 790, 2500 m/s
  • Distance records (9.6k each)  1, 3, 10, 32, 100 km
  • Altitude records (9.6k each) .5, 2, 7, 22, 70 km
  • First launch, First science, etc. ~8k total
     
Kerbin Goodies
  • Reach Space 16k
  • Orbit Kerbin 24k
  • Return from Kerbin orbit 24k
  • Leave Kerbin SOI 26.5k
  • Sun Flyby 38k
Mun Goodies
  • Mun Flyby 26k
  • Mun Orbit 25k
  • Mun Suborbit 26k
  • Mun Landing 31.2k
  • Mun Escape 18.7k
Minmus Goodies
  • Minmus Flyby 29k
  • Minmus Orbit 27.8k
  • Minmus Suborbit 29k
  • Minmus Landing 34.8k
  • Minmus Escape 20.8k


I did one more round of optimizing on the first 3 missions, and finally got to do the first bits of speedrunning.

 

collage_2.jpg 4_crop.jpg


 

The first 3 missions are all variations of the theme: "Hop a bit and grab some science". Mission 4, shown on the right, takes us up to high orbit to get science, accrue funds and clear some pesky storyline missions out of the way.

A low 3 minutes came fairly easily for the first 3 missions, and 7min for the orbit-in/return-from high space. The total for all 4 missions came to 10min - 415science - 540k funds. The
video misses about 30 science by reentering backwards and burning off the wrong bits of the spaceship... Woops! Fortunately, those parts weren't mission critical for now. The times could be shaved down about 20-30% through optimization and practice, but I'm quite happy with it for now.

 

100by576_1.jpg

                                                                                                                                                                                                               

 

100by576_2.jpg

 

 

From here the only way to go is up, up up to the Mun. We have two options: 1 Avoid upgrading more than 1 facility (besides the astronaut complex) to force the game to give us only the missions we want, or 2 Upgrade whatever we want and save/reload our way past the bad missions. Back of the envelope calculations estimate ~2-4minutes of time lost to saving/reloading, or alternatively ~2-4minutes of time lost from not having the upgrades... Which is better? Sounds like a fun experiment.... for next week :) . For now, here's the mission and the possible ships.


Mun Mission: Flyby, Return Science, EVA, Orbit, Land, Walk, Plant Flag, Return.

 

Low Tech

image.jpg

Save a few minutes scraping together science on Mission 4 (high orbit), and use them to upgrade facilities to our hearts content. Mission gathering will require lots of save/reload. Ironically, this ship would net the highest science of all contenders.


 

Low Parts

image.jpg

Only upgrade the launch pad, and strike forward without landing legs, action groups, and other 'essentials'.
 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
 

Low Mass

image.jpg

Only upgrade the VAB and cram the Mun lander into the tiniest ship possible with low science returns, its trial run was well behind the other two.
 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Place your bets, and we'll see the winner next week.

 

 

Link to comment
Share on other sites

your reports are still very pleasent to the eyes. it must take ages to write and "draw" them.

The world first rewards are quite luctrative indeed. Even with the reward slider on 10%! A flyby to the mun would easely yield in 4k of kredits! i wouldn't lose that much money on a trip.  

Link to comment
Share on other sites

Sorry for the late responses, my browser wound up crashing and dumped my post-in-progress. Rewriting took longer than expected... Very nice to hear from you though!

@Benjamin Kerman Totally agreed! The upcoming Minmus mission should make easy work of ~2.5kscience. Until then, each passing mission can make science about 2-4 times faster than the one before, so a bit of gentle scrimping in the beginning is nice. This brings up an interesting question: how much science is best in the end? What tech should we aim for?

@xendelaar Thanks, I'm glad you've been enjoying! I spend longer writing reports than I do actually flying, but I enjoy making these for their own sake so it works out well in the end. World's firsts are great, but for any given World's First you can get about 6-10x the cash for performing the same action through a well chosen mission. It's a little crazy. Good luck with your 10% run- I'm rooting for you! I like the report you've been putting together by the way. My vote's for the sci. jr, but I think you could succeed with anything given what you've made it through already! (see Xendelaar's mission here: All Reward Sliders at 10% Challenge)

 

 

ezgif_2_fb98beab77.gif  We have a mission flag! ezgif_2_fb98beab77.gif

optimized_flag.png

 

Huge thanks to @cratercracker for making these. It's feels grand having such a nice looking flag to use- After all the express intent of the challenge is to run around planting these things everywhere!

 

 

 

Just as a recap, this first part of the speedrun centers around: 1. Getting the Ball Rolling, 2. Making money (2M), 3. Making Science (3k)

   The Mun story missions are a wonderful and tantalizing way to make our money. Tantalizing, that is, in the sense of the Greek comedy of Tantalus straining for his grapes. It seems like every time I finally figured out the story mission system, another unexpected nuance would pop up and ruin the show! This time I have it though... I must. It took 2 weeks but there's finally a good ship and path for making it all work.

   In fact, there's 3 good ways for making it work, and the race between our 3 ships was a surprisingly tight one. Low Science took the early lead with its copious deltaV and 30m/s safe landing speed. Bit by bit though, Low Parts eked its way into the front its 2 minute build time, and silly but effective 2-part lander. But then, as if on cue, KSP seemed to change the likelihood of various missions, and suddenly Low Mass' quicker mission gathering shot it into victory in the first round. It was so neck-and-neck I couldn't help but give each design its due itteration. Much to my amazement, each upgraded nicely and the 3 stayed in a dead lock. How could I choose? Fortunately, I remembered the cautionary tale of the hungry donkey locked with indecision between 2 identical bales of hey- sometimes you gotta just pick... But which one? ...

Low Parts Wins the Mun Mission Money Maker Race! *few!*

What can I say? It was the most fun to fly. Funnily enough, in the final reckoning keeping the VAB un-upgraded (the original point of Low Parts) didn't save that much time, but the general design seemed to work well, and the mission-choice diagram for its path is blissfully simple. Want scary? Check out Low Mass' optimal mission-choice diagram:

Spoiler

 
                                               Flyby                                                                          Flyby + Sci
           .-------------------------.-------|---------------.                                   .---------------------- |
           |                                |                            |                                   |                             |

      Orbit + EVA            Orbit + Sci               Orbit                          Orbit +EVA                Orbit
      (skip Sci)                (skip EVA)                    |                            (skip EVA)               (Skip EVA)
         |                                |                        EVA+Sci                            |                             |
         \                               /                               \                                 /                             |
            Land + 2x Return                                  Land + 1x Return             Land + 2x Return + surface science

The Low Mass route needed to make a bit of extra cash to make up for not getting quite as much science. This is the result! It was very difficult to remember while running, I'm glad I get to abandon it :)
 

 

Here's a video, just for the curious to poke around. I apologize for the blinking while time-warping on Kerbin... no way around it at the moment. There's nothing hugely unexpected, but you may find it interesting to see the updated versions of the first 4 ships (first 3 minutes), which are finally feeling nice and crisp in the latest edition. The Mun mission starts at 10:30, and there's a stitch in the video at 15:00 where I decide to try a different path through the missions, skipping the orbit missions entirely and making the money back on a killer land/return trip. It worked well, and I'll stick with it moving forward.

The music came from Krelez' assembly of uncopywrited (and well credited) chiptunes, in this case Random Chiptune Mix 2. I'm also keeping these videos to only be viewable embedded in the forum to keep it on the up-and-up. I normally do my own royalty-free music mixes, but Krelez' high paced boop-be-doop barrages are downright electrifying to run to, and I hope an enjoyable background for casually watching as well.

 

 

On to Minmus!
KSP's minty-blue science playground has plenty of nice landing spots, but which one is best? I took a caveman adventure to find the best landing place for our expedition, and found this little gem:

 

Laythe_Overview_mini_png.png   minmus_jump.jpg

  Laythe_map_rounded_corners.jpg

(image care of KSP wiki)

These 5 Biomes should handily yield the ~1.5kscience we're yet to collect, and on this route all of them have flat ground in a convenient line. No more squinting trying to line up with the patterns of rocks and catch phantom biomes. These swaths of science are downright football fields. But along those lines, how should we travel between the biomes? A fast-landing lander or a high speed megarover are the obvious answers, but which will work better?  Tune in next week to find out!  *ahem*

 

Edited by Cunjo Carl
Link to comment
Share on other sites

Thanks for yet another interesting report. The layout is again very pleasing to the eyes.

Your ship designs are really cool and differ from my normal designs. I think I will borrow some of your ideas if you don't mind.( I will borrow them with or without your consent :wink:, it's going to happen).

Why did you land on the Mun first? Isn't it easier to land on Minmus instead? Never mind, after watching the video I noticed you've got enough techs to over-design your ships.

How did you pinpoint the clever landing spot on minmus with only the wiki page?

I have to give an update on my own mission, but like you said.. it's a lot of work.... :)

Can't wait to read your next update.

Link to comment
Share on other sites

On 3/18/2017 at 4:31 AM, xendelaar said:

Your ship designs are really cool and differ from my normal designs. I think I will borrow some of your ideas if you don't mind.( I will borrow them with or without your consent :wink:, it's going to happen).

Happy to be of service! ^_^

On 3/18/2017 at 4:31 AM, xendelaar said:

Why did you land on the Mun first? Isn't it easier to land on Minmus instead? Never mind, after watching the video I noticed you've got enough techs to over-design your ships.

This is actually a really involved one. You're absolutely right, Minmus is much (much) much easier to land on first. It has dramatically lower deltaV requirements, better science returns, closer biomes, flatter ground, and lower surface gravity. In short, it's great! Unfortunately we can't quite make all the money we need from World's Firsts alone, so we need to pad them out with money from story contracts, and those go to Mun first (I think it's because the Mun has a much easier SOI to hit, so the developers figure it'll be a better first mission for new folk. ). Going to Minmus first unfortunately seemed to get rid of many of the more lucrative optional Mun missions, which I originally thought I needed. Now I'm not so sure about that... If this could be rerouted through Minmus first (and find some way of not stunting our money making), we could probably save 2-4 minutes! Not a huge amount given my goal for the full run is 6 hours, but the early minutes definitely feel longer :) .

 

On 3/18/2017 at 4:31 AM, xendelaar said:

How did you pinpoint the clever landing spot on minmus with only the wiki page?

I've happened to have spent a fair amount of time bumming around Minmus, which made the hunt a lot easier. Of all places, it turns out to be right next to where I staged the video for this challenge. If you happen to check it out, fair warning, it's very silly:

 

On 3/18/2017 at 4:31 AM, xendelaar said:

I have to give an update on my own mission, but like you said.. it's a lot of work.... :)

Absolutely no pressure or hurry, but I'm totally looking forward to it! Your mission is just on the edge of possible, which makes it completely fascinating to me. I like the format you chose by the way! It's simple, and let's the content do the walking, which is perfect for the material at hand. Just in case it happens to be useful, here's some nice resources I found for image processing:

Spoiler

I don't think you'll need most of these, but I figured I'd just toss in all the resources I've found that are simple / work well.

 

JPG, PNG and GIF compression, color-balancing, bordering, etc:
https://ezgif.com/optijpeg

Image hosting:
https://postimage.io/


Picture editing and cropping:
http://www140.lunapic.com/editor/

Gif Making:
http://gifcreator.me/

 

Collage making
https://www.photocollage.com/

 

 

Edited by Cunjo Carl
Link to comment
Share on other sites

  • 3 weeks later...

Minmus, the toothpaste-green dot in the sky. The secrets it bears will yield science enough to pave our way across the solar system. Yet, as much as I love frolicking in its mellow gravity, it's finally time to finish this leg of the mission and start to put the polish on the first quarter of the speedrun. Many landers were tried, and many worked. But one lander, in particular, shown bright and brilliant for its utility in speed running...
 

image.jpg

And despite my best efforts, it's not ^this^ one :) .

The quest to find the best took three weeks of designing, fiddling, scrapping and redesigning. It's taken nearly as much time and tenacity as mapping the entire labyrinthine story mission system for the Mun launch, but I have to say, speed-lander building has been a lot more fun! I had help along the way as well. The fine folk over in the gameplay help forum left me some great pointers for building biome hoppers, all of which were wonderfully helpful in the learning process. Somehow though, the final version balked at much of the good reasoning and common sense. Speedrunning is full of quirks, I guess!

 

Biome Hopper Trials 1 & 2

image.jpgimage.jpg

The question started as a simple one: Lander or Rover?

The race devolved pretty quickly from the start though, because in the low gravity of Minmus the lines between rover and lander are quite blurred. Both of the above craft used a terrier as the primary power house, and both had plenty of zip in maneuvering. Both could be landed at 25m/s (vertical or lateral) at 2x physwarp with no harm taken, and both were quick and cheap to assemble. Despite being very different they flew in almost exactly the same way!

The lander liked running the mission by turning almost horizontal and accelerating to 50m/s in an aggressively low arc, then flipping backwards halfway to do a gravity turn style deceleration and landing. The rover liked running the mission by traveling 50m/s and bounding in long hops, flipping backwards and pitching down half way to decelerate with the main engine. They were simple enough to fly, and handled mishaps in stride so long as the landing gear springs were turned down and dampers turned up. Neither really liked high physwarps though. At this point, the mission's piloting time was about 4minutes for either.

I took a while to improve on their designs in the hopes either might start to take an obvious lead. However, as is becoming a rolling theme for this thread, they both continued to perform well- The obvious conclusion was that both are fine ways to travel, but I was looking for something a little better. So I appealed for help in the help forums...

 

Biome Hopper Trials 3 (Science Robber) & 4

image.jpgimage.jpg

 

   With a simple question I got an avalanche of good advice! A few key pieces stick out in my mind: @bewing helped me put 2 & 2 together that on Minmus, with gravity so low and terrain so sharp, rovers might as well be landers for how much jumping and flipping they do, and also brought to my attention that KSP's underlying engine Unity will glitch when more than 20 wheels/legs are present. @Aegolius13 taught me that scientists can actually reset experiments on EVA (I had somehow forgotten this). And finally, @Urses gave some great advice for going fast: Make it stout, broad (not long) and use monoprop SAS for thrust.

  Switching out monoprop SAS for vernors created the Science Robber (above, left) named charmingly by @Urses, whose early version were a substantial improvement over landers 1 & 2. Early versions (not shown in pictures) had their legs placed directly on the heavy tanks and worked great! They landed at a quick 25m/s (physwarp x2) while manuevering dramatically faster than landers 1 & 2. They also literally bolted themselves to the ground on touchdown so they could survive skidding at 25m/s and come full stop like a glue ball. Needless to say, this was fantastic for speed running. I was getting flight times of 2 minutes (compared to 4) and so declared morale victory.... but troubles were waiting to strike.

   First though let's digress to the scientist version (above, right). Compared to the other landers it was downright adorable, with the pilot and scientist strapped directly to a tiny core of engine and science! It flew fine, but between  practical difficulties with resetting experiments quickly in a speedrun and the occasional reentry mishaps even on later versions, I decided to let this one be. There's still a chance some of its design elements could improve the run, but since an extra 5 science bays are just about as easy to bring along as a scientist, the impetus just hasn't been there. Plus with something as capable as the Science Robber just waiting to be perfected for build-time, there were other irons in the fire! Little did I know that krakens were afoot.

Biome Hopper Trials 3 (Science Robber) & 4 continued

ddd.jpg

 

   Legs have always been a little wiggly in KSP, which is a huge issue when trying to land fast with high physwap. In this regard, the first version of science robber had legs placed directly on the heaviest parts (or after fuel burned, 2 away) and this worked great! But it was also painfully slow to build, requiring 3 minutes (read: a lifetime).  So, I made some compromises and designed the craft shown in the section above. At first I didn't notice the difference, but it was definitely more twitchy on landing. Then I started exploding on landings I should have made, and sometimes even while sitting on flat ground. This new version could be built in only a minute and half, so even though the legs would kick and wiggle (and even with the reloads) I was saving time... but....

   I wanted the best of both worlds! Without any good ideas, I went with a 'shotgun approach' to problem solving and just tried things at random until they worked, creating the peculiar craft above (collage bottom left). It built fast and landed much more cleanly, so science marches on!... What was up with the legs though? Out of curiosity I continued trying different permutations to see if I could figure out what caused those disastrous wiggles more in some cases but not others. I'm still not 100% sure, but it definitely seemed that legs placed directly on or 2n (even) objects away from the thing they're autostrutted to wiggle much less than legs placed 2n+1 (odd) objects away. I hardly had a clean test environment, or even a good rubric for measurement, but that pattern seemed to arise from the noise. There's a lot of bizarre behavior with those legs, which can (I've heard) be attributed to the elaborate work arounds the devs have gone through to make legs work despite Unity/PhysX being borked.

  Then I realized, my planes can land at high physwarp with even me in the pilot seat! Perhaps there was some anti-kraken magic in those wheels! With that in mind, the work of a few minutes produced a strange but lovely craft that outclassed everything else by far.

  Biome Hopper Trial 5

 

   This thing... Hideous isn't it? It's stocky, bulbous and lanky all at once! And yet it skims along the surface of Minmus like a hungry dragonfly. It can land 25 m/s at even 4x physwarp without flinching (and 35m/s at 3x). Like Science Robber, it uses rcs for its primary thrust, reaching 100m/s over longer hops and making midflight maneuvers/adjustments simple and stress free. It's quick to build, reliable to fly, uses readily available science, and even comes with an after dinner mint. I just....

   I just wish it didn't look like ... that! Eh, for how fast it moves I'm sure it'll grow on me. It regularly hits all 5 landing spots in 1:15!

   The landing gear is a big part of the success of this model. They appear to have an other-worldly ability to absorb the shock of landing. From poking at them for a while, I suspect they're directly programmed to not explode during the moment of contact despite high stresses on them to allow planes to land despite the PhysX/UNITY wheel glitches, which are known to make phantom forces agogo. Normal physics seems to resume for them shortly after touchdown, so they'll still break as intended for wild landings. Still, it explains why more energetic crashes tend to vaporize planes while leaving the landing gear (the forefront of the crash) unharmed and scattering to the wind. These are all just guesses and surmising, of course! Regardless, the end result can be used to great effect for this mission.

   The plane landing gear also can be used to hard-land without the wheels out :) (collage bottom right) . They're much bouncier this way, but are seemingly just as tough. Just a fun aside! Another little tidbit I discovered, the landing legs (small medium and large) are all about equally good at absorbing the energy of landing- the biggest difference seems to be just the distance from their attachment point to the ground.


 

Minmus Launcher

All these landers are great, but they need a way to get to Minmus. Fortunately, It's an easy breezy trip now we have the tier 3 launch pad! This mission's launcher is made by modifying the Mun launcher by upgrading two skippers to special clusters of kickbacks I call "Phlegethons" for their resemblance to the mythological river of flame in flight (collage top right, gif below). Each set is more powerful than a Mammoth so despite these landers being built the size and shape of say, a barn, those phlegethons can push them up no questions asked!

 

ezgif-1-70a343c2db.gif

 


Polishing the Run

In tandem with designing the ultimate speedy biome hopper, I've also been polishing the earlier segments of the run from the start through returning from the Mun. I've wrestled the time down to 25min typical (22min best) and may still push this lower with practice. That's exhilaratingly fast compared to my expectations. But, it comes with a downside. The run is fast because it's complicated, and the complexity is causing much more mental fatigue than expected. By the time I finish the Mun, my brain is cottony like I was 2 hours in to the v1.0.5 run! I'll get better with practice, but without a doubt, this new run will be much harder than the last one.

The Minmus mission is all set to go, so the next order of business is to finish polishing up this first section of the run! In the big picture I'll be happy with any time under an hour, but I'm secretly hoping for a sub-45min. Here's hoping!

 

Edited by Cunjo Carl
Link to comment
Share on other sites

Oh wow, again your designs are awesome! I never tried using monoprops for landing and hopping on minmus. Your hopper is really agile, yet robust! I don’t know why you’re complaining about the way it looks. A effective hopper is a beautiful hopper in my book :D

Where do you test your trail versions? In a sand box game? Or do you revert every time? And how do you measure the actual play-time? Do you have a dedicated stopwatch or something? What happens when you crash a ship when you’re “playing for keeps”, do you just restart and reset the stopwatch? Please enlighten me. :)

I really like this challenge, because it forces you to approach every mission in a very unconventional way.  Just look at Phlegethons! shooting through the sky like a comet! Just amazing haha.

Link to comment
Share on other sites

  • 4 weeks later...

Mission Summary: 40min - 2.3MFunds - 3.7kScience - Flags: Mun, Minmus

including crashes, misclicks and all!

 

image.jpg

image.jpg

 

 

 

Let's go! I finally finished polishing the first leg of the speedrun, and boy it's fun to go fast.

 

The music was my system audio at time of recording, which included 'Random Chiptune Mix 30' assembled by Krelez on Youtube. If you enjoy the music, definitely go give it a 'like'!

With a few more tries, I could hammer the time down another 5 minutes by avoiding crashes and zipping through Minmus with a bit less hesitation... but to be honest I'm really happy with these segments, slowdowns and all! I'm going to call the route for the first part of the speedrun a success, and march on to one of the next challenges: Eve, Jool, or all the rest.

I'm looking forward to doing the hardest one first so with that in mind-

From here I'll be rushing straight into the Jool leg of the mission, with a particular eye towards Laythe. I think once I conquer that old demon, I'll be feeling a bit better. My plan is to launch a double craft, have it separate in enroute to Jool, and have one half do Tylo/Bop/Pol while the other does Laythe/Val. We'll see how it goes! There's a lot to learn.

Perhaps it's surprising, but as much as I love the epic shots of flags, these pictures are some of my fondest memories of the trip.

image.jpgimage.jpg

It's fun to fly fast!

Link to comment
Share on other sites

Very entertaining video! I especially enjoyed the commentary!  With this playstyle, some kind of explanation is vital in order for me to understand why the heck you’re doing stuff in such.. uhm… “ unconventional ” ways hehe.

If this is an easy way of making a video, stick to it! It’s awesome!

It’s cool to see what kind of designs and tactics you’re using to finish as fast as possible. Initially I thought the construction wasn’t part of the challenge, but including that brings a whole other dimension into play!

 

You’re trajectories are awfully inefficient (delta v wise), yet super fast…and that is off course the main objective.

When you wrote you were hoping to complete the challenge within 6 hours I thought it would be doable, even for me. But after watching this video and your previous one, I’m sure I can’t even compete a little bit with you. You’ve got these things figured out in so much detail, that it would be almost impossible to beat. I also like the way you use the glitches to your advantage.

 

It’s really impressive what you are doing!

 

I got one small tip that could help you, but you’ll have to test it for yourself.  You could perhaps speed the process up a bit with using a scientist in your ship. If you dedicate an action group for performing all experiments on board, then you could use the data storage thingy to collect all data (with an action group as well) and after that you can reset the two resettable experiments with the scientist. Those experiments have to be within proximity of the lander can off course to make everything time efficient. If you would add an icky processing lab you would be able to reset all experiments with an action group as well, but that the resetting takes a bit more time.  It could save you some time… perhaps.. I.. I really don’t know. XD This playstyle is sooo much different of what I’m used to haha. Besides that, you already mentioned this idea in the video. My only addition would be the action groups.

 

Also thanks for the shout out. :D You made me blush hehe.

 

Btw.. why are you going to do a Jool 5 mission? That doesn’t sound like a fast thing to do at all. If you just go to Pol, Bop and perhaps Ike, I bet you would be able to completely unlock the tech tree with only those biomes. You would have to know where to land off course and you are playing a caveman (stock. vanilla) career, so pinpointing every biome could be difficult.  

Speaking of which, you’re doing this challenge completely stock? Are you insane haha!? That’s some die hard sh..stuff right there! Why did you chose to do that btw? To make things even more difficult? Or are you a purist? I think I will do a stock game as well in my next career game. It makes everything even more challenging.

 

Anywho, I can’t wait to watch your next update!

Link to comment
Share on other sites

Warning, text wall incoming!!

You bring up a good point about continuing to drive up the tech. Finishing the tech tree by hitting those biomes sounds cool, and it's something I always keep on the table when planning a new section, but the problem I always run into is.... what would I do with it? Would the extra tech save enough time to warrant the time spent gathering it (and the 1.5M for the tier 3 R&D facility)? Well, this time around I have a pre-made science gatherer with the new science container on it... Just as a thumb-in-the-wind guess, it would probably take an extra 10-20 minutes to run up the extra science (and particularly funds) to get some tasty tier3 goodies before going to Jool/Eve. So could the tech bring back the time spent getting it?

Spoiler

This was pretty stream of conscious... In the end, I'm not sure about higher tech. I'll have to give it a closer look!

The aerospike would be pretty fantastic because (in v1.2.2) it covers a tricky gap in engine thrust while having a TWR+Isp combo that grants a 10-20% boost to deltaV for that stage's alternatives (terriers, poodle, hammer, swivel). It's a little awkward because it's actually a bit overpowered for the diameter. Still, it could represent a much more robust central stage. The downside is I'd need to redesign the central stage after going interplanetary, which would cost against it...Probably a cool 5-15 minutes savings from this.

I'm decidedly 'meh' on the main-line tier3 engines. They never quite seem to be what I need for a speedrun. For me the vector winds up being a beefier skipper replacement, but it's a bit awkwardly suited for the task. It would be a nice mainsail replacement for the second stage of the Eve launcher though. Not a critical improvement, but nice. Maybe 3 minutes savings through smaller craft design? The Rhino and Mamoth wind up being pretty equivalent to the twin boar at my TWR + payload requirement for launch, but the Rhino comes out a bit ahead for transfer. Funnily, the Mamoth has slightly inferior cost and parts, which is surprising! (2 twin boar = 1 Mamoth + 2 Kerbodyne-medium tanks) .

The Kerbodyne XXXL tank could save 25 parts on the lifter, which is really nice since craft ability tends to be limited by parts later in the run. It could save a healthy amount of time, but it would mean redesigning the launcher after already going interplanetary. Call it 0-10 minutes savings?

ISRU sounds nice, but it's difficult to do quickly because you need to scan first and absolutely _not mess up_ when getting the parts together because you won't notice problems until halfway through the mission! Also, since the ore locations are randomly seeded, it introduces some more variability and picking landing-sites on the fly which could be out of plane, which I'm not excited about (there's enough drama without it!). I could maybe use it effectively on Eeloo and Moho in a speedrun setting. I think I'd have to try it.

All the other high-tech things (ions, RTG, antenna, rapier, etc) really aren't helpful enough to warrant getting. I guess the RTG could save a minute of solar panel extending over the run?

The bottom line is, I'm not really sure! Like I mentioned, most of the numbers are guesses from running before but nothing for sure. Even then, I could save 5-25 minutes from a ~15 minute time loss...  There would also be a noticeable time drop from the extra complexity....  Still, that aerospike and massive kerbodyne tank would be a treat!

 

Yeah, 6 hours is pretty breakneck fast. I'm on track to hit it, but by the time I'm through I may be cursing myself for not saying '7' instead! One aspect that I've mentioned, but hasn't really become apparent yet is that mental fatigue plays a huge roll in this final time. It's not unpleasant, just noticeably limiting- I often describe it as being 'cottony'. The last time I did a speedrun, I had to play from 8pm-4am while I was already quite sleep deprived. This time, I'll be better slept, but the run is much more complicated because I'm going glitchless v1.2. I think the two will work out about the same fatigue-wise, taking about 40% longer than possible if I was fresh the whole time.

As for stock? That stuff's easy! Go take a check out @ManEatingApe and the rest of the caveman crew for some real diehard KSPers! Silly brashness aside, I've never felt limited by stock KSP. All the abilities and challenges I've wanted from a game, I've been able to find there! I haven't picked up other mods simply because I've never felt the need, with a few exceptions (KER Hyperedit and AeroGui) for fine testing and refinement of craft, and my time programming with KOS which was a wonderful aside. The original reason I made the speedrun challenge 'ideally' mod free was to create a common playing field where people could swap tech and craft without difficulty, but it's never been a big deal for me if folk choose to run with mods set up the way they like it. ... For me though, crashing because you mistimed your suicide burn from interplanetary space is part of the fun :D

As for the level of difficulty, it's certainly different, but there's nothing more difficult to an 8-minute Mun mission than there is in say scouring all Laythe science in one shot on a tight cost budget. It's all just starting with an idea, and incrementally improving it until it shines!

Can you action group experiment resetting with a scientist? I tried earlier on, but couldn't get it to work.

Link to comment
Share on other sites

you're absolutely right about unlocking the tech tree. there is no need for it at all! many high tier tech nodes are pretty much useless. but some high thrust engines could help making faster designs.. i think. 

also. your goal is to plant a flag on every body.. right?  do you need to bring the kerbals back? else you could just hire several kerbals and give them a one way ticket to every celestial body. that would speed up the process right? 

i did the caveman challenge with version 1.0.5 and it was great. the game is completely playable without mods. you've convinced me to do my next career also completely stock.. perhaps a speed run. but ill have to do some trail runs to check if I'm suited for the job! :)

i don't think there is an action group that would reset resetable experiments.  so you would need to place the material bay and goo nearby the capsule. i can imagine that this could be time consuming... 

Link to comment
Share on other sites

@xendelaar For this particular challenge, we gotta bring the Kerbals back. The trip out is fun, but it's that getting them back part that's the real trick!

I didn't know you became a caveman- Congrats! That's a tricky challenge. Come to think of it, you'd probably be able to run some serious caveman missions from your experience in the early parts of your low return runs..

I hope you enjoy the stock playthrough! It's very freeing in a lot of ways, but also very limiting. Maybe.... maybe keep Kerbal alarm clock :D .

Sweet- yeah, give speed running a try and see what you think. My guess is it'll become apparent pretty quick whether the rush happens to make things exciting or just stressful. I'm assuming this will be for after your 10% run, but here's a bite-sized speed running challenge that might be a good barometer: 6 Minutes to Critical. Though are you sure we can't look forward to a 5% returns career mode run? :lol:

I learned recently that the MPL has an action group for resetting experiments, which gobbles up loads of electricity but can reset science probes with the click of a button. Sounds neat! Too bad scientists can't wave the same magic wand...

Link to comment
Share on other sites

  • 2 weeks later...

Video update! It covers the advancements in design for the new general-use launcher. My hope is to use 3 or maybe 4 of them for everything except Eve and Gilly.

I think I'll do a small photo/text update rather than a big video for the next one, I had to stare at this mysterious error a few too many times for my liking.

balls.png

Jeez, youtube...

Here's the synopsis though:

Q: For a vertical launcher what's better: fuel or TWR?
A: Strangely, fuel came out ahead. Still scratching my head on that one.

Q: Since v1.2 is extra draggy for flat-top tanks, should I bother with nose cones?
A: Nah!

Q: What's the fastest we can reenter Kerbin straight down?
A:  6.5km/s is no big deal! So long as you use this one weird trick :wink: (a nearly Ablator-less heat shield backpack)
    .... I've realized in retrospect this is a bit overkill. Even the most reckless Moho-return only has 5km/s at reentry.

Q: Can you reenter with explodey engines still attached?
A: Yep.

Q: How floppy is rocket?
A: Too floppy :/.... :D 

Q: What's the best possible rocket?
A: Just consult the MATH! (I mean, it's actually still super nuanced, but the math definitely helps with guiding some decision making)

Q: Wanna go to Tylo?
A: Pshya!

Well, that was enlightening. Thanks for joining me once again, and for next time please look forward to pictures of craft going all kinds of funny places, and likely my first attempts at making "direct from surface" interplanetary burns. Alright, that gives me 2 weeks to come up with a flashier term for it. :) See you then!

 

Link to comment
Share on other sites

 your latest video was really inspiring! 

i wasn't aware that nose cones aren't essential  for efficient lifting off a craft. i would have figured a great loss of dv with those lumpy parts. 

also the descend vehicle is genius! i will definitely use that design for future designs. you must have had a lot of fun figuring this stuff out.

i was wondering why you don't let the kerbals stay in space of the duration of the speed run.  taking them back just takes more time. but i then realized that having a kerbal in  space that would mean rendezvous and those things cost even more time.

the math you used is also very handy for future work.  I've been thinking of adjusting your excel sheet for optimization when i get back from my vacation. i wonder what the other contributors are working on as we speak.  

thanks again for an interesting post

Edited by xendelaar
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...