Boat Momentum Efficiency Challenge

[seanth]

Boat Momentum Part and/or Fuel Efficiency Challenge

There have been challenges to build the fastest boat, and there have been heavy lift rocket challenges. Welcome to the boat momentum efficiency challenges.

Whether you are considering fuel efficiency or part efficiency (fewest number of parts), efficiency is a tough nut to crack. Arguably, building a very heavy AND fast AND efficient boat is harder than building a very light and fast boat, so let the kerbalizing begin!

Power Fuel Efficiency Scores:

• (foamyesque) Hydrofoil 2: 204,002.24
• (Ezriilc) Orca 6:  153,021.41
• (Ezriilc) Orca 5.3:  99,048.34
• (Ezriilc) Sea Train-depleted fuel: 88,772.11
• (seanth) Franklin My Dear, I Don't Give A Damn: 71,078.48
• (seanth) Crick in My Neck-full throttle: 67,473.30
• (Ezriilc) Sea Train: 63,052.18
• (seanth) Crick in My Neck: 55,997.77
• (foamyesque) We don't need no stinkin' hydrofoils: 55,967.62
• (Ezriilc) Orca 3.4: 54,098.84
• (The_Rocketeer) Soviet R: 46,966.95
• (SpannerMonkey(smce): Challenge Cat: 26,400 (estimate)
• (seanth) Watson Your Mind: 22,542.14
• (foamyesque) Unnamed: 18,543.40
• (Heffy) Make KSP Great Again-minimal fuel: 5,379.70
• (Heffy) Make KSP Great Again: 2,650.18

 

Momentum Part Efficiency Scores:

• (foamyesque) We don't need no stinkin' hydrofoils: 1,651.12
• (Ezriilc) Orca 5.3: 1,268.37
• (Ezriilc) Orca 6:  782.32
• (Ezriilc) Sea Train: 337.82
• (foamyesque) Hydrofoil 2: 305.05
• (Ezriilc) Orca 3.4: 297.89
• (seanth) Franklin My Dear, I Don't Give A Damn: 241.36
• (Ezriilc) Sea Train-depleted fuel: 224.25
• (foamyesque) Unnamed: 155.73
• (The_Rocketeer) Soviet R: 149.9
• (seanth) Crick in My Neck-full throttle: 138.94
• (seanth) Crick in my Neck: 118.08
• (SpannerMonkey(smce)) Challenge Cat: 56.36
• (Heffy) Make KSP Great Again: 25.07
• (seanth) Watson Your Mind: 19.11
• (Heffy) Make KSP Great Again-minimal fuel: 10.05

 

Just for funzies--Momentum per L/s scores ([mass*velocity]/fuel usage per second):

• (foamyesque) Unnamed: 30,411.51
• (Ezriilc) Orca 6:  13,495.18
• (seanth) Franklin My Dear, I Don't Give A Damn: 9,654.28
• (Ezriilc): Orca 5.3:                                    7,756.71
• (The_Rocketeer) Soviet R:                      7,746.6
• (seanth) Crick in my Neck:                      7,039.45
• (Heffy) Make KSP Great Again:               7,391.23
• (foamyesque) Hydrofoil 2:                       6,512.25
• (foamyesque) We don't need no stinkin' hydrofoils: 6,432.94
• (seanth) Crick in My Neck-full throttle:   6,066.44
• (Ezriilc) Orca 3.4:                                     5,906.37
• (Ezriilc) Sea Train:                                    5,570.42
• (Ezriilc) Sea Train-depleted fuel:              3,916.43
• (Heffy) Make KSP Great Again-minimal fuel: 2,964.17
• (SpannerMonkey(smce)) Challenge Cat: 2,479.07
• (seanth) Watson Your Mind:                    1,777.44

 

Momentum Part Efficiency Scoring:

Total mass of your craft, multiplied by its velocity, divided by its part number

(mass of your ship*speed of your ship)/total number of parts

or

(tonne*m s^-1)/part number = Part Efficiency Score.

Momentum Power Efficiency Scoring:

(Thrust of your ship * speed of your ship)/units of fuel per second

or

(kN*m s^-1)/units of fuel per second = Power Efficiency Score.

Momentum Fuel Efficiency Scoring (for funzies):

(mass of your ship * speed of your ship)/units of fuel per second

or

(tonne*m s^-1)/units of fuel per second = Momentum/Fuel Efficiency Score.

General Rules:

• The craft must be carrying at least one Kerbal.
• Hydrofoils are allowed BUT craft must not completely leave the water. That's called a plane.
• MechJeb is allowed since it might make it easier for people to show values in the screen shot (see submission guidelines).
• The intent is to build something using stock parts that works in KSP's normal physics. This means no mods that add parts, alter how stock parts work, or something that alters the aero- or hydro- dynamics (or other physical characteristics) of KSP. Something like MechJeb, even though it adds a part, is allowed since the part in question is essentially massless and can't effectively be used for structure, lift, or buoyancy. 
• When reporting your fuel per second values, remember to add all the fuel being used together. If you are using rocket engines AND ion engines, you would add the liquid fuel, oxidizer, and xenon gas rates together to get one L/s value.
• MechJeb is highly encouraged since it allows you to show part number, mass, etc on the same screen as your craft while it is underway.
• Use of the debug toolbar for cheating is strictly not allowed. Of course, use of the debug menu to visualize aerodynamics when you are testing is allowed AND ENCOURAGED.
• Submissions should be made in the latest version of KSP.
• Submitters are encouraged to share their craft with others so we can improve on designs. 

Submission Guidelines:

Please include images or video of:

1. The boat moving through liquid at that speed.
2. The craft's mass while moving has to be shown. It's not enough to show the mass of the craft in a build area, since fuel might be consumed getting to the water or getting up to speed.

• For the part efficiency challenge, the ship mass, velocity, and parts present must be shown in the image
• For the fuel efficiency challenge, the ship thrust, velocity, and fuel used per second must be shown in the image
• See the following post for examples

 

Craft sites:

• http://www.kerbaltek.com/craftkitchen

Edited July 23, 2016 by seanth
tonne vs kg in part scoring

[seanth]

What do boat momentum and rockets have in common? If you are Theodore von Karman, a lot.

Best know for the Karman Line being named after him, Karman did work looking at the energetic costs of transport. Of specific relation to this challenge, he published “What Price Speed? Specifc Power Required for Propulsion of Vehicles” in 1950 in the journal Mechanical Engineering. In it, he and his coauthor, the Italian aeronautics engineer Giuseppe Gabrielli, looked at “’power per unit weight’ as a function of the ‘maximum velocity’" (Theodore Von Karman and G Gabrielli. 1950. “What Price Speed? Specific Power Required for Propulsion of Vehicles,” Mechanical Engineering. 72: 775–81.)

For this challenge, we’ve tried to keep the spirit of Gabrielli and Karman’s work, but made is a bit more Kerbal.

Here I’ll show you how to get all the information you need to show your score using just the stock game, and then with MechJeb.

 

Stock KSP method

You'll need two screenshots for this: one showing your craft moving, and the other taken in map view.

In these two screenshots we have thrust, mass, velocity, fuel usage per second, and number of parts.

For the momentum part efficiency score, you need the mass, velocity, and number of parts. In this example the score would be:

(35.13tonne * 34.6m/s)/80 which equal 15.19

For the momentum fuel efficiency score, you need the total thrust, the velocity, and the units of fuel per second. Showing the thrust of each of the rockets on the ship lets me sum them all up:

[(98.0kN+98.0kN+98.0kN+98.0kN)*34.6m/s]/(0.76L/s)=

(392kN*34.6m/s)/(0.76L/s) = 17,846

• For craft using rocket engines, remember to add the LiquidFuel and Oxidizer numbers together to get the total fuel consumption.
• You could manually calculate the thrust if you take your craft's mass and multiply it by its acceleration (not gravity's acceleration. The acceleration caused by your engines)

MechJeb Method

If you use MechJeb, showing the values is more compact with a custom window.

All the information is on a single screenshot. This will be my official first entry: Watson Your Mind

For the momentum part efficiency score, you need the mass, velocity, and number of parts. In this example the score would be:

(34.821tonne * 43.9m/s)/80 which equal 19.11

 

For the momentum fuel efficiency score, you need the total thrust, the velocity, and the units of fuel per second. Showing the thrust of each of the rockets on the ship lets me sum them all up:

(441.6kN*43.9m/s)/(0.86L/s) = 22,542.14

Craft file available at https://github.com/kjoenth/KSP-Boat-Momentum-Challenge

 

Note that changes in mass, thrust and fuel consumption all influence the final score. Its an optimization game. Good luck!

 

An addendum note: The thrust value is the mass of the craft times the acceleration to the craft by the engines. That gives you a kN (using the values reported by KSP), and multiplying it the velocity gives you a kN*m/s. If you are wondering, a 1 kN*m/s is equal to 1 kWatt.

"But what does this have to do with momentum?" you say. Momentum is mass*velocity. Another way to calculate a N*m/s is to say momentum*acceleration.

Edited July 13, 2016 by seanth
Redid example submission

[Ezriilc]

19 hours ago, seanth said:

The craft's mass while moving has to be shown.

I'm not sure what the best way to do this might be.  I guess MechJeb?

And that goes for part count and fuel units as well, yea?

[seanth]

11 minutes ago, Ezriilc said:

I'm not sure what the best way to do this might be.  I guess MechJeb?

And that goes for part count and fuel units as well, yea?

Doh. I ended up deciding that the part count was an unnecessary limitation, but left it in the initial description. I just edited the initial description so it's accurate for the rules listed.

In the example 1st entry I posted, I used mechjeb to show the vessel mass, but you should be able to get the ship mass from the map view, right? In the previous Boat Momentum Challenge I had been using the map view.

Edited July 10, 2016 by seanth

[Ezriilc]

18 hours ago, seanth said:

Total score: (34.719 tons * 44.0 m/s)/0.86 L/s = 1776.32 tonnes*m*L^-1

Shouldn't your score result in some sort of arbitrary number rather than a formula?

And why do you use "tons" AND "tonnes" - aren't metric tonnes always spelled thus?

Also, how are you getting your fuel consumption (L/s) and why are we using that in the score?

Your picture is too small to read the details easily.

[seanth]

12 minutes ago, Ezriilc said:

Shouldn't your score result in some sort of arbitrary number rather than a formula?

And why do you use "tons" AND "tonnes" - aren't metric tonnes always spelled thus?

Good catch, re tons vs tonnes. Fixed.

As for the arbitrary number vs a formula: Honestly the units of the score can be ignored. I listed them in the example just to make it clear how the score is calculated.

I spend my work day making sure SI units are correct in what I wrote, so it was just habit. Since this is a momentum efficiency challenge, the SI units of the score would be (kg*m/s)/(L/s), or (kg*m)/L. If we listed the score as (kg*m)/L, the score looks unreasonably big -- 1776320 vs 1776.320 -- so we can use tonnes vs kg.

It doesn't matter too much as long as the entrant makes their units clear somehow.

The score is (mass*velocity)/(units of fuel/time). Long as we know the units being used, we can convert. I just would rather not end up with (pounds*cubit per decade)/(hogsheads per century)

 

Edited July 10, 2016 by seanth

[Ezriilc]

I kinda liked the idea of part count playing into the final score.  I pride myself on making ships with the fewest parts possible.

19 hours ago, seanth said:

Total score: (34.719 tonnes * 44.0 m/s)/0.86 L/s = 1776.32 tonnes*m*L^-1

Where are you getting the "0.86 L/s " value, and why are we using that?

[seanth]

15 minutes ago, Ezriilc said:

Also, how are you getting your fuel consumption (L/s) and why are we using that in the score?

Your picture is too small to read the details easily.

The stock resource gauge in the upper right corner reports the fuel consumption. I guess there is uncertainty about what units liquid fuels are in in KSP, but everything I have looked at indicates that 1 unit of fuel is 1L. If it makes you feel better we can just say "unit" instead of "Liter". The score would have the same numeric value.

The forum isn't letting me post a larger image. I'll see about making the image clickable so people can see a larger image.

[Ezriilc]

11 minutes ago, seanth said:

The stock resource gauge in the upper right corner reports the fuel consumption. I guess there is uncertainty about what units liquid fuels are in in KSP, but everything I have looked at indicates that 1 unit of fuel is 1L. If it makes you feel better we can just say "unit" instead of "Liter". The score would have the same numeric value.

I guess I don't see why we're interested in the rate of fuel consumption.  I thought that dividing by part count made some sense, but I don't get this.

For instance, I've just made a run that gives me this score (I think):  11433.56324

I cut the throttle just before taking the screenie - making the L/s inclusion pointless.  EDIT:  If we use part count instead, my score is 2858.39081.

Edited July 10, 2016 by Ezriilc
Added part count result.

[seanth]

20 minutes ago, Ezriilc said:

I guess I don't see why we're interested in the rate of fuel consumption.  I thought that dividing by part count made some sense, but I don't get this.

For instance, I've just made a run that gives me this score (I think):  11433.56324

I cut the throttle just before taking the screenie - making the L/s inclusion pointless.

The intent is to get at fuel efficiency. We could strap a bunch of very powerful rockets onto a craft and go very fast, but the fuel consumption would be staggering. By taking fuel consumption into account, it means we need to really think about the trade off between power, mass, and drag.

Re your score: your momentum is 11149.6702, but fuel usage is zero. So your score would be (114.473*97.4)/0, which isn't a number. 

This exposes a way to cheat, however. Someone can get their craft up to it's max speed, quickly cut back on the thrust, and take a picture to make it look like the craft is going faster with less thrust.

I can't think of a way to prevent this sort of cheating, so I guess the contest is dead.

It can be restarted as a part efficiency variation, but it's certainly not the original intent.

 

Edited July 10, 2016 by seanth

[Ezriilc]

4 minutes ago, seanth said:

Re your score: your momentum is 11149.6702, but fuel usage is zero. So your score would be (114.473*97.4)/0, which isn't a number.

Divide by zero may not be allowed in the world of computers, but from a practical POV, it certainly works.  If I have 5 apples, and there is zero people to divide them among, then I still have 5 apples.  It's quite the same as dividing by 1, just that no division is happening.

Regardless, as you point out, I could easily put the throttle at some tiny amount to satisfy the arbitrary rule of not dividing by zero.  Hence the part count idea.

But... I don't want this challenge to die.  There must be something here worth doing.

[SSgt Baloo]

Suggestion: how about simplifying scoring by making it strictly about fuel efficiency. Drop an unpropelled buoy in the water. Then see how far your vessel can travel away from it on the fuel you brought with you (make it your target). Then, when you run out, calculate how many kilometers you covered per unit of fuel (starting LF/O gets added together to determine total units of fuel used).

How well would that work?

[Ezriilc]

21 minutes ago, SSgt Baloo said:

Suggestion: how about simplifying scoring by making it strictly about fuel efficiency. Drop an unpropelled buoy in the water. Then see how far your vessel can travel away from it on the fuel you brought with you (make it your target). Then, when you run out, calculate how many kilometers you covered per unit of fuel (starting LF/O gets added together to determine total units of fuel used).

How well would that work?

Not a bad idea.  I once thought of making a challenge to see how fast one could get to the far shore to the East from the KSC beach.  It would require a ship of good efficiency and considerable size.

Edited July 10, 2016 by Ezriilc
typo

[seanth]

The thing with boats is that they are amazingly efficient in the real world if you are patient. You can transport huge masses of goods very slowly, using very little fuel.

SSgt Baloo and Ezriilc have hit on a possible fuel efficiency challenge: how rapidly can you get to the eastern shore (or some specific lat/long) from KSC.

But it also seems like we can have two efficiency challenges going: part efficiency and fuel efficiency.

The part efficiency challenge is just momentum/#parts, so easy to alter the rules. Need to think a little about the distance to travel fuel efficiency challenge rules. I'll alter the first post to make it clear there are two challenges, now.

[Ezriilc]

3 minutes ago, seanth said:

The thing with boats is that they are amazingly efficient in the real world if you are patient. You can transport huge masses of goods very slowly, using very little fuel.

Speed is also a factor in efficiency, even for boats.  Just sayin'.

[seanth]

41 minutes ago, Ezriilc said:

Not a bad idea.  I once thought of making a challenge to see how fast one could get to the far shore to the East from the KSC beach.  It would require a ship of good efficiency and considerable size.

I'm tempted to make the target be the south pole of Kerbin. Heading straight east on a heading of 90° from the KSC hits that peninsula. You need to head out at something around 98° to miss it.

But maybe that's the easy way to state the rules for the fuel efficiency part? Get into the water east of the landing field of KSC, point in a specified heading, and go until you reach land or run out of fuel....

I don't have access to KSP right now so I can't test out how some possible scoring mechanisms might be done.

[Ezriilc]

1 minute ago, seanth said:

I'm tempted to make the target be the south pole of Kerbin. Heading straight east on a heading of 90° from the KSC hits that peninsula. You need to head out at something around 98° to miss it.

The far shore has always been my default target for such distance/speed missions.  I found it to be far enough away to be a challenge, but close enough to be doable without refueling.

More recently, I was thinking that it might be cool to make the destination a place far away that requires some serious maneuvering to get to.  That requires a real ship and not just a missile on water.

It might be best to designate a box of specific coords and require that a run start at KSC Beach and end in that box.

And I still think it would be good to work part count in there somehow.

[seanth]

2 minutes ago, Ezriilc said:

The far shore has always been my default target for such distance/speed missions.  I found it to be far enough away to be a challenge, but close enough to be doable without refueling.

More recently, I was thinking that it might be cool to make the destination a place far away that requires some serious maneuvering to get to.  That requires a real ship and not just a missile on water.

It might be best to designate a box of specific coords and require that a run start at KSC Beach and end in that box.

And I still think it would be good to work part count in there somehow.

I like this idea a lot. It appeals to my no refueling circumnavigation stuff.

I can see there being the part efficiency score, the fuel efficiency score, and a part AND fuel efficiency score. I'm playing around with scoring approaches....

 

[Ezriilc]

Just now, seanth said:

I can see there being the part efficiency score, the fuel efficiency score, and a part AND fuel efficiency score. I'm playing around with scoring approaches....

I think I'd prefer a single score to keep the leaderboard simple - and so I can yell "I win!", without any caveat.

BTW, does anyone else hate this new forum software?  It's like bamboo shoots under my fingernails.

[seanth]

1 minute ago, Ezriilc said:

I think I'd prefer a single score to keep the leaderboard simple - and so I can yell "I win!", without any caveat.

lol. Well, you are definitely winning on the leader board right now.

[Ezriilc]

Just now, seanth said:

lol. Well, you are definitely winning on the leader board right now.

Yea, um... no.  That ship is embarrassingly simple.  I used that as a test and would prefer everyone forgot I submitted it. 

[seanth]

5 minutes ago, Ezriilc said:

Yea, um... no.  That ship is embarrassingly simple.  I used that as a test and would prefer everyone forgot I submitted it. 

Simple, but it elegantly exposed problems in scoring and possible cheats. But, I'll take it off the board.

[Ezriilc]

So, it's been a very long time since I last built a new water craft from scratch.  I'm finding the new hydrodynamics... ahem... less than ideal - terrible even.

I've made a craft that gets fully up on plane, but still can't get out of it's own way.  I find this very frustrating because I'm left to wonder what is needed to go faster.

[seanth]

1 hour ago, Ezriilc said:

I've made a craft that gets fully up on plane, but still can't get out of it's own way.  I find this very frustrating because I'm left to wonder what is needed to go faster.

It's not easy. I found using the aerodynamic overlay arrows helped me squeeze more speed out of designs

[Ezriilc]

Just now, seanth said:

It's not easy. I found using the aerodynamic overlay arrows helped me squeeze more speed out of designs

You ain't kidding.  I used to be able to hit 300 m/s without breaking a sweat, but today I'm lucky if I can get 75 with all my mojo.

So, what about (mass*distance)/time, all measured at a certain destination nice and far away?