Jump to content

Something that rhymes with Schmelta-V


Recommended Posts

I've been away from the KSP scene since about March, and finally got the itch to play again. 1.5 is totally new to me, as is the Making History expansion. Forgive me if I missed earlier discussions on this, but...

The burn indicator now knows when I don't have enough delta-v to complete a maneuver, and furthermore which stages will contribute to the burn, and by how much!?

This is awesome! Wait, that means that stock KSP is calculating delta-v! Finally! It only took 4 years.

So... wait... am I missing something, here? Where's the delta-v readout in the VAB, flight, or map views? I mean, the game is clearly calculating delta-v.  So why make me have to make maneuver nodes to see it? Why doesn't the game just display that information in a window somewhere? I combed through the Settings screen and can't find it. Am I an idiot, or have the devs done all of the hard work to calculate delta-v for each stage and for some reason decided to not display it? So the intent is that players either:

  • Install a mod. KER I presume?
  • Do the calculations themselves. Personally I've been doing this with an Excel spreadsheet.
  • Put a craft in orbit first and then make fake maneuver nodes as a sort of measuring stick to find remaining delta-v in the current stage, and then subsequent stages by doing the subtraction.

That can't be right... Set me straight, guys!

Link to post
Share on other sites

It seems to still have some problems that should be overcome.  

If you plot a maneuver whilst in atmosphere, it seems to use sea-level ISP numbers.  Easily seen if you launch, raise you AP to 80k, then add a maneuver node to circularise.  For me that circularisation is typically on terriers or poodles, and burn time indicator is complaining that the vessel doesn't have enough dV, even though it actually has plenty once it clears the atmosphere.  

If you finish a burn at low throttle it seems to sometimes miscalculate the burn duration (and burn start time) for the next burn.  (It seems to assume that the next burn will also be at low throttle).   

I've also seen a number of reports about buggy behaviour with nervas.

Link to post
Share on other sites

It's not perfect yet, and yeah, there are definitely some bugs to work out.

But it's a massive step in the right direction-- and as long as we're diligent about giving them well-formed bug reports about it, I suspect it'll get fixed in due time.

By "fixed", I mean clear bugs in the burn-time indicator's delta-V display.  Wanting one in the vehicle editor is understandable, but I assume would take significant additional work, because if it were trivial to add, presumably they would have done it already.  So I wouldn't be surprised if they add it at some point, but I also suspect it would be lower down their priority queue than fixing bugs in the existing indicator.

Link to post
Share on other sites

The main issue I assume is that your crafts Δv changes all the time. Not only by burning but by how much throttle you use, how often you steer and so on. How efficient is your transfer? It's a very unreliable number! However, they added it and now they have to also finish what they begun I guess. To me it would make more sense to instead guide the player towards developing rockets which are used multiple times. Away from developing a new rocket for every payload. That's essetnially why you need the Δv number. Because you launch a new rocket pretty much all the time and you have no clue what it can do. If you would instead save every rocket without payload as a subassembly and note how much mass it can get to LKO, you could simply mount an existing rocket that you know can bring your payload to orbit without having to guess. A beginner would accumulate more and more rockets in its fleet and could maybe update them with new versions over time. I made a video a while back, where I show how you can develop a rocket in KSP, find out its capability and then keep reusing it. You can not only develop rockets like that but also upper stages (as part of the payload) which can push a certain amount of mass to the moon and so on. You can go on developing landers for each body, return ships and more. You end up having a rocket module library in your subassemblies you can stack together to get anywhere without having to calculate anything. What KSp needs in order to make this more intuitive is a better subassembliey UI where you can really organize things and not get lost. Here the video that I mentiond:

 

Edited by LukasKerman
Link to post
Share on other sites
21 hours ago, LukasKerman said:

The main issue I assume is that your crafts Δv changes all the time. Not only by burning

Actually, this is incorrect.  It is only by burning.  When we talk about a craft's available delta-V, that's literally all we're talking about:  how much velocity can the ship's engines impart to it when burning all the available fuel.

That's purely the result of applying the Tsiolkovsky rocket equation to the relevant stages.  The only numbers that go into this equation are:

  • What's the rocket's mass without fuel?
  • What's the mass of fuel carried?
  • What's the Isp of the engine that's burning the fuel?

That's it.  It doesn't matter where the rocket is. It doesn't matter how "efficient" your transfer is.  It doesn't matter how many times you burn.  It doesn't matter what your engine's thrust is, so your throttle setting is irrelevant.

That's it.  That's all that the dV burn indicator is tracking.

21 hours ago, LukasKerman said:

how much throttle you use, how often you steer and so on. How efficient is your transfer? It's a very unreliable number!

All of these other factors you're quoting have nothing to do with how much dV does the rocket have.

They're very relevant to play, yes.  But these are things that affect how much dV do you need, not how much do you have.

When you're playing KSP, as a player, you don't actually care about dV itself.  What you care about is results.  For example, you care about "can this rocket successfully land on the Mun and return."  That's what the player actually cares about.

Well, if you want to answer the crucially important question "can it do the mission or not?", then you have to know two very important things:

  1. How much dV does the rocket have?
  2. How much dv does the mission require?

If #1 is bigger than #2, then this rocket can do that mission.  If it's the other way around, then it can't.

The dV meter on the navball is only addressing #1.  It's purely a statement about how much dV the rocket is carrying.

All these other factors you mention, on the other hand-- e.g. the efficiency of your transfer, your throttle, the direction you burn, etc.-- those things are all very important to #2... but not to #1 at all.  Therefore they're irrelevant to the burn indicator on the navball, which in turn means irrelevant to this discussion.  ;)

If you have an inefficient transfer, it doesn't change how much dV the rocket has.  Instead, it changes how much dV the mission requires, which is a different thing.

21 hours ago, LukasKerman said:

However, they added it and now they have to also finish what they begun I guess.

It's already "finished".  It's done.  It works.  It's a complete feature.

That said... it doesn't work perfectly-- it still has some bugs in it.  For example, certain ship designs seem to cause it indigestion.  (I've heard reports of having problems if there are multiple LV-Ns or multiple ion engines on the ship, for example.)  But that's not something "missing", that's just a bug-- i.e. an edge case someone didn't account for.  Presumably they'll fix those bugs so it will correctly report the dV in such cases, and then they'll really be done.

21 hours ago, LukasKerman said:

To me it would make more sense to instead guide the player towards developing rockets which are used multiple times.

Why?  Some players might like to play that way-- others don't.  Pick a hundred KSP players at random, and you'll probably find a hundred different play styles.

For example, I never ever want to play the way you just described.  I like custom-building a rocket for every mission.  It's a major part of the fun of KSP for me.  So, from my personal perspective as a player, I'd be sad if they added a feature like that, because to me it would mean that they "wasted" a bunch of development time working on a feature that doesn't address my needs at all.

That's a hard problem for developers in general-- how to pick what features to implement.  To get the most "bang for their buck"-- that is, to provide the maximum player value for the least investment of engineering effort-- generally they're going to have to target features that appeal to the widest possible number of players.

Putting in a bunch of work towards optimizing the build process for reusable design is certainly something they could do.  It would certainly appeal to some players-- clearly you, yourself, would like to have such a feature.  :)  The real question is... would most KSP players want such a feature?

There's no way to know for sure, of course.  But my impression is that such a feature would not be particularly attractive to most KSP players, mainly because I've been around the forum for a few years now and I rarely hear people asking for that feature.

In comparison, the work they've done with the dV meter on the burn indicator is great, because that's something that helps virtually everybody-- and it's something that many, many players have been asking for, for a really long time.

Of course, that's just for doing burns.  Still conspicuously absent is a dV meter that's usable in the vehicle editor so that players can use it while they're designing a ship.  Lots and lots of players want that, and have wanted it for a long time, and are quite vocal about it.  ;)  So I wouldn't be surprised if Squad ends up adding such a feature in a future update, though of course that's known only to Squad.

21 hours ago, LukasKerman said:

Away from developing a new rocket for every payload.

Why?  Because you want to hurt the players, like me, who love developing a new rocket for every payload?

^ Okay, that's being facetious.  I'm sure that's not really what you mean.  ;) Really all I'm trying to say is, be wary of assuming that other KSP players will want the same things that you do-- we're a very diverse community .  A thing that you think of as a "problem" might be "important fun feature" to someone else, and vice versa.

21 hours ago, LukasKerman said:

That's essetnially why you need the Δv number. Because you launch a new rocket pretty much all the time and you have no clue what it can do.

Yes, that's why the many KSP players who routinely build rockets in the VAB have wanted this feature for a long time.  Because lots of people do this.  Yes.

I would also point out that even for folks who want to re-use their rocket designs a lot... it would still be useful to have a dV indicator in the vehicle editor, because you'd still need to figure out what the rocket's capacity is the first time you use it.

21 hours ago, LukasKerman said:

<Nice description of the various things one can do if one has a rich library of standard launch vehicles for re-use>

 What KSp needs in order to make this more intuitive is a better subassembliey UI where you can really organize things and not get lost.

Oh, I'm not arguing with you there at all.  Certainly there are users who like to build re-usable components, and certainly the current subassembly editor UI in the game is kinda clunky to use and not good for organizing large numbers of things.  Lots of room for improvement there.

And certainly, a better system for creating, managing, and using assemblies like that could be convenient.  (Wouldn't help me, personally, because I have little use for subassemblies most of the time.  But I'm sure a lot of other players, such as yourself, would find it handy.)

So... would it be nice if Squad were to implement that, sometime?  Sure.  But it's a completely different feature serving a completely different purpose than the dV indicator, which means it's not really pertinent to the topic of this thread.

Definitely worth discussing, but it seems like it would be worth talking about in a thread of its own.  Perhaps you'd like to start up a thread in the "Suggestions" forum?  (i.e. something along the lines of "Need a better system for managing subassemblies")  :)

Link to post
Share on other sites
58 minutes ago, Snark said:

It is only by burning.

I hate to nitpick, especially since you correctly address this later in your post, but your delta v can change any time your mass changes, not only when your mass changes by burning fuel. Things like ejecting a payload or fairing also alter the remaining dv of your craft. 

Link to post
Share on other sites
2 hours ago, ZL647 said:

I hate to nitpick, especially since you correctly address this later in your post, but your delta v can change any time your mass changes, not only when your mass changes by burning fuel. Things like ejecting a payload or fairing also alter the remaining dv of your craft. 

Sure. I was basically just including staging-away-stuff as part of "burn". Because I think it's a reasonable assumption that KSP players understand that doing a burn that spans stages will involve... well... staging. ;)

And the stock burn indicator handles staging.

Link to post
Share on other sites
13 hours ago, Snark said:

Actually, this is incorrect.  It is only by burning.  When we talk about a craft's available delta-V, that's literally all we're talking about:  how much velocity can the ship's engines impart to it when burning all the available fuel.

That's purely the result of applying the Tsiolkovsky rocket equation to the relevant stages.  The only numbers that go into this equation are:

  • What's the rocket's mass without fuel?
  • What's the mass of fuel carried?
  • What's the Isp of the engine that's burning the fuel?

While the general idea of your post was correct, there are situations where Isp changes (mainly with atmospheric pressure, but also by differential throttling of 2 or more engine types with different Isp, so that the average Isp changes)... but that wasn't what he was talking about/what you were responding too.

Link to post
Share on other sites
22 hours ago, Snark said:

Sure. I was basically just including staging-away-stuff as part of "burn". Because I think it's a reasonable assumption that KSP players understand that doing a burn that spans stages will involve... well... staging. ;)

And the stock burn indicator handles staging.

Is docking and undocking "Staging"?

Link to post
Share on other sites

First of all thanks for the detailed answer! You're right, Δv is just a number which is only based on the mass and engine efficiency of a craft. However, I was actually refering to the required Δv your craft needs for a mission. Sorry about that! The required Δv changes with everything you do that is off nominal. My point was since the real required Δv is pretty much impossible to predict, you can never be sure that the Δv number of your craft is really sufficient to do the burn. The indicator may show 500 m/s but your vehicle may need 530 m/s because you wobble around too much and have to steer. It would be very frustrating to run out of fuel too early despite having enough Δv on your vehicle based on what the VAB said. However, now that they put so much effort into the indicator, I agree that it makes sense to add it in that regard. If there is a Δv number on the maneuver node there has to be one in the VAB as well.

Quote

Why?  Because you want to hurt the players, like me, who love developing a new rocket for every payload? 

No, you would still be able to do whatever you want of course. Nobody is hurt by proposing to reuse rockets. My point was there is a subassembly system right now that is fairly hidden behind an advanced menue. This subassembly menue is also not very well explained and the parts you put in there are not organized whatsoever. My suggestion is to add a basic folder structure to it and maybe put it up to the front so people can see it more easily, and maybe begin using it for things they frequently re-build from scrach like launch clamps with added light or boosters with added sepatrons and radial decouplers. Things it was designed to make easier.

Quote

But it's a completely different feature serving a completely different purpose than the dV indicator, which means it's not really pertinent to the topic of this thread.

As I have pointed out you can replace Δv with payload-mass-to-destination by using subassemblies and by reusing rockets. Just to have a gameplay element that leads to the same goal as Δv. It's the same purpose but a different take on it. But anyways, my main reason against Δv is not the effort it takes to make it or the lack of accuracy of predicting it. The Δv discussion is as old as KSP and while I don't know Felipe's reasons not to add it, I'm certain KSP would've not gotten where it is had Δv been available right from the start for everyone. I claim had people used Δv they had known their rockets can make it and many funny failures you shared with your friends would've likely never happened. Δv is a double edged sword in that regard. I think new players lose their interest in the game more quickly, if they know their rocket can make it before they even launch. However, I can't prove it of course so it's only speculation but I'm sure Squad has the numbers and the fact they haven't added it yet at least indicates to me that there is a correlation. I could be totally wrong though so I'm glad I'm not the one who has to decide it :) I'm just here to share my opinion on it and I know being anti-Δv is a very controverisal one to have. It's almost like being a teacher who is against using calculators in middle school!

Edited by LukasKerman
Link to post
Share on other sites
On 10/24/2018 at 3:45 PM, LukasKerman said:

My point was since the real required Δv is pretty much impossible to predict

Um.. what? You can very accurately and reproducibly predict the general dV required for almost any given mission, within a reasonable error tolerance.

On 10/24/2018 at 3:45 PM, LukasKerman said:

I'm just here to share my opinion on it and I know being anti-Δv is a very controverisal one to have. It's almost like being a teacher who is against using calculators in middle school!

It's nothing like being a teacher who is against using calculators in middle school because there is a well-reasoned educational and developmental benefit to requiring mental arithmetics, while being anti-dV readout is based on just your personal, and imo fallacious, opinion. 

Link to post
Share on other sites
On 10/24/2018 at 6:45 PM, LukasKerman said:

I claim had people used Δv they had known their rockets can make it and many funny failures you shared with your friends would've likely never happened.

None of the funny stories I have to tell about KSP end abruptly with "And then I didn't have enough fuel to continue so I stopped."

Link to post
Share on other sites

Playstyle A:
Without dV indicator = put some stuff together, throw it at Mun, watch explosion.  Put some more stuff together, throw it at Mun, watch explosion. (Get really good, put some stuff you know well together, show everyone how it's done).
With dV indicator = do exactly the same thing, there's no need to look at the dV indicator (just as I ignore the 'engineer report' in the VAB)

Playstyle B:
Without dV indicator = use a calculator/spreadsheet to work out exactly what to put together, throw it at Mun, watch explosion.  Calculate what went wrong/slap forehead for staging parachutes with engines *again*, tune vehicle to pitch-perfect* capability. throw it at Mun, watch explosion.  (Get really good, put some stuff you know well together, show everyone how it's done).
With dV indicator = make it apparent to anyone who watches that it's your lousy flying that causes explosions but, hey, it's a really good ship in theory.

Point is - you can either learn what vehicles are capable of by trial-and-error or by calculation but the nature of KSP, flight and space mean most of us/our designs mostly fail first time and need practice to build and practice to fly.

(*roll- and yaw- perfect also recommended)

Link to post
Share on other sites

Playstyle B involves A LOT less trial and error. Its mainly related to structural failures/stability issues, etc.

With A you have to solve all those, then find that you don't have enough dV, add more fuel/stages, start all over again dealing with structural and stability issues... and so on

Link to post
Share on other sites

There is also playstyle "C".  Build the return capsule.  Look at its mass (it's in the engineers report).  Add twice that in fuel tanks.  Add an engine.  That stage has around 2k in dv.  Look at the mass of the existing ship.  Add twice it's mass in fuel tanks.  Add an engine.  That is roughly another 2k dv.  Repeat until you have enough dv for the mission.  (Make sure you have enough TWR on lander/takeoff engines.  But the engineers report can help there as well).  No dV readout or spreadsheet required. 

Now obviously different people will prefer different styles.  If I'm playing vanilla, then "C" provides an easy way to design something for a mission with a known dV requirement.  (I typically overdesign, and add a generous margin, and have sometimes landed the Munar landing stage back on Kerbin).  If I'm playing modded, then I'll normally add KER and/or mechjeb, and probably a lifesupport mod and play in x2.56 or x3.2 scales.  (At 2.56 or 3.2 scales, with lifesupport requirements as well, then I'd rather not allow extra margin the way I do in stock).  

Link to post
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...