Realism in KSP

[Motokid600]

I like the difficulty options idea. But.. there shouldn't be too many options. Well... There could be. I want there to be sliders, toggles for all this. But think how complex these forums would get in gameplay discussions.

"Oh what settings do YOU have?", "What settings do YOU have?" "Well I play with these settings." " Oh, well I play those settings."

Next thing you know people will have to post there enabled features in their sigs. Not to mention as the game gets more popular and the community grows... There will be more and more... undesirable users to deal with. ( But more mods to deal with them. ) That will be unavoidable, again as the playerbase grows.

TL;DR - I just hope such optional features don't have the future community breaking up into factions.

Makes me wonder if making these realism features would be better off NOT being optional. Only by disabling with a cfg edit. Or cheat menu..

[Liowen]

Wait, you actually have access to that amount of data? I'd actually like to see that, because then that helps me with development and prioritizing features / fixing compatibility / handling bugs.

No I do not, since Github tends not to show download numbers (or at least I didn't see any when I glanced), but if you have had 14 revision's that is 14 separate downloads for each user. Then you figure that with people who try it and do not like for whatever reason, or have to redownload it because they got a new pc, cleaning up their mods, deleted it, or what have you and that download number goes up, but it is not an accurate view of the total number of players using it as it will eventually be higher than those actually playing the game. While there are a lot on the forums who do use it they do not represent a large majority of the player base playing the game.

As far as not seeing rude comments I have seen plenty of people telling users that "if you are not using FAR you are not playing the game" and such, but it is not all those that use FAR.....just those special ones. Personally I hope SQUAD puts it in as an option that players can pick to use or not, the person who responded to my original post seemingly missed that part of my comment. As for caring what people on this, or any other, forum think or say about me I don't care, in fact I have been insulted by people who mean more to me thn they do and didn't care about them either... so I guess that puts them even lower on my care list.

[ferram4]

No I do not, since Github tends not to show download numbers (or at least I didn't see any when I glanced), but if you have had 14 revision's that is 14 separate downloads for each user. Then you figure that with people who try it and do not like for whatever reason, or have to redownload it because they got a new pc, cleaning up their mods, deleted it, or what have you and that download number goes up, but it is not an accurate view of the total number of players using it as it will eventually be higher than those actually playing the game. While there are a lot on the forums who do use it they do not represent a large majority of the player base playing the game.

I meant total number of players, since surely you would need to know something like that in order to come up with percentages, right? Surely if you're going to argue against something by using the smallness of the number of people that support it that your numbers have some actual data behind them, and not simply make up numbers that support maintaining the stock status quo above all else.

As far as not seeing rude comments I have seen plenty of people telling users that "if you are not using FAR you are not playing the game" and such, but it is not all those that use FAR.

So then surely you can provide a direct link to some examples, right? I've seen FAR users talk about how it makes the game more fun, I've seen them talk about how they enjoy the challenge, I've seen them talk about how they'll never go back to the stock atmosphere. I've never seen them say that anyone not using it isn't playing right.

[thylordroot]

I think that the current solution of enforcing realism through mods is a good approach, personally. I haven't read through all 28 pages of the thread (and that is probably to my detriment), but one argument seems to be that new players have expectations about the way KSP should work and that when it behaves in a manner inconsistent with their experience, the result looks unpolished. What I'm about to say is anecdotal and should be taken in that context, but in my experience, new players don't see it this way because they have rarely have enough experience to know what to expect to begin with, and whatever model of rocketry they've formed in their heads is drastically flawed, anyhow. Let's look at some common mistakes that new players make:

• New players often launch straight up when they are trying to achieve orbit.
  
• New players often build insanely huge rockets (then again, who wouldn't?), even though a smaller rocket would get the job done.
  
• New players often burn radial when they are trying to leave orbit and will burn directly towards targets now instead of trying to find a suitable transfer orbit.
  

I will start off by saying that I think there is merit in shattering misconceptions that arise out of common sense. I think KSP is fun simply because when you fail, it is usually catastrophically and hilariously. With that being said, I would hazard to wager that the new player's experience with respect to orbital mechanics is fundamentally incomplete; I know that mine was when I started playing KSP. As a result of KSP, I've gotten to the point where I really wanted to do my own self-study on the issue. I think we sometimes forget that to these players, even getting into orbit under stock conditions is rewarding (even under stock conditions) because they have no concept of what constitutes a realistic rocket in this case. I now play with FAR, RemoteTech, TAC Life Support, and Deadly Reentry, which I have found can (and often do) punish relatively minor sins against physics with an iron rod. I've gotten to the point where I actually like this, but I am not so sure a new player will. If they really want that experience, our capable and diligent mod community has toiled without end to provide it to them.

Don't misunderstand me: I think there is certainly room for improvement in KSP and some additional realism in stock will help that. I occasionally also find that Squad puts too much emphasis on new players sometimes, and that has occasionally manifested itself in some rather silly ways (like spaceflight preceding the invention of the wheel and manned exploration preceding probes). Spaceplanes are still a chore to design (although they appear to be receiving some much-needed love in 0.25). For all of stock's idiosyncrasies, however, I think that it kept me motivated to keep on playing and to eventually desire a more realistic simulation.

As others have suggested, perhaps the best way to approach this issue is to allow certain aspects of the simulation (such as aerodynamics model) to be chosen at the start of a new save; then, it's just a matter of providing sane defaults. I wouldn't blame them for not focusing on this quite yet, however; I'm sure they have a lot of things they want to get knocked out first, and they seem to know what they are doing (I haven't yet seen a patch that was anything short of awesome). Until then, knowing that mods are available to mitigate this issue has worked wonderfully for me.

[tater]

Too many options is a non-issue. The game is already modded, and people play countless ways. Nothing suggested in terms of "under the hood" realism requires that anyone radically change their gameplay---certainly not more than other projected updates might (contracts, etc, etc). Most could be easily toggled with a difficulty setting, too.

[NathanKell]

I feel like aerodynamics, while perhaps the most obvious area, is perhaps both sucking a bit too much air out of the room (heh), but also not the absolute best example.

Let's consider two examples.

1. Temperature. Right now the *only* reason temperature matters (please correct me if I am wrong; I am unaware of anything else) is because if a part's temperature exceeds its maxTemp, it explodes. A few things produce temperature (not heat, heat is never tracked; *temperature*), notably engines.

Would it really change gameplay for the worse if parts had sane maximum temperatures, and temperature production were proportionality scaled down? I doubt it, since the amount of time until an engine overheated could be made the same, etc.

However, we would no longer have the silly situation of parts which have a higher temperature tolerance than berrylium (a heat shield material) has as its boiling point.

2. An even sillier example: In 99% of the code, g0 is treated as 9.81m/s^2. Fine; that's the usual approximation. However, when an engine module calculates how much fuel it needs (by dividing thrust by Isp and by g0), 9.82 is used. wut?

[DaMichel]

Details ...

I hope reentry heating will eventually be included in the game. It is so very silly to see stuff in flames, to wait for it to finally explode but instead the flames clear and there is no scratch. For extra silliness deploy parachutes while the craft is burning

Not saying details should not be addressed. They should.

Let me add number ...

3.) No more 6 year missions in lawn chairs!

[peadar1987]

Just to lighten the mood a bit:

I play with FAR, and as many of you will know, if your rocket strays too far from the prograde marker in atmosphere, it will tend to tumble unless you have an insane amount of drag at the fiery end, usually breaking themselves to pieces in the process.

Having realised this, I started to build insanely chunky rockets, strutted like crazy, and covered in girders until it looked like a hedgehog, that could survive such a tumble and continue to fly, instead of the much more obvious step of getting better at flying so the tumble didn't happen in the first place.

There's a reason I'm an engineer and not a pilot!

[Teoman2]

I'd like to be the first one to notice (no offence intended, just a statement)that whenever a person says something like this, all the big posters and etc come down and disagree. I just wanted to point out something that I saw.

[Linear]

I'd like to be the first one to notice (no offence intended, just a statement)that whenever a person says something like this, all the big posters and etc come down and disagree. I just wanted to point out something that I saw.

We all are allowed to have differences of opinion so why not? Just because someone has high post count doesnt change the value of what or why they wrote the post.

[Renegrade]

2. An even sillier example: In 99% of the code, g0 is treated as 9.81m/s^2. Fine; that's the usual approximation. However, when an engine module calculates how much fuel it needs (by dividing thrust by Isp and by g0), 9.82 is used. wut?

wat?

WhyyYyYyy!?!

KHAAAA--er, I mean, SQUAAAAAAAAD!

There's a five minute fix right there. Doesn't C# have #defines?

[Cdr_Zeta]

Hello,

Totally disagree with the OP - however yes people can play KSP like a game and it is a game for them at the end of the day - the thread is merely a stance on game and non-realism VS simulation and adding realism to add a fun factor AND a challenge factors; I tend to go with the latter as I have discovered this at the end of the day with Modders putting out great MODS and in essence is what is making this 'game' a 'simulation' for me at the end of the day; if it wasnt for the MODDERS I wouldnt be here.

I see the KSP Devs are adding these things but want to keep some kind of game vs simulation to a minimum maybe. This is a Unity game and is free more or less anyone can write a game like this but have to buy licenses if they want to sell their product; this is a future of gaming allowing MODDERS to participate in gaming development.

Now yes it is simplistic in certain graphical ways; but it is what's on the inside that counts; "...there's alot more going on here folks..." to quote a famous blind fellow ...

It may be a game to some, and a simulation to others; I use food and water and oxygen if Kerbals run out they die SIM over; I like it that way. Thankfully there are different people in this world or it would be a pretty boring place.

Cmdr Zeta

[Ippo]

Doesn't C# have #defines?

Thank god it doesn't

[Xyrus]

Thank god it doesn't

I...I still have nightmares...the macros...the macros won't stop SCREAMING!!!!!!!!

[thylordroot]

I feel like aerodynamics, while perhaps the most obvious area, is perhaps both sucking a bit too much air out of the room (heh), but also not the absolute best example.

I think stock aerodynamics is sort of the elephant in the room and the one that most frequently inhibits gameplay. KSP's drag model works great for rockets, but not so great for spaceplanes; and since few of the stock parts generate lift to begin with, it encourages wing spam.

1. Temperature. Right now the *only* reason temperature matters (please correct me if I am wrong; I am unaware of anything else) is because if a part's temperature exceeds its maxTemp, it explodes. A few things produce temperature (not heat, heat is never tracked; *temperature*), notably engines.

I can see a couple of interesting use cases for where thermal management becomes important. For instance:

• A high temperature environment might kill kerbals and ruin electronics.
  
• Conversely, the opposite extreme could cause kerbals to freeze to death, keep engines from starting, and might even cause probes to cease functioning because they've undershot their minimum operating limits.
  

Squad could also let excessive G-forces rip parts off of your rocket and be a death sentence. I think neither have occurred simply due to gameplay reasons; rocket building might be significantly slowed down by this sort of thing.

Would it really change gameplay for the worse if parts had sane maximum temperatures, and temperature production were proportionality scaled down? I doubt it, since the amount of time until an engine overheated could be made the same, etc.

It really depends on what changes to temperature you propose. If you propose this in isolation, then I doubt it would either (in fact, I'm not sure how it would significantly impact gameplay). Yet, when we start talking about alternate consequences of temperature, then it possibly could, and it goes fairly deep. We have to start considering how temperature propagates throughout the spacecraft, and whether or not a situation like thermal runaway should be permitted.

However, we would no longer have the silly situation of parts which have a higher temperature tolerance than berrylium (a heat shield material) has as its boiling point.

Kerbin is apparently a good source of bizarre material; after all, Kerbin's mean density is some 10x greater than Earth's! And the list doesn't even stop there: just look at the specific impulse of Kerbal engines (especially their jetpacks), or the fact that their helmets can withstand impact at terminal velocity, or that they carry multiple tires many times their own size in their pockets. The Kerbol System is a strange place.

Despite all of this, I think there is some benefit in stock's cartoonish approach to, well, everything. I can crack open a beer and pilot my rocket recklessly. A mistake in stock is usually a blunder caused by the fact that I was not paying any attention to what I was doing. KSP's apparent lack of congruence with reality rarely enters my mind when I'm playing; in fact, I've come to count on it as a source of entertainment. The realistic mods are tremendous fun, too, but they do not share this same sort of gameplay experience, IMHO.

2. An even sillier example: In 99% of the code, g0 is treated as 9.81m/s^2. Fine; that's the usual approximation. However, when an engine module calculates how much fuel it needs (by dividing thrust by Isp and by g0), 9.82 is used. wut?

I absolutely agree with this sentiment: this is a bug and should be fixed. It can do nothing but improve rocket performance.

I'd like to be the first one to notice (no offence intended, just a statement)that whenever a person says something like this, all the big posters and etc come down and disagree. I just wanted to point out something that I saw.

I won't speak for the OP, but I welcome this. The "big posters" have probably been playing for a while and have meaningful input into the discussion. Their disagreement should be welcome (that is, if they're not being contrary for the sake of being contrary; and I haven't seen any evidence of that yet) because it can genuinely improve gameplay. I am very much new to the forums and in comparison to some of the posters, to KSP as well (I really only got into it around June 2013), but I think some good points are being raised in this discussion.

Even better, some of the mod developers have gotten involved. These people are familiar with KSP's internals (at least more familiar than I am) and probably see KSP in a way different way than we do. When NathanKell spoke about what I can only are magic numbers in thrust calculation, that does smack me as a legitimate gripe. Even that .01 m/s^2 error stacks up quickly, and its something that some time with grep would probably fix. This is a bug that could be fixed; it's just a matter of Squad finding the time to fix it. Ferram4 has occasionally cleared up some misconceptions about how stock actually behaves.

There's a five minute fix right there. Doesn't C# have #defines?

Not as such (thought there's nothing preventing you from running C# code through cpp), but I'm pretty sure what you really meant to ask is if it supports constants, and it does for primitive data (c.f. C#'s const keyword). IIRC, these are permitted to be inlined at compile time. It has about the same result as I think you are asking for, and you're absolutely right: magic numbers make a large codebase a veritable PITA.

[Northstar1989]

I like the difficulty options idea. But.. there shouldn't be too many options. Well... There could be. I want there to be sliders, toggles for all this. But think how complex these forums would get in gameplay discussions.

"Oh what settings do YOU have?", "What settings do YOU have?" "Well I play with these settings." " Oh, well I play those settings."

Next thing you know people will have to post there enabled features in their sigs. Not to mention as the game gets more popular and the community grows... There will be more and more... undesirable users to deal with. ( But more mods to deal with them. ) That will be unavoidable, again as the playerbase grows.

TL;DR - I just hope such optional features don't have the future community breaking up into factions.

Makes me wonder if making these realism features would be better off NOT being optional. Only by disabling with a cfg edit. Or cheat menu..

I can see your point about the community possibly breaking into factions, and that wasn't actually something I had thought of before reading that, so +1 for bringing a new concern/idea into the discussion!

I, personally, am nonetheless in favor of expanding the difficulty options to include many related to realism. You see, whatever fractionating of the community might occur from difficulty options- it's already bound to happen as the devs have confirmed difficulty options for 0.25

IMHO, the best way to handle it would be to make SOME realism options part of the base game (like improved aerodynamics, which was always supposed to be a part of the final product), but other realism settings optional (like a more realistic-sized Kerbin, although I think the default should be at least 20%-scale and the current stock >9% scale should be an "easy" setting below that).

Some unrealistic aspects of the game actually *prevent* the devs from expanding KSP in new and interesting directions- like the lack of a life support system inhibiting the development of things like greenhouses and more meaningful permanent base infrastructure (including ISRU systems to obtain Oxygen from the atmosphere and such...), or the lack of a more complex fuels system creating interesting and meaningful choices players can make on what type of rocket engine to use (certain engines could only burn certain fuels, like in RealFuels mod, for instance...) I think these realism factors should be part of the base game, because they create new and more interesting directions to expand the game in.

Other aspects of realism, such as a more realistic-sized solar system, tend to more strongly affect the difficulty of the game- although it must be admitted that having a planet where curvature is visible at sea level also detracts from the aesthetics/visuals and sense of immersion... These options would best be chosen using difficulty settings...

Finally, some aspects of realism (such as working geysers on some of the Joolian moons, which the devs were working on a while ago; or clouds/waves on Kerbin) are more visual/aesthetic than anything else, and aside from the GPU/CPU cost of these features, can only *add* to the gaming experience... These should be defaults, but have the potential to be disabled through the graphics settings screen...

Regards,

Northstar

Edited September 3, 2014 by Northstar1989

[Joshwoo70]

If anyone Want realistic of EVERYTHING then get your self realisim overhaul. Baat going to die

[Sky_walker]

If anyone Want realistic of EVERYTHING then get your self realisim overhaul. Baat going to die

If you'd read through the first page of this topic - you know that NOONE wants to see everything to be realistic.

I can see your point about the community possibly breaking into factions, and that wasn't actually something I had thought of before reading that, so +1 for bringing a new concern/idea into the discussion!

To a degree - it already happened. In a split between Sandbox and Campaign players (it's most obvious is a suggestions forum)

But even if we ignore that - split will happen in 0.25 and it's new "difficulty options". It's pretty much unavoidable.

IMHO, the best way to handle it would be to make SOME realism options part of the base game (like improved aerodynamics, which was always supposed to be a part of the final product), but other realism settings optional (like a more realistic-sized Kerbin, although I think the default should be at least 20%-scale and the current stock >9% scale should be an "easy" setting below that).

IMHO realistic aerodynamics should be mandatory - there is absolutely no reason to have nose cones work as a dead weight. The example you're looking for is probably deadly reentry? That's linked to the areodynamics and more closely related to the difficulty curve (though as far as I seen - majority of new players do expect deadly reentry and already account for that (Eg. descend only a lone capsule) - so it shouldn't be a problem when implemented).

Some unrealistic aspects of the game actually *prevent* the devs from expanding KSP in new and interesting directions- like the lack of a life support system inhibiting the development of things like greenhouses and more meaningful permanent base infrastructure (including ISRU systems to obtain Oxygen from the atmosphere and such...), or the lack of a more complex fuels system creating interesting and meaningful choices players can make on what type of rocket engine to use (certain engines could only burn certain fuels, like in RealFuels mod, for instance...) I think these realism factors should be part of the base game, because they create new and more interesting directions to expand the game in.

Very good points, something I haven't seen mentioned in a topic till now.

[cantab]

To a degree - it already happened. In a split between Sandbox and Campaign players (it's most obvious is a suggestions forum)

But even if we ignore that - split will happen in 0.25 and it's new "difficulty options". It's pretty much unavoidable.

They'll mainly apply to career strategy rather than rocket design though. A ship will work fine regardless, though it might not be economical in career.

Allowing physics settings that are too fine-grained will cause problems. For example "Aerodynamic Failure: On/Off" isn't going to be too much of an issue (though lots of on/off settings will rack up the combinations), but "Aerodynamic Stress Severity: 0%-100%" in 1% steps would be inadvisable IMO.

[Guest]

You could have multipliers, on the other hand. "Hard mode" would have them realistic, while "Medium" would make aerodynamic failure occur at, say, twice the force it does in "Hard". There are already such settings in plugins like Deadly Reentry or FAR.

[tater]

Regarding reentry. I absolutely expected to return just a capsule until I worked up to space-planes. Had I not read stuff in the forum about the lack of reentry being meaningful, I would still be doing so. All the part descriptions for kernels list tolerance to reentry, which tells new players it matters. I even piloted mine in stern-first.

Northstar makes a good point as well. As I said in another post, increased realism can present new problems to solve, and new problems to solve is new gameplay.

[Dodgey]

While I don't have a problem with settings for career mode I don't like the idea of modifiers to the physics engine, such as atmospheric stress on/off, reentry heat on/off. Mostly due to the splitting of the community, could you imagine how confusing the spacecraft exchange would be? I think that done right with the right amount of forgiveness a good reentry heating system and aerodynamic system wouldn't take too much getting used to.

For example, imaging through tweekables you could choose to apply a coat of "reentry paint" to all parts baring some obvious ones, scientific instruments, antennas, solar panels and the such. At the cost of slightly increased mass. Doing so would increase the max temperature of those parts allowing them to not only survive reentry but to potentially shield other parts. Allowing the construction of custom heat shields for individual ships if required. This also means we wouldn't have to limit our construction area to the cone of a heat shield if we want to aero break or the like.

As for aerodynamics I'm all for a revamped system. Perhaps just a little bit more forgiving than FAR but a revamped system non the less.

Edit:

tater, just because there are more problems does mean it is benifical for gameplay. Just because something is more complex does not mean it adds more depth to gameplay. It can be the case that a more complex mechanic increased the depth but it isn't a rule. Squad could add in a new complex electrical system where you had to be sure to have correct transformers between the power generators and the rest of the power system, as well as circuit breakers ect. But would it make for good gameplay? Beliveability is what is really important, simply adding in more complex features does not necessarily increase the depth or improve the gameplay.

Edited September 3, 2014 by Dodgey

[Renegrade]

Thank god it doesn't

9.82.

Granted that a lot of programmers abuse the preprocessor, but the old saw about throwing out babies with the bathwater comes to mind..

Not as such (thought there's nothing preventing you from running C# code through cpp), but I'm pretty sure what you really meant to ask is if it supports constants, and it does for primitive data (c.f. C#'s const keyword). IIRC, these are permitted to be inlined at compile time. It has about the same result as I think you are asking for, and you're absolutely right: magic numbers make a large codebase a veritable PITA.

Yeah, wouldn't have to be an actual preprocessor or anything, but in a physics simulation, you ARE going to want a few global constants like G, or g, or Pi, or e, etc.

Definitely don't want bare magic numbers floating around .. doubly so when they're wrong heh. (#define M_E 3?)

[juanml82]

Cold shouldn't be a factor. If some areas of the ship are supposed to be exposed to it, it can be assumed they are equipped with a Radioactive heat generator (sort of an rtg but for heat). Since that's a passive system, no player interaction is required.

Oxygen for breathing is recycled from co2 (generated by breathing). That's also passive so, like urine, it can perfectly well be abstracted

[regex]

KSP's drag model works great for rockets

It doesn't, it promotes a 10km climb before building up any lateral velocity. It's bad for everything.

Since we're going over little things, KSP should also have correct thrust. Isp does not affect fuel flow, it affects thrust.