Real World Mechanics Simulated By KSP ?

[Dichotomy]

Hello All

FULL DISCLOSURE : I have owned KSP since it first appeared as EA on steam, but have refrained from playing it (other then some brief testing about a year or so ago), until now. Its been a long wait, but finally I will get to play it in anger . The reason for mentioning this is that there is much I am naive about regarding the game (despite being a forum lurker for the past couple of years). So apologies for any such idiocy on my part

I was hoping some of you might be able to answer the question below for me (or point me to a readable link that can help out). I have read much of the wiki (trying to avoid gameplay spoilers, mechanics only) and a few guides here and there but have yet to find a list.

QUESTION

What real-world forces/phenomenons does KSP account for, and of these which are calculated, vs approximated ? (Are there any noticeable absences ?)

My reasons for asking this is to answer questions along the lines of, "what am I designing my ships to account for"?

I am probably going to get way too into this game, but its been a long wait

Thanks in advance to anyone who can spare some time to help me out.

EDIT : Just realised this should probably be in the gameplay/mechanics section. Would it be possible for a helpful mod to shift it for me please. Apologies for the inconvenience.

[Alshain]

Real World:

Orbital Mechanics - Spot on

Aerodynamics - Questionable, realistic enough but not 100%

Reentry Heat - Seems very light since recent changes, but it is there.

Newtonian Physics - No warp drive, no flying from one planet to another in a straight line. This is not Elite: Dangerous.

Parts: Representative of real space flight or theorized and feasibly realistic future tech (SABRE Engine), and sometimes tech that was researched but cancelled (NERVA Engine).

Not So Real World... With Good Reason:

Patched Conics - KSP does not entirely represent real gravitational forces (a.k.a. N-Body) because it allows multiple missions, sattelites, stations, etc to be in orbit at the same time. N-Body would require too much upkeep or your stuff would fall out of orbit. Instead it uses Patched Conics which is a close approximation based on the closest gravitational body, other gravitational bodies (the Mun, the Sun, etc) will not affect your craft orbiting Kerbin.

Star System - Approximately 1/6th the size of Earth and the Solar System.

Hydrodynamics - Prior to 1.0 Water might as well have been concrete. 1.0 softened it a bit, but don't plan on building reliable pontoon planes. Don't get me wrong, people have done it, but usually just to prove they can cobble it together to make it work. This is simply that way because it's not a boat game, so the devs aren't going to waste a lot of time on water.

On-Rails Simulation - This is sort of part of the patched conics but all planets and moons are on rails in a fixed orbit. Craft outside of a defined physics range (2 to 22km depending on the situation) will also be placed on rails due to computer processing abilities. Craft on rails that fall below ~20km in an atmosphere will be removed from the game and assumed lost.

Disclaimer: My comments are for stock. The game is very moddable and if something can be done... well lets just say you can fly the USS Enterprise in this game if you go looking.

Edited May 8, 2015 by Alshain

[Reddeyfish]

real world:

-thermodynamics, (more or less). Engine heat, for example

-1/rsquared solar panel efficiency

not exactly real world, but probably more fun because of it:

-kerbal body physics. (They can survive a 5km drop into water on kerbin)

-rocket structural stability. (KSP rockets are much, much more bendable and flexible than real-world rockets)

-life support (there is none; kerbals can live forever in just a suit)

-probe control (there is no latency [that's a really good thing for gameplay])

[LostOblivion]

Because of the patched conics approximation, there are certain things that are quite useful IRL, but not possible in-game, most noticably the lack of any Lagrange-points, which IRL is used for deep-space observatory satellites.

Question: Although Lagrange points cannot naturally occur on patched conics, is it possible for Squad to implement them by putting a different kind of SOI at these points? And at the naturally unstable ones, kind of reversed SOIs?

[Dichotomy]

Real World:

Orbital Mechanics - Spot on

Aerodynamics - Questionable, realistic enough but not 100%

Reentry Heat - Seems very light since recent changes, but it is there.

Newtonian Physics - No warp drive, no flying from one planet to another in a straight line. This is not Elite: Dangerous.

Parts: Representative of real space flight or theorized and feasibly realistic future tech (SABRE Engine), and sometimes tech that was researched but cancelled (NERVA Engine).

Not So Real World... With Good Reason:

Patched Conics - KSP does not entirely represent real gravitational forces (a.k.a. N-Body) because it allows multiple missions, sattelites, stations, etc to be in orbit at the same time. N-Body would require too much upkeep or your stuff would fall out of orbit. Instead it uses Patched Conics which is a close approximation based on the closest gravitational body, other gravitational bodies (the Mun, the Sun, etc) will not affect your craft orbiting Kerbin.

Star System - Approximately 1/6th the size of Earth and the Solar System.

Hydrodynamics - Prior to 1.0 Water might as well have been concrete. 1.0 softened it a bit, but don't plan on building reliable pontoon planes. Don't get me wrong, people have done it, but usually just to prove they can cobble it together to make it work. This is simply that way because it's not a boat game, so the devs aren't going to waste a lot of time on water.

On-Rails Simulation - This is sort of part of the patched conics but all planets and moons are on rails in a fixed orbit. Craft outside of a defined physics range (2 to 22km depending on the situation) will also be placed on rails due to computer processing abilities. Craft on rails that fall below ~20km in an atmosphere will be removed from the game and assumed lost.

Disclaimer: My comments are for stock. The game is very moddable and if something can be done... well lets just say you can fly the USS Enterprise in this game if you go looking.

Thankyou very much for your comprehensive answer. Exactly what I was looking for.

Sounds good.

The water issues surprised me a little, I thought KSP would modal these a little better, but as you say its not a huge game focus so .......

I do hope the Reentry Heat issues are patched. I don't quite know why but accounting for reentry heat appeals to me. If I may : Does the game account for other forms of heat ie direct solar radiation, superheated atmospheres etc.

The under 20km caveat concerns me a little, but I did know that one existed, I assume its so that discarded stages and space debris don't end up hogging cpu cycles and memory over time. I assume care must be taken when dealing with satellites etc.

Thanks again for your help.

BTW I have been looking at the various mods for KSP and have only settled on one one so far : SCANSAT. Am very happy it has been recently updated to the current version, and that it integrates with the stock scanning parts. Can't really explain it, but the scanning aspect is probably the bit I am looking forward to the most about KSP. I am currently perusing a few of the more realism related MODS, ie FAR, deadly reentry and life-support. Not really sure how far I want to go, or how balanced/complete any of these MODS are yet.

- - - Updated - - -

real world:

-thermodynamics, (more or less). Engine heat, for example

-1/rsquared solar panel efficiency

not exactly real world, but probably more fun because of it:

-kerbal body physics. (They can survive a 5km drop into water on kerbin)

-rocket structural stability. (KSP rockets are much, much more bendable and flexible than real-world rockets)

-life support (there is none; kerbals can live forever in just a suit)

-probe control (there is no latency [that's a really good thing for gameplay])

Excellent ! Am quite stoked about the engine heat modelling. Inverse square solar panel efficiency is nice too, had not really thought about it one way or the other.

1. Ha - 5KM drop into the water Given the nature of the kerbals I can live with this

2. I was kinda assuming the rockets would be more "elastic" then RL, from some of the second hand stories on these forums I am assuming they are still "breakable" enough

3. This one I knew about but do wish that had to be taken into account, even if it was limited and they went into some kind of hibernation when stranded (to open up the possibility of rescues).

4. Hmmm,yes I see your point.Torn on this one, but do agree it is probably better for gameplay without the latency. I guess it is not really different to the lack of latency with manned missions. In those instances I will "rationalize" it by assuming I am controlling the pilot, rather then HQ, and with the probe core I will assume I am playing the probes "ai".

Thanks for the info.

- - - Updated - - -

Because of the patched conics approximation, there are certain things that are quite useful IRL, but not possible in-game, most noticably the lack of any Lagrange-points, which IRL is used for deep-space observatory satellites.

Question: Although Lagrange points cannot naturally occur on patched conics, is it possible for Squad to implement them by putting a different kind of SOI at these points? And at the naturally unstable ones, kind of reversed SOIs?

Yes I was wondering about Lagrange points, shame about those, I was really hoping they existed. I was naively assuming SQUAD had implemented something like what you suggested. I think LP would add many unique gameplay opportunities. I will have to have a hunt around for a MOD, or just hope SQUAD adds the functionality at a later date.

Thanks for your help

[Lukaszenko]

The reaction wheels also don't saturate, although this might be for the best since many real-world methods of de-saturating them probably cannot be used, not to mention that rotational momentum can be cancelled by simply time-warping. They're also overpowered, but then so are other things for the sake of gameplay (such as the EVA suits).

There's other small things that I think can be exploited if you're really bored, such as being able to "walk" a spaceship across space simply by shifting its center of mass around.

[Endersmens]

Also, another big unrealistic thing is command pod torque. (and reaction wheel torque, for that matter)

Place a lander can and launch it with batteries. You can flip it around and roll it for a long time. Doesn't work that way.

[Sky_walker]

Real World:

Orbital Mechanics - Spot on

Aerodynamics - Questionable, realistic enough but not 100%

Reentry Heat - Seems very light since recent changes, but it is there.

Newtonian Physics - No warp drive, no flying from one planet to another in a straight line. This is not Elite: Dangerous.

Parts: Representative of real space flight or theorized and feasibly realistic future tech (SABRE Engine), and sometimes tech that was researched but cancelled (NERVA Engine).

@1 - not really, without n-body physics you can't really get it to it's full extend

@2 - yea... doesn't count either. FAR was probably the closest one to making it sensible.

@3 - Well, it is there, but doesn't really do much... not to mention that many parts of it don't make much sense (eg. for a good while you can keep your landing gear in a hot plasma without any harm or damage o_O)

@4 - Parts are the worst part. There's just so many components that have a totally unrealistic representation that I don't even know what to say. Even NERVA engine you mentioned is screwed, not as badly as pre-1.0, but it still doesn't have almost any flaws it has in real life.

[Esme]

Yes I was wondering about Lagrange points, shame about those, I was really hoping they existed. I was naively assuming SQUAD had implemented something like what you suggested. I think LP would add many unique gameplay opportunities. I will have to have a hunt around for a MOD, or just hope SQUAD adds the functionality at a later date.

Thanks for your help

The subject of the absence of LaGrange points has been talked to death elsewhere - the short answer is that no, they can't sensibly be simulated by adding invisible SOI's with no body at their centre. Which isn't to say that such a thing couldn't potentially be added by a mod if it floats someones boat sufficiently. I wouldn't want anything other than the real thing or a much better approximation of one than faking it with extra SOI's would give. ISTR that someone was seriously looking into whether n-body physics could be introduced to replace the patched conics system. I don;t know how much progress they made, but they certainly seemed to know their stuff on the mathematics.

[Padishar]

In relation to Kerbals surviving a 5km fall into water, they could also survive direct re-entry from orbit in their suit and land on solid ground in earlier versions (more likely to survive if you land them on their heads). I'm not sure if anyone has tried this in 1.0(.2) yet with the heating effects and higher terminal velocity due to decreased drag...

As for n-body physics, the mod you are thinking of is called Principia. It is being developed by a very talented and knowledgeable coder and mathematician and is showing great promise though is quite some way from being generally usable.

Edited May 8, 2015 by Padishar

[Dichotomy]

The reaction wheels also don't saturate, although this might be for the best since many real-world methods of de-saturating them probably cannot be used, not to mention that rotational momentum can be cancelled by simply time-warping. They're also overpowered, but then so are other things for the sake of gameplay (such as the EVA suits).

There's other small things that I think can be exploited if you're really bored, such as being able to "walk" a spaceship across space simply by shifting its center of mass around.

That made me laugh. I wont be using timewarp for anything remotely circumventory (if thats not a word I am coining it!).

I found this thread : http://forum.kerbalspaceprogram.com/threads/86334-KSP-vs-IRL-reaction-wheels and will be reading it after I finish this post ! Thanks for the heads up.

Hey just realised you were the OP!!!! Small world on KSP .....

I am more worried about the opt-out exploits, rather then the opt-in (or should that be the other way around, IDK).

As I intend to play a custom career mode, (no do-overs, deaths on, all the other realism settings on etc, but not sure about where the sliders should be, probably either normal or moderate for the sliders. I don't mind if I have to run out of money and start over.), I want to make sure that I consider the implications of my design and the limitations of the world I am designing for. So I intend to go in with as much mechanic info (not gameplay stuff) as I can. Plus I have a week or so to kill until I can get my hands dirty, so this whets the appetite.

Thanks again

- - - Updated - - -

Also, another big unrealistic thing is command pod torque. (and reaction wheel torque, for that matter)

Place a lander can and launch it with batteries. You can flip it around and roll it for a long time. Doesn't work that way.

Perhaps they have been "rifled"

Hmm I will have to consider this. Assuming I am understanding you, is this a consequence of the atmospheric model they are using ?

Thanks for the reply

[Endersmens]

Perhaps they have been "rifled"

Hmm I will have to consider this. Assuming I am understanding you, is this a consequence of the atmospheric model they are using ?

Thanks for the reply

no no, I mean they have "reaction wheels" inside the command pods, giving them torque, or the ability to turn themselves. Except in KSP they have absolutely absurd amounts of torque, able to rotate themselves in any direction. (you can have a pod on its side and turn it upright.)

[Dichotomy]

@1 - not really, without n-body physics you can't really get it to it's full extend

@2 - yea... doesn't count either. FAR was probably the closest one to making it sensible.

@3 - Well, it is there, but doesn't really do much... not to mention that many parts of it don't make much sense (eg. for a good while you can keep your landing gear in a hot plasma without any harm or damage o_O)

@4 - Parts are the worst part. There's just so many components that have a totally unrealistic representation that I don't even know what to say. Even NERVA engine you mentioned is screwed, not as badly as pre-1.0, but it still doesn't have almost any flaws it has in real life.

Thankyou for your response.

Can I assume you would be suggesting FAR and Deadly Re-Entry then ? I have not used either and I am in two minds about their inclusion.

I am a bit of a realism nut, but my background knowledge is not sufficient to be able to determine what spacecraft parts are not performing in a realistic manner, vs those that are (unless it is egregiously so). So #2 and #3, concern me more then #4 . #1 I had always assumed would be a cut down model, so not really disappointed, when I didn't really consider it a possibility in the first place.

Thankyou again for your insights.

- - - Updated - - -

no no, I mean they have "reaction wheels" inside the command pods, giving them torque, or the ability to turn themselves. Except in KSP they have absolutely absurd amounts of torque, able to rotate themselves in any direction. (you can have a pod on its side and turn it upright.)

Oh OK, (faceplam for me). And here I was thinking I was being funny

I see what you mean now. That does sound more than a little extreme.

Thanks for edifying me.

[Endersmens]

Oh OK, (faceplam for me). And here I was thinking I was being funny

I see what you mean now. That does sound more than a little extreme.

Thanks for edifying me.

its alright, I can understand it being confusing, as it makes no sense in the first place.

The torque is there for when you don't have controllable fins. So it makes sense that they upped the torque, so that it would have an effect. But take away the rest of the ship and it starts to seem a bit much.

[Laie]

Yes I was wondering about Lagrange points, shame about those, I was really hoping they existed. [...] I will have to have a hunt around for a MOD, or just hope SQUAD adds the functionality at a later date.

No way. "Pateched Conics" only mean that at any given time, you will be under the influence of exactly one body. When you go to the Mun, you'll be in Kerbin's gravity until you reach a border, and then *whack* you're in the Mun's gravity and Kerbin doesn't matter anymore. For most space missions, the side effects are negligible. However, Lagrange points are places where the gravities of two bodys even out. They just cannot exist in KSP. This is one of the few cases where "there is a mod for that" does not apply.

Does the game account for other forms of heat ie direct solar radiation, superheated atmospheres etc.

The whole heat thing is underwhelming at this point. The underlying model seems to be mostly fine, but the scheme of how much heat you pick up from where will probably need to go through a few more iterations.

A big problem is that the spaceplane pilots want to ride up at similar speeds as capsules coming down. How can you have heat for one, but not the other?

The under 20km caveat concerns me a little, but I did know that one existed, I assume its so that discarded stages and space debris don't end up hogging cpu cycles and memory over time. I assume care must be taken when dealing with satellites etc.

It's a non-issue most of the time. Keep in mind that physics work nicely for the vessel you're watching -- only unattended stuff is subjected to the simplified physics. Most everything will be in some stable condition before you leave it. The issue is most noticable with spent stages that are either silently destroyed (iow, unrecoverable) or remain in a steady orbit that should by rights be unstable. The latter can be dealt with if you take the time and watch them as they enter the atmosphere.

[Endersmens]

No way. "Pateched Conics" only mean that at any given time, you will be under the influence of exactly one body. When you go to the Mun, you'll be in Kerbin's gravity until you reach a border, and then *whack* you're in the Mun's gravity and Kerbin doesn't matter anymore. For most space missions, the side effects are negligible. However, Lagrange points are places where the gravities of two bodys even out. They just cannot exist in KSP. This is one of the few cases where "there is a mod for that" does not apply.

Except...... http://forum.kerbalspaceprogram.com/threads/68502-WIP-Principia-N-Body-Gravitation-and-Better-Integrators-for-Kerbal-Space-Program

[ColKlonk]

Aerodynamics - Far From it... Unity or KSP just doesn't have the power to get near it -- buttt it's ok for this game

[Hodo]

QUESTION

What real-world forces/phenomenons does KSP account for, and of these which are calculated, vs approximated ? (Are there any noticeable absences ?)

My reasons for asking this is to answer questions along the lines of, "what am I designing my ships to account for"?

I am probably going to get way too into this game, but its been a long wait

Thanks in advance to anyone who can spare some time to help me out.

EDIT : Just realised this should probably be in the gameplay/mechanics section. Would it be possible for a helpful mod to shift it for me please. Apologies for the inconvenience.

Howdy back.

And to answer your question. In short KSP stock doesnt do a whole lot of the real world stuff. But it does do a lot of real world stuff in a way that works for the scale of the environment.

Mind you Kerbin is roughly 1/3rd the size of the real Moon. Which makes it quite small but it still has the gravitation pull of Earth. There are mods that change the scale of everything and make KSP into a pretty good simulation of the Real Earth Solar System including Earths atmosphere. But a great many here dont care for it because of the shear number of additional mods needed to get it to work with in 90% accuracy of the real world.

But if realism is what you are looking for then check on the Realism Overhaul thread. If you just want something close but dont want deal with having to build a rocket with over 9.8km/s of Delta-Velocity to get to orbit then you may want to avoid it.

I personally have both versions of KSP installed on my computer a RO install and a regular install.

[EtherDragon]

Greetings! KSP accounts for most real-world forces/phenomenons, but in a scaled down (1/10th) and simplified way.

Here's a Video (first part of a Three-Part series I'm working on) that gives you a hint about the real world forces in play:

[Dichotomy]

No way. "Pateched Conics" only mean that at any given time, you will be under the influence of exactly one body. When you go to the Mun, you'll be in Kerbin's gravity until you reach a border, and then *whack* you're in the Mun's gravity and Kerbin doesn't matter anymore. For most space missions, the side effects are negligible. However, Lagrange points are places where the gravities of two bodys even out. They just cannot exist in KSP. This is one of the few cases where "there is a mod for that" does not apply.

The whole heat thing is underwhelming at this point. The underlying model seems to be mostly fine, but the scheme of how much heat you pick up from where will probably need to go through a few more iterations.

A big problem is that the spaceplane pilots want to ride up at similar speeds as capsules coming down. How can you have heat for one, but not the other?

It's a non-issue most of the time. Keep in mind that physics work nicely for the vessel you're watching -- only unattended stuff is subjected to the simplified physics. Most everything will be in some stable condition before you leave it. The issue is most noticable with spent stages that are either silently destroyed (iow, unrecoverable) or remain in a steady orbit that should by rights be unstable. The latter can be dealt with if you take the time and watch them as they enter the atmosphere.

In regards to Lagrange points. I would have thought introducing a new SOI with No/Low mass at its centre, would solve the problem. Given you can only be under the effect of a single SOI in KSP. Perhaps this would not work and I am over simplifying the system. I don't know enough about how the SOI integrate with other KSP systems/mechanics to know if a quick and dirty solution like this would work.

A little disappointed by what you said about heat. I don't know why but accounting for heat (in its various forms is a really cool aspect of the game for me , or at least I think it will be). AM seriously looking at Deadly -Reentry ATM, and will look around for other heat MODS.

Your comments on the 20km limit, were heartening. I was a bit concerned by that one. Still makes me nervous about satellites, especially since I may just decouple/launch them at barely newtonian speeds

Thanks for all your responses.

- - - Updated - - -

Except...... http://forum.kerbalspaceprogram.com/threads/68502-WIP-Principia-N-Body-Gravitation-and-Better-Integrators-for-Kerbal-Space-Program

Thanks for the link (even if it wan't intended for me!). Will keep an eye on this.

- - - Updated - - -

Howdy back.

And to answer your question. In short KSP stock doesnt do a whole lot of the real world stuff. But it does do a lot of real world stuff in a way that works for the scale of the environment.

Mind you Kerbin is roughly 1/3rd the size of the real Moon. Which makes it quite small but it still has the gravitation pull of Earth. There are mods that change the scale of everything and make KSP into a pretty good simulation of the Real Earth Solar System including Earths atmosphere. But a great many here dont care for it because of the shear number of additional mods needed to get it to work with in 90% accuracy of the real world.

But if realism is what you are looking for then check on the Realism Overhaul thread. If you just want something close but dont want deal with having to build a rocket with over 9.8km/s of Delta-Velocity to get to orbit then you may want to avoid it.

I personally have both versions of KSP installed on my computer a RO install and a regular install.

Thankyou for your reply and greeting !!!!!!!!!!!!!

Yes I do want realism, but it is more realism of mechanics, rather then just plain realism. That last sentence is confusing even me, i will try to do better ...

For example the re-scaling of the Kerbols diameter, while keeping an earth like Newtonian gravitational force is fine by me.

Its more : given a series of assumed constants/forces/etc does the world behave in a reliable and logical way. Thats more the realism I am after.

I have actually been looking at that exact thread (realism overhaul) over the past few hours. I think I am going to keep my MODS to a minimum to begin with, and really only looking at modding the parts of the main game that are the least realistic/immersive (either due to a poor model, or not being accounted for). Perhaps I also might include a few utility MODS like Kerbol Alarm Clock.

You guys/gals really are a great community. Really expected for this thread to degenerate into a "why are you trying to mod the game when you have so little experience etc etc etc". I really like to savour my first playthrough of a unique game like this, and I have a week or so to kill till I can play so learning about the mechanics and choosing MODS is on the agenda.

On a totally unrelated note : SCANSAT is awesome !!!!!!!!!!!!!!!!

- - - Updated - - -

Greetings! KSP accounts for most real-world forces/phenomenons, but in a scaled down (1/10th) and simplified way.

Here's a Video (first part of a Three-Part series I'm working on) that gives you a hint about the real world forces in play:

Thankyou for this I will give it a look. I have been fairly careful about looking a gameplays videos so far as I didn't want to spoil to much of the "cool moments" ie eclipses, seeing Jool (is that what its called) for the first time etc. This looks pretty safe though

Thankyou for the link and (from newbs like me) for taking the time to create it.

EDIT : Have just watched the link. Very professionally done. Unfortunately for me, none of this was new, but both kudos for the vid and thanks for the heads-up. I am sure it will help many new players.

Good ol' Kepler. I remember doing a Year 5 project on him and Copernicus, way back when......

Edited May 8, 2015 by Dichotomy

[Eric S]

In regards to Lagrange points. I would have thought introducing a new SOI with No/Low mass at its centre, would solve the problem. Given you can only be under the effect of a single SOI in KSP. Perhaps this would not work and I am over simplifying the system. I don't know enough about how the SOI integrate with other KSP systems/mechanics to know if a quick and dirty solution like this would work.

It comes down to that being a poor substitute, for the most part. In particular, if someone accidentally enters the SoI with a periapsis close to the center, rounding errors will tend to slingshot the craft out of the SoI at Kerbin's escape velocity or higher.

You can already fake L4/L5 in patched conics, which are the two you can use without station keeping, and the points that need stationkeeping (stable on two axis, unstable on the third) don't behave close enough to a mini-SoI for that to be an approximation.

Basically, the people that actually care about Lagrange points as something other than "an easy place to park satellites and space stations" wouldn't find anything short of full N-body physics as close enough to be acceptable. There is someone working on an N-body mod for KSP, though last I saw it was still in early alpha.

[Renegrade]

Re: Heat

I'd like to point out that the heat in KSP isn't weak, it's vastly exaggerated for re-entry in terms of the actual speeds encountered. Re-entries from LKO are occurring in the Mach 6-7 range, which while toasty in real life, doesn't involve any REAL heat shielding (like ablatives stuck on the front of a blunt-shaped pod, etc.). Real life re-entries are in the Mach 25+ range, and do involve real, hardcore shielding..

Squad's buffed up the heating somewhat to try to approximate the feel and affect of real life heating in spite of the lower orbital velocities.

[Whirligig Girl]

Star System - Approximately 1/6th the size of Earth and the Solar System.

*Ahem* 1/10 scale, not 1/6 scale.

[sal_vager]

*Ahem* 1/10 scale, not 1/6 scale.

Can confirm, Kerbin is 600,000m while Earth is a little bit bigger, I used to think it was 1/6th as well.

[Laie]

I am more worried about the opt-out exploits, rather then the opt-in

As I intend to play a custom career mode, (no do-overs, deaths on, all the other realism settings on etc

Don't be too eager! Depending on where you're coming from, just getting to space will be a major accomplishment even with all the slight "cheats" like deep throttling and infinite torque from reaction wheels. I heartily recommend either science or career mode for beginners, as the full list of parts available can be quite overwhelming. That doesn't mean that you have to play either mode through to the end, though: but they're a good mean to get a gradual introduction to the game.

But if you're the hardcore realism type, you may consider to quickly move on to Realism Overhaul where you have to play in a real-sized solar system (that is, Earth, mars and so on), using equipment modelled as closely as possible after real-life counterparts. Reaction wheels are nerfed badly, radio transmitters have limited range, almost no engine has a throttle (and most can't be re-ignited either).

One great benefit of stock KSP is that you can try, fail, and try again within two or five minutes, while an ascent in RO takes more like fifteen to twenty.