a way to prevent "skipping" atmosphere during high-speed re-entry

[thereaverofdarkness]

Problem: when a craft makes a very high speed entry into a planet, the craft may reach the ground or get very close to it before the atmosphere begins to decelerate the craft significantly, resulting in impacts at much higher velocities than what is realistic. This occurs because (as I have heard it) the game calculates the craft's current atmospheric drag once per second and maintains that for the entire second. So, for instance, a craft traveling at 75km/s straight down toward Kerbin could have a drag calculation at 70km, then reach the ground before the next calculation, thus experiencing no drag at all before hitting the ground.

Solution: re-calculate drag both once per second and once per scale height unit. This can make rapid entries slow the game more significantly than they already do, but will result in the craft experiencing several stages of increased drag before reaching the surface, no matter how fast it is going. Craft moving much slower won't hardly notice the difference, and craft moving very fast will slow down quickly.

[Alias72]

You do realize that nothing would survive a 75km/s impact with the atmosphere right? In fact let me do the math, a 2 ton spacecraft at 75 m/2 has how much kinetic energy?

ans = 1.3 kilotons TNT. well not quite Hiroshima but not something you want to be near either.

How about not blasting into the atmosphere at mach 220.

In case it isn't obvious I oppose this suggestion for its ludicrous.

[thereaverofdarkness]

You do realize that nothing would survive a 75km/s impact with the atmosphere right?

And it wouldn't survive if one had Deadly Re-entry installed. I'm talking about stock. Having your craft destroyed on entry because you came in too fast is one thing, but having your craft reach the ground at dramatically different speeds entirely dependent on what altitude the physics decides to calculate your drag is very different.

[chaos_forge]

This seems like a rather niche problem . . . honestly if you're slamming into the atmosphere at 75km/s there's nothing that's gonna save you. Also, how in God's name did you get to such a ridiculous speed anyways?

[Greenfire32]

This seems like a rather niche problem . . . honestly if you're slamming into the atmosphere at 75km/s there's nothing that's gonna save you.

Agreed.

Also, how in God's name did you get to such a ridiculous speed anyways?

Moar boosters?

[cwhiz]

LOL dude you need to try slowing down a bit then. Just turn around and decelerate before reaching the atmo. I imagine you are one of a very few number of people playing KSP who has this problem.

[thereaverofdarkness]

I was only using 75km/s and Kerbin as an example. The same glitch is easily achievable coming down to Duna at a much lower velocity. Within 4000 dV from LKO, one could enter Duna orbit using a retrograde Kerbolar slingshot and come in at some 15km/s or so. You won't get much drag until around 15km altitude, so if you're coming straight down you can easily hit the surface with drag orders of magnitude below what you should have experienced on the way down.

And even in a thicker atmosphere in which you aren't going to reach the surface before slowing, you still can experience SUD from large instant changes in drag. A good example of this would be entering Eve from a prograde Hohmann Transfer, which will send you in at some 8km/s. If you come straight down, you can experience atmospheric thickening so suddenly that it can shoot your g-force meter from <1 to >15 in a single frame.

Edited August 16, 2014 by thereaverofdarkness

[BudgetHedgehog]

<snipped Duna things>

Well, it does have an atmosphere of like, what.. 0.2 of Kerbin? That's really not that much. It's a thin atmosphere, it's never going to slow you down as much as Kerbins or Eves. Honestly, the examples you're providing are easily avoidable - are you heading directly towards the centre of the planet at a ridiculous speed? Then burn slight radially when far away for not a lot of fuel and to push your PE out to a sensible height.

You're essentially saying 'I want to slam full on into the atmosphere at Mach 220. Squad, make it so I don't crash'. How about you just.. don't do that and re-enter normally using correct aerobraking and orbital manoeuvres?

I don't understand your example about Eve - it's in a slightly lower orbit than Kerbin so a correct Hohmann Transfer would leave you with very little relative velocity to Kerbin. Even coming from Moho would leave you with even less - you're meeting Kerbin at your solar Ap, when you're travelling slowest. If you still have 8km/s relative speed, you've done something wrong. If you decide to slam head on into the atmosphere at that speed, then you frankly deserve everything that comes to you. That is an insane entry speed and slowing down to like 300m/s at 10,000m just seems so weird. Realism vs believability again - I don't care if it's not realistic, it sure as heck ain't believable which means that, in my eyes, it's not something that needs 'fixing' (even though it's not broke, but hey ho).

Edited August 16, 2014 by ObsessedWithKSP

[Murph]

If you're going for a direct landing on Duna straight from solar orbit, you're not really going for a precision landing site, so I don't see that there's a significant difference of just coming in with say a 15km periapsis to capture you in Duna SoI, then landing on the next orbit.

This is one case where I'm not at all convinced that there's anything worthwhile fixing about KSP aerodynamics. It seems entirely realistic to me that if you hit a thin to moderate atmosphere at hypersonic speed, that you should be generating a suitably spectacular impact with the ground (or just burning up), unless you take some exceptional steps to avoid disaster (such as a load of drogue+normal chutes and/or a big retro braking burn).

[MKI]

Not something to be fixed. Once Re-entry affects are added this problem will no longer exist. You will simply die due to excessive heating as you should in any realistic scenario.

Fixing it to be one way makes no sense in the long run, as re-entry affects will handle the exact circumstance.

[thereaverofdarkness]

Not something to be fixed. Once Re-entry affects are added this problem will no longer exist. You will simply die due to excessive heating as you should in any realistic scenario.

Fixing it to be one way makes no sense in the long run, as re-entry affects will handle the exact circumstance.

WRONG.

All of you are ignoring the problem and focusing on the results. Yes, any craft that catches this glitch dies. My point isn't that is dies, but how it dies. Skipping the atmosphere exactly prevents re-entry damage, and is a glitch that can cause a craft to NOT burn up on re-entry when the atmospheric drag is supposed to burn the ship up.

You guys keep speaking of this as if this is some issue I am having difficulty with. I, just like anyone else here, am fully capable of flying rockets at reasonable speeds if I want to.

WHAT I AM DOING:

* I'm discussing a bug that happens at high speeds and causes unrealistic effects.

* I am suggesting a way to fix these unrealistic effects so that what happens is much more realistic.

WHAT I AM NOT DOING:

* I am not suggesting a way to keep craft alive at ridiculous impact speeds.

* I am not pining for tougher rockets.

* I am not complaining.

* I am not recounting my own experiences.

So far, not one response has been on topic. Please try to keep all further posts on topic.

[cwhiz]

WRONG.

All of you are ignoring the problem and focusing on the results. Yes, any craft that catches this glitch dies. My point isn't that is dies, but how it dies. Skipping the atmosphere exactly prevents re-entry damage, and is a glitch that can cause a craft to NOT burn up on re-entry when the atmospheric drag is supposed to burn the ship up.

You guys keep speaking of this as if this is some issue I am having difficulty with. I, just like anyone else here, am fully capable of flying rockets at reasonable speeds if I want to.

WHAT I AM DOING:

* I'm discussing a bug that happens at high speeds and causes unrealistic effects.

* I am suggesting a way to fix these unrealistic effects so that what happens is much more realistic.

WHAT I AM NOT DOING:

* I am not suggesting a way to keep craft alive at ridiculous impact speeds.

* I am not pining for tougher rockets.

* I am not complaining.

* I am not recounting my own experiences.

So far, not one response has been on topic. Please try to keep all further posts on topic.

The thing is, as someone else mentioned, most people besides yourself and maybe a few others won't really care about such a bug since the end result is that you die whether it be to due to reentry or impact with the surface. I think the fact that you die at such speeds is realistic enough.

In the end, this is an argument for more realism where most don't deem such realism as necessary. Therefore, you probably won't get many responses that attempt to address this bug since most won't care.

[Frozen_Heart]

How on earth did you reach those speeds? Even with a K-Drive I'd struggle with that!

[Murph]

I agree that it's really quite pointless to implement the suggested change right now, at least for the stock game. At a high level, things are working just fine: craft trajectory intersects celestial at high velocity, expected result is craft converted into smouldering wreckage, actual result matches expected result (more or less - it's either an explosion on the surface or auto-deleted). Adding complexity for these circumstances really isn't worth it, in my opinion.

As and when Squad decide to improve stock aerodynamics, that's the time when such things should be looked at. Until then, it's only something for mod authors to worry about.

[Red Iron Crown]

I've had an infinite-fueled ship skip through the atmosphere so quickly that it had no effect at all, which is the same thing the OP is talking about, I think.

The most accurate solution would be to compute the drag more frequently, but that would likely adversely affect performance. Not sure if it would be worth it for what is a bit of a fringe problem, i.e. you have to get your ship up to ludicrous speeds to notice this effect. I don't think I've ever experienced it with a non-cheaty ship.

[thereaverofdarkness]

As and when Squad decide to improve stock aerodynamics, that's the time when such things should be looked at. Until then, it's only something for mod authors to worry about.

Best to start discussing it before they make such an update, to make sure they're aware of the bug and address it during such an update. If they finish fixing the air and this bug is still around, it might stick around forever.

[thereaverofdarkness]

Also, how in God's name did you get to such a ridiculous speed anyways?

I imagine you are one of a very few number of people playing KSP who has this problem.

How on earth did you reach those speeds? Even with a K-Drive I'd struggle with that!

i.e. you have to get your ship up to ludicrous speeds to notice this effect. I don't think I've ever experienced it with a non-cheaty ship.

Several KSPers have achieved these results on my Maximum Impact Velocity challenge. What really made me write this post was that I kept seeing this bug and it has become so bad that it is really skewing the results of the challenge. At first it was just Duna, but someone has already skipped the atmosphere of Jool. Of course, Jool's atmosphere was skipped using infinite fuel, but I've had two challengers skip Duna's atmosphere almost completely (both landing in a basin) completely stock, with no cheats.

[Wanderfound]

Several KSPers have achieved these results on my Maximum Impact Velocity challenge. What really made me write this post was that I kept seeing this bug and it has become so bad that it is really skewing the results of the challenge. At first it was just Duna, but someone has already skipped the atmosphere of Jool. Of course, Jool's atmosphere was skipped using infinite fuel, but I've had two challengers skip Duna's atmosphere almost completely (both landing in a basin) completely stock, with no cheats.

It's a trivial "bug" that only comes up if you're actively attempting to break the physics engine. And most of the time the outcome is going to be appropriate anyway (i.e. instant annihilation). It will never occur in a game that is played sensibly.

I don't think that you're gonna get much support on this.

[thereaverofdarkness]

It's a trivial "bug" that only comes up if you're actively attempting to break the physics engine. And most of the time the outcome is going to be appropriate anyway (i.e. instant annihilation).

Mostly, but not always. Ships that skip the atmosphere do at times also skip the ground and get shot out of the planet completely intact. One who figured out how and why it works could likely set it up intentionally.

[MarvinKitFox]

How on earth did you reach those speeds? Even with a K-Drive I'd struggle with that!

Actually, that was me achieving those speeds, skimming past Kerbol at under 1500m (not km) altitude, and burning some 27500m of delta-v while under maximum Oberth possible.

This resulted in my ship hitting Duna at 71.4km per second, impacting the ground at 1145m altitude at a speed of about 48201m/s (that's 25km/s slowdown in 0.12 seconds)

theravenofdarkness considers the average of 21358g of deceleration (thats kerbin g, not duna) to be unrealistically low.

Because of this, he has branded me a cheater and disqualified my entry into his "maximum impact" challenge.

Problem is, he is full of xxxx. (that which one uses toilet paper to wipe, but may not mention on a forum)

This whole "skipping drag" problem of his is a simple case of denial, because he did not think anyone could achieve that sort of speed.

Here is the natural spped video of my impact. NO timewarp. NO video acceleration. This is live.

(And as you can see for the yellow on the clock, it is actually slower than live)

To throw a bit of mathematics at it.

My 354 kg impactor dissipated 54.5% of its kinetic energy in well under a half a second, before hitting the ground.

Airburst energy is 9.82571E+11 joules

Ground impact is 8.22461E+11 joules

This is 980 thousand sticks of dynamite in the airburst,

and 822000 sticks of dynamite in the ground impact.

And the guy still thinks that air thinner than that 1 km *above* the highest peak of Kerbin, should have stopped my ship dead in its tracks?

P.S.

My ship is pure stock .24.0, plus mechjeb.

It is a rather simplistic design of 28 stages, ion on top of xenon droptanks (7 layers), on top of more ion + plenty xenon(2 layers), on top of nukes, on top of skippers, on top of SLS .

I flew *out* to past jool, then lowered my perigee to 500meters above kerbol (later rectified to 1200m. did you know that you explode at 1162m above kerbol?)

at perigee, with a natural speed of something over 92km/s, uncle Oberth really did a number on my 27500m/s of remaining deltav.

reached Duna at 71.3km/s, with about 1.7km/s still left in the tank.

Edited August 18, 2014 by MarvinKitFox

[Vaporo]

i.e. you have to get your ship up to ludicrous speeds to notice this effect.

Well, I am quite familiar with Ludicrous Speed, so I think I have to speak up for this one (Sorry. I just had to do that). Seriously, even at 75 km/s, you wouldn't entirely "skip" the atmosphere, unless you happen to be running KSP at one frame per second (at which point I would congratulate you for getting a ship that lags your computer down to 1 FPS up to 75 km/s). Even if you are running KSP at 30 FPS, 60 being the maximum on most screens, 75 km/s would mean that you would have at least 30 frames of deceleration if you entered Kerbin's atmosphere vertically. Less, of course, on Duna, but still, my point stands. To fully "skip" the atmosphere, you would have to go 30 times 75 km/s, or 2250 km/s, at 30 FPS. At either speed, though, you would be obliterated by reentry heat, g-forces, and just plain old hitting the ground. In any case, Kerbin's atmosphere does way too much slowing as it is. Here's some experiments:

http://imgur.com/a/SN5gr

With a command pod with a greater drag coefficient.

http://imgur.com/a/K0GLn#0

Didn't press F1 fast enough to get screenshots of actual reentry on either experiment. Still, I spent a significant amount of time in the actual atmosphere, just as MarvinKitFox did.

In both experiments, I was slowed far more than could possibly happen in real life. So, the atmospheric physics did their job a bit too well as far as I'm concerned.

To sum it up, to fully skip the atmosphere you would have to go so insanely fast that you are likely using K-drives or otherwise actively abusing the physics engine, at which point realistic aerodynamic physics would barely slow you anyways.

EDIT: Did some more experiments with FAR installed. trying to reenter at these speeds caused KSP to glitch out in some very fun ways, including causing my command pod turn fully around and head back out of the atmosphere without hitting the ground (probably due to the way it simulates body lift).

Edited August 18, 2014 by Vaporo

[UberFuber]

EDIT: Did some more experiments with FAR installed. trying to reenter at these speeds caused KSP to glitch out in some very fun ways, including causing my command pod turn fully around and head back out of the atmosphere without hitting the ground (probably due to the way it simulates body lift).

Depending on your angle of entry, that may not be a glitch but an expected behaviors. In fact...

http://en.wikipedia.org/wiki/Skip_reentry

In short, when you're entering the atmosphere fast enough, you can actually "bounce" back out the atmosphere completely without any additional external thrust.

[Vaporo]

Depending on your angle of entry, that may not be a glitch but an expected behaviors. In fact...

http://en.wikipedia.org/wiki/Skip_reentry

In short, when you're entering the atmosphere fast enough, you can actually "bounce" back out the atmosphere completely without any additional external thrust.

Errrr no. It was technically a skip reentry, yes, but it was not in any way physically realistic. I came straight down towards Kerbin at 75 km/s, around six times earth's escape velocity. No aerodynamic forces could possibly turn that around 180 degrees.

[Laie]

Honestly, the examples you're providing are easily avoidable - are you heading directly towards the centre of the planet at a ridiculous speed?

Look in the challenges subforum for an explanation why this is even a topic.

Problem: when a craft makes a very high speed entry into a planet, the craft may reach the ground or get very close to it before the atmosphere begins to decelerate the craft significantly, resulting in impacts at much higher velocities than what is realistic.

Not excatly once per second. In my case, I figured that I must have about 16 physics ticks a second. Though the lack of any noticable drag above 30km is intriguing -- it should still be enough to be noticable.

Explainer: this is the last moments of a high-speed impact on Kerbin. X-axis: altitude, y-axis: surface speed. Data was collected using kOS and the simplest possible script. I should have one data point for every tick of the physics clock.

[thereaverofdarkness]

theravenofdarkness considers the average of 21358g of deceleration (thats kerbin g, not duna) to be unrealistically low.

Because of this, he has branded me a cheater and disqualified my entry into his "maximum impact" challenge.

I never disqualified him; in fact he begged me to remove his entry. I never called him a cheater, nor insinuated such. His entry was completely valid and I have said to him several times that there is nothing wrong with it.