[0.24] If a Booster Has Enough Parachutes On It, Count it as "Recovered"

[Andrew Hansen]

Looking at Scott Manley's

about recovery mechanics in 0.24, I saw an idea in the comments that made a lot of sense.

Let's say you launch a rocket that uses boosters attached to the side that are decoupled while the main craft continues into space. If those boosters have parachutes on them, they should be recovered. The problem is that with the 2.5 km physics range limitation, they will be deleted before they land on the ground and won't be recovered.

What if, instead, when a booster has enough parachutes on it, it will be counted as "recovered" when it passes the 2.5 km range? The question that might then be asked is, "What is 'enough' parachutes?" Well, it's easy to figure out the landing speed of something by doing some calculations with its mass and the number of parachutes, so if the landing speed is calculated to be under a certain speed (like 6 m/s), it could be recovered.

[regex]

My only problem with this suggestion is that you have to assume that every launch will be using such tactics (honestly, why not?) so the game will have be balanced around it, which means that every launch will require loads of additional parts (and we're not exactly talking about a difficult design challenge here in stock KSP...) because there is no reason not to recover parts.

100% recovery on spaceplanes is bad enough, I don't think this is a good idea for balancing. Besides, actually recovering those boosters before the 2.5km boundary is far more of a design and gameplay challenge with the current mechanic.

[NovaSilisko]

Could have a "maximumRecoverableMass" value on parachutes as a simple way to check if the debris will survive descent. Do a quick check of the recovery masses on each parachute, compare it to the total mass of the debris, and give a thumbs up or down as to whether it can be brought back. Recovery value shouldn't be very high though, the shuttle SRBs for instance had to be partially reassembled each time, and also refilled with solid fuel.

[cpast]

If this were done (and I'm not sure it should be, for the reasons regex mentioned), they could force it to the lowest possible recovery percentage (10%).

[Woopert]

I would like to see this, actually. I hope the devs take notice of this thread.

[Captain_Brian]

I had the same thought about this and the "maximumRecoverableMass" value, but the problem is all this gets you is having to modify all your builds to put tons of parachutes all over the place like regex said. That said, I like the idea of recovering the parts, but I don't like the idea of all the extra parts to do so.

[NovaSilisko]

I had the same thought about this and the "maximumRecoverableMass" value, but the problem is all this gets you is having to modify all your builds to put tons of parachutes all over the place like regex said. That said, I like the idea of recovering the parts, but I don't like the idea of all the extra parts to do so.

That's just the price of recovery. Most rockets to date have been expendable because it's a pain in the ass to get them back without smashing them like tin cans.

[Captain_Brian]

*SNIP* Making a thread instead as this idea got bigger and bigger in my head.

Edited July 16, 2014 by Captain_Brian

[Jouni]

If you've tried recovering boosters manually, you know that it's much harder than just adding enough parachutes. Some boosters don't survive splashing down in the ocean. Others need level ground, or they fall down and break after landing. Some need to land very slowly or risk losing parts due to the impact.

A mechanism that allows you to recover boosters that you couldn't land safely on your own is just a bad idea.

[Sky_walker]

I'm not a big fan of that. Dropping stages on parachutes is something that basically never happens in real life, yet alone: is applied to every single design. And what you are proposing will do just that: make people attach chutes to every single stage they're dropping.

besides - do an in-game test of dropping some larger booster on a chutes - it explodes most of the time on a contact with ground.

[Andrew Hansen]

If you've tried recovering boosters manually, you know that it's much harder than just adding enough parachutes. Some boosters don't survive splashing down in the ocean. Others need level ground, or they fall down and break after landing. Some need to land very slowly or risk losing parts due to the impact.

A mechanism that allows you to recover boosters that you couldn't land safely on your own is just a bad idea.

Yes, I didn't think about this before....

*thinking*

*more thinking*

Having the boosters load all the way to the ground is really the only way to accurately recover them when they should be recovered. The problem that then arises is lag caused by high part counts. Normally the part count decreases on ascent in a rocket that decouples boosters on the way up, but lag wouldn't reduce if the physics range was extended for boosters.

[Bilfr3d]

You guys know that there is a mod for this right? I've seen it around someone, but I can't remember what it is called (whups ). It calculates the weight and how slow the parachutes can slow it to. If it can be slowed below 7m/s (8m/s is about the highest impact tolerance on some parts) it automatically gets recovered when you stage. Or else you lose it forever

[KaBob]

I think if they combine a low recovery % with a higher than normal parachute to mass requirement it could be fairly balanced. Something like 5% of the cost to imply that while it might not have landed intact the raw materials were at least usable.

[Andrew Hansen]

I think if they combine a low recovery % with a higher than normal parachute to mass requirement it could be fairly balanced. Something like 5% of the cost to imply that while it might not have landed intact the raw materials were at least usable.

This makes sense to me. While it's true that boosters may sometimes fall over after landing and explode in KSP, this isn't really realistic, since in real life stuff doesn't just vaporize into thin air like in KSP. So then this idea could actually work.

EDIT: Some numbers I'd be happy with are: If landing velocity is calculated to be at <6 m/s, the part will be recovered. The recovery percentage could be like 15% or 20% of your money back. 5% seems a bit overkill.

Edited July 16, 2014 by Andrew Hansen

[quasarrgames]

I WANT THIS TO BE A THING!

I also like the idea of having the boosters have their own physics as they fall, deploy their chutes, and land. They could have their own tiny physics bubbles (like 100-200m in diameter) so that they don't cause massive performance drops. you could set a stage to have its own tiny physics bubble when it is decoupled in the VAB.

[GluttonyReaper]

I WANT THIS TO BE A THING!

I also like the idea of having the boosters have their own physics as they fall, deploy their chutes, and land. They could have their own tiny physics bubbles (like 100-200m in diameter) so that they don't cause massive performance drops. you could set a stage to have its own tiny physics bubble when it is decoupled in the VAB.

Does that really solve the issue? I always assumed it was the number of parts being simulated with physics than wouldn't otherwise, rather than the size directly causing performance issues.

[quasarrgames]

Does that really solve the issue? I always assumed it was the number of parts being simulated with physics than wouldn't otherwise, rather than the size directly causing performance issues.

Idk. maybe it would help, maybe not. other than the physics bubble, i don't see any reason why having boosters separate from the main craft would be any worse performance-wise than having those boosters still on the main craft. If you can launch it, you can recover it!

[Justin Kerbice]

Don't forget KSP is not either a simulator nor real life, so talking about damage to some structure, sensible parts, ... is quite irrelevant here.

The main issue is the 2.5km limit which is annoying and ridiculous for such purpose, and it delete everything like an hungry kraken (even sub-vessels with crewed pod !).

Thing can be not too hard and to too CPU intensive by simplifying calculation:

booster mass ok for chutes => do simple math for speed at ground contact,

if speed >= crash tolerance = boom,

else: it's drop gently,

then according to result, put booster near the estimated landing zone or add a log entry like "booster crash on the ground" at the estimated landing time.

[technicalfool]

I would like this idea.

I also don't see why there would need to be a recovery percentage penalty. Yes you can land any of the KSP boosters safely with parachutes. If it explodes on landing, you weren't using enough parachutes. Whether it lands under your control or not, it'll be just as easy or hard to recover.

Quick back-of-a-notepad calculation based on the jetissoned stage's mass, the number and type of parachutes, and get a figure of how fast it should be going when it reaches the floor.

Below 10m/s, it's recovered. From 10-20m/s, it may blow up or be damaged. Roll some virtual dice, use it to determine either a recovery penalty for single-part stages, or a number of parts to be blown up for multi-part stages.

Above 20m/s, it goes pop. Tough luck, add more parachutes next time.

[Jouni]

I also don't see why there would need to be a recovery percentage penalty. Yes you can land any of the KSP boosters safely with parachutes. If it explodes on landing, you weren't using enough parachutes. Whether it lands under your control or not, it'll be just as easy or hard to recover.

You can recover pretty much anything, if you can find a suitable landing zone. The bigger the booster is, the more critical finding a good landing zone becomes, just like with tall, top-heavy landers.

[RidingTheFlow]

Being able to recover anything just by slapping enough parachutes on top is too easy, and effectively eliminates any point of diversity/difficulty balance between reusable and non-reusable designs (since non-reusable designs will be completely pointless if you could just as well add chutes to everything).

[GreenArsenic]

Luckily, there is a biome system in the game. With this system, the game could influence what the speed must be when the parachutes are deployed. For example:

Above water: max speed to succesfully recover item == 10 m/s

Above land: max speed to succesfully recover item == 8 m/s

Above Hills: max speed to succesfully recover item == 4 m/s

Above mountains: max speed to succesfully recover item == 2 m/s

(Also, these numbers could be influenced by the crashtolerance)

This will force the player to carefully pick the spots where the player feels like landing the object.

The safe-parachute-landing-forumula may look like this: (Total Weight)/(Parachutecarry1 + Parachutecarry2)*(crashtolerance of the part directly on the bottom).

If it is bigger than a certain value, it will be recoverable. If it is smaller... Though luck.

[Jouni]

Luckily, there is a biome system in the game. With this system, the game could influence what the speed must be when the parachutes are deployed.

It doesn't work that way. Even if you manage to land safely, the lander may still tip over and break, if its center of mass is not above the base of support.

[GreenArsenic]

It doesn't work that way. Even if you manage to land safely, the lander may still tip over and break, if its center of mass is not above the base of support.

Perhaps that the landing is only succesful if it does NOT land in the hills or mountains. Or a percentage of the time, it does land succesfully there (e.g. 20% of the time). But I do agree that the base of support thing must also be incorporated in the formula, though this could make that extra function kind of bulky.

I still think that even the slightest of additions made to stop things from instant sublimation after reaching 2.5 km, would be a good thing.

[Sky_walker]

Mountains, hills, shores, badlands.... I believe water also breaks up boosters apart if they're parachuted.