Reliability rating

[falconek]

In real life one of the biggest problems in space missions is malfunctioning equipment. Virtually every mission has bigger or smaller technical problems, and it's no wonder, considering what extreme conditions this equipment needs to survive.

How about you copy a nice idea from Buzz Aldrin Race Into Space, and add reliability rating to every part. This rating could be improved by research on certain parts and by using them more (for instance if you have several succesful launches with LV-T30, the next batch of LV-T30 will be more reliable).

Then you can add different types of malfunctions with different effects (premature ignition of a stage, engine didn't fire, leak of propellant... and many more, including malfunctions or science and communication equipment). Some of these would be fatal, and some would let you recover and continue mission with limited scope (like you won't have enough dv to reach that planet AND it's moon, but you can still reach the planet).

Of course this feature could be optional, and adjusted in game options even to the point of competely turning it off. Similar features can be seen in flight sims, like Lock-On, where you can even set probability of colliding with a bird.

[Tex_NL]

So basically you're suggesting random failures?

SQUAD has repeatedly stated it does not want and random events in KSP. Perhaps that's why it's on the What not to suggest list.

[Nuke]

its not a random failure if you launch a space mission with an untested part! it failed because you launched with an untested part. therefore its not random at all. of course once jeb has destroyed it a couple times (and earned some science in the process), it becomes reliable. this would actually be a good way to reward conducting test flights with science. this is up there with levelable parts and trl modeling.

i see "random failures" being the game engine decided to throw a monkey wrench into your mission because of a dice roll. but im all for failures caused by stupidity. so you might unlock a part and it would have the same rating except maybe it overheats, or it is too weak structurally, or the power requirements are ludicrous, or the efficiency is not there. all this stuff would be in the parts information tab, so you can see that something is not right with it. you might build an unmanned test flight (or have a really good abort system) and test the parts. science would be rewarded each time you identify a problem with a part (this would require you fly, have the part fail, and recover it), also the part will be improved. this is part of running a real space program and would be a welcome addition to the game in my book.

Edited February 1, 2014 by Nuke

[jwenting]

it is random, unless you are suggesting that all launches using a part that's not been used in x previously successful launches fail.

Which of course means all launches always fail as there'd never be a successful launch...

[Tex_NL]

it is random, unless you are suggesting that all launches using a part that's not been used in x previously successful launches fail.

Which of course means all launches always fail as there'd never be a successful launch...

Exactly.

If a part may or may not fail at any indeterminate point it is random. That's what random means.

If you would suggest new untested parts could fail similarly to engines overheating or parts breaking due to excessive stress I am all for it. Those kinds of failures are predictable and can be avoided, either by a better design or by piloting. Parts blowing up for no apparent reason would add nothing to the game and would only cause a flood complains on the forum.

[crazyewok]

So basically you're suggesting random failures?

SQUAD has repeatedly stated it does not want and random events in KSP. Perhaps that's why it's on the What not to suggest list.

Well they should make it moddable then so those of us who DO want a challenge can have it.

[Comrade Jenkens]

I like my failures being caused solely by me failing to engineer the craft successfully. Not caused by some random dice roll due to no fault of mine!

[Dunbaratu]

Well they should make it moddable then so those of us who DO want a challenge can have it.

It's incorrect to use the term "challenge" to refer to a failure you have no control over and isn't your fault when it happens. The main reason random reliability failures for parts is a bad idea in KSP is because the design and manufacture of those parts is entirely out of the player's hands. UNLIKE in a real space program. If the manufacturing process is creating parts that fail, there are people who's job it is to fix that. In the real world if a manufactured part fails, there are human beings ultimately responsible somewhere in the chain. In KSP that is not true.

[Kerbart]

Wait, so every time you use a part that hasn't been "tested before" will fail? That's not an incentive to use it in flight. And once it has been tested it will never fail?

What's to stop me from "testing" my parts on the launch pad?

[crazyewok]

What's to stop me from "testing" my parts on the launch pad?

Let me think? Hmmmm CAUSE THATS WHAT THEY DO! If you notice they do test parts on the ground and on unmanned craft before sticking a brand new part on a manned craft.

[falconek]

As I see it, it forces you to engineer simpler and more redundnat solutions. If you take into account, that a part may fail, you may want to consider backup systems, or emergency scenarios. Also, there is risk magement involved. You can run short, easy missions, untill certain part will reach decent level of reliability.

[Nobody]

I am against the random dice rolls to see if a part fails. A mod for that might work for those who want things to go boom for no reason.

Failure due to environmental conditions I could live with for example, having a planet with much higher atmospheric pressure than Kerbin could crush a mk1 lander can(description says it wouldn't do well in atmo) like a sub going to deep in the ocean. That wouldn't be a case of random failure but going beyond the designed operating conditions and encourage the use of probes to test the waters.

[Z-Man]

So, how would "testing" a new part be fun gameplay?

[minerman30]

I have two ideas that could make random failures more realistic to implement.

First, the kind of mission-ruining failures that people are afraid of would be extremely rare. I'm thinking on the order of 0.1%-0.01% chance of failure for things like engines and fuel tanks.

Second, there would be more minor failures that could be repaired by kerbals, such as solar panels (things like the rotation function getting jammed) and reaction wheels. This would mean that a well-designed ship should be able to complete a mission even if parts break down, and it would make manned missions more reliable.

[nholzric]

First, the kind of mission-ruining failures that people are afraid of would be extremely rare. I'm thinking on the order of 0.1%-0.01% chance of failure for things like engines and fuel tanks.

Just because a catastrophic failure is rare doesn't make it any more fun. Plus, from a game design perspective there's a point when you want to make something so rare that you have to wonder if it's worth adding to the game at all.

These random failure threads (besides being on the "do not suggest" list) are the polar opposite of the "simulator" threads http://forum.kerbalspaceprogram.com/threads/66371-Simulator. The folks who want a simulator function are frustrated that they spend so much time flying a mission to Moho only to learn their lander doesn't have enough TWR... Adding bigger engines to your lander will be the least of your problems if your fuel tank randomly fails and starts leaking oxidizer.

[minerman30]

Just because a catastrophic failure is rare doesn't make it any more fun. Plus, from a game design perspective there's a point when you want to make something so rare that you have to wonder if it's worth adding to the game at all.

These random failure threads (besides being on the "do not suggest" list) are the polar opposite of the "simulator" threads http://forum.kerbalspaceprogram.com/threads/66371-Simulator. The folks who want a simulator function are frustrated that they spend so much time flying a mission to Moho only to learn their lander doesn't have enough TWR... Adding bigger engines to your lander will be the least of your problems if your fuel tank randomly fails and starts leaking oxidizer.

I see. So maybe we only have the minor repairable failures.

[MajorThomas]

I'm with Nuke, let new parts have a "testing" phase during which they have a failure rate. Say, initially 10-20%. But after a few missions where the part did not fail, the failure rate drops to zero. And you are told the expected reliability of the part, and how many successful flights it's had, before you use it in the VAB.

And, for those who don't want to deal with part failures (design and piloting failures already being present) have the option to make all parts 100% reliable. But for advanced users, part testing and reliability would be FUN!

[Tex_NL]

Some people have come up with some interesting idea's. I especially like what Nobody said a few posts ago: Failures due to exceeding environmental tolerances. Those kinds of failures are predictable, consistent and avoidable.

I also partially liked what minerman30 suggested: Part failure that can be repaired. The part I didn't like in this is the random factor. If that randomness would be changed by degradation over time it again becomes consistent and predictable. Once the part wears out it fails, not before.

All this got me thinking and then I remembered the recent "Expanding the Kerbal" suggestion: http://forum.kerbalspaceprogram.com/threads/67315-Expanding-the-Kerbal. It talks about different Kerbal jobs; pilot, scientist and of course engineer. See where I am going with this yet? Repairing degraded parts could be an ideal job for the engineer. Regular Kerbals should of course also be able to make repairs. Engineers should be able to repair to the full 100% or perhaps even past that.

[Kasuha]

In real life one of the biggest problems in space missions is malfunctioning equipment.

I don't think KSP lacks this kind of challenge. Launch clamps failing to hold their load, shake on deploying a rocket on launchpad, ships getting destroyed on revert, aircraft being dropped on runway from height... There's plenty of failures involved already. I don't think any more should be added.

[xrayfishx]

If random failures were to be added to the game, I'd see it be something like this:

First time you use of a part there's a 20% chance of failure each second. Each time you use that part, the chance becomes 5 times smaller. So the second use would have a 4% chance of failure each second, then 0.8 etc etc. Eventually the chance would become so small it doesn't really matter. Also, a failure wouldn't mean that the part blows up or sth. What I mean is that nearly all failures would be something you can deal with, for example an engine only being able to put out 70% of rated thrust, or a solar panel not folding up completely. And there should be difficulty settings so you can turn off failures you can't deal with, such as an SRB exploding or an engine turning off. Or turn off all failures, for that matter. Everyone would be happy this way.

I don't think KSP lacks this kind of challenge. Launch clamps failing to hold their load, shake on deploying a rocket on launchpad, ships getting destroyed on revert, aircraft being dropped on runway from height... There's plenty of failures involved already. I don't think any more should be added.

This is different kind of failures, and they always happen right after/before the craft finishes loading and usually end up with reverting, fixing and redoing. They are glitches, real life doesn't have glitches.

Back on topic, though (because Tex_NL's signature perfectly describes this thread), I don't think that in the beginning a part should fail every time, or that not recovering the malfunctioning part should be necessary. Firstly, even if the part does not fail, the post-flight inspection would reveal what _almost_ went wrong and if the part isn't recovered, then since the failure is most likely in atmosphere, at least in the beginning, then the engine blowing to bits leaves behind the bits that can, again, be inspected (that, btw, is how aircraft crashes are investigated).

[Dunbaratu]

Just because a catastrophic failure is rare doesn't make it any more fun.

In fact making them rare makes them LESS fun. Because it then becomes unreasonable to expect players to have experience with it, so that when it does happen players feel as if the game betrayed them unfairly.

Also, think of the tech support nightmare. The game *already* has a few bugs that cause things that should never happen to happen, like space stations exploding the moment you load them from the tracking center because collision detection "turns on" before all the variables got fully populated yet. Now imagine the flood of people reporting "my rocket blew up for no reason" as a bug, and then tech support having to try to sift through which of the complaints are intentionally unreliable behavior and which are bugs? Deliberately creating random failures would mask the behavior of real bugs.

[coolitic]

Random failures will never be implemented.

[Nuke]

So, how would "testing" a new part be fun gameplay?

how many test pilots hate their job?

[Nuke]

I'm with Nuke, let new parts have a "testing" phase during which they have a failure rate. Say, initially 10-20%. But after a few missions where the part did not fail, the failure rate drops to zero. And you are told the expected reliability of the part, and how many successful flights it's had, before you use it in the VAB.

And, for those who don't want to deal with part failures (design and piloting failures already being present) have the option to make all parts 100% reliable. But for advanced users, part testing and reliability would be FUN!

i need to reiterate that im not talking about having random failures. im talking more about having newly unlocked parts in a weakened, unrefined state. you might even have a little gauge or even a bit of comical text to discrive the state of a part. say you get a new engine. this engine might have twice the heat production and half the structural integrity of the refined version. the part info window will display the strength and heat ratings. you may have a trl gauge to indicate level of refinement. you might also have some engineer's notes. first time you fly it:

"brand new part with brand new bugs!"

upon having an overheat (and returning data):

"we fixed the overheating problem, but it still looks a little weak"

overheat problem fixed!

+50 science!

upon having a structural failure (and returning data):

"this part is as good as it can be"

structural problem fixed!

+50 science!

at this point the part is reliable.

every part would have 2 or 3 "bugs" which you can "fix" for science and a better part. the engineer notes will provide hints about what the problems the part may be. a smart player would conduct test flights to exploit these problems for science. it also makes abort procedures important so that data can be returned to the engineers.

Edited February 3, 2014 by Nuke