Part's Failure Feature

[Onyx_Pheonix]

A recommendation for people looking for more difficulty in the game, add the potential for parts failure as an option. Engine failures, control surfaces sticking, gimbal lock, partial damage states for parts overstressed by aerodynamic or gravitational forces. Maybe for each of those add a way to mitigate the potential for these problems too, balanced against extra weight for reinforcement or reduced speed for an aircraft. Several mods out there already based on realism and equipment limitations. 

[The Aziz]

And it should remain in a mod territory. Mission failures need to be caused by player's error (so there's always an option to prevent them) not some RNG that can ruin 30 year long interstellar mission near its end.

[jastrone]

1 hour ago, Onyx_Pheonix said:

partial damage states for parts overstressed by aerodynamic or gravitational forces

this is the only one i see working. the other ones are basically 'go out and fix something because you where unlucky'. it would however be nice if you fly your plane a bit to hard you have to repair it. or maybe when you go through reentry something breaks and you have to fix it. and maybe rovers are fixable if you crash them

[regex]

One thing that could be fun is the (low) possibility of an engine failure during, say, the first five minutes of a launch. Quite frankly I have zero reason to ever install a launch escape system aside from the looks. It's honestly kind of meaningless without production and rollout times, and the ability to simply revert to launch, but it gives you a reason to actually have those sorts of safety features.

[mcwaffles2003]

3 hours ago, The Aziz said:

And it should remain in a mod territory. Mission failures need to be caused by player's error (so there's always an option to prevent them) not some RNG that can ruin 30 year long interstellar mission near its end.

How would you feel if it wasn't RNG though? For instance, some parts could have a use life that is fully deterministic & monitored and require maintenance via something as simple as having enough engineers on board based on part count/repair frequency, similar to how we need scientists to repack some experiments? 

 

You could still have those parts on rockets without engineers, but you would watch their maintenance bars get low and eventually develop a malfunction that was wholly foreseeable and predictable.

Edited October 24, 2022 by mcwaffles2003

[razark]

2 hours ago, regex said:

One thing that could be fun is the (low) possibility of an engine failure during, say, the first five minutes of a launch.

I've enjoyed the Kerbal Launch Failure mod.  It's made having a LES worthwhile.

It's also led to some interesting quick thinking to save a mission when one of your stages suddenly isn't there anymore.

 

Any failures post-launch should be something to make missions interesting, but not game killing, and provide a way to conduct repairs.

[Vl3d]

17 hours ago, Onyx_Pheonix said:

Engine failures, control surfaces sticking, gimbal lock, partial damage states for parts overstressed by aerodynamic or gravitational forces.

Sounds so good - random part failure would make for great improv moments. But.. parts wearing out should be the main cause of failure. Randomness should be statistically transparent.

15 hours ago, regex said:

Quite frankly I have zero reason to ever install a launch escape system aside from the looks.

Always hated that I don't really have a reason to use a launch escape outside of paying without reverting. But then you can get kraken attacks and failures should be a separate thing from bugs.

I want more emphasis on careful unit and integration testing - but when you launch you should also know the percentage of possible random failure and take a chance.

Edited October 24, 2022 by Vl3d

[The Aziz]

Well.. LES may be finally useful when MP arrives and a friend accidentally sabotages your launch.

[Vl3d]

30 minutes ago, The Aziz said:

a friend accidentally sabotages your launch

Yeah but we need a coop VAB designer for that.

[regex]

38 minutes ago, The Aziz said:

Well.. LES may be finally useful when MP arrives and a friend accidentally sabotages your launch.

And for those of us who probably won't play MP?  "It should just remain in modded territory" doesn't cut it for a piece of gear that's been useless since day 1.

[pandaman]

16 hours ago, mcwaffles2003 said:

How would you feel if it wasn't RNG though? For instance, some parts could have a use life that is fully deterministic & monitored and require maintenance via something as simple as having enough engineers on board based on part count/repair frequency, similar to how we need scientists to repack some experiments? 

 

You could still have those parts on rockets without engineers, but you would watch their maintenance bars get low and eventually develop a malfunction that was wholly foreseeable and predictable.

That's all well and good, but imagine 100 plus ships all with multiple parts that could be 'jolly inconvenient' if they stopped working.  

An automated 'wear and tear' alert system would use lots of computing power to track everything, as well as the player being constantly bombarded with 'this part needs fixing' messages at inconvenient moments.

Fine as a mod, but not as even a toggleable option in stock IMHO.

[Onyx_Pheonix]

Thanks for the thoughts, folks!

3 hours ago, pandaman said:

as well as the player being constantly bombarded with 'this part needs fixing' messages at inconvenient moments.

My idea isn't to have this as an issue that would present itself in every mission. In fact, I would like it to only be rarely, as a proverbial spice to the game, not a main ingredient. Ways to mitigate that could include gaining engine data, as I believe a few mods already utilize. This would mean you can't go for a deep interstellar mission with an untested engine without risking failure, but maybe with x amount of flights then the chance of failure is eventually eliminated, or the base failure chance is mitigated as you go up in tech tiers.

23 hours ago, The Aziz said:

it should remain in a mod territory

 

3 hours ago, pandaman said:

Fine as a mod, but not as even a toggleable option in stock IMHO

I think this should be an option in the game, much as reentry heating or permanent Kerbal death. In multiplayer it could be a server setting where the server pop all abides by the same rules. Sure modders can do it, but this gives players more options for re-playability in the base game. While there is a term, "Kerbal" denoting the inherent chaos in the game, the developers also did a great job capturing the spirit of astronavigation, physics, and aerodynamics. Some people crave a more realistic feeling to the classic KSP which is where mods like Realism Overhaul come into play. This would allow the game devs to optimize and market the game in another potentially profitable way if they add it as an option to base and not just allow it as a mod. 

[The Aziz]

6 minutes ago, Onyx_Pheonix said:

reentry heating

You want to make literal physics optional?  May as well remove the need for engines as an option.

10 minutes ago, Onyx_Pheonix said:

Some people crave a more realistic feeling to the classic KSP which is where mods like Realism Overhaul come into play. This would allow the game devs to optimize and market the game in another potentially profitable way if they add it as an option to base and not just allow it as a mod.

Game is moddable, mods are usually free. It's probably not worth assigning workman hours for a feature that would please only a bunch of hardcore players.

Also, have RO people figured out how to deal with interstellar travel yet?

[Onyx_Pheonix]

2 minutes ago, The Aziz said:

You want to make literal physics optional?  May as well remove the need for engines as an option.

Reentry heating already is optional in KSP, I'm just using that as an example of an option that can be toggled.  

3 minutes ago, The Aziz said:

Game is moddable, mods are usually free. It's probably not worth assigning workman hours for a feature that would please only a bunch of hardcore players.

The benefit would be to market that as a part of the game. Including it gives the studio the credit and rights to it, as well as includes support for content so it wouldn't need to be updated as a mod after every iteration released in development. This also isn't just a hard core player thing, most KSP players I've talked to in my friend groups love the idea of mechanics like this. If all we can get is mods, that's what we'll take(because these mods will happen), but I'm recommending this be more than a mod because it's a popular mechanic. 

[mcwaffles2003]

4 hours ago, pandaman said:

That's all well and good, but imagine 100 plus ships all with multiple parts that could be 'jolly inconvenient' if they stopped working.  

An automated 'wear and tear' alert system would use lots of computing power to track everything, as well as the player being constantly bombarded with 'this part needs fixing' messages at inconvenient moments.

Fine as a mod, but not as even a toggleable option in stock IMHO.

I didn't mean the player would have to manually deal with the problem. Merely having the required amount of engineers would suffice to maintain the ship and so long as that threshold is maintained larger scale stat tracking would be unnecessary.

[t_v]

1 hour ago, The Aziz said:

You want to make literal physics optional?  May as well remove the need for engines as an option.

Just fyi, you can disable re-entry heating, and there are ways to remove the need for engines (“set orbit” comes to mind). I don’t think that the at was the important part though, the OP is essentially saying that part failures are just as deserving of developer time as other key aspects of space flight, like heat management.
 

What I have to say about it: first, part failure is already in the game to an extent. Wings shear off, wheels break, engines overheat, and rockets oftentimes blow up getting to orbit. These failures can be amplified by mods such as FAR and FFT, but the concept is there in the base game. However, these don’t feel like failures because the stress tolerances of parts are constant, the amount of heat produced by engines is predictable (usually), and when a rocket blows up on launch, chances are it is because of the design and not chance.

Another thing that dampens these failures is quick loading; when you are on a mission to Eve and your mothership explodes its engine on a long burn, it gives the same sense of being stranded as when that engine fails randomly.

Essentially, these deterministic (except for the kraken) failures are better than random, even mitigatible random, failures, and there are definitely opportunities where they can happen more often, as anyone who has played with large ships and larger engines can state. 

[The Aziz]

2 hours ago, Onyx_Pheonix said:

Reentry heating already is optional in KSP

 

52 minutes ago, t_v said:

Just fyi, you can disable re-entry heating, and there are ways to remove the need for engines (“set orbit” comes to mind)

If it's in the settings (I forgot if it is) it's only there because once the game was without reentry heating. The toggle was there for save compatibility - wouldn't want your pre-1.0 craft suddenly burn up on reentry only because it was made before heatshields were invented implemented. Same situation with commnet.

Set orbit is literally the cheat menu, not part of standard mechanics.

59 minutes ago, t_v said:

Wings shear off, wheels break, engines overheat, and rockets oftentimes blow up getting to orbit. These failures can be amplified by mods such as FAR and FFT, but the concept is there in the base game. However, these don’t feel like failures because the stress tolerances of parts are constant, the amount of heat produced by engines is predictable (usually), and when a rocket blows up on launch, chances are it is because of the design and not chance.

My whole point. Don't add anything you cannot prevent. All the examples you provided can be avoided by more careful driving/flying, proper heat management. If something bad happens, it's the player's fault, and it should stay that way. Why? Nobody is learning through random engine failure and rocket falling into fiery end in 26seconds into the mission, but everyone is, or should be learning through bad design or careless steering through the atmo, when a rocket flips. If I wanted RNG, I'd get back to Mars Horizon.

[Vl3d]

In real life rocket science mitigating the possibility of failure is a very important topic, right? I mean, most decisions are taken after considering parts and systems failure (and weather) probabilities. This types of considerations should be in the game.

[pandaman]

19 hours ago, mcwaffles2003 said:

I didn't mean the player would have to manually deal with the problem. Merely having the required amount of engineers would suffice to maintain the ship and so long as that threshold is maintained larger scale stat tracking would be unnecessary.

I can see why random part failure appeals to many players, just not me personally for a game of this size and scope.   We will just have to agree to disagree on this one.

To me ensuring that I maintain 'the required amount of engineers' is 'manually dealing' with it,  just not at the point of failure.

Also, from a realism standpoint, if you haven't got the right parts with you no amount of engineers can fix it.  The current 'use this many repair kits' method works, but it’s very much a 'fudge' if you want realism.

[mcwaffles2003]

53 minutes ago, pandaman said:

I can see why random part failure appeals to many players, just not me personally for a game of this size and scope.   We will just have to agree to disagree on this one.

Dude, the whole time I have been saying not random though... see:

On 10/24/2022 at 2:10 AM, mcwaffles2003 said:

How would you feel if it wasn't RNG though?

I am talking about a completely deterministic system.

 

54 minutes ago, pandaman said:

To me ensuring that I maintain 'the required amount of engineers' is 'manually dealing' with it,  just not at the point of failure.

It isn't much different from ensuring you have enough solar panels on a craft though. 

55 minutes ago, pandaman said:

Also, from a realism standpoint, if you haven't got the right parts with you no amount of engineers can fix it.  The current 'use this many repair kits' method works, but it’s very much a 'fudge' if you want realism.

That would actually be a good idea, make sure the required amount of personnel and materials exist.

 

At the end of the day I feel like it could add a cost analysis structure in kerballed vs probe craft. Probes would have a shelf life but be lightweight and easy to launch while craft with kerbals could be maintained indefinitely at the cost of mass.

[pandaman]

21 minutes ago, mcwaffles2003 said:

Dude, the whole time I have been saying not random though... see:

Ok, you got me there .

So if I understand correctly, what you propose is essentially natural lifespans of parts, which varies according to how much and how hard they get used and abused. 

Certainly not illogical, and I can see the appeal, but for me at least, it is an extra layer of complexity I don't really want to deal with.  Yes a 'toggle'  or 'difficulty setting' would  sort that out game play wise, but the game would still need to keep track of the wear level for all relevant parts.   Not an insignificant amount of data.

[mcwaffles2003]

2 hours ago, pandaman said:

So if I understand correctly, what you propose is essentially natural lifespans of parts, which varies according to how much and how hard they get used and abused. 

Essentially, yes.

2 hours ago, pandaman said:

Certainly not illogical, and I can see the appeal, but for me at least, it is an extra layer of complexity I don't really want to deal with.  Yes a 'toggle'  or 'difficulty setting' would  sort that out game play wise, but the game would still need to keep track of the wear level for all relevant parts.   Not an insignificant amount of data.

To each their own, just happy were on the same wavelength now

[pandaman]

9 hours ago, mcwaffles2003 said:

To each their own, just happy were on the same wavelength now

I'm always happy to accept and understand others' points of view and opinions if I can.  All opinions are equally valid unless established facts prove otherwise.

Edited October 26, 2022 by pandaman

[Pthigrivi]

Also we’re going to have probes and other unmanned vessels that need to make it to Jool or even other star systems, potentially networks of communications satellites that will be up there on their own for years or decades. Even if its deterministic I just don’t think manually maintaining those is something most players actually want to deal with. It is a cool idea for a mod though. 

[Vl3d]

I agree that parts wear and failures should always be optional as a difficulty setting.

Parts failure is fun in the design phase because it forces you to think about redundancy and it increases craft complexity a little bit, which forces you to reuse your designs more, which increases the sense of accomplishment. It's also fun when redundancy pays off and you save the day like a hero.

I play with parts failure but when it really causes a disaster I usually revert to launch or load a previous save. So the game design trick is how to introduce a mechanic that has the benefits mentioned but is not frustrating.

One solution is to have parts failure only after parts wear. This way there's no randomness. But then it introduces the concept of maintenance and re-doing missions, and worst of all - you can't time warp for long periods at will because everything breaks down meanwhile and it kills your progress. So how would you circumvent this? Simple.

Have parts failure only after wear and only on the currently controlled craft. For manned craft part maintenance should be done automatically if you have enough repair kits and engineers. For unmanned (or craft without repair kits) you have to include part redundancy in the design.

I think this solves most issues.