Random events might be fun!

[minerman30]

I think that, with regards to random events such as space weather or breakdowns, some means of anticipating them would help balance the fun vs. frustration factor. I think we can live with randomness if we can see it coming and at least take appropriate action. Likewise, we'd need to be sure that such random events are not "one-hit kills" for spacecraft. Random events should also be rare - like, happening only once every 1,000 game-days or so on.

Take solar flares, for example. If I was to implement them in the game, I'd probably limit space weather to three states, with one state - "calm" - being the normal, and the other two happening only occasionally. Two classes of solar flares would be enough. Minor solar flares would mean temporary loss of communication for ships not shielded by planetary bodies (i.e. science transmissions won't go through), but no system malfunctions. Major solar flares would introduce system malfunctions, but at a small chance of them happening. I would NOT include Coronal Mass Ejections or Carrington-level solar storms because, well, any spacecraft caught in the open when such a thing happens is fried. Period.

While ground weather would be too complex to model, perhaps ground conditions that change occasionally might help - with a weather report that can show what said conditions will be in a region. Again, for example, we can apply this to launching a mission from the KSC. When you go to launch a spacecraft from the launchpad or runway, you get, not just a list of ships and kerbonauts to fly them, but a "five day forecast" with some events (thunderstorms, tropical storms, hurricanes) having a chance of launch failure. You might still lose a ship, but only if you don't heed the forecast. Knowing it's coming in advance would help to make sure players don't toss their computers through a window if their 200 ton Duna mission is torn apart by Tropical Storm Jeb on ascent.

Eve might have a chance of malfunctions due to heat for missions that land on the surface. Duna might have dust-storms that pop up in one biome or another every so often that block communications or reduce efficiency of solar cells or rover wheels, but not so bad as to tear landers and rovers apart. Again, events that can be anticipated.

As for system malfunctions themselves, if an event triggers a malfunction (e.g. a probe gets hit by a major flare), I'd probably have the malfunction be a one-off thing, with permanent damage limited only to non-critical hardware. So you might lose a temperature sensor or a comm antenna, but not your lander motor or RCS system. Some permanent damage could be repairable by kerbonauts - like rover wheels are now. One-off malfunctions would mean the part in question would not work, but if you tried to use it again, the malfunction would be gone. E.g. A comm antenna gets zapped by a flare but not permanent knocked out. Attempting to transmit with it once means the transmission is lost, but use the antenna gain and it works as normal (for those who would need an in-game reason, we can imagine the ground controllers finding a workaround to get that antenna working again, as has happened with so many missions in reality). Of course, having redundant systems - like extra antennas and sensors - would also offset the permanent malfunction issue. Some measure of prevention can also help: e.g. if your Duna lander is about to get hit by a dust storm, you could fold up solar arrays to protect them.

In the case of malfunctions affecting probe cores and command pods, we can imagine different cores having different levels of robustness. A Stayputnik core for example might be cheap but vulnerable to flares, but an OKTO or HEX core could be hard as nails, letting you laugh at major flares even if you have the bad luck of having two or three major flares in a row.

Lastly, improving prediction of random events can also be a feature: space and ground weather can be predicted by KSC, but putting a weather satellite (with say a special weather camera part) in orbit around Kerbin or Duna can improve prediction of events there, while a satellite in Kerbol orbit with, say, a magnetometer part or a solar telescope part can help improve flare prediction, so that forecasts can go from just "what's happening today" to "what will happen 5 days from now". These, of course, can also deliver science points.

Just my own thoughts on the matter.

This is exactly what I want in KSP.

[Sirrobert]

Then make sure your payload is properly shielded.

Some of you are just afraid of being forced to think, I think.

Why is it that some people asume you are afraid of a change, just because you don't want it?

Maybe, I don't know, I actually don't LIKE to be forced

I think that, with regards to random events such as space weather or breakdowns, some means of anticipating them would help balance the fun vs. frustration factor. I think we can live with randomness if we can see it coming and at least take appropriate action. Likewise, we'd need to be sure that such random events are not "one-hit kills" for spacecraft. Random events should also be rare - like, happening only once every 1,000 game-days or so on.

I have a better idea: How about we 'balance' the fun/frustration factor to 1. NO frustration Why the blody diarrhea would someone want to intruduce MORE furstration?

While ground weather would be too complex to model, perhaps ground conditions that change occasionally might help - with a weather report that can show what said conditions will be in a region. Again, for example, we can apply this to launching a mission from the KSC. When you go to launch a spacecraft from the launchpad or runway, you get, not just a list of ships and kerbonauts to fly them, but a "five day forecast" with some events (thunderstorms, tropical storms, hurricanes) having a chance of launch failure. You might still lose a ship, but only if you don't heed the forecast. Knowing it's coming in advance would help to make sure players don't toss their computers through a window if their 200 ton Duna mission is torn apart by Tropical Storm Jeb on ascent.

"O sweet, a launch window to Duna is coming up, let's prepare.

Sweet almost done. Now just connect the engine module, and we can begin our journey. O look, a tropical storm. Gues I can launch my egines in a few days than. SO MUCH FUN, TIMEWARPING. Wait what do you mean the launch window has passed? Now I have to wait another 200 days to get to Duna"

Sounds like fun...

In the case of malfunctions affecting probe cores and command pods, we can imagine different cores having different levels of robustness. A Stayputnik core for example might be cheap but vulnerable to flares, but an OKTO or HEX core could be hard as nails, letting you laugh at major flares even if you have the bad luck of having two or three major flares in a row.

Great. So at the start of my career, I might aswel NOT send out probes, as they'll get completly dissabled at the very first random event. And when I finally DO unlock the shielded ones, the RNG might still deceide to screw me over and turn my extremely expencive (remember, we'll have money at this point) science probe into a very big expencive piece of debris. SUCH FUN

Lastly, improving prediction of random events can also be a feature: space and ground weather can be predicted by KSC, but putting a weather satellite (with say a special weather camera part) in orbit around Kerbin or Duna can improve prediction of events there, while a satellite in Kerbol orbit with, say, a magnetometer part or a solar telescope part can help improve flare prediction, so that forecasts can go from just "what's happening today" to "what will happen 5 days from now". These, of course, can also deliver science points.

Great, I'm already looking forward to reading weather reports in my game. Cause there's nothing more fun than watching the weather channel right?

[Xeldrak]

My luck, rocket building skills and flying skills usually collaborate to provide me with a whole buch of random failures I have to overcome while doing a mission. Like "Time to dock to my Duna lander back to it's interplanetary stage"...."what do you mean, they forgot to install the RCS-System?!"

[Galileo Kerbonaut]

If random events means things blow up without any reason, no thanks. I've tried the April Fool joke and it quickly became frustrating. On the other hand, if it means certain functions would break occasionally (antennas, lights, solar panels etc.) and could be repaired on site, that would be very cool and adds another depth-layer to the gameplay.

[badpandabear]

If random events means things blow up without any reason, no thanks. I've tried the April Fool joke and it quickly became frustrating. On the other hand, if it means certain functions would break occasionally (antennas, lights, solar panels etc.) and could be repaired on site, that would be very cool and adds another depth-layer to the gameplay.

This is similar to a mod I was thinking of writing. Basically there would be three types of failures. Minor failures might include lights getting stuck on/off or a probe going into safe mode. These could be fixed via a menu click. Major errors would be less frequent but more dangerous. Like a fuel tank leaking, a solar panel producing less energy, or a command pod losing torque. These could be fixed by a Kerbal on EVA but could be serious if they happened during a critical phase of the mission. Catastrophic failures would generally involve explosions or things not working at all. These would be irreparable and the only way to deal with them would be redundancy (like extra parachutes) or rescue missions.

What I wasn't sure about was how to handle the probabilities of a part failing. A simple way would be to assign a probability/unit time for each severity of failure, compute when the next failure will happen, and then when that time comes around randomly pick a part to fail. These probabilities would decrease when certain technologies are unlocked simulating the Kerbals getting better at testing and manufacturing.

A more complex way would be to assign a probability to each part. A newly unlocked part would be more likely to fail as the Kerbals would have less experience with it. (I think I got this idea from a recent Scott Manley video). As above certain technologies would cause a overall lowering of failure rates as the space program matures. This would tend to discourage more complex shops with new parts early in the game, which might be more realistic but could be annoying.

[Sparker]

I couldn't manage to read the whole thread, sorry. I've got an idea to share with you, guys.

For career mode, what if all parts had such parameter as reliability? If you have just develeped, for example, an engine, It's reliability is low and it can easily fail in different cases. But the more you use the new part the more reliable it becomes. If you log the data of the failure and send it back, then you will help the engineers to find the trouble with the engine and increase it's reliability. If you manage to return the broken part home, it will add even more reliability to it than if you simply send the data.

It would be reallistic as in real life there were very many failures with new spacecrafts and probes and it took some time to make them reliable.

What do you think about it?

[Luckless]

I would have no issues with an RPG expansion to go along with career mode. Optional features that you can configure on and off.

Give the kerbals skills, and have components have various subsystems. More highly trained and skilled pilots have a better chance of being able to fix issues, or spot them before they happen.

Customize your systems based on various choices with pros and cons. Cover every surface in a ship with loop fabric for hook and loop fastening, and you reduce the chances of a loose object causing a problem (Pilot lets his mug drift into contact with the 'vent waste' button and causes the craft to spin unexpected.), but now you've potentially raised cabin fire hazard.

Maybe your RCS glitches out, you have roll and pitch, but no yaw control. Missions can still carry on, but you must roll the craft to switch between yaw and pitch.

There are lots of options that can be interesting if the system is well designed along side a far more complex mission program.

Random chance being ship goes boom with no warning or reason is bad. Random chance of something interesting and different happening when you have possible options and force you to make unexpected choices? That sounds fun.

[KerbMav]

The only random events I would want to see are shooting stars, harmless solar flares, meteorite impacts on the moon a good way from anything, ...

[AndrewBCrisp]

I have a better idea: How about we 'balance' the fun/frustration factor to 1. NO frustration Why the blody diarrhea would someone want to intruduce MORE furstration?

Are you saying all of your launches have been perfect from day 1? If so you have my envy.

Frustration is part of the game whether we like it or not - it's part of anything where learning is involved. Whether or not random events are introduced there will still be frustrating circumstances. I'm trying to envision a way where we can make the random events challenging but not crippling. Think "speed bump" - not "brick wall".

"O sweet, a launch window to Duna is coming up, let's prepare.

Sweet almost done. Now just connect the engine module, and we can begin our journey. O look, a tropical storm. Gues I can launch my egines in a few days than. SO MUCH FUN, TIMEWARPING. Wait what do you mean the launch window has passed? Now I have to wait another 200 days to get to Duna"

Sounds like fun...

Hence the reason I mentioned the ability to forecast - so that one is not caught unawares. And really, using your example above, nothing is stopping you from launching the engine module (or other modules) EARLY. Plus, I have yet to meet a launch window that was so tight that one couldn't still make the transfer a few hours ahead or behind the optimal moment.

Besides, the way I envision it, a launch-preventing storm would be a truly rare event - something you might get once every thousand days or so. Unless one wants to suggest that random number generators are sentient and malevolent, the odds of your fun-killing scenario happening would be infinitesimal.

Great. So at the start of my career, I might aswel NOT send out probes, as they'll get completly dissabled at the very first random event. And when I finally DO unlock the shielded ones, the RNG might still deceide to screw me over and turn my extremely expencive (remember, we'll have money at this point) science probe into a very big expencive piece of debris. SUCH FUN

Please re-read what I said about how system malfunctions might work. Summing up: mission critical hardware may get hit by one-off malfunctions, requiring a part restart/reset (think of how we can reset Goo canisters now as an example) or (if kerbals are present) maybe a repair. Only non-critical hardware runs the risk of permanent malfunctions. Again, I am imagining this as a speed bump to a mission, not a brick wall.

Again, random events would be rare (something I mentioned in my previous post, I believe) - so please stop thinking of this like some malevolent Q-entity sitting gleefully in orbit, ready to snap his fingers and turn your Kerbals into gerbils or your RCS monopropellant into butterscotch pudding. Besides, if we are going to have solar weather, we will probably also have planetary magnetic fields. There will still be safe places and uses for Stayputniks even if the RNG truly hates you and there are solar flares every game-week. I would imagine the Mk 1 command pod would be more robust than the Stayputnik, and the Mk1-2 more robust than the Mk 1.

Great, I'm already looking forward to reading weather reports in my game. Cause there's nothing more fun than watching the weather channel right?

It can be argued that learning the Tsiolkovsky Rocket Equation is Not Fun, or calculating thrust-to-weight ratios, or finding transfer windows to the various planets. We've certainly had it grilled into us by a moribund school system that Math and Science Are Boring. As a result, some of us don't bother. Others use calculators to do the work for them (I use a spreadsheet, myself). Still others do find it fun and I'm sure there may be one or two people out there that can do it all in their head in an instant. We all approach this game in different ways, because we all have different approaches to what is fun.

Taking a few seconds to read a game forecast may not be Fun, but it would be less Fun to be blindsided by an event you could have known was coming and could have prepared for. I think that, if this system or something like it is implemented as a mod, then having the ability to anticipate events is useful. And anything that's useful in a game helps to preserve the fun factor, I've found.

Whatever mod finally emerges from this may not look anything like what I've suggested. At this stage, I'm trying to suggest possible ways that such a system can be implemented so it adds a challenge, rather than makes players want to flip tables or burn Squad staffers in effigy or something. You're welcome to disagree, but I would appreciate it if you didn't just issue a blanket condemnation - that's never fun for anyone. Thank you.

[NathanKell]

Some people seem to be under the misapprehension that this is the "you will be forced to use it" forum rather than the Addon forum. If you're not interested in a mod like this, fine, we get it, more power to ya. You don't have to keep telling us.

[Holo]

I love the idea of this, since unlike many people I like encountering problems I have to think creatively to solve. The most fun I've ever had in Kerbal Space Program was when a mission went wrong and I didn't predict it, and recovering from it successfully.

[Xat]

Would be nice to have some randomness in KSP as long as it wouldn't destroy whole flight without any way to fix it.

[kiwiak]

What about scenarius with random failures?

You fly to the moon, and you know that your angine might fail in one point.

[peadar1987]

I like the idea of random weather, so long as it's not mission-ending (like Tropical Storm Jeb). Launching through clouds, flying a plane on a windy day, landing on Duna in a dust storm, getting tossed about by strong winds as your probe drops into Jool's atmosphere... I think all of those would be really fun to play (and of course, you'd have the option of simply turning weather off altogether)

I've gone off the idea of random component failures though, think they'd just be too frustrating if they were common enough to affect gameplay at all.

[Bacterius]

How about we make it, you know, optional. So people who don't like it, can turn that feature off, and those who like it, can turn it on. There, how difficult was that?