Weather?

[Tristen Simon]

Shouldn't you think that KSP2 should have weather? It would put an interesting twist on the game, but only would be available in the Normal and Advanced Difficulty. (Of course, clear weather is possible.) In Normal, you could get low winds, medium winds, rain, partially cloudy skies, and snow. In Advanced, you would get low winds, medium winds, high winds, rain, rainstorms, thunderstorms, partially cloudy, cloudy, heavily cloudy, snow, and hail. I think weather would be a good addition to the game.

[Bej Kerman]

If I recall correctly, don't quote me on this; take this with a grain of salt, KSP 2 suggestions and stuff don't go in the pile with KSP 1 stuff. Just don't quote me on this, might be wrong.

[Angelo Kerman]

Moved to the KSP 2 discussion forum.

[K^2]

Yeah, I would very much like to have simulated weather. Doesn't have to be anything terribly complex, but it would add a lot of variety both to Kerbin and to some of the other worlds. Looking at what they're doing with atmospherics, it looks like they're planning to have stock clouds, so this already gets you part of the way to having weather.

I also think that physics effects on craft should be a difficulty toggle. I don't know if making severity of visual weather a difficulty setting is worth the hassle. All-or-nothing is definitely an option, and things like fog and other low-visibility conditions might be easy enough to turn on and off, but for weather overall, I'd prefer a more global approach, where a very rudimentary 2D fluid sim runs across the planet's surface to determine the weather. You can make that work with a mesh of just a few hundred points and it will give you realistic cyclone systems, predominant wind patterns, etc. I think that'd be worth the effort, but the downside is that if the sim says you're having a thunderstorm, you're having a thunderstorm. But so long as you can toggle severity of the wind effect on your craft and how bad visibility gets, I don't think that's a big deal.

[Jack Mcslay]

I think you're overestimating how hard it would be to implement and how much it would actually add to the challenge. This isn't minecraft, you can't just  if (random()>x) rain(); doing weather patterns on a game in planetary scale needs atmospheric modeling in planetary scale, currently only flight sims on fairly big budget simulate weather patterns and I don't think they have actual atmospheric models but just create storms at location X.

And I say that it doesn't add that much to the challenge because unless the players want to fly through storms on purpose they will just do what they do in real life: postpone the launch until weather conditions are favorable. In the end it's too much effort in a game where you don't really spend all that much time flying in the atmosphere.

I wouldn't really expect anything beyond aesthetic rain, maybe clouds blocking sunlight to solar panels.

[K^2]

5 minutes ago, Jack Mcslay said:

I think you're overestimating how hard it would be to implement and how much it would actually add to the challenge.

It's a very simple fluid sim. So long as you aren't going for meteorological precision of predicting weather a week ahead, you don't need to run it with particularly high resolution or too many parameters. An isosphere with a few hundred points over the planet is adequate for such simulation, and you just need to track pressure, temperature, and humidity for each. Final step is faking vorticity by simply having wind blow along isobars at speed proportional to the pressure gradient, and you generate rain wherever relative humidity hits 100%. There are ways to improve on that by tracking these parameters at several altitudes. That would let you detect unstable atmosphere - which is a more reliable indicator of a thunderstorm, and identify areas that should contain fog. But it's not strictly necessary for basic sim. I'd see how it works with pure 2D first, and then go to 2.5D model if necessary.

[Guest]

Last time this was discussed I was against it because it would just be a waiting simulator for Kerbin and a random failure mode for other atmospheric planets/moons.

That pretty much remains my opinion, but I see a window of opportunity to have it as a meaningful and engaging gameplay loop, not as a system on its own, but as a consequence/reason to launch and do weather analysis with satellites and probes.

[Guest]

I don't think it would be particularly hard. I do think it would be pointless unless integrated with other gameplay systems to create possibilities for interesting, emergent gameplay.  Cosmetic weather would be nice though.

[Incarnation of Chaos]

My opinion, probably not stock. Could it be? Probably, but given the current state of KSP2 i'd doubt their biggest issue on the tracker is "Implement wind".

Could be a potential DLC addition though, and i don't think most would mind if it was made configurable. And it could be modded in at some point, though that might be slow going considering such a mod doesn't really exist for KSP right now (Only ones are well EOL or All Rights Reserved or Both sadly).

[Vanamonde]

The subject of this thread is the possibility of weather effects in KSP2. A number of posts about development delays have been removed, as they are off-topic. 

[K^2]

On 12/27/2020 at 12:51 AM, Master39 said:

Last time this was discussed I was against it because it would just be a waiting simulator for Kerbin and a random failure mode for other atmospheric planets/moons.

I can definitely see how it can be, but I can make the same broad argument about having atmosphere on other bodies in general, like Eve. Anyone who did the Eve mission going in cold probably ended up with some stranded Kerbals at best. Likewise, one could say that inclined orbits are just a waiting simulator and/or random failure mode if you arrive at a planetary system and then realize that inclinations of the moons are all wrong relative to your arrival trajectory.

I think the trick is balancing it out so that you have options of planning around weather or pushing through it by being over-prepared. If inclement weather is punishing but not catastrophically so on its own, it could be a great gameplay element. Do you take the risk landing on Duna as soon as you can? Or do you drop some probes first to generate a forecast? Maybe if you have a backup for solar power, some extra fuel to land against the wind, and a more rugged lander, then you don't need to wait the sand storm out, and you can just go for the landing. And I don't think the weather on Kerbin should ever get so bad that you can't launch. But maybe in a strong wind you aren't going to get that nice 0° inclination orbit just by holding the rocket steady at 90° heading during gravity turn. So do you wait for better weather or do you eat the time and fuel cost of fixing your orbit once you clear the atmosphere? The biggest impact could be visibility for airplanes, but even there, simply adding a glide slope indicator to the navball when you're near a runway would basically let you do everything IFR. Given how instruments-heavy orbital transfers and even docking in the game are, I don't think having players learn to fly instruments if they want ability to fly planes in any weather is a huge ask. Likewise, learning a crawl landing is one more skill to tackle crosswinds, but it's not actually more advanced than anything else you have to do in the game. And nearly all of it is emergent gameplay you get simply for having weather.

I think there's a way to do it right, where it's purely an addition to the game. But your concerns are definitely valid as far as it being very easy to screw up and make weather into something that's just an annoyance. We've seen Squad bounce a bit back and forward between a more relaxed game and more punishing sim, especially as far as interaction with atmosphere goes. Even parachutes started out as one-button problem solver, then became a huge headache, and now we have them more or less balanced where you have to put some thought into it, but aren't likely to crash just because your re-entry wasn't quite perfect. It sounds like Intercept is taking this as a lesson and are going to work on balancing all these aspects out, and I think if they were to implement the weather, they'd be able to tune the severity of it to be a kind of challenge that will reward you for planning around or powering through as fits your play style or choice for particular mission.

[Jack Mcslay]

13 hours ago, K^2 said:

Likewise, one could say that inclined orbits are just a waiting simulator and/or random failure mode if you arrive at a planetary system and then realize that inclinations of the moons are all wrong relative to your arrival trajectory.

Glitches aside, that's entire the player's fault for not planning orbits correctly. Even some form of weather forecast isn't going to do much help if you're encountering Duna straight into an aerobraking trajectory

[K^2]

9 hours ago, Jack Mcslay said:

Glitches aside, that's entire the player's fault for not planning orbits correctly. Even some form of weather forecast isn't going to do much help if you're encountering Duna straight into an aerobraking trajectory

Say you're planning a direct mission to Bop. Your most fuel-efficient trajectory involves aerobraking at Laythe. You can hit Jool SoI at pretty much any angle you want for free, so your encounter with Laythe is a guarantee, but your final inclination in Joolian orbit is going to be a function of when you arrive. So you have to time that if you want to end up in a plane similar to that of Bop. And that's on top of timing the Kerbin-Jool transfer, because who wants to make an inclination change in interplanetary?

So your options for a Kerbin-Bop are either to time positions of Kerbin, Jool, Laythe, and Bop all to be perfect with respect to each other for your specific flight plan, so you need ability to even see what Joolian system looks like at your arrival. You can bring a bit of extra fuel, timing just Kerbin-Jool window, and then hope your final orbit around Jool is good enough that you can fix with your reserves, being fully prepared to scrub your mission or strand some kerbals if you end up in particularly bad plane. Or you can build a much bigger rocket and bring a lot of extra fuel to brute-force it.

You either prepare and wait for the good window, you cross your fingers and accept random chance of failure, or you overprepare for any contingency. So how exactly is this different from having to either plan around, rolling dice, or preparing for anything in terms of weather?