0.90 Kerbal Weather Systems! Alpha 0.5.3 WIP! (Jan 2)

[CommanderCoye]

Um... Is the mod still in development? Or will we never have... Rain...

[silverfox8124]

Yes, the mod is still in development. There's a lot of sorting out we have to do with numbers. I'm not going to ship out something that doesn't work, and I'm going to make sure it works at least 99% before shipping it. We still have a lot of testing to go through. Sorry to say but yeah, this will take some time.

[silverfox8124]

Dev Update!: Progress is coming along, we are tackling errors down one by one, tweaking the values that are fed to the sim and making sure it can handle variances in inputs as well.

[silverfox8124]

Dev Update: We're working on fixing stuff up, mainly correcting calculations and stufferies.

Fun Fact: If you were to have a cell resolution of ~1cm^3 it would result in 1.0995116277762E13 cells per layer, and would mean about 8,815,220 layers, which in total memory usage would add up to: ~46.525ZettaBytes. It would also take (at default 50 cells/tick, 50 ticks/s)

3.876974756562125506×10^16 seconds,

4.48723930157653415×10^11 days

1.4752567566826961589×10^10 months

1.2293806305689134658 billion years

12.285643925820610006 million centuries

1.2285643925820610006 million millennia

1.2293806305689134658 eons

0.08781290218349382 universe ages

 

Mind you the age of the universe is: 14 eons

So if you have more data than exists, a tenth of the lifespan of the universe, go HARDCORE and run grid level 20 at 0.01m layer resolution. HARDCORE USERS ONLY

[Buster Charlie]

14 minutes ago, silverfox8124 said:

Dev Update: We're working on fixing stuff up, mainly correcting calculations and stufferies.

Fun Fact: If you were to have a cell resolution of ~1cm^3 it would result in 1.0995116277762E13 cells per layer, and would mean about 8,815,220 layers, which in total memory usage would add up to: ~46.525ZettaBytes. It would also take (at default 50 cells/tick, 50 ticks/s)

3.876974756562125506×10^16 seconds,

4.48723930157653415×10^11 days

1.4752567566826961589×10^10 months

1.2293806305689134658 billion years

12.285643925820610006 million centuries

1.2285643925820610006 million millennia

1.2293806305689134658 eons

0.08781290218349382 universe ages

 

Mind you the age of the universe is: 14 eons

So if you have more data than exists, a tenth of the lifespan of the universe, go HARDCORE and run grid level 20 at 0.01m layer resolution. HARDCORE USERS ONLY

In event of heat death of the universe, how will I submit bug reports?

[silverfox8124]

1 minute ago, Buster Charlie said:

In event of heat death of the universe, how will I submit bug reports?

Ezpz just send all the relevant data to [email protected]

[silverfox8124]

Dev Update! Progress is being made, one by one bugs are falling and we're tackling the issues as they come, like not only ONE, but TWO methods of Squads which give erroneous values which I found out the hard way.... So yup, fun day today....

On another note, I may or may not in the future once the simulation is properly modelled, build a "standalone" version whereas it's networked attached and will do the processing and such for you, while you run KSP on another machine. Also ontop of this, I am personally developing a cluster computer, and may or may not implement a KWS version for cluster computers.

[wasml]

Appreciate the effort your taking to do this - eagerly (but patiently) awaiting the release of this mod. This will take a large step towards making the planets more interesting and not just a goal to get to.

[silverfox8124]

1 hour ago, wasml said:

Appreciate the effort your taking to do this - eagerly (but patiently) awaiting the release of this mod. This will take a large step towards making the planets more interesting and not just a goal to get to.

Oh trust me, I'm just as eager to get it out, but I can't put out something that doesn't work. And also note that when the sim does release, there's  a high chance there will only be weather for Kerbin, at least to begin with. We don't have the necessary elemental composition for Eve or Laythe or Jool  to make a reliable sim out of. When we do get that information, we will try and incorporate it in.

[silverfox8124]

Dev Update! We are making good progress, we are onto the winds recently, and we're tackling issues as they come, we had a sit down and had to bring together our views on how the simulation should work, so that there aren't any discrepencies and differences. We are still working those out, but things are coming along. Don't get your hopes up for a 1.0.5 release

[silverfox8124]

Dev Update! FOG! We have fog conditions happening in the sim, now all we need to do is handle it and patch the sim accordingly, however don't get your hopes up for a fog shader just yet. It'll just be a weather condition status for now.

[silverfox8124]

Dev Update! We are still working on the sim, don't worry. we've switched to 1.1 development officially and it's coming along. However, as many simulations turn out, it's making sure inputs are right and many things come together and that values are in range and equations are correct, and that's what we've been doing for the past 3 months. We are inching closer to a finished project, don't worry!

[quarrion]

Keep up the good work

[silverfox8124]

Dev Update! We're getting closer and closer, the simulation is growing more and more balanced every day. And I'm even designing it with laptops like my own in mind. And on a 2.4Ghz laptop, it runs smoothly, I can update the planet in about 4 seconds. So performance so far is nice. It'll happen eventually guys, thanks for being so patient <3

[Tortoise]

Keep up the great work!

[ThatOneBritishGuy...]

Don't give up! Sounds like development is coming along nicely. It does trouble me however that I have little visual stimulus so far. Do you happen to have any screenshots from the sim?

[silverfox8124]

I'm sorry to say, but there is no visual stimulus for the sim so far. we're mainly working out equations and code. which you can find under the current working name of "Simple Weather" on my github and you can see code happen. but thats about it until we get the sim all working proper.

[silverfox8124]

Dev Update! We're continuing to move along. Today we worked on some reflection stuff with sunlight and focused more on the cloud end of things. We will be testing some things soon to make sure we have certain equations all done proper. We added a new variable to the cloud struct which will give us more accuracy to the real world with having seperate values for the amount of solid and liquid contents in a cloud, on kerbin it would be ice and water. This control and flexibility will allow us to better manage what falls to the ground.

[DasBananenbrot]

The progress reports sound very promising. Keep up the good work

[diomedea]

Sure equations and code alone don't tell much. One thing with this mod, sure, is that proper handling of weather comes first, visual effects later. With code now running without major problems, is possible to experiment what happens after a consistent interval of in-game time. While each test run takes longer (we can't yet use timewarp in game, that feature will come later), unexpected behaviour is revealed and needs be corrected. Sometimes, it's just a single error in how an equation was entered in code, so takes very little to fix after the issue is discovered (oh, how many little bugs come each day). But the simulation increasingly shows to need more precision too, so more or better equations have to be used. Weather modelling is really a multi-disciplinary subject, and one never knows where research will land.

E.g., now the effort is with laws about how relative humidity really goes, as it keeps increasing more than expected. Till now, used those good equations every meteorologist knows about, about evaporation, saturation and condensation. But those aren't enough, so the effort goes about some way to include thermal velocity of water molecules and relative speed of diffusion of the different gases in the mix. Surprisingly, most good documents found on the subject were not from meteorology nor physics, but about insulation and building materials....

[silverfox8124]

Dev Update. For the past few days I've been working on some visual stuff to aid in debugging and for later use in release to you guys. Here's what I got so far:

 

Now what this doesn't show you is that the window is following your mouse position. So you can hover your mouse over a spot and get the info for that spot.

(Feature thanks to Kethane~)

[Pilotmaster]

not updated

[silverfox8124]

8 hours ago, Pilotmaster said:

not updated

Yeah, that's because I haven't updated the mod yet. That's what we're currently working on...

[ss8913]

On 5/12/2016 at 8:43 AM, silverfox8124 said:

Yeah, that's because I haven't updated the mod yet. That's what we're currently working on...

lol... "it's not updated because it's not updated" - it's both funny *and* correct

 

Question though, @silverfox8124, how does this interact with EVE clouds, and is it better/worse/no-different-than say, Kerbal Wind?  Do I need both or just wait for yours to update?

[silverfox8124]

4 minutes ago, ss8913 said:

lol... "it's not updated because it's not updated" - it's both funny *and* correct

 

Question though, @silverfox8124, how does this interact with EVE clouds, and is it better/worse/no-different-than say, Kerbal Wind?  Do I need both or just wait for yours to update?

At first release for 1.1 it wont have eve installed just yet. And kerbal wind is someone ripping off the very first version of KWS and running with it. Kerbal Wind wouldnt strictly conflict with KWS but it is not recommended to have with it. Kerbal Wind is just forces being applied how you want them, like a wind chamber, last i checked it didnt have a random component to it so it was just you telling wind which way to hit the craft. I suggest waiting on the update for KWS as it's getting closer and closer to release.