Issues with weather in Command

[Mgellis]

Two issues...

First, this is minor, but it has always bothered me that the temperature range for Command is so extreme. When I look up weather for various parts of the world when I'm designing a scenario, few places regularly have a 20-degree C (36-degree F) change between midnight and noon. A more realistic temperature range would be +5 C for day, baseline for dawn/dusk, -5 C for night.

Better yet, why not simply allow people to set the high and low temperatures for the scenario? (Dusk/dawn would be the average temperature between the high and low temperatures.)

Please consider these changes.

Second, is there an explanation anywhere of what the different types of cloud cover mean? I get clear and solid, but what do light and moderate mean? 25% and 50%? Some other values?

Thanks.

[thewood1]

The Boston area yesterday 5 C at 5AM and 28 C at 3PM. Most of the NE US has similar temp swings in Spring and Fall. Even in winter and summer a 15 C swing is common.

[Mgellis]

I'm not saying it doesn't happen (I used to live in South Dakota, so I know some places do have extreme temperature changes), but it's not what I normally find when I research temperatures, weather, etc. Anyway, I wish there was a way to do something about it. It's a very minor issue, but it annoys me sometimes.

(I wonder if there is a way to do this with lua...change the temperature range somehow? If it was just a matter of adding one Trigger: Scenario starts --> Action: lua command to change weather range Event, that would be a bit clunky, but it would be all right.)

The Boston area yesterday 5 C at 5AM and 28 C at 3PM. Most of the NE US has similar temp swings in Spring and Fall. Even in winter and summer a 15 C swing is common.

[thewood1]

Does it have impact on the game itself?

[SunlitZelkova]

From the manual:

Temperature affects several sensors but primarily infrared ones. During the night, man-made and mechanical objects stand out more clearly against the background on infrared sensors. Conversely, during daytime (and particularly in hot climates) IR sensor range shrinks dramatically. Imaging sensors are less severely affected than older ones. Temperature differs from the “average” setting depending on region of the world-the equator is a lot warmer than the Arctic, obviously

And Mark, I found that cloud cover is also detailed in the manual too.

Cloud coverage affects line-of-sight for visual and IR sensors. Light cloud coverage usually means scattered clouds at high altitude; aircraft can get under the cloud base and still visually acquire targets below, illuminate for LGBs etc. Thick cloud coverage means thick clouds at all altitudes; LGB guidance may be difficult or impossible unless the aircraft descends to low altitude, and spotting is likewise degraded.

I myself did not know this either, I've always assumed light clouds were equivalent to what is referred to as "partial cloudy" in weather reporting. I find it fun how this game is always teaching me new things, whether about itself or different weapon systems.

I think more detailed weather settings would be interesting, especially because the "weather modeling levels" is still present in the editor UI (but inactive, with nothing more than level 0 being available). Of course, I don't know what kind of priority they should take over other new features.

[Mgellis]

I tend to treat "light" cloud cover as "scattered" and "moderate" as "broken," but I admit I don't really know exactly what the terms mean or how they affect sensors, etc. Mostly, I'm just trying to get the conditions in the game reasonably close to what things are like in terms of real world weather. As an example, if I look up the weather in a location and it says "scattered clouds," I use a "light" cloud setting in the scenario editor. I think I'm pretty close here, but I don't really know.

The temperatures aren't really important, but every once in a while I look at a temperature in a game and I'll think, "Wow, that's really hot for that time of year," or something like that. I don't think it affects gameplay (at least, not right now--it will be interesting to see how weather is developed in Command as the years go on), but it's one of those little annoying things I wish there was a way to fix. Mostly, I'm just griping.

Cloud coverage affects line-of-sight for visual and IR sensors. Light cloud coverage usually means scattered clouds at high altitude; aircraft can get under the cloud base and still visually acquire targets below, illuminate for LGBs etc. Thick cloud coverage means thick clouds at all altitudes; LGB guidance may be difficult or impossible unless the aircraft descends to low altitude, and spotting is likewise degraded.

I myself did not know this either, I've always assumed light clouds were equivalent to what is referred to as "partial cloudy" in weather reporting. I find it fun how this game is always teaching me new things, whether about itself or different weapon systems.

I think more detailed weather settings would be interesting, especially because the "weather modeling levels" is still present in the editor UI (but inactive, with nothing more than level 0 being available). Of course, I don't know what kind of priority they should take over other new features.

[thewood1]

From the manual:

Temperature affects several sensors but primarily infrared ones. During the night, man-made and mechanical objects stand out more clearly against the background on infrared sensors. Conversely, during daytime (and particularly in hot climates) IR sensor range shrinks dramatically. Imaging sensors are less severely affected than older ones. Temperature differs from the “average” setting depending on region of the world-the equator is a lot warmer than the Arctic, obviously

And Mark, I found that cloud cover is also detailed in the manual too.

Cloud coverage affects line-of-sight for visual and IR sensors. Light cloud coverage usually means scattered clouds at high altitude; aircraft can get under the cloud base and still visually acquire targets below, illuminate for LGBs etc. Thick cloud coverage means thick clouds at all altitudes; LGB guidance may be difficult or impossible unless the aircraft descends to low altitude, and spotting is likewise degraded.

I myself did not know this either, I've always assumed light clouds were equivalent to what is referred to as "partial cloudy" in weather reporting. I find it fun how this game is always teaching me new things, whether about itself or different weapon systems.

I think more detailed weather settings would be interesting, especially because the "weather modeling levels" is still present in the editor UI (but inactive, with nothing more than level 0 being available). Of course, I don't know what kind of priority they should take over other new features.

Yes, I knew that. I mean to the level of detail that a 20 deg swing has any effect in-game. Setting an exact temp would seem to have no impact. It looks to me like they are reading off average temps in a region. So that means if you want to see an impact from temp, it might be more than just setting the temp.

[SunlitZelkova]

Given its prominence in the manual, I would imagine there is a visible effect, but it would be extremely difficult to discern its exact impact in a given engagement due to the presence of multiple other factors (aircraft type, weapon type, pilot proficiency, likelihood of weapon failure, etc.). Temperature's influence on the engagement could be easily mistaken for another factor, and vice versa.

So perhaps it isn't that important after all.

The temperatures aren't really important, but every once in a while I look at a temperature in a game and I'll think, "Wow, that's really hot for that time of year," or something like that. I don't think it affects gameplay (at least, not right now--it will be interesting to see how weather is developed in Command as the years go on), but it's one of those little annoying things I wish there was a way to fix. Mostly, I'm just griping.

Going back to the question about modifying the temperature through Lua, wouldn't it be possible to do this by using a simple change weather event, but having it at the same time the temperature switches from day to night within the game? The automatic temperature shift would basically be preempted, so when it drops, the baseline setting is switched higher. And then when it becomes day again, the setting is shifted back down to avoid it climbing absurdly high.

[thewood1]

Yes, you can do it through zones.

[Nikel]

(I wonder if there is a way to do this with lua...change the temperature range somehow? If it was just a matter of adding one Trigger: Scenario starts --> Action: lua command to change weather range Event, that would be a bit clunky, but it would be all right.)

It is according to several AIs I tried, however each one creates a diferent code and when you run it in the console there is always an error.

When you ask for the error in the code the same AI created, it generates a more complicated code.

The new code displays an error when you run it in the console.

I bet you may continue ad infinitum


Conclusion, we must learn lua. But I do not think I can, could not learn BASIC when I was young...

[thewood1]

The attempts I have seen to use generative AI with CMO lua have all appeared to create more work than they save.

[Uzabit]

I’ve found that it can work quite well as long as you include excerpts from the Command Lua reference and don’t just ask for code out of the blue. However, having some coding experience is crucial for spotting bugs in the generated code. LLMs aren’t magic, all-knowing code generators - it often takes some back-and-forth before you end up with code that actually works and does exactly what you want.

By the way, I don’t think you can adjust the temperature range via Lua - at least I couldn’t find anything about that in the reference, and it seems to be hard-coded.

[Nikel]

Yes, I suppose If you know lua it will save you a lot of time typing the code, and you "just" have to detect and fix the error(s).


This is the code created by ChatGPT Open AI. The first try that conduced to a endless rabbit hole

Code: Select all

-- Daily temperature simulation script
local baseTemp = 15        -- Average temperature (°C)
local amplitude = 5        -- Half of the total daily range
local peakHour = 15        -- Hour of peak temperature (3 PM)

-- Get current scenario time
local scenTime = ScenEdit_CurrentTime()
local hour = tonumber(os.date("!%H", scenTime))

-- Calculate sinusoidal temperature
local pi = math.pi
local temp = baseTemp + amplitude * math.sin((2 * pi / 24) * (hour - peakHour))

-- Get current weather
local currentWeather = ScenEdit_GetWeather()

-- Set new temperature while keeping other weather values
ScenEdit_SetWeather({
    temperature = temp,
    cloud = currentWeather.cloud,
    rain = currentWeather.rain,
    fog = currentWeather.fog,
    windDir = currentWeather.windDir,
    windSpeed = currentWeather.windSpeed,
    seaState = currentWeather.seaState
})

-- Optional: Log the update
print(string.format("Hour: %02d | Temp set to: %.2f°C", hour, temp))

[Uzabit]

Yeah - from the script you posted, I can already spot some hallucinations:

Weather Lua.png (20.19 KiB) Viewed 226 times

Weather Lua 2.png (33.59 KiB) Viewed 226 times

Your script tries to set the following parameters, which don't exist:

ScenEdit_SetWeather({
temperature = temp,
cloud = currentWeather.cloud,
rain = currentWeather.rain,
fog = currentWeather.fog,
windDir = currentWeather.windDir,
windSpeed = currentWeather.windSpeed,
seaState = currentWeather.seaState

[Nikel]

Yes, I deleted those additions, and still it will not work.

But perhaps, did not do it correctly

Not mine, it is ChatGPT OpenAI.

[Uzabit]

Have you checked the fields you're trying to retrieve with ScenEdit_GetWeather ()?

I would have to debug the code line by line...

[Nikel]

This is the error, with or without the 3 extra lines.

ERROR: [string "Console"]:18: Invalid arguments to method call

Line 18 is this

ScenEdit_SetWeather({

[Uzabit]

The idea is interesting, but this script seems to work against CMO’s built-in temperature handling rather than with it.

CMO already simulates day/night and latitude-based temperature changes on its own - I’m just not sure exactly how it calculates those from the base value. By manually setting the average temperature here, the script overrides that internal behavior, which could lead to unrealistic or conflicting results.

Another issue is that the script only runs once, so unless it’s triggered periodically (e.g., every 10–15 minutes via a repeatable event), the average temperature will stay fixed at the value from the initial run.

[Nikel]

Yes, the AI recommended running it every hour by event.

I created the trigger and the lua action for the event, but as the code is wrong, could not progress.

[Uzabit]

Now, as a thought experiment: The script would set the maximum average temperature during the day at the peak of the sine curve. Additionally, the internal CMO temperature handling would also set a higher temperature at the same time. The effects would therefore overlap. I'm not sure if this would be desirable behavior.