More Variable Weather??!!

[Stiebler]

I was unnecessarily gloomy about progress in a previous PM to Subtype Zero, hence his pessimism above. My apologies for that.

Actually, we have made good progress, and we hope to have a new mod 'RealWeatherFix' available after further testing.

Stiebler.

[Blacklight]

Nothing new to report yet but with further testing, I've noticed that leaving from Wilhelmshaven, about 90% of the time, when I cross up into the grids (longitude) of where Scapa Flow is, the winds almost always change directly to 15 meters per seccond and all hell breaks loose (After reading the above post about the windspeed increment multipliers, it's possible that I always reach these coordinates just as the change kicks in. To test this, I'll change the windinterval or whaterver that number is a little and see what happens). I think I've narrrowed down somewhat the location of one of the weather change latitudes. What I'm going to try to do is try to drive above and below it to see if there's a line that produces a change.

It could be that the weather is set to be the same along groups of grids also. Something to consider.

This is going to be tough to really test well.

[snwcrsh]

I can confirm this observations. Left two patrols from Wilhelmshaven with calm winds (4 or 5 knots). Changed to 15 after Scapa and stayed this way for two weeks all over the atlantic (no change whatsoever)

[WolfOfCampscapel]

Gnørf. Especially annoying, seeing as 15 m/s is no big deal in the north sea. 20, or 25 (steady for an hour or two) as we had yesterday on the coast, is more like it.

It can stay at 15 m/s for days at a time. That's commonplace. 30+ steady is exceptional. 20+ is uncommon, but seen several times every autumn/winter.

... west coast of norway.

[Blacklight]

I think something may be broken with the randomised die roll for the weather. I've noticed also that the weather tends to either be 15 meters per seccond or near dead calm up to 5 meters per seccond. It's rarely in between that.

Either that or I have another theory. Perhaps it's possible that each zone has it's own set of chances or modifiers to the die roll for different kind of weather that play into the factor. Up by Scapa Flow may have a high modifier to the die roll to make it tend more toward stormy weather whereas down south near Gibraltar, It seems to be about even between calm lightly stormy.

I can definitely say that the zone around Scapa seems to produce the most storms in my experiences.

[Dekessey]

Originally Posted by Stiebler

I was unnecessarily gloomy about progress in a previous PM to Subtype Zero, hence his pessimism above. My apologies for that.

Actually, we have made good progress, and we hope to have a new mod 'RealWeatherFix' available after further testing.

Stiebler.

Excellent news!!

[GlobalExplorer]

Originally Posted by Stiebler

I was unnecessarily gloomy about progress in a previous PM to Subtype Zero, hence his pessimism above. My apologies for that.

Actually, we have made good progress, and we hope to have a new mod 'RealWeatherFix' available after further testing.

Stiebler.

Excellent stuff

[Yoriyn]

I make some tests on the WeatherRndInterval variable. A use different numbers each time when i start new career (12 tests). I start each campaign 1 september 1939 in first flotilla in Kiel.
Everytime first weather changes have a place after 32h, second after another 33h, third 34h, fourth 35h...

Test raport:

I random save the game during tests.
Saving during patrol do not affect to weather changes, but after save and quit the game, when you load the patrol weather timer is reset (you must wait 32h before weather change)
example:
last weather change have a place 2:00am 5/09/1939 so another shut be (32h) 10:00am 6/09/1939. But when you save and exit, lets say 18:00 5/09/1939, weather changes will be 2:00am 7/09/1939.

WeatherRndInterval variable, I use numbers from -11 to 96.
Weather allways change after 32h, it's not dependant on the WeatherRndInterval value.
WeatherRndInterval < 0 weather do not change for lots of days (stay the same), only wind changes observed. For this value weather was allways rainy and foggy.
0<WeatherRndInterval<12 weather changes are most realistic (no hard weather changes)
WeatherRndInterval>12 drastic weather change everytime (from big storm to wind 0m/s, and again big storm).

I suppouse after tests, the WeatherRndInterval variable respond for:
"How many weather checks the weather will be the same. The biger value the less chance the weather stay the same in the next weather check (32h)."
And "Biger value of the WeatherRndInterval, the weather change more drastic"

[irish1958]

Yoriyn,
Great testing and a lot of work.
I really think the only way to control the weather is by cracking the saved game file and writing new values that make sense. The weather module in this game appears to be badly broken: storms lasting for weeks or months; wind going from 0m/s to 15m/s in an instant, and then staying that way for months; heavy fog with 15m/s winds; no fog when calm; pretty much the same weather in the arctic and in lower latitudes; no snow in the winter in the arctic; etc. etc. etc.

[difool2]

Dunno why you always got a change after 32 hours, because as I said above running RWI at 1 I typically get changes every 16-21 hours, including from a reload. This is departing from Kiel in the 7th/13th flotilla 1940.

I feel like we are a bunch of blind men groping an elephant trying to figure out exactly what the hey it is, and everyone seems to have a different perspective, whether because of some more subtle factor which has gone unnoticed, or whatever.

[Hanomag]

And here I thought after making the original changes I was done with the SH3 weather calamity.

I thought there was a change and the weather was not stuck on one pattern too long. But after catching up on this thread I cant possibly be right.

So keep up the good work!

Im sure eventually you guys working on this will figure it out....

[Subtype Zero]

Yoriyn--Great information! I almost always get my first weather change after 32h out of Kiel.

Saving during patrol do not affect to weather changes, but after save and quit the game, when you load the patrol weather timer is reset (you must wait 32h before weather change) example: last weather change have a place 2:00am 5/09/1939 so another shut be (32h) 10:00am 6/09/1939. But when you save and exit, lets say 18:00 5/09/1939, weather changes will be 2:00am 7/09/1939.

Good catch! I knew the counter reset after some kind of save operation, good to know it is only after a save AND exit that the reset occurs. I have also noted that WeatherRndInterval has no effect on the frequency of the weather change.

WeatherRndInterval < 0 weather do not change for lots of days (stay the same), only wind changes observed. For this value weather was allways rainy and foggy.
0<WeatherRndInterval<12 weather changes are most realistic (no hard weather changes)
WeatherRndInterval>12 drastic weather change everytime (from big storm to wind 0m/s, and again big storm).

Also very interesting! Could you be more specific about the changes you observed with WeatherRndInterval > 12? Were there always dramatic shifts in the weather? What was the effect of really high numbers, like 96? Finally, what were your other settings in the campaign.mis files? Stiebler and I are very curious about the effect of the campaign.mis files on the scene.dat file (which is getting closer to release, by the way)

[Turm]

Just to throw more confusion into the mix, I have seen the weather change within an hour (game time) of loading a saved game.

In fact, I was so amazed that the weather changed after just an hour (especially after reading previously on the forum about the timer being reset) that I quit to the menu and loaded it again, and lo and behold the weather indeed changed again after an hour.

So, after reading this thread it makes me think it's one of these two things:
1) The weather was 'in the process' of changing when I saved that game, and was simply finishing it's change (because sometimes the wind can change several times in a short period of time, then no changes will happen for days),
2) Crossing some latitude or longitude possibly triggered it,
3) Or it was a freak accident and there was something odd about my save game.

It was in stock SH3 when it happened. Very strange, but backs up the reasoning that the WeatherRndInterval parameter has no effect on the frequency, but rather the amount of change that's possible each time the weather changes.

Watching this thread closely, let's hope for some more promising info

[difool2]

All I know is, using the settings that I have now, that 3 week long storms are a thing of the past, and that's all I ever really wanted. I get changes once every day or two or three. If I knew why I get a 16 hour window while everyone else gets 32 I'd tell ya.

[irish1958]

And your current settings are?: