Realism- and gameplay-related hardcode fixes for SH3.EXE

[h.sie]

yes, weather is in EnvSim.act, at least some important routines.

with the patched EnvSim.act, I run Sh3 in TC=128 and watch the weather. looks good. every 2 hours a random weather change is triggered, but in some cases weather stays constant (that's just random). weather changes from bad to good to bad to good all combinations are there. rain, fog, storm, thunderstorm, sun.

I see no need for further modifications, except for fine-adjusting the weather change interval, which I chose to 2 hours for demonstration purposes.

only issue: I once saw rain without clouds. but that cannot be related to my fix!!(??)
maybe noone saw such a situation, because weather changed so seldom?

[irish1958]

I can't find the download link for this mod. When I pull up the mediafire link, it is not listed. (I am using Chrome.)

[h.sie]

Hi irish, see here:

http://www.subsim.com/radioroom/show...&postcount=524

Should also work with unpatched sh3.exe.

[Stiebler]

@H/sie,

yes, weather is in EnvSim.act, at least some important routines.

with the patched EnvSim.act, I run Sh3 in TC=128 and watch the weather. looks good. every 2 hours a random weather change is triggered, but in some cases weather stays constant (that's just random). weather changes from bad to good to bad to good all combinations are there. rain, fog, storm, thunderstorm, sun.

I see no need for further modifications, except for fine-adjusting the weather change interval, which I chose to 2 hours for demonstration purposes.

Well, brilliant work! It shows the power of imagination, it never occurred to me to look for the weather routines outside the main SH3.exe code.

However... it will take more than a change to the time updating to repair the weather routines. What I found, for the Real Weather Fix, was that the faster the weather changes, the more it finally sticks at 15m/s, heavy clouds, heavy rain.

[Edit: Click on this link for my account of how SH3 handles the weather:
http://www.subsim.com/mods1/nygm/nygmweatherreport.zip ]

Stiebler.

[h.sie]

@Stiebler: Thanks. I' ve run SH3 the whole afternoon in TC=128 and I discovered a balanced weather mix - too good in my opinion for atlantic theatre. no sticking on windspeed = 15.

Edith: Ok, in the last time I discovered that windspeed=15 occurs more often than other windspeeds......

[Magic1111]

Originally Posted by h.sie

If I were you, I would not play a career with a beta. V15C is stable.

h.sie

Aaaah, okay, thanks for clarify this h.sie ! Yes, I continue my career with V15C, because during career I need a stable working MOD !

Best regards,
Magic

[h.sie]

After my 1st day working with weather the following is clear now:

1) There is a seconds counter that counts up to a certain maximum value (about 160000 seconds = 44 hrs). This max value represents the weather changing time period.

2) This max. value varies a little bit and surely depends on the settings in scene.dat and campaign files.

3) If the seconds counter reaches the max. value, the counter is reset to 0 and a certain routine (which I already have located) is called which sets new weather settings (windspeed and so on).

4) Then the current weather slowly adjusts to fit the new weather settings.

In the next time I'll set the weather change interval to 1 h or less in order to long-time-record the windspeeds generated in order to see if really windspeed=15 dominates.

If so, I'll try to patch the routine which periodically generates new weather and reduce windspeed a little bit so that we have a homogenuous distribution of windspeed.

[Draka]

Ya'll SURE ya'll don't have a copy o' th' SDK in yer possession??!!?? Ya'll are workin' WONDERS - keep this up and we'll really have a true simulation!

[Magic1111]

Originally Posted by h.sie

After my 1st day working with weather the following is clear now:

1) There is a seconds counter that counts up to a certain maximum value (about 160000 seconds = 44 hrs). This max value represents the weather changing time period.

2) This max. value varies a little bit and surely depends on the settings in scene.dat and campaign files.

3) If the seconds counter reaches the max. value, the counter is reset to 0 and a certain routine (which I already have located) is called which sets new weather settings (windspeed and so on).

4) Then the current weather slowly adjusts to fit the new weather settings.

In the next time I'll set the weather change interval to 1 h or less in order to long-time-record the windspeeds generated in order to see if really windspeed=15 dominates.

If so, I'll try to patch the routine which periodically generates new weather and reduce windspeed a little bit so that we have a homogenuous distribution of windspeed.

Wonderful, simply WONDERFUL !!!

[Yoriyn]

h.sie: What setting/configuration/mods you using, cos your modified envsim.act cause no weather change even after week in my game

[h.sie]

clean (hopefully) gwx . currently no mods

have you moved the original file away from the sh3 install dir? this is important.

[Yoriyn]

Yes I made one copy of old one to new folder and overwrite it with new one.
Are you doing tests in campaign mode or single mission?

[h.sie]

only single mission. but I tested campaign, too. and it works.

[h.sie]

Here are some windspeeds generated at the beginning of the weather change interval:

Indian Ocean , March 1944
15/10/14/15/15/11/15/8/12/13/0/0/8/15/4/15/15/14/15/5/15/15/11/
0/15/15/11/10/15/11/14/15/12/0/7/4/3/15/9/12/7/1/8

Arctic Ocean, December 1943
7/7/15/15/15/15/15/7/15/15/7/15/7/15/15/15/15/15/15/14/
15/15/15/7/11/15/7/7/7/7/9/15/15/15/7/10/15

Koenigsberg, March 1941
0/4/0/6/6/0/6/6/0/0/0/6/6/6/6/6/6/0/0/6/5/6

North Atlantic, April 1940
15/8/0/0/15/12/2/15/15/12/15/13/15/14/9/15/15/14/11/4/0/11/
15/8/15/5/8/12/8/15/4/3/0/10/15/7/15/15/14/5/15/8/15/4/
8/13/15/15/15/15/13/11/0/0/0


Windspeed=15 seems to be too often. Maybe one should replace 50% of the 15 by an randomly chosen value.

[Hitman]

Here are some windspeeds generated at the beginning of the weather change interval:

Indian Ocean , March 1944
15/10/14/15/15/11/15/8/12/13/0/0/8/15/4/15/15/14/15/5/15/15/11/
0/15/15/11/10/15/11/14/15/12/0/7/4/3/15/9/12/7/1/8

Arctic Ocean, December 1943
7/7/15/15/15/15/15/7/15/15/7/15/7/15/15/15/15/15/15/14/
15/15/15/7/11/15/7/7/7/7/9/15/15/15/7/10/15

Koenigsberg, March 1941
0/4/0/6/6/0/6/6/0/0/0/6/6/6/6/6/6/0/0/6/5/6

North Atlantic, April 1940
15/8/0/0/15/12/2/15/15/12/15/13/15/14/9/15/15/14/11/4/0/11/
15/8/15/5/8/12/8/15/4/3/0/10/15/7/15/15/14/5/15/8/15/4/
8/13/15/15/15/15/13/11/0/0/0


Windspeed=15 seems to be too often. Maybe one should replace 50% of the 15 by an randomly chosen value.

It depends on the zone and season, of course. Frequent 15 winds for the Arctic, North Sea and northern part of the North Atlantic from October to May is not really that bad. In fact it is quite realistic, and you must remember that 15m/s is not the strongest possible wind by far. In fact it is the wind for a quite rough sea, but not for a huge storm.

Looking at the numbers you posted, it seems that the logarythm is working more or less well, only that it sometimes gives too strong wind in the Indian Ocean, f.e.

The problem comes when rain and fog are added to that 15 m/s wind, as happens usually. With strong winds and high waves (Waves can be easily modded and many environmental mods have detuned consideraby their size) you can still attack, but fog and rain kill the visibility and make it almost impossible.

I don't know if you would be able to fix that, making fog and rain less frequent. Specially, the heavy fog should only appear in areas where warm/cold water meets warm/cold air, like Northern Sea, Ireland, Terranova, etc. and not the centre of the Indian Ocean or the Atlantic!

Can you post the statistics for the generated fog/rain in those areas?