|
berto -> Berto's Weathers Mod: NOTES (1/30/2012 7:10:50 PM)
|
From the Weathers Mod main NOTES.txt file:
quote:
My weathers mods -- why bother? For some answers, see the NOTES.txt
files in the weathers mod game folders (for example, the weathers mod
ACW/NOTES.txt file).
In my weathers mods, I try to base weathers on real-world climate and
weather patterns, for example
* climate (weather) and terrain maps, such as:
* http://www.architecture.uwaterloo.ca/faculty_projects/terri/carbon-aia/images/climate-eere.jpg
* http://www.bamboobotanicals.ca/images/care/canada-climate-zone-lg.jpg
* http://www2m.biglobe.ne.jp/%257eZenTech/English/Climate/Russia/Russia_Climate_Map.gif
* http://jb-hdnp.org/Sarver/Maps/WC/wc09_ruseurasclimm.jpg
* http://pikt.org/ageod/ref/ClimateZonesEurope.jpg
* http://pikt.org/ageod/rop/weathers/ROPGameMapExtent.jpg
* climate graphs, such as:
* http://www.climatemps.com/graph/tbilisi-georgia_files/image001.gif
* climate data and summary descriptions, such as:
* http://en.wikipedia.org/wiki/Astrakhan#Climate
* http://en.wikipedia.org/wiki/Richmond,_Virginia#Climate
I also base things on my understanding of how climate and weather
work, from personal experience (residing and traveling), but also a
lifetime of casual study (geography classes, news accounts, global
warming blogs, histories, etc.).
Within reason -- given limitations in the AGE engine weather
mechanism, and limits on time and effort; I am not a meteorologist or
climatologist, and the weather mods are not a scientific study or PhD
thesis! -- within reason, you can expect a fairly good weather model
and representation of the real physical world.
In my weathers mods, there will be much less variability, fewer
weather patterns, than in the unmodded, official games.
In my opinion, AGEOD has in some ways over-engineered its weather
model (in other ways, under-engineered); the weather is too granular,
with too many weather areas and terrain types. This is especially
true of the earlier games (AACW, WIA, NCP). Simplifying is key.
But climate and weather are complicated and unpredictable, you might
say. You might say; and I would respond:
* day-to-day weather differences; but game turns are not one day,
rather 15 (or 30) days
* freak snowstorms or cold snaps, especially to the south; in real
life, they do happen, but rarely, and usually dissipate quickly
* inverted continental temperature patterns (warm to the north,
cold to the south; or warm to the west, cold to the east); but in
real life, these are atypical, are not at all usual
* the "Little Ice Age" -- roughly the 16th through the 19th Centuries
-- and the generally somewhat colder weather toward the end of that
age compared to our current weather; but the differences are
"somewhat" (in real life, real history), not drastic (as sometimes
in the official games)
We are looking to average normal weather over time (15 days, or 30
days) and space (the game map). Variety may be the spice of life, and
for some people make for more interesting game play; but my preference
is for realism, and plausibility. These are serious historical
simulations, not anything-goes fantasy games. My opinion.
So game weather will be blander and more boring than what you are used
to.
In my weathers mods, I strive to follow these general principles:
* colder to the north, warmer to the south
* colder in the uplands (hills and mountains, and depending on the
game, also wilderness, taiga, etc.); warmer in the lowlands and
toward the coasts
* consistency east to west (not north to south)
* broad banding, averaging out weather over the 15-day (or 30-day,
in the case of WIA) game turns
* no "crazy quilts" or "peek-a-boo" (now you see it, now you don't))
weather patterns, or as little as possible
In order to achieve the desired effects, I have to
* in all games, re-assign some regions from one weather area to
another
* in some games, reduce the number of weather areas
* in all games, constrain each region to just one or two "adjacent"
weather possibilities in any given month
Region re-assignments:
In the official game data files, regions are sometimes mistakenly
assigned from one adjoining weather area to another, leading to
"orphans", "holes", and "fingers". At other times, there are outright
goofs (for example, in RUS assigning the southern Steppe region
$Alexandrovsk to the northern Subarctic weather area). These must be
corrected.
Reducing the number of weather areas:
Here are some typical weather areas (in this case, from the
WIA/GameData/Areas.ini file):
---------------------------------------------------------------------
// Area #81, Western_Great_Lakes
{Area}
UID = $area_Weather_Western_Great_Lakes
Name = Western_Great_Lakes
Hierarchy = -1
ListParams = $Lac Huron Sud|$Lac Huron Nord|$Chutes aux Saubles|$Saginaw Bay|$Lac Huron|$Sauts des Francais|$Anse au Tonnerre|$Sault|$Lac Superieur Sud|$Lac Superieur Ouest|$Anse de Michipicoton|$Grande Baie|$Baie des Puans|$Lac Michigan Centre|$Lac Michigan Sud|$Lac Superieur Est|$Saint Joseph|$Muskoka|$Huron|$Nunda|$Ojibwa|$Buffalo|$Quinatowa|$Portage|$Saint Clair|$La Grande|$Ganago|$Keweenaw|$Laurium|$Marinette|$Menominee|$Laona|$Wausau|$Oconto|$Shawano|$Wautama|$Dubuque|$Oshkosh|$Kewaunee|$Manitowoc|$Mauwatosa|$Ocanomowoc|$Mascoutins|$Maquaketa|$Neenah|$Savanna|$Kitchigaming|$Green Lake|$Pottewatomis|$Le Detroit|$Saint Denis|$Sandusky|$Sebawaing|$Lapeer|$Levallon|$Flin|$Saginaw|$Standish|$Sable|$Mackinaw|$Cadillac|$Clare|$Alma|$Kalkaska|$Manistee|$Manton|$Muskegon|$Grand Havre|$Owosso|$Grand Rapides|$Fontain|$Maumee|$Saint Ignace|$Sault Sainte Marie|$Munising|$Manistique|$Marquette|$Ispheming|$Escanaba|$Saginaw river|$Elkhart|$Dawagiac|$Kenosha|$La Porte|
{/Area}
// Area #82, Eastern_Great_Lakes
{Area}
UID = $area_Weather_Eastern_Great_Lakes
Name = Eastern_Great_Lakes
Hierarchy = -1
ListParams = $Milles Iles|$Lake Ontario|$Niagara Mouth|$Long Point Bay|$Lake Erie|$Bass Islands|$Lake St Clair|$Nioure|$Famine|$Gameydoes|$Cataraqui|$Quinte|$Ganaraske|$Oshawa|$Rouille|$Mississauga|$Niagara|$Sippee|$Urse|$Cedar|$Utanvas|$Blanc|$Sandosky|$Lorain|$Miamee|$Ashtabula|$Cataraugus Creek|$Tegynagerunte|$Irondequoit|$Gannagaro|$Osceola|$Oneida|$Oswego|
{/Area}
---------------------------------------------------------------------
I effectively reduce the number of weather areas by emptying the
weather area members list, as in
---------------------------------------------------------------------
// Area #81, Western_Great_Lakes
{Area}
UID = $area_Weather_Western_Great_Lakes
Name = Western_Great_Lakes
Hierarchy = -1
ListParams =
{/Area}
---------------------------------------------------------------------
and reassigning the member regions to another weather area (in the
example, to Eastern_Great_Lakes).
(The consolidated regions are therefore sometimes misnomers. In the
above example, I would like to rename Eastern_Great_Lakes as simply
"Great_Lakes". But that would necessitate changing too many game data
files. To ease mod installation, I try to keep things simple, and
just live with some of the misnomers. But this is behind-the-scenes
stuff that shouldn't matter to you, the gamer.)
In the most extreme case -- WIA -- I have reduced the number of active
weather areas from the default 19 down to the mod 7. In other cases
-- ROP & RUS -- the number of active weather areas is unchanged (but
still with region re-assignments among areas).
More tightly constraining weather possibilities:
Here are some typical weather specifications (in this case, from the
ROP/GameData/Weathers/WeatherPatterns_Interior_Uplands.ini file):
---------------------------------------------------------------------
Interior_Uplands|January|normal|Fair|0
Interior_Uplands|January|normal|Mud|0
Interior_Uplands|January|normal|Snow|70
Interior_Uplands|January|normal|HarshWeather|30
Interior_Uplands|January|normal|VHarshWeather|0
Interior_Uplands|January|low|Clear|0
Interior_Uplands|January|low|Mud|0
Interior_Uplands|January|low|Snow|70
Interior_Uplands|January|low|HarshWeather|30
Interior_Uplands|January|low|VHarshWeather|0
Interior_Uplands|January|high|Clear|0
Interior_Uplands|January|high|Mud|0
Interior_Uplands|January|high|Snow|20
Interior_Uplands|January|high|HarshWeather|80
Interior_Uplands|January|high|VHarshWeather|0
Interior_Uplands|January|coldwater|Clear|0
Interior_Uplands|January|coldwater|Mud|0
Interior_Uplands|January|coldwater|Snow|70
Interior_Uplands|January|coldwater|HarshWeather|30
Interior_Uplands|January|coldwater|VHarshWeather|0
Interior_Uplands|January|water|Clear|0
Interior_Uplands|January|water|Mud|20
Interior_Uplands|January|water|Snow|80
Interior_Uplands|January|water|HarshWeather|0
Interior_Uplands|January|water|VHarshWeather|0
Interior_Uplands|February|normal|Fair|0
Interior_Uplands|February|normal|Mud|0
Interior_Uplands|February|normal|Snow|100
Interior_Uplands|February|normal|HarshWeather|0
Interior_Uplands|February|normal|VHarshWeather|0
Interior_Uplands|February|low|Clear|0
Interior_Uplands|February|low|Mud|0
Interior_Uplands|February|low|Snow|100
Interior_Uplands|February|low|HarshWeather|0
Interior_Uplands|February|low|VHarshWeather|0
Interior_Uplands|February|high|Clear|0
Interior_Uplands|February|high|Mud|0
Interior_Uplands|February|high|Snow|50
Interior_Uplands|February|high|HarshWeather|50
Interior_Uplands|February|high|VHarshWeather|0
Interior_Uplands|February|coldwater|Clear|0
Interior_Uplands|February|coldwater|Mud|0
Interior_Uplands|February|coldwater|Snow|100
Interior_Uplands|February|coldwater|HarshWeather|0
Interior_Uplands|February|coldwater|VHarshWeather|0
Interior_Uplands|February|water|Clear|0
Interior_Uplands|February|water|Mud|50
Interior_Uplands|February|water|Snow|50
Interior_Uplands|February|water|HarshWeather|0
Interior_Uplands|February|water|VHarshWeather|0
[...]
---------------------------------------------------------------------
As you can see, each terrain type is limited to just one or two
"adjacent" weather possibilities in any given month. For example:
100% chance Snow; or 50% chance Mud, 50% chance Snow. To allow three
or more possible weathers for a given region and month is to invite
inverted weather/temperature patterns, "checker board" patterns, or
worse.
Contrast the above with some typical standard defaults (in this case,
from the official
ACW/GameData/Weathers/WeatherPatterns_Weather_Upland_South.ini file):
---------------------------------------------------------------------
Weather_Upland_South|January|Other|Fair|20
Weather_Upland_South|January|Other|Mud|40
Weather_Upland_South|January|Other|Snow|25
Weather_Upland_South|January|Other|Frozen|15
Weather_Upland_South|January|Other|Blizzard|0
Weather_Upland_South|January|Marsh|Fair|60
Weather_Upland_South|January|Marsh|Mud|25
Weather_Upland_South|January|Marsh|Snow|5
Weather_Upland_South|January|Marsh|Frozen|10
Weather_Upland_South|January|Marsh|Blizzard|0
[...]
---------------------------------------------------------------------
For all the craziness that results, too much variation!
I use a Perl script (program), mkweathers.pl, to generate the
month-to-month weather parameters (chances) for each weather area and
terrain type.
You don't need to run this script, or to understand its operation;
that's my job. Just know that the script helps to ensure consistent
treatment across all climate types, across all terrain types, across
all games. The various Weathers *.ini files are a great mass of data.
mkweathers.pl helps to make sense of it all, to manage all the
complexity, and to reduce outliers, typos, and other mistakes (such as
the combined weather chances exceeding 100% for any given month and
terrain type).
How do my modded weather parameters compare to the AGEOD official
defaults? To answer that question, I use another Perl script,
ageweathers.pl, to compute concise, concise, easy-to-see cross
references of one game's weather specifications to another's. In my
weathers mods, you can expect many changes, but I try not to stray
unnecessarily far from the official AGEOD standards and defaults.
For the latest version of mkweathers.pl & ageweathers.pl, (also the
latest version of the Weathers Mods, other scripts, and possibly other
useful stuff), visit
http://pikt.org/ageod/weathers/
After all the region corrections, re-assignments, consolidations;
after specifying the weather parameters (chances); after performing
some rigorous, comprehensive QA (using AGElint, and by other means);
after all of that, I tested, tested, tested. Based on my tests, I am
fairly well satisfied with the AACW, ROP, RUS & WIA Weathers Mods
results. Not perfect. But good enough. Certainly better than the
standard defaults.
There is much more I could say, but let's keep this short (ha!). For
more, much more discussion, please see any of
* http://www.matrixgames.com/forums/tt.asp?forumid=460 (AACW)
* http://www.ageod-forum.com/showthread.php?22114-Berto-s-ROP-Weathers-Mod (ROP) [older, but still interesting discussion]
* http://www.matrixgames.com/forums/tt.asp?forumid=898 (RUS)
* http://www.matrixgames.com/forums/tm.asp?m=3172568 (WIA)
* http://www.ageod-forum.com/showthread.php?21844-Berto-s-WIA-Weathers-Mod (WIA) [older, but still interesting discussion]
Some disclaimers:
* I am not a climatologist. (I only play one on TV.)
* My criticisms of AGEOD games weather modeling may be out of date.
The official game weathers may have changed recently, or since I
have written this.
* The AGE weather model is imperfect (what isn't?), and limits what
we can achieve. There are some maddening quirks and gotchas in
the AGE weathers system, also in the game data files. We try to do
the best we can, given all the constraints, and the mod objectives.
* I am diligent, and careful, and to the best of my ability I QA my
work and strive to give you the best product reasonably possible.
I make my share of mistakes, but except for honest differences of
opinion, I correct my goofs!
So, that in a nutshell (well, a bagful of nutshells) summarizes my
Weathers Mods.
Enjoy!
Robert Osterlund, aka "Berto"
Oak Park, Illinois, USA
2012/09/12
|
|
|
|
