Random weather too random?

[JAMiAM]

ORIGINAL: redmarkus4

When you "quoted" me above you omitted my key point which is that what is needed is a more granular weather grid. I have a visualisation of how this might work in my head and will create some mocked up screens over the weekend to explain...

I only touched upon it, in my last sentence, because I think that we are in complete agreement on that aspect of the weather model. I think we need more grids, or zones, and more localized variation of effect within them. However, I don't think that level of simulation is likely given some developer statements.

[Redmarkus5]

ORIGINAL: JAMiAM

ORIGINAL: redmarkus4

When you "quoted" me above you omitted my key point which is that what is needed is a more granular weather grid. I have a visualisation of how this might work in my head and will create some mocked up screens over the weekend to explain...

I only touched upon it, in my last sentence, because I think that we are in complete agreement on that aspect of the weather model. I think we need more grids, or zones, and more localized variation of effect within them. However, I don't think that level of simulation is likely given some developer statements.

OK, thanks.

Here is a 'simple' solution to the weather model. Fix the weather, by season, in the top right square. Base the weather on adjacent squares on a random die roll, but limit the possibilities based on the season AND the adjacent weather conditions. This would allow bad weather to spread west and south in the autumn, but with a degree of randomness, and retreat east and north in the spring, again with variability.

Given that we are talking about the decisive conflict of World War Two (if not in all of modern history) and that its outcome was decided as much by the weather as by other factors, not to make some attempt to model it realistically is simply unacceptable IMO.

(Remember - this is a concept image. Please - no posts asking 'Why does it say mud in the Baltic States?')

[Redmarkus5]

Autumn

[Redmarkus5]

Spring

[JAMiAM]

RedMarkus,

Interesting concept. Well thought out. Since we're brainstorming here, I would further suggest that the die-roll of the zones should be based not only on the seasonal weighting, but also by the conditions in adjacent zones. Thus, you would tend to create actual front conditions. Given your proposal of how the weather should 'flow' from or to the upper right corner, the weighting algorithm would need to take this directability into account.

Good discussion!

[Magnum88]

I think a granular weather system and one with some sort of pattern sound like excellent improvements but also sound like a fairly large changes that will not be quick to implement.

One thing that I think would help the current arrangement is a forecast system for each zone's next week weather. I understand weather forecasting even today is not completely accurate but the forecast would be only for the next week and is a big picture forecast for a huge region. It also helps overcome the massive scale of the current weather zones allowing you to prepare for the Mud at least a little before being 100 MPs from your railheads and isolated (and having your panzer divisons getting their asses handed to them by some half-strength, half-trained rifle division).

[FredSanford3]

ORIGINAL: Magnum88

I think a granular weather system and one with some sort of pattern sound like excellent improvements but also sound like a fairly large changes that will not be quick to implement.

One thing that I think would help the current arrangement is a forecast system for each zone's next week weather. I understand weather forecasting even today is not completely accurate but the forecast would be only for the next week and is a big picture forecast for a huge region. It also helps overcome the massive scale of the current weather zones allowing you to prepare for the Mud at least a little before being 100 MPs from your railheads and isolated (and having your panzer divisons getting their asses handed to them by some half-strength, half-trained rifle division).

+1 on the forecast. Maybe even put some uncertainty- 60% clear, 40% mud for a spring or fall 'typical' forecast, to say 90-100% clear, 0-10% mud in the summer. So on for all seasons and zones. Combined with Redmarkus's maps and 'shifting' concept, that'd be a pretty cool weather model. I wonder what the level of effort to implement would be?

[Redmarkus5]

ORIGINAL: JAMiAM

RedMarkus,

Interesting concept. Well thought out. Since we're brainstorming here, I would further suggest that the die-roll of the zones should be based not only on the seasonal weighting, but also by the conditions in adjacent zones. Thus, you would tend to create actual front conditions. Given your proposal of how the weather should 'flow' from or to the upper right corner, the weighting algorithm would need to take this directability into account.

Good discussion!

Agreed! The adjacent zone is a definite factor in what happens in any given zone each turn. Sorry if I didn't make that clear.

So, if zone x is clear at the moment, but the adjacent zones to the east have mud, you better look out for mud in the next turn or two...

[Redmarkus5]

ORIGINAL: Magnum88

I think a granular weather system and one with some sort of pattern sound like excellent improvements but also sound like a fairly large changes that will not be quick to implement.

One thing that I think would help the current arrangement is a forecast system for each zone's next week weather. I understand weather forecasting even today is not completely accurate but the forecast would be only for the next week and is a big picture forecast for a huge region. It also helps overcome the massive scale of the current weather zones allowing you to prepare for the Mud at least a little before being 100 MPs from your railheads and isolated (and having your panzer divisons getting their asses handed to them by some half-strength, half-trained rifle division).

With the suggestion above you can do your own forecasts - what's coming down the line in weather terms can be assessed from what is happening next door.

[Redmarkus5]

ORIGINAL: Franklin Nimitz

ORIGINAL: Magnum88

I think a granular weather system and one with some sort of pattern sound like excellent improvements but also sound like a fairly large changes that will not be quick to implement.

One thing that I think would help the current arrangement is a forecast system for each zone's next week weather. I understand weather forecasting even today is not completely accurate but the forecast would be only for the next week and is a big picture forecast for a huge region. It also helps overcome the massive scale of the current weather zones allowing you to prepare for the Mud at least a little before being 100 MPs from your railheads and isolated (and having your panzer divisons getting their asses handed to them by some half-strength, half-trained rifle division).

+1 on the forecast. Maybe even put some uncertainty- 60% clear, 40% mud for a spring or fall 'typical' forecast, to say 90-100% clear, 0-10% mud in the summer. So on for all seasons and zones. Combined with Redmarkus's maps and 'shifting' concept, that'd be a pretty cool weather model. I wonder what the level of effort to implement would be?

Where it says "Based on die roll, weighted by season" in my diagram, that's the random factor. What is meant is that no die roll will give you snow in the summer or mud following blizzard without a snow turn in between, but you could have mud in just one square and clear in all the surrounding ones...

[edub180]

I really like redmarkus4's idea and hope something like that it will be implemented at some point.

[vlcz]

I "resurrect" this topic in a somewhat whinnig way..

Lsst week , sick of my red-army demi-godless capacity to exactly know when to launch a clear time offensive that can not be countered thanks to sure following rains, I started a friendly Grand Campaign against a close friend.

turn three, July 1941 MUD.... Game Over.

I have readed in some other topic there is a coded amend to "force" Blizzard in december 41, even in "random" weather. Real or not it seems apropiate. For the sake of mental sanity and avoiding "false openings" should it not be made an arrangemente to assure clear in July-August 41??????

[Blubel]

It would be great if an option could be coded in.

[Baelfiin]

ORIGINAL: vlcz

I "resurrect" this topic in a somewhat whinnig way..

Lsst week , sick of my red-army demi-godless capacity to exactly know when to launch a clear time offensive that can not be countered thanks to sure following rains, I started a friendly Grand Campaign against a close friend.

turn three, July 1941 MUD.... Game Over.

I have readed in some other topic there is a coded amend to "force" Blizzard in december 41, even in "random" weather. Real or not it seems apropiate. For the sake of mental sanity and avoiding "false openings" should it not be made an arrangemente to assure clear in July-August 41??????

only 1 mud per weather zone max in the summer, be glad its out of the way so early. If you get lucky you will have mud in a weather zone you have not gotten to yet !! 8)

[vlcz]

ORIGINAL: Baelfiin
only 1 mud per weather zone max in the summer, be glad its out of the way so early. If you get lucky you will have mud in a weather zone you have not gotten to yet !! 8)

I wasn´t aware that limit[X(] it is a GREAT notice indeed (although I really feel there should be no mud posibility before september 41) , I´m sure you can imagine the dificulties of feeling lucky with a red bordered 3rd pz group next to the landbridge[:@][:(]

[Schmart]

I'm pretty sure that with random weather, just as there is one mud turn guaranteed in summer 1941, there is one snow turn guaranteed during the first blizzard. This can stop a soviet winter counter-offensive just as much as the German summer advance.

In one of my games vs Axis AI with random weather, snow continued for ALL of April 1942 (obviously got lots of rare die-rolls), so the Axis spring counter-offensive lasted twice as long as usual. In the same game (more lucky die-rolls), there was clear weather for all of June 1942. Both were major advantages for the Axis. So it can work both ways.

[Joel Billings]

From the manual, 22.2.2:

The following exceptions
will occur when using random weather:
»» There will be at most one snow turn per weather
zone in December and January.
»» There will be at most one mud turn per weather zone
during turns from 19 June to 30 September.
»» There will be at most one mud turn per weather zone
during turns from 01 December to 31 March.

From readme file:

7. Missing Rule – Section 22.2 should contain an additional exception that states there will be no snow or mud during December 1941 and January 1942 (only Blizzard).


The chance of Mud in the summer was added in to account for times when major rain would come in and significantly slow operations. Although usually for just a few days, we felt they added up enough to throw in 1 mud turn per zone per summer. It's part of the variability that you get with random weather. You have to expect that you will get hit with it at some point. I have't gotten the sense from posts that random weather favors one side or the other assuming the weather gods play fair over the long run.

[morganbj]

I still say a lot of the weather "problems" could be solved if there were a few other types of weather. For example, why not have a "Summer Rain" type that slows only certain units down and only in semi-random places from June to September. You could also have not just Snow and Blizzard, but add a "Heavy Snow" and maybe even a "Light Snow." Maybe a "Freeze." I don't know what all of them should be, but it just seems limiting to have four types with which to simulate actual weather effects into the game.

And yes, I know it add a bit of programming, but it sure seems like it will have to be done for the Western front game anyway.

[Q-Ball]

Random Weather isn't perfect, but I think it's basically OK within the engine. Over time, it favors the side that has the initiative, but over time that evens out since the initiative passes from the Germans to the Russians.

My only gripe is Ice Levels. They don't really make sense; I think they are too "Icy". You can have MUD, but have Rivers still frozen solid.

[JAMiAM]

ORIGINAL: Q-Ball

Random Weather isn't perfect, but I think it's basically OK within the engine. Over time, it favors the side that has the initiative, but over time that evens out since the initiative passes from the Germans to the Russians.

My only gripe is Ice Levels. They don't really make sense; I think they are too "Icy". You can have MUD, but have Rivers still frozen solid.

Agreed. Even given the "Icy"ness of the rivers, there should still be some residual movement penalties and combat effects, even if they are frozen "solid". Nobody is going to drive a Pz Korps across a frozen river without giving a good testing of the ice pack, and it will change (weaken) as the amount of vehicles cross over it. Likewise, attacking across a frozen river is difficult. The embankments, and vegetation along the river offer good defensive position, and the clear iced section is a killing ground for ground fire, and artillery. Once the area has been shelled, it would be extremely foolhardy to expect that the ice hasn't weakened and gotten more dangerous.