Weather

[johnmcd]

I’d like to see weather play a greater, or at least more comprehensible role. Wind shouldn’t just make arrows less accurate, it should also give them greater or less range depending on wind direction, men and horses marching or galloping into the wind should tire more quickly and travel more slowly.
Also, when its wet it should also be possible for it to be windy and so on. I dunno, I just think that for something which makes a big difference to the game it is also remarkably underdeveloped.

[Anssi Hakkinen]

STW's weather is underdeveloped? Now there's a statement. There aren't many games out there that model weather at all, let alone have an engine that dynamically changes the weather conditions.

Wind direction etc. would require complex models for defining which arrows are deflected where etc. - I'm not saying it can't be done, and I'm not saying it wouldn't be cool if they did it, but it's not the most crucial addition they could make considering how much time it would take. And wind + rain is called a storm; so what if they're all thunderstorms.

------------------
"If you start a fight, you must win it. Fighting, however, is not your objective. The most important thing is to win without fighting."
- Risuke Otake Sensei

[Buccaneer]

As an occassional game designer/programmer, I perceive the engine that is used in Shogun is not as precise as you described. You talk about modeling each arrow/unit/etc. I don't think that what's happening. In a game like Combat Mission, you are dealing with true 3D units in which the velocity/mass/angle/etc. of an HE shell is calculated for its impact as it hits the Panzer. Such modeling, with today's PC technology, can only be done on a handful of units. In a non-3D units game like Shogun, you would know these things about an archery assault:
# of archers, therefore, # of arrows,
Least-squares coordinates of the archers unit,
wind velocity/direction,
unit strength modifier,
plus the similar attributes for the enemy defensive unit.

You run those numbers through a formula, apply some random number generator and you get: the percentage of arrows calculated for the various levels of casualties. For example, of the 100 total arrows: 17 killed, 13 reduced morale, 7 did something else. If there were severe weather, then the numbers would 9, 5, 1 if all other things were equal. You just then apply those results randomly to the units and call up the dead unit sprites for those got killed. What you see on the screen with unit sprites and pixelated arrows are just cosmetics. The results are already calculated each time they 'shoot'.

My point in all of this rambling is that to you can easily adjust for weather as johnmcd suggested. All you have to do is calculate the various modifiers and apply them to numbers.

[Hosakawa Tito]

Doesn't the weather have these affects in the game already?I seem to notice that during
rain storms and snow a greater degree of fatigue on my units,especially cavalry and naginata.One of my favourite bridge crossing tactics is to wait for a thunderstorm to lessen the effectiveness of the enemy archers.I don't know of any other games that I have played that have had this feature,at least to this degree.I think there could be more windy days,without rain,or how about very hot weather with the effects of thirst on the warriors especially the horses?Isn't there a thread somewhere,question times 3 or something like that where these ideas can be presented to the developers?
Tito

[johnmcd]

yes weather features, yes it is better than in most games, but no its not perfect and if we are allowed to wish then this is something I'll wish for improved weather modelling with things like wind direction carrying arrows and hills providing shelter.

[BanzaiZAP]

And point in fact, I believe it was Target (an actual CA programmer) pointed out that Shogun DOES track each arrow, etc! That's how forests work - it looks to see if the arrow hits a tree. Basically, it takes the angle needed to hit intended target, adds modifers for weather, honor, etc, and makes that a margin around the intended target. Fire arrow using modifiers to determine trajectory, then see if anything was in the flight path. This is how an arrow that misses a target may hit the guy next him. Hence Loose formation making it less likely to lose men (near misses only hit the ground). You can even follow an arrow in flight, from archer to target! (Judicious use of the Pause key helps in this.)

Doing it generically like Buccaneer describes would have problems with things like overlapping units. If a unit of spearmen walks through the archer unit receiving fire, an open calucation may give results like not having misses hit the second unit, or having more kills than arrows-shot type of thing. Not to mention - how would it know how to draw the arrow's flight?

Back on subject, I think that weather just increases the miss-margin used by archers. That's one of the many reasons some low-end machines are bogged down in archer-heavy engagements. It would be great to have wind directly affect the arrow flight. Someone could get a height advantage, but then the attacker moves upwind to increase his own archer range.....

-- B)

[Erado San]

Buccaneer,

I am sure that each arrow fired (also each bullit from a musket or arq) becomes a 3D object moving on its own. And yes, it was indeed Target who explained it all to us. The trajectory path is then calculated as it moves along. When it hits anything, the hit chances are calculated then. So for wind to have this effect the wind parameters should be included in the formulas for calculating its movement. And the wind speed might effect the speed at which the arrow hits an opponent, thus perhaps influencing the 'to kill' formulas.

[This message has been edited by Erado San (edited 12-29-2000).]

[johnmcd]

I'm not fussed about how its modelled so long as its consistant, but it doesn't seem to me that either way would have an unendurable burden placed on it by allowing the things I suggest.

Another neat effect you could do with weather would be to have fog in the valleys, so you would be invisible to those on high (and vice versa) and low cloud so those on high would be hidden from those lower down. That would surely allow a whole new set of tactical manuevers. more so than the current fog. Also, snow drifts would be cool! running a man down a hill and seeing him disappear up to his neck would be smart!

[Buccaneer]

I did some tests last night and I really believe that there is no way that the trajectory/impact of each arrow is calculated. I set up a Quick Battle of 1 archers and 1 ashis, held the positions of the archers and saw that there were some pattern of the arrow sprites flying towards the ashis. I can't tell for sure since I can't freeze the frames, but I did see a repeated pattern of 8 types of arrow sprites depending on distance.

I am still convinced that there are some generic algorithms in calculating the impacts of the projectiles. It is clever enough to make it appear that each arrow is calculated, but there is not enough compute power to do that in the real-time battles of Shogun (or any wargame for that matter).

Therefore, any additional weather modifiers can be done fairly easily. For exampe with snow drifts, just have variables of snow depth/slope/facing and then recalculate the reductions in speed by unit type and then calculate the percentage of men that will get stuck. Then all you do is call up the stuck_in_snow sprites to show that some of the men got stuck.

[Anssi Hakkinen]

I won't get deeply involved in this, as I have never seen Target-sama and therefore cannot vouch for what he did or did not say, but I say this:

STW is a far more complex game than you think, Buccaneer-san. There is clear and tangible evidence in your \logfiles subdirectory that the game tracks each and every soldier individually: that alone adds a whole new level of complexity to any calculation.

This is not Kessen, nor is this C&C. This is STW.

------------------
"If you start a fight, you must win it. Fighting, however, is not your objective. The most important thing is to win without fighting."
- Risuke Otake Sensei

[horatio]

I just have one thing to sa about the weather: I WOULD LIKE TO SEE AN OPTION WHERE PREDICTING THE WEATHER IS NOT VERY ACCURATE. I love this game, but my god, if the weathermen can't do it with doppler radar, how in heck did they do it back then?!

[BakaGaijin]

Nobody's predicted the weather in Shogun. When you click that "Wait" button, you're waiting until the next morning to decide if that day's weather makes you happy. If you were "predicting" the weather, wouldn't that mean that you could choose any day and not have to "wait" for the next day in order to see the weather?

If you mean that the weather should change more dynamically, I suggest you watch a little more closely. Sometimes, the weather stays fairly constant throughout the day, but this also happens in real life. I've fought battles which have started clear, but ended up with drenching, but intermittent rain showers.

[Buccaneer]

Anssi, I have no doubt that STW is more complex than those games you listed. However, just because the logfiles show calculations for each man, don't automatically assume that it is calculating every man. Let's take my hypothetically scenario one step further. You know the percentage of men that must be killed (let's say 17 out 100). All you have to do is have the program randomly pick 17 men (say #5,7,12,15,26,31,etc.) and fill in the KILLED string into the logfile. We have no way of tracking it back to the sprites, so it is another clever of folling you into thinking it is more complex than it is. I can sit here and reverse engineer what you are seeing on the screen and in the logiles just by a few random number generators and various calculations. You wouldn't be able to tell the difference.

In no way am I criticizing STW. It is an engaging strategy with great battles gameplay. My point is that ideas like weather, etc. should be brought up because I believe that the game engine is not that complex (not like Combat Mission which uses true 3D modeling instead 2D sprites in a 3D world).

[BakaGaijin]

Then go play Combat Mission.

Better yet, go MARRY IT if you love it so much! YEESH!

Spoofing logfiles seems like a righteous buttload of trouble to go through just to convince us that calculations are being done on an individual basis when they really aren't. There's certainly no shame in CA saying that they calculate combat on a unit-by-unit basis. After all, everyone else does it and it doesn't change the fact that Shogun is tons of fun to play.

[Anssi Hakkinen]

BakaGaijin, please be civil. After all, Buccaneer-san specifically said that he is NOT criticizing STW.

But, unfortunately, there is an abundance of evidence that the game really tracks individual soldiers, and no real evidence to the contrary. The logfiles show that units succesfully defending narrow passes or bridges have some soldiers that have dozens of kills (the front row ones) and some who have no kills at all (the back row ones). Features like this would require extremely deliberate attempts at misleading the users, and, like Baka said, why would CA bother?

I don't know about the arrows etc., but the soldiers *are* individuals, and you may quote me on that.

By the way, why should STW be less complex because the units are sprites? Granted, that gives them less opportunity to interact with the 3D environment, but 3D-modelled troops would hog up immense amounts of resources and accomplish nothing: the cosmetic advantage is negligible and features like hit locations aren't modelled anyway. That the units are sprites doesn't mean they don't have a definite location.

I always thought Combat Mission had 2D infantry graphics; furthermore, the game models a squad of troops with 3 guys. I wouldn't call that exceptionally realistic either: it's inferior to STW, in fact. But we're really not here to discuss that game.

------------------
"If you start a fight, you must win it. Fighting, however, is not your objective. The most important thing is to win without fighting."
- Risuke Otake Sensei

[Buccaneer]

Ok, ok, I withdraw from the battlefield. I go along with the fact that each unit is modeled. Just because we are not controlling individual units, it doesn't mean the movements and actions are not calculated. However, IIRC, this whole discussion got started on arrows. I doubt that arrows are modeled individually (Anssi-san is not sure either), so the point about weather effects is still a good and doable idea.

Anssi-san, it seems like my knowledge of STW is about the same as your knowledge of CM...which isn't very much. http://www.totalwar.org/ubb/smile.gif

BTW, I like STW a little better than I do CM, but neither game, for me, can come close to the genius of Civ2. It's not the eye-candy (obviously) but the gameplay (and replayability).

[BakaGaijin]

Weather (get it? GET IT?) or not the arrows are modelled individually, directional wind effects would be a breeze (God, I'm ON A ROLL!) to implement, I'd think. The wind effect would be a constant vector on a 2-dimensional plane. I'm not up on the programming aspect of that, but the mathematics would be pretty simple.

A problem arises, of course, if the arrows currently use a "random strike" algorithm rather than direct path computation. I'm fairly certain that the arrows are computed individually, but that doesn't mean that their flight paths are. It's quite likely (since computing dozens of flight paths would be a huge processor load) that the game is programmed to compute impact points as a random deviation from a central target. To create the "missed shots might hit others in the formation" effect, the computer simply needs to know how spread apart the formation is. If this random deviation is the case, perhaps the only way to implement directional effects of wind would be to fudge. Crosswinds randomly affect accuracy, headwinds decrease range and potency while decreasing accuracy, tailwinds increase range and potency while decreasing accuracy. Of course, I imagine that this would also be a fairly small programming task.

[This message has been edited by BakaGaijin (edited 01-01-2001).]

[BakaGaijin]

By the way, Bucc, I'm sorry for going off on you a bit. http://www.totalwar.org/ubb/wink.gif

I actually liked the Combat Mission demo alot, if that means anything.