Clouds & Wind option / interaction

[zooloo75 834]

Just out of curiosity, ACE seems to handle this pretty well in A2, couldn´t the same techniques be applied in A3?

I have a feeling that people don't read :c

It is possible - I have already done it.

[metalcraze 290]

Can you make an addon version of your script though? That will apply to all missions played.

[zooloo75 834]

Can you make an addon version of your script though? That will apply to all missions played.

I don't know how to do that yet, but a developer of Combat Space Enhancement has contacted me to use my BWIS/AWIS system in their mod. So look forward to their release :)

[NeuroFunker 11]

i mean if 1 single man, could implement this already, i can't believe, BIS wouldn't make this by themselves later.

[Pulverizer 1]

i mean if 1 single man, could implement this already, i can't believe, BIS wouldn't make this by themselves later.

A simplistic, unrealistic and inconsistant scripting workaround is not up to the quality required for a standard game feature IMO. Better leave it for mods.

It should be far more complicated than simply adding a bit of wind to bullets over time as the mods and scripts do it.

[metalcraze 290]

Every mod idea that makes it into ArmA was a script workaround in one way or another.

ACE does wind well enough and to a player it's transparent. Which actual issue can you name that it has apart from being a workaround?

Adding less than real-world-perfect bullet wind simulation is a lot better than not having it at all.

[Pulverizer 1]

ACE does wind well enough and to a player it's transparent. Which actual issue can you name that it has apart from being a workaround?

For instance:

-MP syncing issues that Dwarden already mentioned

-Wind should generally be stronger in the open, mountains or shores than say woods and densely-built areas, valleys or indoors

-Wind shouldn't be a single, global vector all across the land, not in force and not in direction. it should also be unpredictable and have fast microvariation.

-AI should also have to compensate for wind

-Wind should probably affect a .50cal sniper rifle bullet quite differently compared to 9mm subsonic pistol round or a cannon shell or rocket, m203 grenade etc. There might be not enough real world data available to base all this on.

Adding less than real-world-perfect bullet wind simulation is a lot better than not having it at all.

Those who are happy with the simplest imaginable wind "simulation" can very easily add it as a script or mod to their game.

Edited April 30, 2013 by Pulverizer

[dmarkwick 261]

For instance:

-MP syncing issues that Dwarden already mentioned

The only consideration IMO, the other considerations seem to be overkill IMO:

-Wind should generally be stronger in the open, mountains or shores than say woods and densely-built areas, valleys or indoors

Well, OK this one might be workable :) but it'd need to be simplified down to say altitude-dependant?

-Wind shouldn't be a single, global vector all across the land, not in force and not in direction. it should also be unpredictable and have fast microvariation.

Which amounts to simple randomisation, which kind of negates MP synchronicity.

-AI should also have to compensate for wind

I don't believe so, I don't see the value in implementing wind offset to AI ballistics only then to have AI have to reverse that offset via processing. Better to have wind not affect AI at all, which in simplistic terms would be "simulating" AI compensating for wind. As a concession, I would scale the wind offset proportionally to the AI skill level.

-Wind should probably affect a .50cal sniper rifle bullet quite differently compared to 9mm subsonic pistol round or a cannon shell or rocket, m203 grenade etc. There might be not enough real world data available to base all this on.

I guess applying wind offset on a per-ammo type basis wouldn't be too cumbersome, but I wonder if it'd be worth it.

[Pulverizer 1]

Well, OK this one might be workable :) but it'd need to be simplified down to say altitude-dependant?

It's more about the type and shape of the land, rather than mere altitude.

Which amounts to simple randomisation, which kind of negates MP synchronicity.

Au contraire. It would require even more MP syncing because the people in the same general area should witness same effects. This would probably require some kind of repetetive pseudo-random patterns that would lessen the amount of syncing needed.

I don't believe so, I don't see the value in implementing wind offset to AI ballistics only then to have AI have to reverse that offset via processing. Better to have wind not affect AI at all, which in simplistic terms would be "simulating" AI compensating for wind. As a concession, I would scale the wind offset proportionally to the AI skill level.

I'd say it would be visually very inconsistent to see (thus the more difficult for me to judge wind based on) AI next to me have its tracer rounds fly straight and true while mine would curve of a meter to the side in the same exact conditions.

I guess applying wind offset on a per-ammo type basis wouldn't be too cumbersome, but I wonder if it'd be worth it.

Having no real world data, I can't tell for a fact wether it would be worthwhile either. However, due to the huge relative size, weight and speed differences of ammunition, I would think the effects of wind are also significantly different on them IRL.

Edited April 30, 2013 by Pulverizer

[adonai 1]

http://forums.bistudio.com/showthread.php?153076-Bullet-Wind-Interaction-System

I've added wind interaction with aerial vehicles. The effects are subtle, but they are present.

THANKSSSS!!!!!!!! Ive got it working with some vehicles with all in arma and all i can say is this really adds to vehicle immersion; like crazy. I 've been flying 100kt crosswind landings all night lol

Thanks again for this it works perfectly

[dmarkwick 261]

It's more about the type and shape of the land, rather than mere altitude.

Au contraire. It would require even more MP syncing because the people in the same general area should witness same effects. This would probably require some kind of repetetive pseudo-random patterns that would lessen the amount of syncing needed.

I think these are basically granularity issues, altitude dependent would not be ideal but would be easier than some terrain-judging & object analysis. As such I think a single global constant wind sync would probably be the most pragmatic.

I'd say it would be visually very inconsistent to see (thus the more difficult for me to judge wind based on) AI next to me have its tracer rounds fly straight and true while mine would curve of a meter to the side in the same exact conditions.

Yes it would, but I seem to remember BIS replying that AI wind cancelling calculations would be the difficult area, and so I just propose a simple, pragmatic solution. Perhaps an improved version of that would be: AI have no wind applied to ballistic calculations but the round is artificially curved at the mid-point to simulate. Relatively easy to implement and a whole lot less processing than AI wind calculations :)

Having no real world data, I can't tell for a fact wether it would be worthwhile either. However, due to the huge relative size, weight and speed differences of ammunition, I would think the effects of wind are also significantly different on them IRL.

Perhaps, but again, pragmatism would lean me toward the simple approach of constant effects on rounds.

[Pulverizer 1]

...

Well to simplify everything is of course more practical than to make wind behave as you would intuitively expect it to. But again, if you're happy with physics from Bizarro world, I suggest you download it as a simple mod instead of forcing it as a standard feature.

[metalcraze 290]

For instance:

-MP syncing issues that Dwarden already mentioned

There are no MP syncing issues in ACE that I've noticed

Otherwise you simply would not hit.

-Wind should generally be stronger in the open, mountains or shores than say woods and densely-built areas, valleys or indoors

-Wind shouldn't be a single, global vector all across the land, not in force and not in direction. it should also be unpredictable and have fast microvariation.

-AI should also have to compensate for wind

-Wind should probably affect a .50cal sniper rifle bullet quite differently compared to 9mm subsonic pistol round or a cannon shell or rocket, m203 grenade etc. There might be not enough real world data available to base all this on.

Those who are happy with the simplest imaginable wind "simulation" can very easily add it as a script or mod to their game.

Yeah you say that but you don't seem to be bothered by a 5 seconds full-heal medic system of ArmA3. Why have it if it's so simple and unrealistic?

Of course it can be better. But not having it is a lot worse than having a simple implementation.

[dmarkwick 261]

Well to simplify everything is of course more practical than to make wind behave as you would intuitively expect it to. But again, if you're happy with physics from Bizarro world, I suggest you download it as a simple mod instead of forcing it as a standard feature.

I would intuitively expect an ingame wind system to act exactly as I described it - the overkill things you mentioned would be entirely lost detail IMO but at a high processing cost. Appealing to more complex simulation = must be better is unhelpful when that complexity is simply invisible, or even unnecessary. Not that either of us can "force" anything :)

[Dwarden 1125]

what I meant is correct low level MP syncing of all weather related events ...

also if you implement things like that, you would want to improve the way things behave in game too

clouds to cast dynamic shadows from Sun and Moon to terrain (thanks to proper MP sync, the position of clouds and the casted shadow is same for everyone)

wind to affect physics of vegetation (bushes and trees) in proper direction (so everyone sees the branches, leaves, bushes, grass affected in same direction)

all particles being affected by the wind (again correct sync is crucial)

water waves size and direction affected by wind thus needs to be in sync for everyone

ye sure, it all looks simple and easy to add ... but it isn't

[metalcraze 290]

Obviously adding proper wind simulation to everything is a lot of work - but adding it only to bullets will be a good start.

Gameplay first, eye candy second.

All or nothing is not a choice.

[NeuroFunker 11]

what I meant is correct low level MP syncing of all weather related events ...

also if you implement things like that, you would want to improve the way things behave in game too

clouds to cast dynamic shadows from Sun and Moon to terrain (thanks to proper MP sync, the position of clouds and the casted shadow is same for everyone)

wind to affect physics of vegetation (bushes and trees) in proper direction (so everyone sees the branches, leaves, bushes, grass affected in same direction)

all particles being affected by the wind (again correct sync is crucial)

water waves size and direction affected by wind thus needs to be in sync for everyone

ye sure, it all looks simple and easy to add ... but it isn't

why not sync bullets first, and anything else later? Who cares much about bushes being affected by the wind for everyone the same, or the waves? Are syncing bushes trees and waves, more significant for gameplay, then the bullets?

[dmarkwick 261]

Obviously adding proper wind simulation to everything is a lot of work - but adding it only to bullets will be a good start.

Gameplay first, eye candy second.

All or nothing is not a choice.

I might guess that particles will come under "quite important" too - for smokeshell viewblock reasons.

[metalcraze 290]

As far as I know smokeshells are already affected by wind, aren't they?

[dmarkwick 261]

As far as I know smokeshells are already affected by wind, aren't they?

Yes they are, but this should mean we should have synchronised wind :)

[Dwarden 1125]

why not sync bullets first, and anything else later? Who cares much about bushes being affected by the wind for everyone the same, or the waves? Are syncing bushes trees and waves, more significant for gameplay, then the bullets?

sure projectiles affected by wind shall be first (but remember the AI problem it involves for aiming), but then someone start argue that wind affects FDM (flight dynamic model) and VDM (vehicle driving model) and SLM (Ship floating model) and ye he is right :)

[dmarkwick 261]

As a slight aside, I notice that the default wind direction change period is entirely too fast (IMO natch ;)) which is manifested mostly by clouds wandering all over the sky.

[Wolfstriked 11]

They are adding wind into Dayz and looks real good so far.

[Tonci87 163]

Hmm, DayZ will have a different Server architecture/Client<->Server relationship. I guess their MMO style approach makes synching easier.

[metalcraze 290]

The video is from mod version though.

Syncing shouldn't be the problem at all. We already have script commands for its direction and speed, all server has to do is send them to all clients if there's no in-built sync (which is how it is in A2). I doubt that official solution will be any more complex considering the same conditions.

Edited April 30, 2013 by metalcraze