[Mechanics of the game] Physics of arrows

[Cheap]

Hi all,


When watching the Teutoburg walkthrough, I loved to see that arrows seem to have a new kind of physics, as they either get stuck in shield/flesh or they rebound on shields and armor.


Does anyone have some details about the mechanics under it? Does it react on zones differently, is it random whether it rebound/get stuck or does it depends on several factors like angle of arrival, distance, arrow type (eastern/western design, wooden/stone/iron head, flaming or not)?


And in the case of the Macedonian phalanx, will they rebound on sarissas also? Will it be possible to tighten the ranks depending on the context or will the phalangites tighten automatically as do the Roman with the testudo?


In RTW it seems that there was some kind of "halo" around the pike wall which made arrows weaker or sometimes disappear in flight, but it was not linked to the contact with the sarissas.
If anyone has any detail on it I'd love to read about it

[torongill]

RTW's phalanx formation had increased defense against projectiles.

Jack Lusted has stated that arrows stuck in killed characters. That's all he has said, so maybe if the arrow doesn't kill it will bounce back from the character or his shield.

I cannot answer to the other questions, but I can add something - in Rome 2 the higher ground won't add range, because range is static and can be seen by the "rainbow" of the range. However projectiles shot/thrown downhill will have cause more damage than those on even ground or uphill.

[B-DizL]

I doubt its that complex. there are probably only a few animations for the arrows depending on the situation but i doubt they take into account angle and speed etc. It either sticks or doesnt.

[|Sith|Galvanized Iron]

They stuck in Shogun 2 also...

[Cheap]

In a conference in late 2012, James Russell said they were heading toward a more simulation-oriented physical engine that would allow less approximation and more simulation, I wanted to know how it evolved for R2:TW.
For example, a guy hidden in rocks can be either protected by the collision of the missiles with the rocks, or by a bonus given in the database allowing his "defence points" to be raised.

In the example of last Jack Lusted's exclusive content thread, we know that if a missile coming from front or left of the unit is currently considered hitting the shield and thus, it's damages are reduced.

But maybe this could be more precise and the area covered by the shield could be fully protected against missiles while a bit under and a bit over the shield, it would be completely uncovered by it. If the trajectory is precisely calculated, then the engine knows if the arrow hits the shield, the soldier (armour or leather joints or flesh).
And the same for sarrisas: will there be collision detection with them?

I wonder how hard it can be to develop and how CPU consuming it can be. Is it feasable nowadays for a RTS?

[Ichon]

I cannot answer to the other questions, but I can add something - in Rome 2 the higher ground won't add range, because range is static and can be seen by the "rainbow" of the range. However projectiles shot/thrown downhill will have cause more damage than those on even ground or uphill.

Where was that stated? Disappointing if true.

In a conference in late 2012, James Russell said they were heading toward a more simulation-oriented physical engine that would allow less approximation and more simulation, I wanted to know how it evolved for R2:TW.
For example, a guy hidden in rocks can be either protected by the collision of the missiles with the rocks, or by a bonus given in the database allowing his "defence points" to be raised.

In the example of last Jack Lusted's exclusive content thread, we know that if a missile coming from front or left of the unit is currently considered hitting the shield and thus, it's damages are reduced.

But maybe this could be more precise and the area covered by the shield could be fully protected against missiles while a bit under and a bit over the shield, it would be completely uncovered by it. If the trajectory is precisely calculated, then the engine knows if the arrow hits the shield, the soldier (armour or leather joints or flesh).
And the same for sarrisas: will there be collision detection with them?

I wonder how hard it can be to develop and how CPU consuming it can be. Is it feasable nowadays for a RTS?

It seemed Russel was hinting at something more advanced but with newer details it seems it is an improvement but not quite so huge. Probably it will be simply bonuses or penalties to hit or damage in the calculation. Not actual physics of flight where you can follow arrow leaving bow and embedding into a shield which in reality with 100s of arrows at the same time would be quite a feat.

Hopefully there is more varied types of vegetation providing cover- from thick forest making arrow shots nearly impossible to hit to light brush or thin forest where arrows get a penalty to hit but can still cause damage over time.

[Sar1n]

Where was that stated? Disappointing if true.

http://www.twcenter.net/forums/showt...tal-War-Center

In last update, Jack gave some details about combat in R2, including the bit about ranged combat. Disappointing indeed.

[Jack Lusted]

The deflecting of arrows/them getting stuck in shields is done in a very simple manner. If an arrow hits someone but does not kill it has a chance to deflect or stick in the shield if it hits near the shield bone on the units skeleton.

Projectiles are fully simulated, we do calculate their accuracy and trajectory before they are fired and they will follow that once in the air, and can be blocked by trees/buildings.

[Ichon]

The deflecting of arrows/them getting stuck in shields is done in a very simple manner. If an arrow hits someone but does not kill it has a chance to deflect or stick in the shield if it hits near the shield bone on the units skeleton.

Projectiles are fully simulated, we do calculate their accuracy and trajectory before they are fired and they will follow that once in the air, and can be blocked by trees/buildings.

Interesting info there, thanks.

Is accuracy still a radius from edge of target unit or something different now? Even if calculating the radius from center of target unit that might be improvement but I'm not sure how that would affect accuracy vs moving units.

BTW- is there ever any chance of more detailed explanations of the engine after launch?

[Jack Lusted]

Interesting info there, thanks.

Is accuracy still a radius from edge of target unit or something different now? Even if calculating the radius from center of target unit that might be improvement but I'm not sure how that would affect accuracy vs moving units.

I'm not sure what you are asking there.

[Ichon]

I'm not sure what you are asking there.

Unless I've misunderstood earlier TW in warscape the accuracy was a measurement from the front facing edge of a targeted unit with a radius around that edge that was wider with lower accuracy and narrower with higher accuracy so more arrows would tend to hit with the higher accuracy. If unit was moving however it was easy to dodge the arrows due to this as well as the arrow speed vs unit speed.

Unless you are saying individual skeletons are targeted in a formation by arrows from individuals in the archer formation? In that case dodging even vs slow arrows would be significantly more difficult as the arrow volley is far more diffused over a targeted formation and accuracy is not vs entire unit but on individual skeletons within a unit.

In relation to cover before if most of a unit was under cover but the front edge was not then the cover would not apply to the entire unit since accuracy calculated from the front edge of the unit was not under any cover. If individual skeletons are targeted then part of the unit under a tree is protected completely (unless the to hit calculation is merely altered by some amount reflecting cover making hit lower chance but not impossible). Compared to another part of the unit not behind trees but standing in open since accuracy is per individual skeleton those in the open are likely suffering much higher rate of hits. It would be more realistic this way and encourage more careful use of terrain features.

[Jack Lusted]

Unless I've misunderstood earlier TW in warscape the accuracy was a measurement from the front facing edge of a targeted unit with a radius around that edge that was wider with lower accuracy and narrower with higher accuracy so more arrows would tend to hit with the higher accuracy. If unit was moving however it was easy to dodge the arrows due to this as well as the arrow speed vs unit speed.

Unless you are saying individual skeletons are targeted in a formation by arrows from individuals in the archer formation? In that case dodging even vs slow arrows would be significantly more difficult as the arrow volley is far more diffused over a targeted formation.

Individuals will target individuals in an enemy unit when firing projectiles, and accuracy is worked out based on distance and values of the projectile etc. Simple path projection is also used based on the current speed and direction of the unit being fired at. Of course once the projectile has been fired there is no guarantee that it will hit the individual it was fired at, and combat calculations are done based on when a projectile hits an enemy.

[Ichon]

Individuals will target individuals in an enemy unit when firing projectiles, and accuracy is worked out based on distance and values of the projectile etc. Simple path projection is also used based on the current speed and direction of the unit being fired at. Of course once the projectile has been fired there is no guarantee that it will hit the individual it was fired at, and combat calculations are done based on when a projectile hits an enemy.

Awesome- appreciate your taking time to answer this! It makes sense and should be interesting differences from previous TW games especially with LoS in sieges.

Ohdamn- the one question I've had about LoS totally forgot while had your attention! Can cavalry or elephants or even siege towers (anything large) block LoS? So putting archers behind friendly cavalry would lower/stop those archers from targeting certain enemy units? Or disguise some units behind? Particularly the siege works evident in the Carthage video made me wonder about this and walls/buildings in sieges affects on LoS.

[Jack Lusted]

In relation to cover before if most of a unit was under cover but the front edge was not then the cover would not apply to the entire unit since accuracy calculated from the front edge of the unit was not under any cover. If individual skeletons are targeted then part of the unit under a tree is protected completely (unless the to hit calculation is merely altered by some amount reflecting cover making hit lower chance but not impossible). Compared to another part of the unit not behind trees but standing in open since accuracy is per individual skeleton those in the open are likely suffering much higher rate of hits. It would be more realistic this way and encourage more careful use of terrain features.

Trees and stuff will block projectiles, so having a unit in trees will reduce the amount of projectiles that will hit them.

Awesome- appreciate your taking time to answer this! It makes sense and should be interesting differences from previous TW games especially with LoS in sieges.

The individual targeting is how it has been done since Empire.

Ohdamn- the one question I've had about LoS totally forgot while had your attention! Can cavalry or elephants or even siege towers (anything large) block LoS? So putting archers behind friendly cavalry would lower/stop those archers from targeting certain enemy units? Or disguise some units behind? Particularly the siege works evident in the Carthage video made me wonder about this and walls/buildings in sieges affects on LoS.

The LoS system is still being worked on and balanced but units do not block you from seeing units behind them currently.

[torongill]

There was some information a while ago in a Spanish game magazine that the higher the difficulty, the less can be seen. One example was that on the hardest difficulty enemy units don't show on the mini-map or the tactical map at all. But it would be nice if for example on hardest difficulty you can see only the first enemy line, unless you have a unit that is at elevation.
I am nor explaining it very well, my point is that there are quite a few historical examples(it was in fact in Xenophon's book on cavalry iirc) of generals using their cavalry as a screen behind which to hide infantry. Caesar won at Pharsalus in that way. But I've read the reverse has been done - that Alexander would hide his cavalry behind the phalanx, taking advantage of the elevated sarissas and the dust going up from the movement of the units. Oh well, this can be simulated even on medium level - you just need to use "general's view"

[|Sith|Galvanized Iron]

Then hopefully archers firing out from trees will also get same penalty this time?

[Jack Lusted]

Then hopefully archers firing out from trees will also get same penalty this time?

All projectiles can be blocked by trees if they hit them. That has been true since Empire.

[|Sith|Galvanized Iron]

All projectiles can be blocked by trees if they hit them. That has been true since Empire.

Well actually in even in earlier Total Wars arrows were blocked by objects, right? I remember how the central flag in RTW barbarian settlement worked partly as a shield against arrow fire.

[Mithridates II the Great]

Will arrows be useful against heavy armored units such as cataphracts ?

[Paedric]

What about artillery?
Can the projectiles be blocked by trees too?