descr_projectile.txt research

[Ecthelion]

So in my great personal crusade to make gunpowder units even remotely resemble the actual thing property wise (and thus give a whole new layer to strategy) I've been fooling around with the projectiles file and I've found some interesting things:

Here is the entry for a musket bullet:

projectile musket_bullet

effect bullet_model_set
end_effect bullet_impact_ground_set
end_man_effect man_impact_tiny_set
end_package_effect bullet_impact_wall_set
end_shatter_effect bullet_impact_ground_set
end_shatter_man_effect man_impact_tiny_set
end_shatter_package_effect bullet_impact_wall_set

damage 0
radius 0
mass 0.05
accuracy_vs_units 0.001
min_angle -60
max_angle 70
velocity 60 85
display aimed
effect_only

And here is one for a arrow:

projectile arrow

effect arrows_new_set
end_effect arrow_impact_ground_set
end_man_effect man_impact_tiny_set
end_package_effect arrow_impact_wall_set
end_shatter_effect arrow_impact_ground_set
end_shatter_man_effect man_impact_tiny_set
end_shatter_package_effect arrow_broken_impact_wall_set

effect_offset -1.5
damage 0
radius 0.1
mass 0.05
affected_by_rain
min_angle -75
max_angle 65
velocity 20 48
display aimed
effect_only


The arrow one comes from the original RTW, the musket bullet is listed under the "new" section.

Anyway, the most interesting things are:

affected_by_rain

and

accuracy_vs_units 0.001

I tried putting the rain attribute on the musket, I got a crash

Then I changed the accuracy value to something smaller, and all I got was MORE accuracy. It seems that the closer the accuracy value gets to 1 the LESS accurate it gets, I've test this and it works.

This would be a major step forward were it not for one little thing. For some reason, the missile units do not seem to improve their accuracy with proximity to target, one would think that being closer to a target would improve accuracy, but no, it makes absolutely no difference.

Had anyone found a way around this obvious bit of oversight?

[Re Berengario I]

I'm boggling about thse values for musket projectiles...

max_angle 70

Are they shooting at birds? Muskets have to shot at straight angle (like crossbows) while bows can be used with an arched shot.

And no "affected_by_rain"... untile late 1800 when they enclosed the gunpowder into a metallic shell, muskets (and about every firearm) were HEAVILY affected by rain and humidity. I fail to see an arrow affected by rain unless it's a visibility problem for the archer.

So swapping the things round bring to a CTD... bad news indeed!

[Ecthelion]

More than anything else, I am EXTREMELY disappointed by the accuracy to range thing. This is just ridiculous! It strips a whole layer of tactics out! And I really can't think of a way to mod it back in.

[Valakov]

suddenly have a perverse desire to add the rocket model to a merc gun unit.. shoulder mounted RPG's anyone?

[wohoo]

lol, valakov, i've done that for at least 1-2 weeks now.

all you needed to do is change musket_bullet to something like 'rocket' or 'basilisk_shot' to make your musketeers more... interesting.

After intial round of testing, however, i believe that the code "firey" makes siege weapons more potent in killing units as supposed to have them pinned down and standing up again when they got shot by basilisk cannon ball most of the time. o_O

[Doberman_Pharaoh]

I am pretty interested in the "effect_offset -1.5" line. have you found what it does? it seems that "fiery arrows" have a standard of -0.85 offset - could that be an additional "accuracy" model? what do you know about it?

here is my bodkin arrows from Americas campaign descr_projectile.txt


projectile bodkin_arrow

effect arrows_new_set
end_effect arrow_impact_ground_set
end_man_effect man_impact_tiny_set
end_package_effect arrow_impact_wall_set
end_shatter_effect arrow_impact_ground_set
end_shatter_man_effect man_impact_tiny_set
end_shatter_package_effect arrow_broken_impact_wall_set

effect_offset -1.5
damage 0
radius 0.1
mass 0.05
accuracy_vs_units 0.025
affected_by_rain
min_angle -75
max_angle 65
velocity 20 48
display aimed
effect_only

projectile bodkin_arrow_fiery

flaming bodkin_arrow
effect arrows_fire_new_set
end_effect arrow_flaming_impact_ground_set
end_man_effect man_impact_tiny_set
end_package_effect arrow_flaming_impact_wall_set
end_shatter_effect arrow_flaming_impact_ground_set
end_shatter_man_effect man_impact_tiny_set
end_shatter_package_effect arrow_flaming_broken_impact_wall_set

effect_offset -0.85
damage 0
radius 0.1
mass 0.05
accuracy_vs_units 0.05
affected_by_rain
fiery
min_angle -60
max_angle 65
velocity 20 48
display aimed
effect_only

[alhoon]

I'm not sure but I think the effect offset for the flaming arrow tells the AI where the fire is.

[Condottiere SOG]

I'm boggling about thse values for musket projectiles...

max_angle 70

Are they shooting at birds? Muskets have to shot at straight angle (like crossbows) while bows can be used with an arched shot.

And no "affected_by_rain"... untile late 1800 when they enclosed the gunpowder into a metallic shell, muskets (and about every firearm) were HEAVILY affected by rain and humidity. I fail to see an arrow affected by rain unless it's a visibility problem for the archer.

So swapping the things round bring to a CTD... bad news indeed!

Flax and sinew bowstrings lose tension in wet and humid weather.

[Tyre]

Mongols used a dried bone paste to make their composite bows which would totally dissolve in wet conditions.

[JaM]

Actually if you keep your match dry and burning (that is why musketeers used those wide hats...) then it is possible to fire during rain with exeption of heavy rain. in light rain it would be even better that with no rain, because it will cool down a barrell faster.

And at the other side, rain was a huge enemy of bows and crossbows - composite bows can break (water can damage their structure), or loose its flexibility. string can crack, etc... and regarding balistic effect of rain - bullet will be much less affected by rain and wind than bolts or arrows.

Overally - rain was not ideal fighting condition... but dont forget, that Spanish and Dutch musketeers were able operate in humid Flanders terain during 80 years war.

[Beiss]

display aimed

Eh? Can the arrows be made visible when aiming? I've always wanted this. Been trying to install Real Arrow mod but without success too.

[Heinz]

I began digging into the projectile.txt myself and found out a little more:

radius seems to be the size of the projectile the game uses for collusion calculation. therefore a higher value means the projectile is larger. i actually changed it to about one million, which enabled the arrow to hit even when it should miss by 10ft.

effect_offset describes the place where the arrow appears on the battlefield. if changed to -25 the fired arrows appear a couple yards behind the archers and disappear the same distance short of the target (they still hit, they are just invisible)

projectile delay seems to have no effect at all or maybe i just didn’t notice

display aimed / spin /invert_model_z / particle_trail these seem to be ineffective too, since first adding them to arrows doesn’t change anything and even deleting aimed goes unnoticed. apparently they only work in combination with the right projectile, e.g. spin only with the rocket

accuracy: in order to make archers more/less accurate it seems to work to just add the "accuracy_vs_units" line. it looks like all unites without the line are set to some default value, adding it works perfectly fine.

effect_only this one is a little unresponsive on first glance, but it simply says: "hey engine, there aint no special projectile model, just use the effect from line one to display it." my guess is, that arrows and bolts are only effects, which means they aren't real objects or it is simply a matter of performance, since the models appear to have different levels of detail:
model data/models_missile/missile_ballista_bolt_high.CAS, 40.0
model data/models_missile/missile_ballista_bolt_med.CAS, 80.0
model data/models_missile/missile_ballista_bolt_low.CAS, max
e.g: if you want your archers to shoot ballistabolt, you'll have to delete "effect_only" and add one or more of these model lines instead additionally to changing the mere effect line.


i don't know if any of this is new to anybody around here. i didn't find it in another thread, so i thought it might not hurt to share what little i figured out lately.

[alhoon]

As a note for accuracy, the less the number the greater the accuracy somehow.

[Vaz]

Well, just to add to the research - and adding on to the closer you get to 0 the more accurate you are - If you have 0 as the accuracy, Ranged units won't fire. For the DaC mod, we're having some units with *hopefully* perfect accuracy - and having tested between 0.05 and 0.0000000000001, there is a huge, marked difference. Whereas out of 70 shots, there would have been perhaps 6-7 kills (on open order), there would now be 20-30 kills.

[Beiss]

Do you think there's a limit to how many decimals you can add? I do wonder. How about a negative value?

[Taiji]

I wonder about the value of the difference between 0.001 and 0.0001... I have doubts it would be worth much.

I would expect a very noticeable visual difference between 0.01 and 0.05 at 150m, but actual results would depend on the armor of the target and the attack of the bowmen, and of course the hit-rate.

One point to note would be that, with more direct munitions like bullets, extreme accuracy can cause several missiles to land in the same soldier. When a single bullet kills most of the time this works out as a lot of redundant bullets.

The final thing I can think of that's worth mentioning, 'cantabrian_circle'. The more accurate the weapon (especially bows) the better this formation works. Probably a very fast bow projectile could negate some of this... if it was very fast, so fast that it hit the horse almost the instant it left the bow, then perhaps it could hit them. Archers predict to aim, just like in Doom (you know them monsters that throw the fireballs? brown dudes with spikes? learning the value of strafe?...), which means that they are always shooting at the wrong thing when they fire at a cantabrian circle. Only those with faster projectiles or less than perfect accuracy cope well.

Have you tested this last point Vaz?

[Theodotos I]

Hmm, does this accuracy change as the range gets closer, or is that not possible in Med2?

[Taiji]

Troops seem to me to get more accurate the closer they are... they're certainly more deadly...

I think accuracy is affected by rain, experience and line of sight also.

[Vaz]

Sorry about the length of time it took to answer.

I wonder about the value of the difference between 0.001 and 0.0001... I have doubts it would be worth much.

The main benefits come from when vertical firing. There's a rather noticeable difference between 0.001, and 0.0001 when firing over the heads. It's slight, but it's noticeable. Using this, I lost 150 Soldiers in the time taken to kill 132, with the 0.0001 when firing in arcs overhead at a moving target.

[quote]I would expect a very noticeable visual difference between 0.01 and 0.05 at 150m, but actual results would depend on the armor of the target and the attack of the bowmen, and of course the hit-rate.[quote]
You'd be correct.

[quote]One point to note would be that, with more direct munitions like bullets, extreme accuracy can cause several missiles to land in the same soldier. When a single bullet kills most of the time this works out as a lot of redundant bullets. [quote]
I had noticed this - this was intended to be a limitation factor - only only a single unit with the "perfect" accuracy has BP to offset this, and even then 75 shots resulted in only 50 deaths (63 attack, AP, BP, Gunpowder) when fighting on Grassy Plain.

[quote]The final thing I can think of that's worth mentioning, 'cantabrian_circle'. The more accurate the weapon (especially bows) the better this formation works. Probably a very fast bow projectile could negate some of this... if it was very fast, so fast that it hit the horse almost the instant it left the bow, then perhaps it could hit them. Archers predict to aim, just like in Doom (you know them monsters that throw the fireballs? brown dudes with spikes? learning the value of strafe?...), which means that they are always shooting at the wrong thing when they fire at a cantabrian circle. Only those with faster projectiles or less than perfect accuracy cope well.

Again, a limitation factor against the otherwise overwhelming superior firepower of these archers. However, I can confirm that the "instant" velocity works, with spectacular effects if I may say so as well.

Have you tested this last point Vaz?

[chikubi]

How can mod musketeers shoot fast as spanish dragoons?