Artillery tests

[JMM]

Very interesting...

Generally speaking, the first phase of battle is a duel between the 2 ART...

First of all : I am ready to modify the model if you can give me several examples about behaviours which are wrong in the game.

I don't recall reading of any historical instances where artillery alone was sufficient to defeat an attack, so it can hardly have been a regular occurrence. Here it happened 100% of the time.

Did you read an instance where an Infantry unit tried to attack in front of a company of ART?

Question to Holdit : could you give us the duration of attack.

Leaving aside the question of how powerful artillery is, I find it very interesting that the tipping point for infantry to rout is when casualties reach, or are about to reach, 10%. I wonder if this is one of the basic rules in LG - take 10% casualties without inflicting any in a single attack and your morale drops below the point at which you wlll continue to obey orders.  This seems to be a sensible rule but I wonder how realistic it is? My studies of the English Civil War suggest that troops would break well before getting to 10%, but then both the level of training and the practical combat experience of the troops (at least in the early stages of the war) was way lower than in Napoleonic times.  What do historians of other periods think?

In the game, a unit routs when the casualties are 3% to 7% during an attack... so I think in the very interesting tests made by  Holdit, the casualties continues during the flight.

After several shoots, the visibility decreases and the guns must halt the fire during some minutes.
Each gun chooses its target..

JMM

[HarryInk]

Hmm... some time back JasonC produced some tutorials for playing Russians vs Germans in Combat Mission.  I remember the difference between approaching a machine gun nest with a single platoon vs a company of 3 platoons.  You see the flexibility & power of the larger formation.  It will be interesting to watch the difference between a regimental advance, a brigade advance and a divisional advance.

[Holdit]

I think the tests were useful, and the results were what we'd expect from the way the code is written.  We didn't know how close the infantry could get in a simple frontal assault, and you've found out. 

I'm assuming that a frontal assault the way you've set it up is impossible.  But that's not new.  Your results do tell us HOW impossible it is, which is something we needed to know.

Is there any way to get that lone infantry regiment to advance all the way to the guns and possibly attack them?  I don't know, but it's a good next test.

Consider the limitations of the cannons.  They have a fixed amount of ammunition.  They have a fixed arc they can fire to each side, and don't turn the battery to fire at units outside that arc, far as I know.  After they've fired for a while, their visibility goes down, and they lose effectiveness.  They may even move if this happens.  They have their own morale issues.  Can you exploit any of these?  Is there anything I've missed in this list?  What is the difference in the effectiveness of the various gun sizes?

Does the formation of the attacking infantry have any effect at all?  We need to try them all.  Does sending out skirmishers make any difference?  If line infantry can't do the job, can guards handle it?  Does it help to have the corps commander or the army commander nearby?  Does it help to have other infantry deployed nearby behind the attacking regiment?

Can you give the artillery something else to shoot at while your infantry moves  close enough to be a threat? 

Once you've determined that one regiment can't do the job, if that's the case, what happens if you send two?  Using the best case tactics from the tests above, can you get a second regiment out of the arc of fire and close enough to the guns to attack them?  If two doesn't work, can you do it with three?

Once you've figured out how to defeat that artillery battery, we can analyze the results and decide if artillery is too powerful or not, and if it's too powerful, how much too powerful it is, and figure out what can be done from there.

Hook

All good points, some of which had already occurred to me. I have some homework to do on this, so I'll (try to) refrain from further comment until I have completed it.

Holdit

[Hook]

After several shoots, the visibility decreases and the guns must halt the fire during some minutes.
Each gun chooses its target..

Excellent... these may be the key to defeating artillery with infantry.  I wasn't aware that each gun chose a target individually.  That's great news.

All good points, some of which had already occurred to me. I have some homework to do on this, so I'll (try to) refrain from further comment until I have completed it.

Ok.  I'm keeping an open mind for now, which I hope my post showed.  Once we've got more information we can figure out where we need to go.

BTW, I'd expect cavalry to be a harder to stop than infantry, if for no other reason than you'll get off fewer shots at them as they advance.

Hook

[Holdit]

Very interesting...

Generally speaking, the first phase of battle is a duel between the 2 ART...

First of all : I am ready to modify the model if you can give me several examples about behaviours which are wrong in the game.

Thank you JMM - and if I can't provide examples then I wouldn't expect you to change anything. As I've said before, I have no objection to being proved wrong. I may even prove myself wrong. No problem.

Did you read an instance where an Infantry unit tried to attack in front of a company of ART?

So far, no. And I'm not convinced that I will. Not 1 v 1 anyway.

In the game, a unit routs when the casualties are 3% to 7% during an attack... so I think in the very interesting tests made by  Holdit, the casualties continues during the flight.

Yes they do, but I ignored losses inflicted after the unit began to run away. I did notice that these retreats did start off as "unit flight", but continued artillery fire into the backs of the infantry turned that into a straighforward "rout".

Each gun chooses its target..

Excellent.

Straight away that suggests that attacking with two regiments side-by-side, with their junction centred on the centre of the battery, should halve the fire against each regiment.

Holdit

[CBR]

As I spotted this thread I decided to have another go at the demo with a third pc that actually managed to work.

Using the Polish 8x12pounder battery I got similar results as Holdit posted:

-Line Infantry sent to attack the artillery re-deployed into a massed loose ordered formation and advanced at about 45(49?) meters/minute
-It routed at around the 5% as expected at a distance of about 1200 meters and after the battery had spent 7 ammo of roundshot
-Artillery then started using canister and spent about 5 ammo and produced considerable losses (must be a bug and did not see such behavior when I tried sending multiple units against the artillery)
-On average it seems 12 pounders manage about two casualties per round fired.

I also tried a big artillery v artillery fight at about 1200 meters distance. Within an hour it was over with near 100% loss for one side and the winning side had spent an average of about 30 ammo.

So what to conclude from that?

The best possible test that I know of gave about 50% hits with ricochet rounds at 1500 paces. Other tests I know of were about 20%. If we assume that each hit would cut down 3 men, as the target is a 3 rank line, then real life tests done under perfect conditions would give 0.6 to 1.5 casualties per round and in LG it is about 2.

For counter battery effect then IIRC a gun would on average take out 0.1 to 0.2 enemy guns per hour (something like that as I have not written the stuff down) and in LG is seems more like .7 to 1.0 per hour if I look at that single test I did.

So overall I'd say LG long range artillery fire is at least 5 times better than it should be.


CBR

[Gunner24]

So overall I'd say LG long range artillery fire is at least 5 times better than it should be.

I will leave it to others to say if that's right or wrong, I have no clue, but I hope it's wrong !.

I'm sure I remember reading that on AVERAGE Napoleonic Artillery killed (or killed and wounded) 1.5 men per round.

[Sean E]

Very interesting...

Generally speaking, the first phase of battle is a duel between the 2 ART...

First of all : I am ready to modify the model if you can give me several examples about behaviours which are wrong in the game.

Did you read an instance where an Infantry unit tried to attack in front of a company of ART?

Question to Holdit : could you give us the duration of attack.

In the game, a unit routs when the casualties are 3% to 7% during an attack... so I think in the very interesting tests made by  Holdit, the casualties continues during the flight.

After several shoots, the visibility decreases and the guns must halt the fire during some minutes.
Each gun chooses its target..

JMM

Actually all of Morand’s 1st Division advance during the battle of Borodino was against Raievski’s redoubt under heavy artillery fire but closed with the guns.

Another example although I know from another period was Pickett’s Charge during the battle of Gettysburg. This was an advance of 12,500 men against roundshot, shell, canister and musket fire. They still managed to close with and melee the union position.

Of course this assault failed, but they didn’t withdraw after 3% to 7% casualties.

[CBR]

I will leave it to others to say if that's right or wrong, I have no clue, but I hope it's wrong !.

I'm sure I remember reading that on AVERAGE Napoleonic Artillery killed (or killed and wounded) 1.5 men per round.

That average is not correct. We have stats for ammo consumption from several battles and they are below that average.

It obviously depends on target, distance and formation. It would be something like cavalry, infantry and artillery from highest to lowest kills per round (although cavalry would be in 2 rank formation only it is a higher target and therefore easier to hit) Column, line and skirmisher in same order and obliviously shorter range better than longer range

[CBR]

Of course this assault failed, but they didn’t withdraw after 3% to 7% casualties.

Units in general did not rout when taking only 5% losses. I do not understand where such a low number should come from as there is little historical justification for it. If it was say 25% +/- for situational modifiers it would be a lot closer.

[Gunner24]

Ops, I got it the wrong way around :  

Contemporay and modern analysis has concluded for every one and a half artillery rounds fired on a Napoleonic battlefield, about one casualty would result.

Page 269, The Waterloo Companion : Mark Adkin

So that means 15,000 rounds fire = 10,000 casualities.

[Gunner24]

Units in general did not rout when taking only 5% losses. I do not understand where such a low number should come from as there is little historical justification for it. If it was say 25% +/- for situational modifiers it would be a lot closer.

If it were 5%, were there no Officers there !......I would have thought it would be higher than 5%, but a lot lower than 25 %.

[Ras]

Units in general did not rout when taking only 5% losses. I do not understand where such a low number should come from as there is little historical justification for it. If it was say 25% +/- for situational modifiers it would be a lot closer.

That's my impression from reading literature on the period, too. And I find the constant routing and rallying (in a short time) to be totally annoying. It doesn't seem possible to attack anything with infantry without routing at least once. Also, the advance on an artillery position should be made a bit faster and in a more open formation. It's a pity that you can't order a higher march pace or put the whole regiment of light infantry into open order.

[CBR]

Unfortunately Mark Adkin does not provide a source for that statement. Also bear in mind that it includes canister fire, short range fire as well as fire against columns, yes even overshooting of ricochets hitting formations in the rear of the target. The grand battery were located IIRC 800 meters from the Anglo-Allied line.

The casualties I saw in my tests were done against line formations (although they changed into that loose formation instead when rounds started hitting them) and at 1200 meters or more.

[Hook]

So overall I'd say LG long range artillery fire is at least 5 times better than it should be.

If you dropped long range artillery fire by 80%, the result would be that it would be pretty much totally ineffective.  Halting for artillery preparation would be the height of folly.  Regimental guns would be detrimental, as they wouldn't have enough effect to bother with and they'd slow down the regiment.

Small changes here have large results on the battlefield.  You could still ignore enemy artillery with a 50% reduction.  20% would be more reasonable.

However, I don't believe it's the strength that's the problem here.  Long range fire seems quite accurate to me, especially against moving troops.  I'm not sure how accuracy is calculated, but I know individual guns are tracked as well as individual cannonballs, and we see dirt sprayed up where the balls hit.  Even the smallest change to the accuracy of the individual guns here would have major effects, so caution is needed.  If the accuracy is adjusted, the strength doesn't need to be, especially if every round doesn't result on a hit on a stationary target after it's gotten the first hit.

The other part of this equation is the morale effects on the target and what actions they should take under various conditions.

Long range counter-battery fire hits more than just the physical cannons. It also takes out crew and support troops.  You can render a cannon hors-de-combat just as easily by killing the crews as destroying the physical cannon, although the only effect we see is destroyed cannons, which I would consider an acceptable abstraction.  An accuracy reduction would make a difference here as well.

Hook