Demonstration of the artillery retreating glitch

[Garnier]

Here are some screenshots demonstrating this bug. In the first screenshot, the CS infantry regiment initiated a charge against the selected cannon. Just before the charge reached the gun, I clicked the retreat button on the gun. Instantly the charge stopped, and the cannon began limbering to retreat. The CS troops were able to fire on the retreating gun but were unable to charge to capture it. They did not hit any of the gunners, so the cannon was able to limber and escape without any damage.



The second screenshot shows that the gun was able to retreat 20 yards, then unlimber and continue firing. I repeated the test with the second gun as well, with the same results.



What should happen is the CS infantry should be able to charge the gun even when it is "retreating" as long as it is not yet limbered, and capture it. Otherwise, by using the retreat button, guns can escape capture almost 100% of the time.

Here's a save shortly after these tests. (160kb)

[GShock]

Excellent, Garnier.

[RebBugler]

Good points here and with our discussion earlier today on Vent. There is much to consider to address this issue of unrealistic behavior and outcome.

First, consider that this game is played much different from the MP standpoint. With TC2M and with an AI controlled enemy, this design works fine and is fundamentally realistic...you overrun the guns and take them, but even with AI control, there are almost always a few that automatically retreat and you can not capture them.

On the other hand, with MP, players can hit the retreat button at the last second and save all their guns...agreed, highly unrealistic.

So...What to Do..What to Do

Norb may have a trick or two in his bag of codes, this is the easiest path I see.

Otherwise, I'm thinking a drastic design concept such as having separate gun and caisson sprites. Then the primary way to capture a gun would be to eliminate the caissons with musket fire, this is historic and realistic. Rarely did infantry charge into double canister range when they could more safely eliminate the gun crews and caissons (hate to say it, but, shooting the horses was typical of gun capturing).

Anyway, I know you guys want justice, err...realism. Lets stay open minded and look for a good historically realistic approach here.

[GShock]

What to do is the EASIEST of all things. When servers ROUT the gun must be left in place for capture and the limber and ammo as well. In case of retreat, they limber the gun but the limbering time should be coherent with the number of servers who are working that gun.

Limbering and unlimbering under fire is a hard predicament. Servers can't shoot back, their morale should drop amazingly when taking losses. Hence the rout. Arty is not supposed to retreat easy like this when engaged in ranged combat with infantry... actually I guess they shot one maybe two cannister rounds but then servers would rout even without taking losses.

It's what you would expect anyone but the X-men to do when 15 face 300+ : R O U T.

[Garnier]

What to do is the EASIEST of all things

Splitting a unit that has always acted as a single entity into two entities isn't the easiest thing to do.

[RebBugler]

Agreed, splitting entities would be a drastic and difficult fix, Norb would have a cow if forced to do this. Besides, the game is practically maxed out sprite wise, additional sprites would be last resort.

How About?

Enabling capture as guns are limbering. Player hits retreat at the last second but still can't save guns from capture because they must go through limbering mode before they can retreat to safety.

Don't know if it's possible, but Norb knows, what sayeth you guys? Would this 'just' the cause? ...if Norb can code it.

[Garnier]

Yes that's what I proposed in ventrilo and at the end of my post. I imagine that would be the easiest way to code it, and the simplest way for it to work.

[RebBugler]

Garnier wrote:

Yes that's what I proposed in ventrilo and at the end of my post. I imagine that would be the easiest way to code it, and the simplest way for it to work.

Need more feedback here before it's proposed. Hey AP, you were passionate about this area, what do you think?

Anyone...what are the negatives with this proposal?

[RebBugler]

Hmmm...

Just thought of a BIG negative for SP, not MP. Guns are too of an easy prey, they can be wiped out completely. As is, the AI takes care of this, having some guns retreat while limbering thus saving their hide. Some get captured, some escape, a fair balance.

In order for this proposal to work for both modes, I think artillery would have to automatically limber and retreat when the infantry gets within a certain range. Thereby in SP, some guns will have a chance to escape...keeping that good balance...no wiping out of multiple batteries in a single play.

OK, I found an issue, and possible solution...more coding and BIG artillery behavioral change :woohoo: , anymore?

[Shirkon]

RebBugler wrote:

Hmmm...

Just thought of a BIG negative for SP, not MP. Guns are too of an easy prey, they can be wiped out completely. As is, the AI takes care of this, having some guns retreat while limbering thus saving their hide. Some get captured, some escape, a fair balance.

In order for this proposal to work for both modes, I think artillery would have to automatically limber and retreat when the infantry gets within a certain range. Thereby in SP, some guns will have a chance to escape...keeping that good balance...no wiping out of multiple batteries in a single play.

OK, I found an issue, and possible solution...more coding and BIG artillery behavioral change :woohoo: , anymore?

Unfortunately this wouldn't be historical. Look at Capt Cowan's 1st Battery, New York Light Artillery during Pickett's Charge. He fired double canister at 10 yds then began to retire by prolong. Or the hand to hand fighting in the batteries the evening of the second day when Early's Division attacked though the brickyard against Cemetary Hill. Artillery often fought until the infantry was within the battery itself before attempting to pull out, whether by recoil, prolong or limbering up. I would be against any change that made artillery automatically limber up if the enemy came within a certain range.