Is Barrage broken?
- Thread starter OTAKUjbski
- Start date
slobberinbear
Ursine Skald
I'd say anything over 50%. But I'm pretty aggressive. If the odds are worse and I'm out of siege units, I'll send in some cannon fodder (green troops) to soften up the tough defender(s) until the odds improve.
Gliese 581
Your average civ junkie
When I attack with CR: It depends on how many I have.. two dozen can last a while. 
There are few times that I use the barrage promotion. But when I use it I find it very useful.
1. On a catapult that is attacking an approaching SOD. Should be barrage2 for the vs melee bonus. Usually this is when advancing enemy are within striking range of my mounted troops. It damages the spears/pikes and gives my mounted a better chance to survive. Thus damaging all the enemy siege units.
2. When a treb/cannon/artillry has been around for a bit it might live to get a 4th promotion. A little extra damage doesn't hurt. And usually if it has survived that long, many others haven't so i might be running low on siege.
3. Intercontinental invasions when I only have the first city or two and am limited in the number of bombers I can base because I do not have enough cities or an airport. Or if I have split off my attack stacks and share the bombers for reducing city defenses. I will typically assign 1 barrage promotion and the rest CR promotions.
4. tanks facing large stacks of antitank infantry get drill1 pinch barrage and suicide themselves to damage as many of the antitanks as possible. Unless there is a huge mob of antitank infantry usually I stop those attacks when retreat odds exceed 60%. Or not, depending on how the RNG has been screwing me.
But for city assaults, it's CR all the way when available.
1. On a catapult that is attacking an approaching SOD. Should be barrage2 for the vs melee bonus. Usually this is when advancing enemy are within striking range of my mounted troops. It damages the spears/pikes and gives my mounted a better chance to survive. Thus damaging all the enemy siege units.
2. When a treb/cannon/artillry has been around for a bit it might live to get a 4th promotion. A little extra damage doesn't hurt. And usually if it has survived that long, many others haven't so i might be running low on siege.
3. Intercontinental invasions when I only have the first city or two and am limited in the number of bombers I can base because I do not have enough cities or an airport. Or if I have split off my attack stacks and share the bombers for reducing city defenses. I will typically assign 1 barrage promotion and the rest CR promotions.
4. tanks facing large stacks of antitank infantry get drill1 pinch barrage and suicide themselves to damage as many of the antitanks as possible. Unless there is a huge mob of antitank infantry usually I stop those attacks when retreat odds exceed 60%. Or not, depending on how the RNG has been screwing me.
But for city assaults, it's CR all the way when available.
Artichoker
Emperor
- Joined
- Dec 21, 2007
- Messages
- 1,711
First of all, I congratulate you on your thorough effort...
But your comparisons are biased.
The key problems with your comparisons are:
1) The Trebuchet has a base strength of 4, which is used to compute collateral damage. But this is insignificant compared to the Longbowman's strength of 6. If you use a Combat I Catapult, you might get better results.
2) The Artillery's base strength is 18, which is insignificant compared to Mechanized Infantry's base strength of 32. What you should have done was use Infantry as the defending unit.
A much better comparison would be to take Cannons and Riflemen and perform the same tests. I bet you the results will look better. Also, give CR promotions to either Cannons, Grenadiers, or Riflemen to see the different results.
OTAKUjbski
TK421
I think you misunderstood the numbers.
A CR3 Maceman has 28% Combat Odds versus a healthy Drill I, CG3 Longbowman.
A CR3 Trebuchet has 74.7% Combat Odds and will do 8 HPs of collateral damage (5.5/6.0) -- increasing the CR3 Maceman's Combat Odds to 57.6%.
A B3 Trebuchet has 18.6% Combat Odds and will do 11 HPs of collateral damage (5.3/6.0) -- increasing the CR3 Maceman's Combat Odds to 59.9%.
----
The increase in combat odds to the Maceman is not +18% ... it's only +2.3%!!!
Sure ... now that I understand how collateral damage works.
----
I don't have time tonight to do complete tests, but I did rerun the above comparison with Catapults instead of Trebuchets, and the results are the same ... a trivial increase in Maceman odds in exchange for a drastic difference in Catapult survival:
A CR3 Maceman has 28% Combat Odds versus a healthy Drill I, CG3 Longbowman.
A CR3 Catapult has 18.6% Combat Odds and will do 9 HPs of collateral damage (5.5/6.0) -- increasing the CR3 Maceman's Combat Odds to 58.3%.
A B3 Catapult has 1.1% Combat Odds and will do 12 HPs of collateral damage (5.3/6.0) -- increasing the CR3 Maceman's Combat Odds to 60.7%.
So even when using Catapults instead of Trebuchets, the increase in the CR3 Maceman's Combat Odds is still only a trivial +2.4% at the expense of a -17.5% difference in the Catapult's Combat Odds.
----
And with a Cannon:
A CR3 Maceman has 28% Combat Odds versus a healthy Drill I, CG3 Longbowman.
A CR3 Cannon has 93.1% Combat Odds and will do 14 HPs of collateral damage (5.2/6.0) -- increasing the CR3 Maceman's Combat Odds to 61.9%.
A B3 Cannon has 53.1% Combat Odds and will do 18 HPs of collateral damage (4.9/6.0) -- increasing the CR3 Maceman's Combat Odds to 64.8%.
So even when using a Cannon instead of a Catapult, the increase in the CR3 Maceman's Combat Odds is yet still only a trivial +2.9% at the expense of a -40% difference in the Catapult's Combat Odds.
vicawoo
Chieftain
- Joined
- Feb 12, 2007
- Messages
- 3,226
I think you misunderstood the numbers.
A CR3 Maceman has 28% Combat Odds versus a healthy Drill I, CG3 Longbowman.
A CR3 Trebuchet has 74.7% Combat Odds and will do 8 HPs of collateral damage (5.5/6.0) -- increasing the CR3 Maceman's Combat Odds to 57.6%.
A B3 Trebuchet has 18.6% Combat Odds and will do 11 HPs of collateral damage (5.3/6.0) -- increasing the CR3 Maceman's Combat Odds to 59.9%.
----
The increase in combat odds to the Maceman is not +18% ... it's only +2.3%!!!
Sure ... now that I understand how collateral damage works.
----
I don't have time tonight to do complete tests, but I did rerun the above comparison with Catapults instead of Trebuchets, and the results are the same ... a trivial increase in Maceman odds in exchange for a drastic difference in Catapult survival:
A CR3 Maceman has 28% Combat Odds versus a healthy Drill I, CG3 Longbowman.
A CR3 Catapult has 18.6% Combat Odds and will do 9 HPs of collateral damage (5.5/6.0) -- increasing the CR3 Maceman's Combat Odds to 58.3%.
A B3 Catapult has 1.1% Combat Odds and will do 12 HPs of collateral damage (5.3/6.0) -- increasing the CR3 Maceman's Combat Odds to 60.7%.
So even when using Catapults instead of Trebuchets, the increase in the CR3 Maceman's Combat Odds is still only a trivial +2.4% at the expense of a -17.5% difference in the Catapult's Combat Odds.
----
And with a Cannon:
A CR3 Maceman has 28% Combat Odds versus a healthy Drill I, CG3 Longbowman.
A CR3 Cannon has 93.1% Combat Odds and will do 14 HPs of collateral damage (5.2/6.0) -- increasing the CR3 Maceman's Combat Odds to 61.9%.
A B3 Cannon has 53.1% Combat Odds and will do 18 HPs of collateral damage (4.9/6.0) -- increasing the CR3 Maceman's Combat Odds to 64.8%.
So even when using a Cannon instead of a Catapult, the increase in the CR3 Maceman's Combat Odds is yet still only a trivial +2.9% at the expense of a -40% difference in the Catapult's Combat Odds.
I misread this:
#
# Damaged once by lvl1: 5.5, 57.6%
# Damaged once by B1: 5.5, 57.6%
# Damaged once by B2: 5.4, 59.1%
# Damaged once by B3: 5.3, 59.9%
# Damaged twice by lvl1: 5.0, 63.4%
# Damaged twice by B1: 5.0, 63.4%
# Damaged twice by B2: 4.8, 77.2%
# Damaged twice by B3: 4.7, 78.6%
and was comparing barrage 3 1 hit versus barrage 3 2 hit. Anyway, the difference between 2 barrage 3's collateral versus 2 city raider trebs is a 15.2% difference. I'm also wondering, since barrage 3 should be +100% total, why the longbow isn't down 1 strength per hit. Shouldn't it be barrage 1 to be +20%, barrage 2 to be 50% total instead of 30%, and 3 to be 100% instead of +50%?
siggboy
Monarch/Epic
Fair enough, but if the siege has done a thorough job most defenders will be so low in HP that you don't even need counter-promotion micromanagement to mop up the defenders at overwhelming odds.
Also, due to the paper-scissors-stone nature of Civ IV warfare I strongly doubt that juggling with counter-promotions gives better results than just going in with City Raiders (except for some special cases like Landsknechte which can be jack-of-all-trades).
I think Roxlimn's point is that the cleanup troops will have minimal casualties anyway, so they might as well be promoted for optimal stack protection rather than cleanup efficiency.
Since after the first 1 or 2 rounds of siege, the defenders should be too damaged to be much of a threat, something that improves initial survivability seems best... hence CR for every siege unit.
This is vastly different for tanks, which usually have decent odds as it is. Here, softening up the enemy stack seems more important... and for cleanup, I prefer Drill. City Raider is reserved for can openers that need to crack especially well defended cities.
Since after the first 1 or 2 rounds of siege, the defenders should be too damaged to be much of a threat, something that improves initial survivability seems best... hence CR for every siege unit.
This is vastly different for tanks, which usually have decent odds as it is. Here, softening up the enemy stack seems more important... and for cleanup, I prefer Drill. City Raider is reserved for can openers that need to crack especially well defended cities.
JBossch
King
Exactly. The math geeks don't seem to understand that cannons wreck LB's. Spearmen have no problem mopping up the scraps. The same applies to any generational advantage in siege. Why debate irrelevant details like "barrage vs. CR?" Most high-level players seem to hold off on promotions until a decision is necessary. (Barrage vs. units in field, as softeners, etc.; CR vs. fortified city garrisons) Problem solved?
TheMeInTeam
If A implies B...
- Joined
- Jan 26, 2008
- Messages
- 27,989
In theory, but I bet the thread continues
. I'll contribute a bit:I've tried some crazy stuff at monarch/emperor and liked one strategy: CR siege with drill units. As you pointed out, siege, especially when up a tech generation, destroys defenders. You can make a little bit extra in siege and run drill, where high levels of drill drastically lower the odds of taking any meaningful damage. This plus a medic III type unit generally keeps an army marching after one turn of healing...pretty effective. The limiting factor is when you run out of siege, so promo allocations of siege matters there. I liked CR siege in my runs of this, because it survives a lot better (and siege that survives can heal in the captured city and be recycled to the front line once healed).
The above is especially relevant for protective, which is otherwise a weak trait in many situations. Crossbows are underestimated in medieval times and with drill promos can cover everything but knights...drill not only gives them strong odds against melee but also a high chance of coming out in good shape. CR siege to soften defense means that you can even break longbows without too much damage.
IMO CR is the most superior siege promotion in most cases, but as you've already pointed out, strength is better in the field and barrage is better as the AI opposing stack count grows. I don't know where the cutoff is but if I'm facing a stack of 50 knights with 12 cannons and some other units I'd bet barrage performs a bit better than in the math posts shown here!
This thread shows what many have known for a long time. Barrage is essentially a useless promotion for siege. I have 2 simple rules that seem to work extremely well and cover all likely situations that seige is needed for.
a) attacking a city, use CR line of promotion
b) attacking a stack outside a city (or fort), use combat line of promotion
I usually have half my siege promoted with CR and half with combat. Early on catapults only get the combat promotion and trebs get the CR for obvious reasons, but later cannons and artillery are promoted half and half.
Against a city the CR siege goes first and the combat siege follow against the already collateral damaged defenders. Just as other people have found with the ground troops that wipe out the shattered remains it doesn't really matter what is used after defenders are damaged. Once the first few specialised seige have done some collateral damage the ones that follow on don't need to be specialised or highly promoted. Against a target outside a city then the combat siege are used first and then the CR siege are used if more collateral is needed.
Barrage promoted siege just don't have the survival chances of the CR + combat double act. With cannons I find my combat promoted ones are so successful and survive so well that they get to 17 exp and I give them the March promotion for faster healing. It works a treat.
a) attacking a city, use CR line of promotion
b) attacking a stack outside a city (or fort), use combat line of promotion
I usually have half my siege promoted with CR and half with combat. Early on catapults only get the combat promotion and trebs get the CR for obvious reasons, but later cannons and artillery are promoted half and half.
Against a city the CR siege goes first and the combat siege follow against the already collateral damaged defenders. Just as other people have found with the ground troops that wipe out the shattered remains it doesn't really matter what is used after defenders are damaged. Once the first few specialised seige have done some collateral damage the ones that follow on don't need to be specialised or highly promoted. Against a target outside a city then the combat siege are used first and then the CR siege are used if more collateral is needed.
Barrage promoted siege just don't have the survival chances of the CR + combat double act. With cannons I find my combat promoted ones are so successful and survive so well that they get to 17 exp and I give them the March promotion for faster healing. It works a treat.
Iranon:
It's a little bit of both, actually. Much of the stack should be composed of siege weaponry and anti-cavalry units like counter-cavalry (love those Cossacks) and Formation Pikes or Rifles. Since I don't really need CR, I prioritize stack defense with non-siege units, just as you said.
That said, a substantial part of the stack is unpromoted so as to save the promos for attacking the city. This makes my stack defense units better because these units don't defend until after the rest of my stack defenses (which ideally should be enough to absorb all attacks). When necessary, you promote these guys with anti-promos as needed for additional stack defense on the field. I've had occasion to promote a Pikeman with Shock to advantage.
In addition to that, once you're mopping up, stack defense manipulation can allow you to save non CR siege from having to attack at bad odds. If, for instance, you're at that point where you've run out of CR siege for shock attacks and have bad odds for your Combat siege, you can try out the odds of your mounteds and see how that goes, just for eliminating the few tougher defenders remaining before your siege gets to work once again. If you expect your mounteds to do well on shock attacks but have a Pikeman in the way, a Cover promoted Maceman could just edge that Pikeman over to the point where he's the toughest defender, which would still generate favorable odds, of course, and later on open up the target stack to mounteds to eliminate the slightly tougher LBs, followed by siege.
I call this process "picking out the Pikeman," and is so effective that I had simply figured that everyone around here did that as a matter of course. The key concept here, of course, is that the mounteds starting from the Medieval Knight unit are exceedingly powerful units with only 1 unit weakness - Pikemen. HAs can also be used earlier, of course, but pikeman picking is more effective later on. As long as you can selectively eliminate Pikemen from an enemy stack, your Knights should be able to make short work of them.
So how do you do that? CR or Combat Cannons go first to level collateral damage across the entire front line up. You don't want those guys to be healthy enough to present themselves as "best defenders." If significant opposition is left over, you use Cover Macemen or Cover Musketmen to fish out the Pikemen to render them useless as counterunits, hopefully killing them in the process. Macemen are better at killing Pikemen, but Musketmen are better at fishing them out from the bottom of the defender stack.
Once the enemy Pikemen are useless, go to town with Knights. Or Cuirassers. Or Cavalry.
It's a little bit of both, actually. Much of the stack should be composed of siege weaponry and anti-cavalry units like counter-cavalry (love those Cossacks) and Formation Pikes or Rifles. Since I don't really need CR, I prioritize stack defense with non-siege units, just as you said.
That said, a substantial part of the stack is unpromoted so as to save the promos for attacking the city. This makes my stack defense units better because these units don't defend until after the rest of my stack defenses (which ideally should be enough to absorb all attacks). When necessary, you promote these guys with anti-promos as needed for additional stack defense on the field. I've had occasion to promote a Pikeman with Shock to advantage.
In addition to that, once you're mopping up, stack defense manipulation can allow you to save non CR siege from having to attack at bad odds. If, for instance, you're at that point where you've run out of CR siege for shock attacks and have bad odds for your Combat siege, you can try out the odds of your mounteds and see how that goes, just for eliminating the few tougher defenders remaining before your siege gets to work once again. If you expect your mounteds to do well on shock attacks but have a Pikeman in the way, a Cover promoted Maceman could just edge that Pikeman over to the point where he's the toughest defender, which would still generate favorable odds, of course, and later on open up the target stack to mounteds to eliminate the slightly tougher LBs, followed by siege.
I call this process "picking out the Pikeman," and is so effective that I had simply figured that everyone around here did that as a matter of course. The key concept here, of course, is that the mounteds starting from the Medieval Knight unit are exceedingly powerful units with only 1 unit weakness - Pikemen. HAs can also be used earlier, of course, but pikeman picking is more effective later on. As long as you can selectively eliminate Pikemen from an enemy stack, your Knights should be able to make short work of them.
So how do you do that? CR or Combat Cannons go first to level collateral damage across the entire front line up. You don't want those guys to be healthy enough to present themselves as "best defenders." If significant opposition is left over, you use Cover Macemen or Cover Musketmen to fish out the Pikemen to render them useless as counterunits, hopefully killing them in the process. Macemen are better at killing Pikemen, but Musketmen are better at fishing them out from the bottom of the defender stack.
Once the enemy Pikemen are useless, go to town with Knights. Or Cuirassers. Or Cavalry.
lol I haven't played Civ for a While... I've been so busy but from memory I also tend to Promote down the CR line with Seige units becuase of higher Survival rates and to counter the tough CG defenders.
I've wondered why the AI doesn't mix up the promotions a little bit, have a combination of CG and Drill Promoted Defenders in a City, Drill Minimizes/decreases Collateral Damage Correct?
So the AI would have the CG defender counter the the CR Attacker and the Drill units counter the Collateral Damage so that may force you to mix up your seige promotions but I'm only talking theory...
I'm too lazy to test it out. Besides the mixed combintion of Drill and CG defenders would probably only be useful with Protective leaders, and possibly Charismatic Leaders.
Can someone test this out for me, I'm too lazy, Can a Unpromoted seige unit cause collateral Damage to a Drill 4 Defender? and if they do cause collateral damge, how much?
I've wondered why the AI doesn't mix up the promotions a little bit, have a combination of CG and Drill Promoted Defenders in a City, Drill Minimizes/decreases Collateral Damage Correct?
So the AI would have the CG defender counter the the CR Attacker and the Drill units counter the Collateral Damage so that may force you to mix up your seige promotions but I'm only talking theory...
I'm too lazy to test it out. Besides the mixed combintion of Drill and CG defenders would probably only be useful with Protective leaders, and possibly Charismatic Leaders.
Can someone test this out for me, I'm too lazy, Can a Unpromoted seige unit cause collateral Damage to a Drill 4 Defender? and if they do cause collateral damge, how much?
AndrewN
Prince
This is an interesting question, and I would like to thank the OP for bringing it up. First of all I would note that I have always promoted my siege units with barrage.
However this thread got me thinking so I have done a quick and dirty test using worldbuilder to see if barrage is the correct route to take.
In the first set of experiments I used Catapults vs a number of Swordsmen (2 to 8) all on flat grassland. A Catapult will damage the top defender and cause collateral damage to up to 6 further defenders. If there are more than 7 defenders in the stack the remaining will be undamaged.
With an unpromoted Catapult, I did 0.5 collateral damage to the Swordsmen in the stack, that is the damaged Swords had strength 5.5
With a Barrage I Catapult, I did 0.6 collateral damage to the Swordsmen in the stack
With a Barrage III Catapult (it had Barrage I and II as well, I didn't try Barrage I and II on their own), I did 0.7 damage to the Swordsmen in the stack. A straight reading of the description for Barrage III (+100% collateral damage) would indicate collateral damage of 1 point not 0.7, but without knowing the maths behind the calculations I can't definitely say it is a bug.
Next I tried a variety of siege weapons against Swordsmen.
Unpromoted Catapult does 0.5 collateral.
Unpromoted Treb does 0.5 collateral; I assume this is a rounding problem
Unpromoted Cannon does 0.8 collateral.
Unpromoted Artillery does 1.0 collateral.
Unpromoted Mobile Artillery does 1.1 collateral
Artillery promoted with Barrage I, II and III does 1.3 collateral
Mobile Artillery promoted with Barrage I, II and III dose 1.4 collateral.
I didn't try and combat with a city, but I assume the collateral damage figures will be similar.
This is more than enough to convince me that the Barrage line of promotions is a waste of time. Using the City Raider line so that more Siege weapons survive is probably more useful a small amount of extra damage. For Catapults vs Swords 1 extra unpromoted Catapult will do more damage than 2 Catapults promoted to Barrage III, so having a few more survive is useful.
The downside is you cant use a single line of promotions fore both city work and open field, but as most of the large stacks will be found defending cities this is not too serious a problem.
However this thread got me thinking so I have done a quick and dirty test using worldbuilder to see if barrage is the correct route to take.
In the first set of experiments I used Catapults vs a number of Swordsmen (2 to 8) all on flat grassland. A Catapult will damage the top defender and cause collateral damage to up to 6 further defenders. If there are more than 7 defenders in the stack the remaining will be undamaged.
With an unpromoted Catapult, I did 0.5 collateral damage to the Swordsmen in the stack, that is the damaged Swords had strength 5.5
With a Barrage I Catapult, I did 0.6 collateral damage to the Swordsmen in the stack
With a Barrage III Catapult (it had Barrage I and II as well, I didn't try Barrage I and II on their own), I did 0.7 damage to the Swordsmen in the stack. A straight reading of the description for Barrage III (+100% collateral damage) would indicate collateral damage of 1 point not 0.7, but without knowing the maths behind the calculations I can't definitely say it is a bug.
Next I tried a variety of siege weapons against Swordsmen.
Unpromoted Catapult does 0.5 collateral.
Unpromoted Treb does 0.5 collateral; I assume this is a rounding problem
Unpromoted Cannon does 0.8 collateral.
Unpromoted Artillery does 1.0 collateral.
Unpromoted Mobile Artillery does 1.1 collateral
Artillery promoted with Barrage I, II and III does 1.3 collateral
Mobile Artillery promoted with Barrage I, II and III dose 1.4 collateral.
I didn't try and combat with a city, but I assume the collateral damage figures will be similar.
This is more than enough to convince me that the Barrage line of promotions is a waste of time. Using the City Raider line so that more Siege weapons survive is probably more useful a small amount of extra damage. For Catapults vs Swords 1 extra unpromoted Catapult will do more damage than 2 Catapults promoted to Barrage III, so having a few more survive is useful.
The downside is you cant use a single line of promotions fore both city work and open field, but as most of the large stacks will be found defending cities this is not too serious a problem.
The CR promotion also opens up Accuracy. In the very early game I will sometimes promote one of my CR1 catapults to CR1, Acc if it's lucky enough to get to 5 exp. That doubles its effectiveness at lowering cultural defences or walls and that's useful when you only have 3 or 4 catapults. Later in the game when trebs are available there is no need for Accuracy as your stack of seige should be big enough to lower defences in 1 or 2 turns anyway, if it can't, why are you attacking?
