Is Barrage broken?

[Alsn]

Fezzik, im not really sure what youre looking for but i conducted a test with CR2 vs B2 cats a few posts up. Not sure if you wanted something else or if you just missed it, but anyway.

 

[pigswill]

There's another complication. Very few wars are decided in one battle. Therefore victory chance = survival chance = promotion chance. After half a dozen attacks you're going to have a lot more CR2 seige surviving to get to CR3 than barrage2 seige surviving to get to barrage3.

 

[Fezzik]

Fezzik, im not really sure what youre looking for but i conducted a test with CR2 vs B2 cats a few posts up. Not sure if you wanted something else or if you just missed it, but anyway.

Ahh, sorry. I guess I did miss it when I read back through OTAKUjbski's tests. Was it a mix of both, starting with Barrage 2 and finishing with CR2? It looks like they are separate tests.

My hypothesis is any cat that would die anyway (the throwaways, fresh from the barracks) should be Barrage 2, not City Raider 2. Each increases the chances of your CR2 cats surviving. 0.2 extra dmg x 6 defenders x 4 cats adds up. Even 1 additional CR2 cat surviving is an advantage. Just an idea, but it needs testing.

 

[kazapp]

Could someone post a summary, or is the current status "unknown"?

 

[siggboy]

Even if the percentage is deceiving, they may have some value.

"Deceiving" is a weak expression considering what one should expect going strictly by the tooltip wording:

"+100% collateral DAMAGE" directly refers to the damage actually done in terms of unit hit points. In order to have the promotion comply with the tooltip information, the calculation would have to be done as follows:

1. Let a = base strength of attacker, d = base strength of defender
2. Calculate damage = floor(10*(3*a+d)/(3*d+a)) [this is the formula currently used, ignoring any Barrage promotions]; this value will be between 3 (if a = 0) and 30 (if d = 0) and is the amount of HP subtracted from the target. In effect, a collateral attack takes away between 3% and 30% of the victim's base strength.
3. Apply the Barrage bonus percentage to damage: actual_damage = damage * ( 1 + bonus / 100 )

Maybe that would make Barrage too powerful, but it would certainly be a lot more consistent with the tooltip and much easier to comprehend by the casual player.

 

[siggboy]

Could someone post a summary, or is the current status "unknown"?

Executive Summary:

• Barrage is a very bad promotion overall, and should not be used for city raiding in particular.
• City Raider is the promotion line of choice for siege weapons that are on the offense (capturing cities).
• Combat is the promotion line of choice for siege weapons that are on the defense.

 

[Fezzik]

I tested the idea of using B2 cats first, then CR2. It didn't work out as well as I expected.

Defenders:
3x CG3 Longbows
3x CG2 Longbows

Attackers (group 1):
6x CR2 Cats

Attackers (group 2):
4x B2 Cats
2x CR2 Cats

Results:
Average combat odds (3 trials):

1) CR2  3.5%  B2  1.4%
2) CR2  9.5%  B2  2.5%
3) CR2 21.9%  B2  6.3%
4) CR2 47.2%  B2 19.0%
5) CR2 69.8% CR2 77.7%
6) CR2 82.8% CR2 86.4%

You basically guarantee the 4 B2 cats will die, in exchange for 4-8% greater chance of the last 2 CR2 cats living. Not a good trade. So please ignore my theory. Barrage is absolutely as bad as has been suggested. It was even worse with Trebs.

 

[OTAKUjbski]

I'm sorry for the walls of text, but it keeps me (sort of) from having to come back to the computer every 15 minutes ... it's all I can do given the IRL chores and stuff I have to do to keep my house in working order ...

RE: Executive Summary: That's the summary, but some people seem to still disagree -- which I guess means we haven't come up with sufficient proof to support our claims.

Right - but I'm not at all following why it is that assortments of various contrived combinations of attackers and defenders is going to tell you anything about what the intended results were.

I'm trying to find a situation where Barrage has a clear advantage over City Raider.

I feel like that situation will likely be the one the developers intended Barrage for -- which would justify it's current mechanic.

EDITOR's NOTE:

I originally surmised Barrage was intended for situations where the majority of defenders are very strong -- such as Macemen vs Sitting Bull's Longbowmen. Barrage Siege have a similar survival rate than CR Siege, so the extra collateral damage should justify the sacrifice (the sacrifice that'll be made either way).

However, because of the reduction in the defender's strength by even the CR Siege's 'level 1' collateral damage, follow up CR Siege have successively better Combat Odds -- especially by comparison to the Combat Odds of successive follow up Barrage Siege.

Conversely, Barrage Siege is less likely to significantly damage the top uber defender (if at all in some cases), so the likelihood of follow up Barrage Siege having similar or insignificantly better odds is higher than when using CR.

So in the end, more Siege units die for only a minimal increase in Combat Odds to the non-Siege city takers -- thus making Barrage cost more in overall production , WW and reduced GG :gpp: acquisition.

----

This does have me thinking, though ... maybe the first 1 or 2 Siege should be CR while the follow up are then Barrage ... perhaps the developers intent is in finding some sort of mix or balance.? (I'll have to test that too somehow, I guess.)​

The "problem", if there is one, certainly isn't in the rounding (because they implemented that same problem everywhere).

It is one of the problems but not the most substantial by any means.

... it's worth noting they've also alleviated it in most everywhere else (I particularly gravitate to the Finances fix here).

If I ran the zoo, the first comparison I would look at is the effect that using a damage modifier, rather than a strength modifier, implementation would do.

At the current +20%, +50% and +100% 'intent', I think using a strict damage modifier (like the tooltip implies) would definitely be overpowered, so a tweak to the actual values applied would likely be necessary -- possibly to +15%, +30% and +60%.

Even if the percentage is deceiving, they may have some value. Barrage 2 does cause more collateral damage, correct?

"Deceiving" is a gross understatement, IMO.

Barrage II doesn't always cause more collateral damage than Barrage I. In fact, Barrage doesn't always cause more collateral damage than it's previous promotion level!

That is most prevalent in weaker Siege units and when combating units with similar base strength as that of the Siege unit.

This is the Rounding Error referred to by VoU and is just one of the many issues with Barrage.

IMO, the most considerable 'broken' aspect is the amount of extra collateral damage done.

 

[UncleJJ]

It was even worse with Trebs.

Actually it doesn't make a lot of sense to be attacking a city defended by 6 CG longbows with only catapults and swords. They are an expensive way to beat it. I might occassionally attack a city with say 2 longbows if the AI has recently gained Feudalism, using just catapults and swords, but that would be just a transitional thing at the end of a war.

As soon as the AI has large stacks of longbows, and especially if they are CG ones, I wait until I have trebuchets and maybe macemen to crack the best defenders. So I would be interested in seeing your figures for trebuchets.

There doesn't seem to be much doubt that the CR promotion is better than the barrage one against defenders in a city as long as the siege has a small chance of winning and is not pure suicide. But I am most interested in the comparison of barrage catapults versus combat catapults both used against a large stack outside of a city, maybe on the flat. The AI produces quite a few barrage promoted siege and it would be interesting to see if that is a good move.

 

[VoiceOfUnreason]

This does have me thinking, though ... maybe the first 1 or 2 Siege should be CR while the follow up are then Barrage ... perhaps the developers intent is in finding some sort of mix or balance.? (I'll have to test that too somehow, I guess.)[/INDENT]

Again - I don't think the experiment is going to tell you anything new; we already know that collateral damage inflicted on wounded units is equal to the collateral damage inflicted on their healthy comrades.

BUT - it's probably worth remembering that this damage calculation is the same formula that was originally used for direct combat, which is not the formula used today. (Note: I can't prove that - but if you dig around for the change list on 1.61 you may find a note like "firepower is now average of curr and max strength"). So perhaps this damage calculation ought to have been changed in a parallel fashion.

Edit: further research hints that the current combat model is the third variation - 1.52 included an intermediate variation, which seems to be the closest match to the current collateral implementation.

 

[futurehermit]

In my anecdotal experience I have found barrage 1 to be pretty much completely useless. Barrage 3 cannons seem effective, but I'm sure CR3 cannons are also effective

 

[obsolete]

My favourite promotions still are Barrage I, II & III.

My favourite toys, are Barrage level III Modern Armours. And yes, I will also use them to penetrate into cities.

If you have a SoD in a MP game, no matter how big it is, the one thing that scares you most is a brigade of Barrage-III-Modern-Armours (but Stealth bombers are scary too).

 

[DaveMcW]

Combat is the promotion line of choice for siege weapons that are on the defense.

There's some pretty good evidence that the best promotion to raid cities with is... city raider.

But I haven't seen any proof that barrage is useless in the field.

 

[pi-r8]

My favourite promotions still are Barrage I, II & III.

My favourite toys, are Barrage level III Modern Armours. And yes, I will also use them to penetrate into cities.

If you have a SoD in a MP game, no matter how big it is, the one thing that scares you most is a brigade of Barrage-III-Modern-Armours (but Stealth bombers are scary too).

i'll bet you could still take em down with enough Cho Ko Nu's...

 

[CivCorpse]

Ok, it's pretty simple. And everyone is making it into a huge debate involving hours of tests and what ifs, and why would you be using cats instead of trebs. blah blah blah.

Ask yourself this question. If you have a unit attacking a city that can receive a CR promotion, is there any other promotion you would pick? (@ obsolete Try CR1 Barrage2 tanks for hitting cities or even CR2 barrage1 MA's if attacking CG3Mechs. I also like barrage on tanks but have much better success with a mixture)
Barrage is meant for attacking units in the open field not behind walls. We are on page five of a comparrison between apples and oranges. Are we that bored with Civ4?

It is like this.
I have a siege unit that is going to attack cities. I give it CR promotions. There is a reason it is called "City Raider"
If i have a siege unit that is going to attack troops in the open, I use barrage.

You will notice a lack of test numbers. A. Because we have 1000000000 tests already and B. Playing the game has given me ample testing already.

CR for cities. Barrage for anything non-ctiy. Simple

@ Uncle JJ. There are the few times I would attack LB's with cats and swords. Usually it is when my neighbor attacks his neighbor on his other flank and gobbles up a huge tract of land. We have tech parity, but his superior land mass will mean once he recovers he will out tech and out produce me. I might stab him in the back to grab a few cities to narrow the gap. Maybe allowing me to attack later with trebs/maces before he techs so far ahead i will never be able to.

One last time. CR for attacking cities. barrage for anything else

 

[VoiceOfUnreason]

But I haven't seen any proof that barrage is useless in the field.

Really? Haven't you been paying attention?

Collateral Damage: Catapults vs Strength 5 units (Axes, War Chariots, etc)
Unpromoted: 10 hp
Barrage I: 10hp
Barrage II: 11hp
Barrage III: 13hp

Collateral Damage: Catapults vs Strength 6 units (Longbows, Swords, Vultures, etc)
Unpromoted: 9 hp
Barrage I: 10hp
Barrage II: 10hp
Barrage III: 12hp

Collateral Damage: Catapults vs Strength 8 units (Maces, Praets)
Unpromoted: 8 hp
Barrage I: 8hp
Barrage II: 9hp
Barrage III: 11hp

Collateral Damage: Catapults vs Strength 10 units (Knights)
Unpromoted: 7 hp
Barrage I: 7hp
Barrage II: 8hp
Barrage III: 10hp

Two promotions to get 1 extra point of collateral damage seems like a decent approximation of useless. If your plan is to redline the defenders, then use your attacking pieces, the promotions are worthless in practice.

For situations where you won't be redlining the defenders, if that 1hp should happen to cross a jump point, then you might be getting some real value out of it. If you wanted to figure that out, you'd need to determine what combination of strengths and bonuses of your next attacking piece.

Example: a small incoming stack includes a couple of spears, your counter attack includes horse archers. Your unpromoted catapult will do 11hp of damage to the spears, the first two barrage promotions add 1hp each. Therefore, depending on the promotions you use, the spears will be at 89%/88%/87% strength when the HA ride in. Problem: is there ANY combination of promotions on the HA and promotions on the spears that places a jump point between 89% and 87%?

Maybe, maybe not. The width of that gap does scale linearly though, until it hits the cap - I'm guessing that you have a much better chance of finding some combination of bonuses that gives you a jump point between 67% and 61%, for example.

CR for cities. Barrage for anything non-ctiy. Simple

Yup. Simple. Wrong, but simple.

 

[vicawoo]

I think barrage 2 versus catapults should be 11 damage. Civ 4 keeps damage out of 100, it's just that due to how it displays unit hp, it looks the same as barrage 1.

And people are looking at this wrong. Say we're considering this 6 longbow situation. You're wasting catapults if you attack with more than you need to hit the 50% collateral damage cap (and for the record, according to combat explained, chinese crossbows have a collateral cap too). You need fewer barrage catapults to hit the collateral cap. Barrage 2 catapults are 25% more effective against 6 longbows than no barrage catapults since you need 1 less.

The proper thing to do, would be to attack with 5 cr catapults and 13 swords or 4 barrage 2 catapults and 14 swords. Otherwise, you are using catapults in a role (attacking without doing collateral damage) where there is a strictly better choice (swords, axes).

Formulaic calculations for barrage values (defender strength, barrage 1, ...)
5, 10, 10, 12, 14
6, 9, 10, 11, 12
7, 8, 9, 10, 11
8, 7, 8, 9, 11,
9, 7, 8, 9, 10,
10, 7, 7, 8, 10,
11, 6, 7, 8, 9,
12, 6, 7, 7, 9,
13, 6, 6, 7, 8,
14, 6, 6, 7, 8,
15, 6, 6, 7, 8,
16, 5, 6, 6, 7,
17, 5, 6, 6, 7,
18, 5, 6, 6, 7,
19, 5, 5, 6, 7,
20, 5, 5, 6, 7

 

[CivCorpse]

Really? Haven't you been paying attention?

Collateral Damage: Catapults vs Strength 5 units (Axes, War Chariots, etc)
Unpromoted: 10 hp
Barrage I: 10hp
Barrage II: 11hp
Barrage III: 13hp

Collateral Damage: Catapults vs Strength 6 units (Longbows, Swords, Vultures, etc)
Unpromoted: 9 hp
Barrage I: 10hp
Barrage II: 10hp
Barrage III: 12hp

Collateral Damage: Catapults vs Strength 8 units (Maces, Praets)
Unpromoted: 8 hp
Barrage I: 8hp
Barrage II: 9hp
Barrage III: 11hp

Collateral Damage: Catapults vs Strength 10 units (Knights)
Unpromoted: 7 hp
Barrage I: 7hp
Barrage II: 8hp
Barrage III: 10hp

Two promotions to get 1 extra point of collateral damage seems like a decent approximation of useless. If your plan is to redline the defenders, then use your attacking pieces, the promotions are worthless in practice.

For situations where you won't be redlining the defenders, if that 1hp should happen to cross a jump point, then you might be getting some real value out of it. If you wanted to figure that out, you'd need to determine what combination of strengths and bonuses of your next attacking piece.

Example: a small incoming stack includes a couple of spears, your counter attack includes horse archers. Your unpromoted catapult will do 11hp of damage to the spears, the first two barrage promotions add 1hp each. Therefore, depending on the promotions you use, the spears will be at 89%/88%/87% strength when the HA ride in. Problem: is there ANY combination of promotions on the HA and promotions on the spears that places a jump point between 89% and 87%?

Maybe, maybe not. The width of that gap does scale linearly though, until it hits the cap - I'm guessing that you have a much better chance of finding some combination of bonuses that gives you a jump point between 67% and 61%, for example.






Yup. Simple. Wrong, but simple.

Yup, Complicated. Wrong, but complicated.

You're assuming a single catapult attack. Try 2. Does this take the defending units down below the jump point? Or 3-4 if the enemy SoD is large. Do you normally attack cities with a single CR cat/treb to soften them up? Then why use only one catapult to soften an SoD. And you don't need to redline them. Just get them low enough your HA or knights can survive. I did a test and it takes 6-7 HA's to wipe out a full stack of cats. Or knights for trebs or cavalry for cannons. Taking out your enemies siege is the first thing you concentrate on when dealing with an SoD. It preserves city defenses and protects your own troops from collateral damage. What promotion would give better results?

What promotion do YOU use instead of CR for attacking cities?
When do you use CR in the open field?
What promotion will be more effective for dealing collateral damage in the open field?

 

[Ibian]

Basic siege is so powerful that you could leave them unpromoted and barely even care about it. If they were more balanced it would be painfully obvious to anyone how worthless barrage is.

What promotion will be more effective for dealing collateral damage in the open field?

Combat. Im sure someone will get around to testing this soon, but it should be obvious than surviving siege that does a borderline meaningless lower amount of damage is better than dead siege that did ever so slightly more.

vica, what are you yappin about? Its never worthless to weaken defenders as much as possible before killing them off, and its better to have a small army of foot soldiers and a huge army of siege than the other way around. With the first option you win more fights and get more promotions and GG points and higher promoted troops, and can drop defenses much faster. With the second option you had better hope your troops are stronger than the enemys (which will be harder since promotions are spread out) and be very careful about keeping your siege alive.

 

[CivCorpse]

Basic siege is so powerful that you could leave them unpromoted and barely even care about it. If they were more balanced it would be painfully obvious to anyone how worthless barrage is.


Combat. Im sure someone will get around to testing this soon, but it should be obvious than surviving siege that does a borderline meaningless lower amount of damage is better than dead siege that did ever so slightly more.

Well someone got around to wasting an hour or so testing Combat vs Barrage. And the results are pretty interesting.

First these are the parameters of the test. I used nonaggressive nonprotective civs for both sides of the equation. Then I asked myself what is a real SoD comprised of. So I created a well balanced attack force with no weakness. I tried to be realistic in my promotions for the enemy. Each SoD consists of the Following:

2 Knights combat 2
5Macemen CR2
2longbows CG2
2 crossbows drill2
2Pikemen. combat2
8Trebs CR2
a reasonable stack that is well balanced for defensive purposes and designed to take and hold a normal size city.

A reasonable stack well balanced stack that you could realistically face in the field. Your goals in attacking with siege are A. To weaken as many units as possible with emphasis on pthe Pikes. Your first assault with killing troops is probably going to be your knights. In addition to being your strongest troops. You want to do as much damage as possible to the enemy siege units. And damage to the crossbows as well so you can follow up with maces of your own.

I used stacks of 5catapults each. One stack had combat2 the other had barrage2. The difference between barrage 1 and 2 is greater than the difference between combat1 and combat 2 because of the 10% vs melle with barrage2. It would be interesting to see how barrage3 stacks up vs combat3. Though I would rarely have 10exp units at home for defense unless I was playing a charasmatic leader.

With regard to the value of live catapults after the fight vs dead ones. it became somewhat moot since in most cases all the catapults on both sides died. Occasionally one would survive on either side and very very rarely (like twice in all the testing) would there be two. of the occasional survivor it was a 60/40 split with combat having the advantage. I ran 20 attacks per stack and then averaged them out. I used remaining Pike strength as a measure because they are our focus. If they are fairly healthy they can really mangle the knights.

The results are as follows. I list both pike strengths for a reason.

Combat2 average pike strengths are 5.25 for the healthiest and 4.59 for the weakest. Both are dangerous to attacking knights. With a realistic possibility of killing a knight each and possibly two with the healthier pike.

Barrage2 average Pike strengths are 4.56 for the strongest and 4.26 for the weakest. The chances of knights surviving are much higher.

An odd thing I noticed was that after the knights were weakened enough to not be the defenders. If the combat2 catapults were attacking they drew relatively healthy maces as defenders and usually died. The barrage drew relatively healthy crossbows. i think this had 2 effects. One it is a weaker opponent to begin with. And it is not in the main stack to absorb collateral damage at a reduced rate. Thus increasing the damage suffered by the remaining troops. Also the remaining maces in the barrage opponent were well damaged. Non were redline or close but all were down to the 5-6 1/2 range. The crossbows faired worse as direct opponents probably because they lose there anticollateral bonus.

The combat opponents had generally healthier crossbows. And they usually had 1-2 banged up maces while the remaining maces were in the mid to low 7 range.

If the catapults were surviving with any regularity at a rate of 2-3 per attack then maybe that would offset what turned out to be a rather nice amount of damage by team barrage.
But in reality it was maybe one extra catapult about 1/3 of the time. i would rather lose a catapult for 50hammers than lose a knight for 90.

Results: Barrage is better in the open field, atleast with cats. With cannons it might be different because with C2 they actually do survive.

Now i go back to my statement that CR is better for cities and barrage is better in the open field.