Are trebs really worth it?

AFAIK the bombard damge is based on your relative strength after bonuses, hence trebs would do more colat damage attacking cities. Or maybe I'm wrong

4 trebs will stay in my stack all the way until i get bombers for 1 turn bombardment, or eh, my stack gets wiped out due to, eh, inclement weather or something

Their can be no argument, that for the same hammer outlay, cats > trebs, going forward. n the defence, the difference is marginal.

However, except for a few attack minded trebs for colat damage of big city stacks, i'll most use city raider maces, on account of how they become city raider grenadiers, and the lucky few survivors, city raider infantry

shadow2k said:

CR3 = 75% city attack

A Cat attacking a city with CR3 is 8.75str

A Treb attacking a city with CR1 is 9str
with CR3, it's 11str

Click to expand...

This is incorrect. Cats attack cities at 5, Trebs at 4. Right-click-hold over the next city you plan on attacking and see for yourself.

All promotions, except for Combat promotions, affect the strength of the defender. They do not augment the attacker's strength.
See here for details.

Good point, gunkulator.

Anyone who plays a lot of MP knows that cats are much better. Apart from the fact that you can get to construction early, and enigineering is much later, cats are very effective. with the collarteral damage promotion, you can take almost no losses after softening up a stack with 3-4 cats. Killing your opponent's 1000 hammer SOD by suiciding 200 hammers worth of catapults is really worth it!

Dnomal said:

When warlords first came out I though they were great, I'd use the charismatic trait and upgrade them like mad, but recently I've moved back towards catas and I started to wonder if trebuchets were really worth the effort?

Trebuchet: 60 production 4 strength 100% city attack, -25% bombard
Catapult: 40 production 5 strength -15% bombard
(both have a 25% withdrawl chance)

Meaning you can get 3 catapults for the price of 2 trebs.

Click to expand...

At cost 60, trebuchets were much better than catapults at 40. But now the trebuchets cost 80 (twice as much as catapults), which makes it much less clear.

The point about collateral damage is valid. I'd point out, though, that I usually want to take multiple cities.

If I wanted to take 1 city, suicide cats would be the better choice.

I I wanted to take multiple cities, Trebuchets would be the choice for me since I'd like my unit to win its battle, survive the attack and take another city after it's healed for a bit.

Suicide cats are useful. Not having to commit suicide is better.

All this is valid for single player only. In multiplayer, you'll be fighting in the field a lot more often. In single player, I find that I only attack cities while I defend in the field. Maybe that's because I move 1 in the opponent's territory while he moves 2 or 3 or more. Hmmm....

Trebuchets for the win. Catapults for the stalemate.

gunkulator said:

This is incorrect. Cats attack cities at 5, Trebs at 4. Right-click-hold over the next city you plan on attacking and see for yourself.

All promotions, except for Combat promotions, affect the strength of the defender. They do not augment the attacker's strength.
See here for details.

Click to expand...

Bah, forgot that's how they did that nonsense.

So basically, take an unpromoted Longbow, fortified in a city on a hill.

I decided to just hit up the world builder for it, since the first strike will throw me off.

CR 2 Cat vs unpromoted Longbow fortified in a city on a hill with 0 culture defense:

5 vs 7.8 6.4% odds, 23.4% retreat chance

CR2 Treb vs same unpromoted LB in same situation:

4 vs 3.52 64.3% odds, 8.9% retreat chance


6.4% compared to 64.3%. That's a huge difference, well worth the extra cost of a Treb over a Cat

The catapult has a higher retreat chance because it's chance of winning is so low. If they both have to roll a chance to retreat, they have the same odds to do so.

If you give that same Longbow City Garrison 2, the Cat is at .4% odds of winning, while the Treb is at 21.9%.

Catapults will cause slightly more collateral damage, but will die far more often in trying, and will cause much less damage to the top defender. The Treb *may* be able to damage more units at once, or take a units total HP's slight lower overall...but I'd have to get into the XML file to find that out.

I personally want my Treb to damage that top defender considerably, if not flat out kill it. I'm far less worried about the crappy defenders sitting underneath it, which is what the cats are better at damaging.

Trebs > Cats

Dnomal said:

When warlords first came out I though they were great, I'd use the charismatic trait and upgrade them like mad, but recently I've moved back towards catas and I started to wonder if trebuchets were really worth the effort?

Trebuchet: 60 production 4 strength 100% city attack, -25% bombard
Catapult: 40 production 5 strength -15% bombard
(both have a 25% withdrawl chance)

Meaning you can get 3 catapults for the price of 2 trebs.

Assuming that you group catas or tebs together then you will have 15 strength worth of catas, with the ability to take down a cities defence 45% a turn.

While two trebs, would give 8 strength, with 16 IF you were attacking a city, able to bombard a city by 50% a turn.

Now on the power graph you would get more points for having 3 catas than 2 trebs. Admitadely you'd have to pay one extra gold for the upkeep (or two for pacifism), but you would have you second cata finished while the first treb would only just have been finished.

Thoughts?

Click to expand...

You shouldn't compare the unit(s) by summing their total STR dude.

I will take 1 Modern Armor over 40 warrior.

Maybe I should try to create this scenario and see how many Barb warrior can this lone MOdern Armor beat

I think it's clear that trebs, once they become available, are better at taking cities than cats. One advantage that hasn't been pointed out is that the higher survival rate means less WW. Cats are certainly not obsolete, but once I have Engineering I prefer to build mostly trebs.

However, often that means that I build very few trebs. I generally pursue the Liberalism tech route, so unless I have excellent trading opportunities, I may not get Engineering until just before Chemistry. At that point, I'd rather focus my builds on grenadiers and wait for cannons. If I do have Engineering early, I'll gladly build trebs, but I don't usually prioritize the tech.

peace,
lilnev

Galileo44 said:

Anyone who plays a lot of MP knows that cats are much better. Apart from the fact that you can get to construction early, and enigineering is much later, cats are very effective. with the collarteral damage promotion, you can take almost no losses after softening up a stack with 3-4 cats. Killing your opponent's 1000 hammer SOD by suiciding 200 hammers worth of catapults is really worth it!

Click to expand...

Multiplayer =! Single Player

If you play a multi game the same way you'd play a single game, you'd be dead before the birth of Christ.

For MP, I'm sure what you say is true and cats are much better - after all, in most MP games you're rushing to achieve military dominance as fast as you can, and that implies more field battles and less city-taking than in a SP game, where the AI's field activities are still kind of weak. If you're just trying to take down a human opponent's SoD, you'd be using Trebs for something they're not designed for, but something cats ARE designed for. But if you're opponent put that SoD in a city, with its defense bonuses, you'd need a lot more than a couple hundred hammers' worth of cats - and trebs would have an advantage at that point.

It's very situational, really, and depends on playing to the particular unit's strengths rather than trying to blindly throw units at an enemy.

Based on a review of the SDK, and a couple of tests within worldbuilder....

1) As far as collateral combat is concerned, damage is centered around base strength, not combat strength. So current hit points, terrain advantages, combat bonuses... those are all irrelevent. Barrage I, II, III boost the attacker's base strength. Drill II,III,IV scale down the amount of damage done to the defenders

The remaining calculations consider unpromoted seige units against unpromoted longbows.

2) Collateral damage from catapults can take a longbow down to 55HP (3.3). Trebs can take a longbow down to 59HP (3.54). Not what you expected after playing Vanilla? They changed the calculation of the limit to the collateral damage in Warlords - the relative strength of the defender and attacker have been added to the calculation.

No matter how much you distort the relative strengths of the units, you cannot take the enemy down past the CollateralDamageLimit of the attacker (50% for both catapults and trebs).

3) In this scenario, the cats are doing 9HP of collateral damage each turn, the trebs 8HP. Because of the rounding in the display, you may not be able to detect the difference during the first round of combat, but it is there.

It is a bit more obvious if the defending units are bigger (vs knights, the damage is 7HP vs 6HP)

4) I spend way too much time investigating this stuff.

Could you expand that a bit more, VoU?

Is the catapult doing 9HP collateral damage to as many units as a catapult can cause collateral on in a single suicide? Is the number of units that will be damaged different for cat and treb? Will Barrage affect the amount of damage done per unit, or also the number of units that will be damaged?

If a city has 4 longbows, only three will take collateral damage, regardless of whether the cat could damage 4 or not? And if there are a dozen, how will the damage be spread when the second cat suicides - will undamaged units take collateral, or can damaged units take more collateral again?

Based on 9HP and 55HP limit, it seems to me that five cats (or 6 trebs) is the maximum useful number to throw at a city unless the number of defenders is large enough that not all will be taking damage per suicide - although most of the time less than maximum number of suiciders (or less suicidal trebs) is needed before CR-troops can handle the rest.

Elandal said:

Could you expand that a bit more, VoU?

Click to expand...

Is the catapult doing 9HP collateral damage to as many units as a catapult can cause collateral on in a single suicide? Yes. But remember that 9HP figure is specific to a catapult dishing collateral damage onto a base strength 6 unit. Bigger units will take less damage, smaller units will take more.

Is the number of units that will be damaged different for cat and treb? No. The limit on the number of units subjected to collateral damage is defined by iCollateralDamageMaxUnits in CIV4UnitInfos.xml, so modding in a difference is easy - but out of the box they are the same.

Will Barrage affect the amount of damage done per unit Yes, by modifying iCollateralStrength (the attacking unit's strength factor).

or also the number of units that will be damaged? No.

If a city has 4 longbows, only three will take collateral damage, regardless of whether the cat could damage 4 or not? Yes. The defending unit is exempt from collateral damage. Defenders receive at most one dose of collateral damage per combat, so you have "wasted" half of your potential collateral strikes by attacking a smaller stack.

And if there are a dozen, how will the damage be spread when the second cat suicides - will undamaged units take collateral, or can damaged units take more collateral again?

Previously damaged units can take damage again. At the beginning of the collateral round, each non-exempt defender is assigned a priority. The units at the top of the priority queue are the ones that take the damage. Priority is determined by a die roll, but loaded in favor of fully healed units (so it is HP based, not strength based, priority).


And just to be clear - because the question comes up a lot - the collateralCombat calculation is only done once per attack, immediately prior to the "first strike rounds". The amount of collateral damage inflicted is completely independent of the number of combat rounds your unit wins, for example.

Thanks for the answers, VoU.
One more question: armor units (and at least battleships) can get Barrage promotions even if they don't normally cause collateral damage. In this case, the unit strength for collateral is not based on the normal unit strength + Barrage modifier, but only the Barrage modifier: eg. Tank has S28, Barrage I iirc +20%, so tank with Barrage I would have collateral strength of 5.6 - just a bit higher than a standard catapult?

Elandal said:

One more question: armor units (and at least battleships) can get Barrage promotions even if they don't normally cause collateral damage. In this case, the unit strength for collateral is not based on the normal unit strength + Barrage modifier, but only the Barrage modifier

Click to expand...

Yes, but it's part of the rule, rather than an exception.

The actual calculation of iCollateralStrength is BaseStrength multiplied by a collateralDamage factor. collateralDamage is calculated by adding iCollateralDamage of the unit type to the extra collateral damage available to that particular unit, based on promotions.

For catapults and trebs, iCollateralDamage is 100%. For a catapult with Barrage I, the result is:

iCollateralStrength = 5 * ( 100% + 20% ) = 6.

Tanks have iCollateralDamage = 0% out of the box. When you add in the Barrage I promotion, the result is
iCollateralStrength = 28 * ( 0% + 20% ) = 5

Notice the implications of the rounding.

No promotions
Catapult: 5
Trebuchet: 4
Tank: 0

Barrage I
Catapult: 6
Trebuchet: 4
Tank: 5

Barrage II
Catapult: 7
Trebuchet: 6
Tank: 14

Barrage III
Catapult: 10
Trebuchet: 8
Tank: 28

They're rounded to integers? That's weird - when the integers are small, rounding causes massive effect (Barrage I is useless for treb, Barrage II causes large jump in collateral strength).

I generally give barrage proms to cats, also to any siege that walks the CR line and gets to L5. CR3 B3 arties are nasty. That said, I don't really weep when I lose them - they're there to go first so that there's but some pulp left for mopup..

Elandal said:

They're rounded to integers? That's weird

Click to expand...

Not really, most of the game is set up that way - one of the reasons that micromanagement is so prevalent in the game. Almost everything except the display seems to use fixed point match of some sort. Something of a pity, really - given that the DLL is implemented in a language that allows you to sweep a lot of those issues under the rug.

Elandal said:

when the integers are small, rounding causes massive effect (Barrage I is useless for treb, Barrage II causes large jump in collateral strength).

Click to expand...

The integers are always going to be on the order of the strengths of the units, so the rounding gets messy.

I've posted the Collateral Damage Calculator I've been using - it's an excel spreadsheet with a few examples and no documentation, usability considerations, or testing.

Some findings:
Barrage I tanks are unpromoted catapults. Barrage II tanks are Barrage I Cannon. Barrage III tanks are a shade better than Barrage II Artillery.

Barrage I is useless for Trebs.

Drill II-IV on Longbows (or crossbows, for that matter; str 6 being the key here) completely nullify Barrage I and II promotions for catapults. Barrage III does make things better.