Should You Target Strong or Weak Units First?
Вставка
- Опубліковано 15 жов 2024
- Here's an AoE2 puzzle: should your archers target archers or skirmishers first? Does it change the outcome? Let's check it out.
1:45 The answer
2:16 Confirming the result by alternating and not focus firing
3:08 Castle age (Crossbows and Elite Skirmishers)
3:33 Second example with Galleons and Turtle Ships
4:04 An exception to the rule
Facebook:
www.facebook.c...
Patreon: / spiritofthelaw
Full intro song: • Spirit of the Law Chan...
Background music from Epidemic Sound: www.epidemicsou...
Game: Age of Empires 2 HD with Forgotten Empires, African Kingdoms, and Rise the Rajas expansions - Ігри
![Lp. Сердце Вселенной #30 ДВА В ОДНОМ [Один человек]• Майнкрафт](http://i.ytimg.com/vi/nrXy15orKhY/mqdefault.jpg)
Generaly, in any game with a decision which units to kill first, the ones who reduce the damage output of the enemy the quickest are the best target. That means you need to get the highest damage per second reduction per second. To get that we have to divide the damage per second by the time to eliminate that damage source. So the best targets are high damage per second and low time to kill, and the worst ones are the low damage per second and high time to kill.
madin1510 it reminds me how fast a team can melt once you eliminate their tank (damage sponge) in some competitive online games.
A video detailing the best 3 tank and glass cannons units in AOE would be nice
MafiousBJ when your tank got eliminated without your DPS capitalizing his tanking, well your team deserved to lose.
This is the clearest explanation in the comment section, as far as I can see.
For tanks, siege ram if arrows or pikes are involved, elephants for everything else. As for most easily dispatched while putting out high damage, It's gotta be siege onagers
weylin6 those are the first that come to mind. Specially rams in king of the hill games. But AOE2 is so situational that i would love to see Spirit to take a deeper look at it, specially at some glass cannons like gbetos that when massed melt buildings
@strela yes indeed they deserve to lose
I thought this is some sort of military strategy video. But no! This is a genuine AOE video!
At 2017!
@American Restoration Initiative you want a game like aoe or just a good game?
This applies in WoW battlegrounds lol and using these techniques help with the pvp
It would apply to military strategy too. It has to do with the same principle as "force concentration" in military strategy - you want to deny the enemy as much damage output as quickly as possible. This shouldn't really change whether it's a video game or military strategy. (After all, games are trying to follow that to some degree.)
For actual military strategy, it's a lot more difficult and can't reasonably be answered within five minutes. You can't make a general rule for that, you need a lot of information to make a close approximation and you will most likely lack information as well (supplies, ammunition, for example). It's a lot easier if you have units that do a certain amount of damage and take a certain amount of hits, but in real combat, that is never the case and you won't get a replay to analyse different scenarios either. And even if you do consider all of the information that is available to you, there will still be some "random elements" like human error, simply bad luck and what have you.
Floofy shibe It certainly is. That's one of the main reasons why wars are even carried out to begin with. If you could simply calculate your chances of victory, almost no one would even consider going into a losing war. Surrender would most likely be preferable to a 1% chance of victory.
There are ways to attempt that, definitive measures of your own side like ammunition, supplies and weapons, as well as relative measures, such as how well your troops were trained, their mental state and such compared to what you expect from the enemy.
But even if your troops are superior when it comes to all of that, that still doesn't guarantee a victory. Your enemy can get unexpected reinforcements or sabotage you, the weather or terrain can turn the tides of battle, your own troops make fatal errors or even raise a mutiny... war can't be calculated. It would require far too much information to even try that. There are many things I didn't even take into account with what I just used as examples, not to mention what it looks like when you get into the specifics.
Strategy is incredibly important, but it relies on information. To have a chance at winning just about any war, information is the key. How many troops does the enemy have here, how are they equipped, do they have fortifications or a favorable position, from where could reinforcements arrive? Many such questions have to be answered before you can work out a good strategy.
So basically:
-Units that do lower damage get a lower priority
-Units that do higher damage get higher priority
-Units that have low health get higher priority
-Units that have high health get higher priority
So targeting order:
#1 high damage, low health troops
#2 high damage, high health troops
#3 low damage, low health troops
#4 low damage, high health troops
Reason for putting low damage, low health under high damage high health is to minimise wasting ammo
I don't see how that conserves ammo at all. You still have to spend time and ammo shooting at those units regardless of order or priority. Maybe in a case where your unit or team does enough damage per round or unit of time to justify the higher prioritization that would make sense, but generally you still want to choose to eliminate the enemies' damage output as quickly as possible.
That's not accounting for range, but okay
@@ionarevamp that is true. In this game, generally #2 and #3 are switched. You indeed need to look at how quickly you drop their DPS.
I just tried to make a general rule, but #2 and #3 are very interchangable
@@spartanwar1185 yeah, of course it isn't accounting for range, if it was accounting for range, I would have put range there.
I don't see how range changes the priority list all that much. I think the way range affects things is way too dependent on many other factors
I believe the video clearly demonstrated the priority of low health, low dmg over high hp, high dmg.
Your goal should always be to reduce your opponents dps as quickly as possible. so it's not really about killing weak or strong units first, but instead all about dps/health(corrected for armour, so basically tdk) of a unit. the higher that fraction the more priority you should give said unit in targeting.
@Anonymous 12 yes, but strong and weak should not just take their survivability into account. if the enemy has so called "glass cannon" troops (in AoE stuff like Shotel Warriors in the case of melee troops) they are obvioulsy strong enemies if you take both into account. but if you only look at survivability, they are pretty weak.
obviously you do not have to do that math every time. but some troops combinations are pretty standard and you should know by hard which one of them has more dps and dies quicker. for instance archer plus skirms. OBVIOUSLY you have to kill the archers first.
@@jurgnobs1308 yes, kill the archers first...save the onagers for last.
@@jaelsonnen5750 that made zero sense
But if you target weaker enemy's first it's you get the dps reduction faster (most of the time with every hit). So you can say that in _generell_ it makes more sense to target weak enemy's first.
@@a_mage_as_old_as_joe1598 No, it's exactly how OP said it to be. If a unit can be killed twice as fast but deals less than half the damage, it should not be prioritized.
Take it to the extreme if you find it hard to understand conceptually:
• A *weak unit* deals 10 dps and has 10 health.
• A *strong unit* deals 2000 dps (x200) and has 50 health (x5).
You can kill 5 weak units or one strong unit in the same time.
5 weak units deal 50 dps.
1 strong unit deals 2000 dps. That is more.
It's about the quotient of damage reduction you achieve per time spent killing stuff. If a unit's ratio of offense to defense is high, kill it. If it is low, kill other things first.
Before watching: Answer is always attack what reduces enemies dps/time the quickest. Glasscannon units like archers, onagers, shotels etc should be killed first, tanky units like elephants, rams last.
The Huskarl example also follows this rule, without using exakt numbers, lets just say Huskarls do about 20x more dmg to the archers than the militia but are only 15x more tanky than the militia, so by focusing Huskarls you get rid of more enemy DPS quicker, resulting in more of your own units surviving.
I have worked as a Strategy Consultant in the StarCraft Pro Scene and there is quite some advanced math and analysis you can do on RTS games. Ive recently come back to my old kid love AoE2 and from what im seeing there is barely a true pro scene which i find sad because its an amazing game. Lets improve the metagame of AoE together!
Patrick Fame unless they are fuking your base up
Patrick Fame is the pro scene any better yet? (2020)
onager can be done later if ur a micro nerd :p
@@jamesaltonfilms can't say, haven't followed it lately. watched quite a bit of it around when i wrote the initial comments, but it wasnt something that kept me interested for an extended period of time. i suppose it's still a hobbyist scene without any real money in it, so naturally the pro scene won't be nearly as strong as a fulltime pro scene like SC2. you get out what you put in.
I thought I'd make a "puzzle" video where you have a situation and think about what you'd do - then I try it out both ways and you can see if your choice was right ;)
Maybe a fun future series if there are some other good topics?
if an lone enemy monk is converting a unit of yours, should you try to attack the monk and kill him or run away to prevent the conversion? the answer will probably vary with infantry, cavalry and archers
We already put a strain on our brains with every video. Now you makes us use it more? This is agony
What about increasing damage by garrisoning 20 units in castles? eg. 20 arbalest in a castle will (maybe) get a dmg increase while 20 arambai will get a decrease, but higher accuracy since it's the castles accuracy.
My hyoptheses is that garrisoned castles is very overpowered but very underused.
what about a more realistic unit comp like knights and pikemen
Spirit Of The Law i want see you deafening some top players. ..... becomeing a legendary player. ...!!!!!!!!!
If video games have taught me anything, it's to pick off the weaker enemies first then go for the stronger ones. Though if it takes 8 shots to kill an archer in feudal from full hp if you had more than 9 archers you could set 9 to focus down the archers and then the rest on a skirm. And the same for castle age.
>1 day ago
A wizard!!
lol what's up with that?
Supporters on patreon gets the videos early.
CallofBear patreon
Icewind Dale taught me to take out the stronger enemies first lol
a huskarl is just a love child of an elite skirmisher and a woad raider
NeverFinished 3Digits aren't all units supposed to be the child of something and a female villager?
(Remembers a bunch of male villagers can repopulate..... Shrivers.... xD)
NeverFinished, it's gotta be a supremacy villager. Have you seen the attack stats on those gbetos?
MafiousBJ Wait, so does this mean that the battle elephant is the child of a war elephant and a female villager?
Toast i was thinking how petards can be childs of villagers and demo ships, no explaination for the elephants tough
MafiousBJ Oh my. The female villager sure made that ship explode. lmao
You should always attack the one unit that has the highest attack/defence relation. You want to reduce the enemy attack output the fastest way possible to reduce accumelated damage over time, therefore you if you attack the one with the highest attack/defence relation, that will decrease the enemy damage output the most, fastest.
Even if the difference is small, the effect will be large as your relative advantage will increase relative to your advantage (if one archer survives two volleys longer, that means he might deal enough damage to delay your next archers loss for another volley).
A good example is 1 siege onager and 1 siege ram vs 20 arbalests. Naturally you would target the SO first, as the SR's ATT/DFC is really low to units, while the SO's is high. However, if it was the same enemy units against 4 close castles. Now, it would be smarter to take out the SR, since it's ATT/DFC just became higher than the SO's (I'm not completely sure if this is true, but it's certainly a completely changed scenario from the one with the units).
Do agree there. The tricky thing of course is in the end to make a quick judgement of what unit that is. Units with hidden bonuses and thing like armour play a huge role in this. But you goal is always remove the damage dealers quickly. In some games where there is more emphasis on support units then they might also be a prioritized target. But that is because there ability to ether buff or de-buff. Generally there squishy to. In AoE is mainly the monk that fulfil this role and not much else.
AoE3 has a variety of those annoying buff units, added in the Asian Dynasties expansion. The Japanese Daimyo, which can increase attack enough to make even the overpriced, underpowered Ashigaru really special. The Indian Mansabdars, which buff units of their type in the vicinity (like the Sepoy variant boosting your Sepoys).
I would also want to go on a huge rant about how the Asian civs are so OP given their buffs etc, the shipments, the fact that they advance by building 'Wonders' for every age (costing the same as a regular advance too, so by being able to create villagers in the meanwhile, this is just a huge advantage with no drawback)... I could go on forever.
I think strong vs weak units is quite a broad term and can vary. What you are looking for i DPS to Tank ratio in an inverse relation. In the example you used, skirmishers have a slightly higher DPS, so one would assume they should be the first target. Their pierce armour makes it so they have the equivalent of about 5 times more tankiness however, so the archers, even with the lower DPS, become the better target.
It's why in the militia / huscarl example, it was better to attack the huscarls. The militia DPS was so low, they barely posed a threat compared to the huscarls. Add to that the fact that the wasted shots served the purpose of essentially making the militia tankier than they were makes huscarls the optimum target in this scenario.
My thoughts exactly, this was kind of a dumb video.
Before watching, I already had a simple rule in mind: the glasscannon first, the tank last.
I was going to say something to this effect. Nicely put.
it makes me kind of sad that spirit didn't think of this since his videos are usually so robust and well though through. still an interesting video though but as you said the unit with the highest effectiveDPS / effectiveHP should be the first target. provided there is no "dps waste" that could come from to many arrows or to much movement.
Yeah man, I made a similar comment before I saw yours. Maybe if we say it's "not mathy enough" Spirit will be forced to make a follow up...
WEEE SPIRIT OF THE LAW UPLOADED A NEW VIDEO!
NOW I CAN UPGRADE MY INTELLIGENCE
Charielity Knowledge +4
Time -5
Penis +10
Watch *Spirit of the Law Video* (Cost: 1000 food, 200 gold)
Increases the ability of the user to win AOE games.
"You see, killbots have a preset kill limit. Knowing their weakness, I sent wave after wave of my own men at them until they reached their limit and shut down."
-Zapp Brannigan
I live for spirits intro...
At the end I was expecting some kind of graph, something like a curve showing when you should start focusing on stronger units first
Guess this channel has spoiled me into expecting graphs and maths everywhere
I just send in 50 20 damage zerglings and hope he has no air units
Lol
Zerg rush
Dark templar 6 min rush
Even if he does have air units a few thousand zerglings will eventually win the game
You should make a bigger video where you can talk about the dps/hp relationship in units and how it affect optimal focus. Great job on this one still.
Very simple - attack the unit with highest ratio of damage to defense.
No
@@pob-4810 He's right. In general attack-to-life ratio is what matters.
Example:
You have 10 archers each with 15 life and 1 attack vs two enemies:
Enemy 1: 30 life, 3 attack
Enemy 2: 10 life, 1 attack
They both have the same attack-to-life ratio, so it doesn't matter which one you target. If you target down Enemy 1 first you will receive 4 + 4 + 4 + 1 = 13 damage total. If you target down Enemy 2 you will receive 4 + 3 + 3 + 3 = 13 damage total.
I see it as a DnD perspective where players often can hit well above their weight in terms of challenge rating when fighting a boss because of action economy where the players will have several actions per round (1 per player assuming no magic shenanigans) and the big bad boss only gets 1 (assuming no magic shenanigans). So eliminating the archers first is taking away action economy from the enemy
I though this thing was obvious but good point with the militia bait tho.
It is a tactic That I use in other video games where I use cavalry or sometimes shield walls of infantry to distract my AI oponent while the archers/crossbowmen volley fire on them.
I will bring you the counter question:
Say, youre in a 1vs1, and you saw an archer rush coming from your enemy coming from miles away, and you decide to greet him with skirmishers.
You have 6 skirms and your enemy has 3 archers and 3 skirms of his own, identical upgrades on both sides.
In this case, you should be focusing down the enemy skirmishers first, for their bonus against AR units piercee your skirmisher armor, while your armor comfortable shrugs off archer arrows
As a guy that likes rushing a lot and mainly plays infantry civs, such as aztecs and celts, in a skirmisher battle, you should be prioritizing the enemy skirmishers first to avoid greater casualties
It is kinda answered already in the militia/huskarl vs arbalest fight. the archers act like the militia, while the skirmishers act like the Huskarls.
Whether you should choose unit A or B is merely a function of A's attack and armor, B's attack and armor, and your attack and armor. It just so happens that in this case, the overwhelming armor of B (skirmishers) causes it to fall in the favor of killing archers.
It'd be cool to try to derive a closed form equation that tells you how many archers you'd have left after the fight - perhaps using Lanchester's laws.
Nicely presented. In games I usually go the route of prioritising the enemies which have the highest damage output vs HP ratio (or instead of high damage output could be any crippling status effect that can really mess up your strategy). It's easier to see this in turn-based games, where you ideally want to beat your enemies in as few turns as possible to minimise how many free hits they deal to your team.
im neither sure what game this is, nor what hes really talking about, but somehow i still really enjoyed this video.
shoutout to anyone who got this video in their recommended too
It's Age of Empires 2
in (the art of war) written by Sun Tzu in the 6th century BC, he does argue that you should focus your strength to target the enemy weaknesses, while trying to make up for your own weaknesses.
I wish you could research something that would make troops automatically focus fire, but only fire as many arrows as necessary to kill an enemy.
that would be taskmanagers like officers or fire-team leaders
I propose a metric for use in this scenario, "Damage per time to kill" or DPTTK. This would be defined per enemy unit as:
[unit DPS / [ [unit effective hp / your total damage (rounded up to whole number) ] / your attack speed ] ] if accounting for overkill. Or simply [ unit DPS / [unit effective hp / your total DPS ] ] ignoring overkill or if using mixed attack speed army.
Using this, you simply attack the unit with the highest DPTTK
You could make the same comparisons with just:
UnitDPS / EffectiveHP
UnitDPS / AttacksToKill
@@8bit_pineapple I do have a bit of a tendency to over-define things... But yeah that's a much more concise way of putting it
answer : just do what Biper does
micro everything and kills their army without losing any of your archer
It's mathematically simple isn't it? DPS of enemy unit / time to kill it. The higher the value, the better the target.
In addition to the formula, slightly more weight should actually be given to lower HP enemy units since killing those will decrease enemy DPS earlier, resulting in a higher survival rate of your units.
You don't need to target lower HP enemy units first. I made the same mistake but then I worked out an example.
Let's say you have 10 archers each with 15 life and 1 attack vs two enemies:
Enemy 1: 30 life, 3 attack
Enemy 2: 10 life, 1 attack
They both have the same attack-to-life ratio, so it doesn't matter which one you target. If you target down Enemy 1 first you will receive 4 + 4 + 4 + 1 = 13 damage total. If you target down Enemy 2 you will receive 4 + 3 + 3 + 3 = 13 damage total.
You picked a very specific case here. If you had the critical mass to kill the skirms with one volley, you would be better off killing them first.
True, but how often do you have 30 archers in Feudal age?
Also, in that case, you may want to split your fire and take out the archers quickly.
I that case, patroling in would be better.
And then you misjudged or one of the shots missed and the skirm is left at 1 hp.
If you have a big army just let your units auto target.
So as a general rule you would look at something like time to kill (ttk) divided by dps.
When one of the enemy units takes double the ttk, but has 3 times the dps than their other units, focus it. If it is the other way around, focus the other units.
Where it get's really complicated, is when you yourself have a mix of units, so the enemy dps varies depending on what of your units are attacked first.
It is bad to generalise this one. Use a formula instead.
Target's dps / amount of hits to kill target
If this calculation give high result, flag the unit as glass cannon. If it give low, flag the unit as tank.
Always focus glass cannons before tanks - any gamer in any game
There are many situations where focusing a frontline tank can be the optimal play compared to diving past to get to the backline dps. If you can safely deal damage to the tank while avoiding the damage from the dps through positioning, you can eliminate the tank before the real battle begins and you will have an easy time dispatching the squishies.
An example of this would be mangudai vs hussar (lower dps/tankiness)+onager (higher dps/tankiness). You can (at least some times) be able to hit the hussars from outside of the onager range and keep retreating so that you are always out of range. Then, as you whittle down their hussars, you can move in on the onagers. (this kind of situation is more clearer in mobas or mmos where you can get a clear pick on their tank and kill him before the dps can react properly)
This is really helpful! I've always been one to target the really tanky units first, believing that would be most effective. Thank you for this!
Before I watch this, I'm guessing weak.
The rule is generally that you target the easier targets to pick off. In most games- whether it it first person shooters, RPGs, real time strategy, etc- having more units means more actions. More actions means they can murder (or depending on the game, debuff, stun, etc) you. Less opponents means less actions they can throw at you. They are killing you slower, which gives you more time to kill them. Of course, other considerations come into play- ranged units over melee if you have the distance, AoE over others because their actions just count for more, etc.
I need to go to get food for the next week. Then I saw you uploaded a new video. Seems like I'll be starving next week.
Just find some berry bushes.
Steal a sheep
Toast not realy,the decay rate os pretty big...unles he has a big family
Florinache Florinache Well I have a "Big Mama"
Florinache Florinache I'm sure he has refrigeration researched.
“It never hurts to see a counter-example” is solid wisdom
Except in this case it makes no sense. What is an actual example of a sponge unit that does not take many shots AND does very little damage? There is none which is why he picked an example so silly not only would you never see it but would be literally impossible in a 1 v 1 game.
0:48 Just use Kratos with a lightsaber and you'll win every fight.
I learned this years ago in as it is a valuable skill in strategy games you always kill the high dpm yet squishy units first, or another way to put it "smash the glass cannons first".
*In short: Always attack the weakest first, unless that is overkill (wasted damage).*
Is that a correct summary?
yes
I think it is all about attack Vs tankiness ratio. Fragile units with high attack should always be prioritized in order to reduce enemy firepower, while tanky units with a low attack (such as skirmishers) should be low priority targets.
You missed a very crucial element in your example, which is offense vs defense, or in other words why is unit A strong against unit B. While it's true that Skirmishers have bonus damage vs archers, they're also significantly tankier thanks to their extra pierce armor. You can ask the question of which of these 2 elements is more significant in the Skirmisher vs Archer scenario, and your test proved that the extra armor is more significant than the extra damage. Same deal with the Huskarl, the extra pierce armor is more signifcant than the extra damage. The problem is that you used Militias alongside the Huskarls rather than Champions, with Champions I suspect you'd have seen the same result as with the Skirmishers.
However these are 2 rather extreme examples of counter units having both an offensive and a defensive advantage against their counter. In the majority of cases counter units only deal extra damage with no extra defenses. For example the Spear line vs Cavalry, the Archer line vs the Spear line, etc. Thus when you take extra defenses out of the equation it suddenly makes much more sense to go after the counter units first as they're significantly more threatening yet die just as quickly.
TL : DR stronger and weaker are too generic of terms to be of any significance here. It's mixing 2 entirely different concepts, 1 of which is tanks vs damage dealers like in classic MMOs and the other is counter units. Tanks, as their name suggests, are supposed to soak up damage and are therefore the last units you should target. Counter units on the other hand are the best damage dealers against your own troops, therefore they're the first units you should target. Skirmishers and Huskarls happen to be among the anomalies of units that are both counter units and tanks. To isolate the tank issue you'd need a test of Champions vs fewer Champions and a War Elephant for example, in which case it's obvious the War Elephant should be left for last. To isolate the counter unit issue you'd need a test of Paladins vs Champions and Halberdiers for example in which case it's obvious the Halberdiers should be targeted first. Mixing the 2 issues together simply makes no mathematical sense and in fact answer a completely different question than the one you were asking which is which is more significant: offense or defense, nothing to do with stronger and weaker units.
So this is why I was so good at Empire at War, I always focus fired the weak units and turrets on stations first.
Can your next video be a Malians overview?
I'd love to see a Malay overview from his tactical and mathematical standpoint, but unfortunately he needs to finish the African Kingdom's civs first.
Awesome video! I picked up the convention of targeting weak enemies before strong ones from my time playing SWTOR, barring special mechanics, but it's good to see that hold true in with the numbers in AoE2 as well.
Numbers advantage is no joke - there's a multiplicative factor in force concentration such that 2:1 is roughly a 4:1 advantage. When the enemy elements' offensive and defensive capacities are scaling roughly equally, that is when there aren't glass cannons or bullet sponges to consider, you want to be focusing the weakest elements first to reduce their strength relative to yours in the fastest way possible.
Road to 1.8k?
Something I learned a long time while playing Halo at higher difficulties: you want to reduce the number of guns pointing in your direction as quickly as possible. This improves your overall chance of survival, and this video helps to illustrate that.
I think with +1 defense skirms you're better off just running away, a much better question imo would be with skirms/archers with only fletching, so the archers deal 2 damage per hit (twice as much) meaning it might be better to focus fire down the skirms first because of their higher bonus damage.
Archers still deal more damage per second than Skirmishers, so you should still focus fire the Archers.
Tocaraca, skirmishers have 6 total attack (2+1+3 bonus) while archers have only 5 (4+1).
Yeah but once you focus fire an archer they no longer gives 5 dmg, if you focus fire the Skirm, the enemy will still be able to deal 11 dmg while possibly reducing your dmg output by killing your archer. It is dps/tank ratio that should be the video main focus
Charles Jackson yes... bit archers attack every 2 seconds, and skirmishers only attack every 3 seconds!
Sun Tzu says to attack where they are weakest and to avoid what is strongest.
But if you have walls and castles then you should still make sure you have the means to attack their siege weapons and hope that they are stupid enough to try and storm your positions without them.
Always kill weaker characters first in any strategy game (assuming we're talking pve). The faster you take out enemies the less damage you take. Bottom line.
I could actually see somebody like you running into this kind of problem against bots, saving the game, doing all the calculations, come back three days and hundreds of tests later and finish that encounter.
you are usually pretty good with these calculations, but this time you messed up heavily.
first you find out that it is good to kill the ones first that go down quickly.
then you say that the example with the ridiculously weak militia went the other way and think it is because of the wasted arrows.
well, the wasted arrows make up for a bit of the reason, sure. but the actual exeplenation is far simpler. because it is not about killing as many enemies as you can as quickly as possible, but about getting their damage output on your troops down as quickly as possible.
so, if it is skirmishes and archers against range, you have one unit type that dies quickly and deals a lot of damage and a second unit type that survives longer and also does less damage. the obvious choice is to kill the archers to get the damage output of the enemy down as quickly as possible.
then in the example with the militias: the militias do die quickly but they also do not deal very much damage. hence, killing a militia does not siginificantly drop the enemys damage output.
the huskarls do survive longer but a dead huskarl actually significantly drops the enemys damage output. that is why it makes sense to ignore them first.
you basically just looked at defensive stats instead of taking the attack into consideration. and that makes all the difference.
I really do not know why you did not think of that, as you usually tend to think of everything relevant.
to be perfectly honest, if I were you, I would redo this video
skirmishers deal a lot of damage too, they have bonus damage against archers, so its not "obvious" is it.... u kill the archers first because they die quicker and do comparable damage to skirmishers
Have always done this in every single game I've played. From RTS to FPS.
Thanks for actually doing the math!
The only exception I can think of right now is when the "strong unit" does Area Damage. As long as the difference in HP between it and the other enemy units isn't like Boss unit vs its minions. It all comes down to the rate you reduce the enemy's Damage per Second.
rule of thumb is take the easy DPS off the field. an enemy with 1 hp in games can still typically still deal full damage.
As a general rule of thumb, one should always look at the damage (potential damage out put over time) and if enemy units have a very high damage, prioritize these. If the damage is either similar, or the thread to your forces is very low (comparing the numbers) one should always go on taking out the easiest groups of enemys to eliminate, as their damage (over time, while taking out tankier enemys) will be relatively high.
(Not accounting for different support units, bonuses and range related tactics)
That's because there is a ratio between Damage Per Second and Time To Die to consider when choosing which ennemies to attack. If you increased the Skirmishers attack bonus against Archers, they would eventually reach a point where it would be better to target the them first. The overkill damage due to focus firing can also play a role when you have a big army. But when you have a big army, you rarely focus fire anyway...
I haven’t seen this game before, but this video looks interesting enough for me to look for more
surely it's a matter of a. how much extra armour it has compared with b. how much extra damage it does. If a is bigger than b then leave it till later, but if b is bigger than a you have to take it out first since it will be damaging you a lot and you can stop that damage easily
The big differences is the DPS from the tanky units. The Skirmishers and Turtle ship's rate of fire is low while the Huskarl is the opposite so it's threat level is way higher.
Excellent video. Very informative. Keep it up. I am sure that many thousands eagerly await your next video. :)
I don't even know which game this is and I haven't watched the video yet, but in my personal experience it's better to take out the weakest enemies first so they aren't pestering you.
After watching, this is actually accurate to my experiences of pvp in WoW
I'm pretty sure the conclusive answer is to target the one with the highest (damage)/(shots it takes to kill) ratio but then only enough to kill in one volley. If you have more, focus fire the extra on the next unit, but this could result in 2 or 3 groups (like in the arbelest case) and would require too much micro and would be unhelpful.
I guess it depends on the defense to damage ratio of the units. The one with low defens/damage ratio are the first to target.
hey spirit, i subscribed when you only had 1500 subs. last year when you started focusing more on skylines i unsubscribed along with stopping play of aoe2, i was happy to have stumbled back to your channel and saw your new vids. made me instantly re-sub and play a game of aoe2 on hd! keep up the good work! oh and if you could talk to the powers that be to finally fix the beserker that would be great =)
Thank you for adding the counter example. It makes me think about hallucinated units in SC2.
Love it when you put the minutes where the answer come out in the video description
You have to consider both the enemies damage output and its tankiness to figure out which units to take out first.
The priority can be easily figured out with the following simple formula:
Priority = DPS/Tankiness
Where DPS would be the amount of time one single enemy units needs to kill one of your units and Tankiness would be the amount of time one of your units needs to kill one of the enemy units.
Ignoring the overkill problem and different unit ranges (short range units lose some of their damage potential due to having to walk more) we can easilly figure out the best way to reduce the enemy damageoutput vs time spent.
One of the more practical episodes! Well done :)
I did the maths and considering a theoritical situation where you have to choose between two type of ennemies (no matter how many there are) the option where you loose the least amount of life is the one where you attack the type of ennemy with highest ratio dps/life.
dps (damage per second) is kind of the force of the unit and the life represent how much time you have to attack it before it stops attacking you.
I suppose we can generally demonstrate that in more general situation with more than 2 types of ennemies, you still want to attack the one with highest dps/life ratio.
Focus glass cannons, before tanks. It is the ratio of attack/defense, not the total value of attack+defense. I use the terms "attack" and "defense" loosely (e.g. defense would account for hp also). In this scenario, archers were more glassy than skirmishers.
The most important thing you left out is the dmg/hp ratio. If the target is tanky, but deals huge damage, it might still be best to focus it
This is actually boss strat for most rts games.
Good vid.
oh snap, looks like I just found a great youtuber! Look forward to scouring through your vids now
I love Spirit's intro montage, so awesome!!!!
I guess the rule of thumb is to target units with the most "damage per health" first.
(To account for most things, damage here is defined as number of hit it takes one ennemy unit to kill one of yours, and health is the symetric)
Skirmisher is a counter unit which counters more by having a big armor than with damage bonus.
I would like to see the test for a mixture of sword and pike against cavalry, with the huge damage bonus of pikes and there not so shining HP, It might be worth targetting them first.
When encountering situations that degrade your situation if you don't do anything, attacking opposition regardless of what their capabilities are at least provide information to you as to the effectiveness of your attacks. Once you get data as to degradation levels inflicted upon you and you upon the opposition, you can then adjust your attacks to see what changes.
I'm surprised the entire video didn't talk about a hits to kill vs hits to be killed product. This product makes sense as a comparison because more hits required to kill an enemy makes them less useful of a target, and the enemy requiring more hits to kill you makes them a lower priority as well.
The Archer takes 8 hits to kill and kills your units in 8 hits. That is a product of 64.
Skirmishes take 30 hits to kill and requires 6 hits to kill one of your archers. That is a product of 180 (and high numbers are bad).
Thus you are dramatically better off killing his archers (as shown in the video, above).
This formula is simple and works for almost any of the situations described above.
The relation should be something similar to this:
(assuming all younits have the same attack speed) To target the tankier unit the amount of damage it puts out compared to the squishier unit must be the same factor or a higher.
So if you have 1 Volley Vs 3 Volleys, It starts to make sense to target the beefier Unit when it does triple the damage or more than the faster to kill unit. But that is probably rarely the case.
If the units have a differing attackspeed the problem gets more complex. We need to convert our numbers to time to kill needed the enemy units (probably in seconds) and convert their damage into dps (Damage per second)
1dps and 1 s to kill vs 1dps and 3stk
The beefier unit deals 3 dmg from the point of targeting to its death, the weaker only 1.
this means targetting the squishy unit creates a loss of 1 damage every second, targeting the tank results in reducing the recieved damage by 1/3 of a point each second. Or better phrased the sum of the dps gets reduces by 1 0r 0.33 per second.
After that short thought experiment my proposedmathematical approach is to:
1. calculate enemy dps and ttk (time to kill)
2. calculate the loss of sum dps by killing the units via dps/ttk ( I think my units will get messed up resulting in dps²)
dropping the units we get a number that is a fraction of the dps like in my beforehand example 1 or 0.33
3. compare all available dps² numbers and work them from highest to lowest.
Someone with more time and enthusiasm towards math can polish that up a bit I guess XD What do you think about that approach? It seems ... a bit ... *mathier* than your (incomplete) practical testing.
However, the more complex army composition gets, the harder it gets to calculate every number and work accordingly, especially when you end up with a range of different ttk AND dps numbers depending on which unit attacks which.
Although the time to kill is highly dependent on the amount of units and the damage they do is. Having enough archers or extremely powerfull units don't necessarily mean you should target the skirmishers first. If you can kill the skirmisher in one volley, you could probably kill 3 Archers in one volley changing the numbers to 3dps² and to 1dps². Just to be aware of that.
This is the same strategy Napoleon used. He would join 2 battalions to attack 1 enemy battalion. This would result only in 20% loss per battalion compared to around 80% loss if he would go 1 on 1. The trick is as enemy forces are getting smaller, the less of a chance to get your soldiers killed.
This just confirms the cardinal rule of damage-dealing I have ever followed: target the weak first unless they are the tank.
This is because of unit balance in AoEII favours defense over offense, the rock-paper-scissors relationship is based on bonus defense against damage type rather than bonus attack against unit/armour type.
It’s all about what is more efficient at reducing the dps. You want avoid tanky enemies early , target the highest dmg/hp characters. The quicker you reduce the dps the longer you stay in the fight snd dish out more damage
This is actually quite simple, you want to take out the one that has higher damage output to damage resistance ratio. So always target the unit that has highest damage per second relative to the time it takes to take it down. Thus you lower opponents damage output the the most and don't let them tank with damage resistance. How to define that is the hard part and comes with experience in the game. I have no idea what dps-dmg resistance ratio an elite teutonic knight has when compared to let's say man-at-arms if I have paladins for example. Also the opposite is true, I definitely want enemy to target my low dps high resistance units and leave my hand cannons alone
I was thinking the same thing. Not a complete solution but I was thinking if you used the time it took for each unit to each the other as a measure of relative strength, like what he did when he revisited the Paladins vs Cavaliers in the Lanchester's Law Video. Once you can attach numbers to the relative strengths I would think you should be able to predict not only which units you should target first, but be able to generalize the answer to whether an engagement is winnable (in theory) or how terrain bonuses affect the answer. Reminds me of the villager gather rate formula video. I want to see this revisited.
Theoretically, it also makes sense to remove the enemy archers first and have a different unit drop in for support against the skirmishers later. I found the video interesting because it clearly demonstrates the value of numbers on the field, unit composition aside. I'm willing to bet if the red units were 5 skirms + 5 archers for a total of 10 instead of 6, the red units would've won every exchange.
I would also imagine any unit bonuses against other units is clearly more noticeable with larger squads. Reminds me of the cavaliers vs paladins video, where you get better value out of more cavaliers vs fewer paladins.
The question is what is week, because is not the same a unit with low pH, low attack but too much armour that other with too high damage, but few armour and pH. So the point is whom you can kill first to reduce more faster possible their damage.
As presented this is a highly deceptive question.
The units presented are not in equal numbers. You are evaluating the strength of individual fighters, rather than evaluating cumulative strength. There are more enemy archers present, so even if they are individually weaker, they are cumulatively stronger, so you are attacking the stronger force first. You are simply defining "stronger" as suits the case you want to make. And defining it wrongly.
This is pure Lanchester's Law in action, Spirit of the Law would you be so kind to discuss that sort of real life strategy in future ? I'd love to see you use your scientific approach, and use of Age of Empires units to explain real life battle tactics
Basically to sum up the video, target the ones that are not tankly so that the base damage doesn’t weaken ur army, unless they are they high damaging or both, so you want to take out that threat first, yet have someone take care of the least tanky or low damage.
Interesting recommendation, UA-cam 🤔
Great video ☺👍 It takes me back to the mid-2000s when I was introduced to AOE by my math teacher in high school. Never played seriously, though. Nostalgiaaaa 😄 AOE games are so fun. And I just remembered introducing my cousin to the game too.
I'm just blown away by how many people STILL play this fav of my childhood....
The mathematical approach to identify the optimal targeting solution is to find the "Effective DPS" / "Effective Health" ratios and attack the highest valued ones first. These values take into account pierce, defenses, bonuses, etc. and obviously will vary depending on both attacking and defending units. This becomes more difficult when multiple unit types are involved, but the game is built on numbers, and taking a purely mathematical approach will yield the "ideal" solution... ignoring tactics and micro skills.
Year late but there's another strat not employed here. In your vs huskarls and milita set up you do a full focus fire with all units. You also put forth that a lot of arrows get wasted. It makes me wonder what a _partial_ focus fire would be like where you allow a portion of archers (near the bottom of your mob) to pick their own targets and whittle down militia efficiently while the majority of your archerers focus fire huskarls.
Note: I am not particularly good at AoEII and Im just applying general RTS gut feeling.
Spirit of the Law intros are the only one I do not skip on youtube.
I'm sure there's an easy mathematical way to figure this out. You just have to calculate the relation between enemy DPM and own DPM against the respective units.
It always makes sense to reduce the number of damage-sources on the battlefield, therefore I would intuitively say that you should focus the weakest units first with the smallest amount of own units firing at them (to not waste any shots on weak opponents).
Usually we should take out the weaker units first unless we encounter that situation with the huskarls where you should focus fire the huskarls first, but usually they would have upgraded their militia line and would have the blacksmith upgrades they used for the huskarls on the the militia so you should have done that instead of dark age militia. It is usually best to go back to our town and avoid the fight altogether.
I Did stay confuse.
If no bônus? Like paladins vs paladin + Champion. Paladins must focus paladins or champion first?
That explains why some rts games you have tanks that can taunt enemies into attacking it for a moment, preventing them from picking off the weak ones
the important point is actually how quickly you can reduce the enemy damage output, that's why the last test got you different results, the militia were barely damaging your units so killing them wasn't a good use of time.
Of course, most of the time the weaker units will still be doing at least moderate damage so it'0s usually better to focus on them
It is about reducing the incoming dps. The faster you kill the enemy the faster their dps drops which means your dps stays up and theirs go down. Numerical superiority will quickly escalate to a quick win once one side starts doing less dps reduction. And typically dps is reduced quickest when you focus the fastest to kill units first. However if the weakest enemy is too weak to actually hurt you then killing them first does not reduce the incoming dps. The biggest thing to understand is that not only does focusing down the correct units decide the battle but also that the early reduction of incoming dps is massively important. Once one side has numerical advantage it is escalates quickly. Make the right choice but make it quickly.
Hi. I'm coming from a various sided view. The principle I'm following is something as follows: reduce enemie's damage output to be lower than yours so that the following damage trading will be in your favour.
Your first example: skirmishers have high dmg but also a lot tankier than archers. So if you can sustain while killing their archers you have won the "trade". In the following trade you will have a higher damage output and therefore outpower them by force/numbers.
Basically im trying to imply that you should reconsider the definition "strong" as in term of attack/defence ratio relative to your attacking force. When you tip the scale you have won the battle. And sometimes it's about taking out numbers. However, calculating this in the midst of a battle should be done quickly and just by a rough estimation, which usually is led by on-hand experience. This i think can applied to any game or martial encounter.
What's your thoughts on this? Fill me in or better - make me see another principle.
It's simple. For very fight you have to calculate sums of integrals of damage for every possible decision, and choose the one giving the smallest.