The Other Worlds Shrine

Your place for discussion about RPGs, gaming, music, movies, anime, computers, sports, and any other stuff we care to talk about... 

  • Strategy game concepts: countering

  • Because playing them is not enough, we have to bitch about them daily, too. We had a Gameplay forum, but it got replaced by GameFAQs.
Because playing them is not enough, we have to bitch about them daily, too. We had a Gameplay forum, but it got replaced by GameFAQs.
 #134951  by Don
 Fri Apr 10, 2009 3:28 pm
Again this is inspired by the article I read on Starcraft which roughly says:

If enemy has 12 Dragoons maybe you counter it with 14 Zealots, but it's more complicated than that because if they're a micro god, they might beat you anyway and then you lose.

Well, it's not a bad idea, but it's like saying Dragoons counter Knights in FFT unless the Knight was Orlandu.

Again I'll go over with example from ROTK 11 which seems to be one of the most balanced strategy games out there. We'll take a simple unit matchup: Cavalry versus Halberdier. The game has a basic rock paper scissors model: Cavalry beats Halberdier which beats Spearman which beats Cavalry. At least in theory.

The game's model works like this. Cavalry has an attack rating of 100 and defense of 80, while Halberdier has an attack rating of 80 and defense of 100. There's a relatively small (10-20%?) bonus for the rock paper scissors advantage since Cavalry counters Halberiders. Cavalry are also faster, so it's almost certain they'll get the first hit in. When you attack, the attacker first inflicts damage, and then defender attacks back with whoever that survived the attack. So if you just mindlessly attack back and forth with unupgraded units, Cavalry always counters Halberdier. So far so good.

Now each unit has 4 level of upgrades. Throwing out the stuff that just cancels each other out, you have the following relevent upgrades:

Cavalry level 2 - increase movement by 1 hex.
Cavalry level 3 - allows cavalry to range attack (2 hex range). This attack is done at roughly 80% the strength of a normal melee attack.

Halberider level 2 - 30% chance to take no damage from normal ranged attack on defense
Halberdier level 3 - 30% chance to take no damage from normal melee attack on defense

So let's just assume you have all your units totally upgraded. Well first thing you notice is that the Cavalry can now range attack the Halberdier with no risk of retaliation, but mathematically it works out it will inflict .8 * .7 = 56% of its normal damage. If a Cavalry makes a normal attack, it will inflict on average 70% of its normal damage while eating a full retaliation from the Halberdiers, so a normal melee attack is almost always worst than a ranged one. Now I mentioned normal attacks. There are also special attacks that generally inflicts about 150%-250% of your normal damage (depending on the game difficulty setting) that bypasses the defensive specials. Special attacks require willpower which is a limited resource. Once a unit runs out, it tends to stay run out unless they resupply or go back to a city. For all practical purposes, a unit never gets its special attacks back once it uses them up unless it withdraws from combat.

Now it seems pretty obvious the solution is just use your special attacks on the Halberdier. That works for normal mode where special attacks inflict 250% of normal damage, but not on Super where special attacks only inflicts 150% of normal damage. In a well fortified area, both unit is likely to exhaust its special attacks before doing meaningful damage to each other. Good players will also combo willpower draining attacks as much as possible which gives you even less special attacks to work with.
So what happens when both units are out of specials?

Well let's look at the abilities again. Halberdier's special only works on defense. If it goes up to Cavalry and attack, it is going to eat 100% of the Cavalry's retaliation with no chance of damage prevention. This means it will almost certainly lose even if it is the attacker in every fight despite attacking first. Likewise the Cavalry will not be attacking in melee range since the defensive bonus of a Halberdier means it will win. So, what is going to happen is that the Cavalry will probably just range attack while inflicting low damage, while the Halberdier will not be attacking at all.

In a real game situation, the fight might look something like this. Let's say you started with 1 unit of 20K Cavalry versus 1 unit of 20K Halberdier. After all your special attacks are used up, you might be looking at say 16K Cavalry versus 14K Halberdier. They will fight some more and you will end up with say 10K Cavalry versus 7K Halberider. At this point the Halberdiers will probably stop attacking and just defend (remember most of the Halberdier losses comes from trying to attack Cavalry, not being attacked), or simply retreat if this is an option. If retreating is not possible, given enough time you'd probably see the Halberdier unit completely destroyed while the Cavalry hover around ~10K strength.

Now does that mean Cavalry countered Halberdier? I mean you started with the same number of troops, and the Halberdier all died while Cavalry still has half of its original strength. But now we get into the economics. For most players in ROTK, horses are roughly twice as expensive as any other unit, because the specials to produce horses cheaply is rare while the specials to produce other weapons are common. Further, Cavalry are the strongest unit in the game by a mile. If you tried to fight Cavalry with their supposed counter, Spearman, most of the time you'd see Cavalry destroying their supposed counter by even greater margins (depends a lot on terrain). Indeed, the counter to Cavalry is Cavalry, not Spearman.

Now you ask why not just have all Cavalry? Well that's what you usually want except horses are expensive. Also, each city can only hold at most 100K of every weapon type. If you sent 100K horses, and enemy defended with 30K Calvary, 30K Archer, and 40K Spear, and managed to trade resources. That means in a few months you're going to see the lowly Halberdiers + 2 Catapults taking out your city for basically no loss because Halberdiers will just form a protective front around Catapults that takes out your city. If you had any Cavalry left, this wouldn't work because Cavalry can break through defensive parameters very easily and then destroy the Catapults before they get into range.

Therefore, if someone is fighting your Cavalry with Halberdier, that means they have a resource advantage on you all else being equal. Of course, if you can make cheap horses so that it's not a resource disadvantage, then your opposition is at a huge disadvantage. If you simply keep pressing your attack, what's likely to happen is that the enemy will either send out 20K worth of Cavalry or some combination of 20K in Spear/Archer/Halberider in an attempt to trap your remaining 10K Cavalry, and your 10K will quickly get killed, which means at the end you lose 20K horses while they lose no horses. They could lose more than 20K in other resources but again, these other resources are not as expensive as horses. And remember, once your horse supply goes below about 20K, you'll start seeing Catapults coming your way, and Catapults obviously counter everything if not killed before they're in siege range since you can't fight a war without a city.

So now it sounds like Halberdier always counter Cavalry? Not really. Let's say you attacked with 5 units of 20K Cavalry versus the same thing on Halberdier. First thing that will happen is all 5 Cavalry will focus all their special on one Halberdier unit and kill it, while Halberdier have no chance to do the same. Even if they got one of the Cavalry units low, since Cavalry are the fastest unit that unit wll almost certainly withdraw. After the Cavalry are out of specials, they'll simply move around the Halberdiers and sack your city. Remember, Halberdiers can never attack Cavalry normally without taking massive retaliation. The defender will simply lose his city, and at that point the Cavalry will go into the city to resupply finish off the Halberdiers, who have no chance of escaping because Cavalry are much faster.

So what exactly counters what? The answer is that it depends. A Cavalry will almost never lose to a Halberdier whether it's 1 on 1 or 5 on 5. But if you know what you're doing you might be able to use a Halberdier unit enough to slow down a Cavalry to the point where it is effectively countered. In the previous example, almost certainly the 10K Cavalry side will withdraw their forces because you don't want to get trapped. So the attacker will have an edge in troop numbers, but the defender now has an edge in weapon numbers. Depending on what these numbers are either side could've won. A fight in ROTK is never as simple as "okay they have 100K of troop Y so we bring 100K of troop X".

Catapults gets countered by Cavalry (technically anything can counter it since it cannot fight back, but Cavalry can get to them the fastest, and they do most damage per attack). Why would you bring a Catapult if you know enemy has Cavalry? Maybe you do it so you can draw out their Cavalry and attempt to kill them if you know that city cannot make more horses and is only getting them through transports. Of course it can easily backfire if the enemy Cavalry manages to destroy your Catapults without getting trapped. But if you're able to trap the Cavalry that will no doubt attempt to destroy the Catapults, then you can say your Catapult really countered the Cavalry, even if the Catapults got destroyed.

That should be what countering is about. It does not always need to be a strict unit advantage. You can actually lose a fight and still come out ahead if you consider other dimensions like mobility or resources.

 #134952  by Mental
 Fri Apr 10, 2009 3:42 pm
Play me a few games of Warcraft 3 and I will teach you all you like to know about wargaming strategy. :D

We'd see exactly how much these 5000-word essays help you when your line of Abominations and pathetic support contingent of Crypt Fiends is having interference run by upped-armor Treants that cost me nothing while being distracted by Taunts from my Mountain Giants on the front line and getting outflanked by my leftover Huntresses and hit from range by Dryads (with all but the dryads Roared from two Druids of the Claw giving support), with an Immolated and Evading Demon Hunter running around behind your lines causing damage and virtually untouchable due to a high moverate combined with Boots of Speed. The screams of your dying as your rotting corpses feed the forest floor for my Ancients, THAT is strategy.

 #134953  by Don
 Fri Apr 10, 2009 4:23 pm
Warcraft 3 is just a game about memorizing and clicking really fast. That is not strategy. Everything you described can be countered by 'clicking very fast'. At a strategic level the counter to what you describe is absolutely trivial. In Warcraft 3 all the counters are absolute. You got a DH with Immolate? I position my unit exactly far enough so you can only immolate one of them at a time. You cast Roar of the Wild, I dispel it. You have Dryads then I use a unit that is designed to kill Dryads. You have a Mountain Giant with taunt? I just click attack on the right target very fast to counter it.

Now can I actually do this? Probably not. That means my APM skill is lacking. And if you go against someone with an APM of 1000 they can probably just do everything I described and then you'd be the one getting owned.

Strategy and countering is not memorizing some chart because it says A always beats B. In the siege of Washington in Panzer General I cannot get close to a Major Victory until a friend pointed out you use your bombers to counter the air defenses around Washington. But wait, air defenses are supposed to counter bombers, not the other way around? Any guy can tell you you're supposed to take out air defense with either artillery or some kind of ground assault! But Washington is so well fortified that that you'll never just find a situation where you consult your unit charge and says A always beats B. Every unit in the city area will always beat anything you have if you're not willing to make some sacrifices.

The whole series of event actually goes like this:

1. You use strategic bombers to suppress anti air. This merely stops them from counter attacking, and you can't even kill them like this. You will lose a lot of bombers this way.

2. With anti-air suppressed, you use your fighter advantage to kill whatever fighter/bomber the enemy still has.

3. Your remaining strategic/tactical bombers can now actually bomb the artilleries defending Washington without getting instantly killed.

4. With artillery destroyed/suppressed you can use your infantry/tanks to assault the main city. Since their units are as good as yours, expect to lose a lot of stuff here.

5. Once you actually find a hole in the defenses, you can use your tanks or infantry to kill the air defense that is still well protected.

 #134964  by Mental
 Fri Apr 10, 2009 7:20 pm
Don wrote:Warcraft 3 is just a game about memorizing and clicking really fast. That is not strategy. Everything you described can be countered by 'clicking very fast'. At a strategic level the counter to what you describe is absolutely trivial. In Warcraft 3 all the counters are absolute. You got a DH with Immolate? I position my unit exactly far enough so you can only immolate one of them at a time. You cast Roar of the Wild, I dispel it. You have Dryads then I use a unit that is designed to kill Dryads. You have a Mountain Giant with taunt? I just click attack on the right target very fast to counter it.
No, it's not even close to a "game about memorizing and clicking really fast", Don. It's incredibly deep. Every single thing you just mentioned as a counter has another counter to it as well, but I don't have the patience or desire to sit here trading essays with you.

I mean, I'm actually offended. I've put probably a year of VERY active play into that game (probably an average of two or three hours a day) over the course of the two years I've actively played it - one when it came out, plus the last one - practicing that game and I don't even consider myself more than a journeyman. I have so far to go that it's incredible just in terms of strategy, and that doesn't even count learning what items and creeps are on each of several dozen maps and where and at the highest levels of play not knowing even that shit will lose you the game. You are full of it. I think the shit you're posting is just about useless in a real match.

Play me. I will kick your over-analyzing ass so far under a table that you'll finally realize that you're just wasting bandwidth with these 5000-word policy papers.

 #134968  by Don
 Fri Apr 10, 2009 8:00 pm
So you being better at a game you spent a lot of time playing that is known to be heavily dependent on fast clicking is supposed to prove something? I like how you even mentioned Mountain Giant. That unit basically says 'countered by killing this unit last', ala clicking very fast. There is zero strategic depth to the unit Mountain Giant. All it does give you an APM edge because it takes one click of taunt for you but you need to click several times for the opponent to counter the effect of taunt. It is incapable of doing any meaningful amount of damage for its cost. All it can do is take damage. That is not the same as using a Halberdier defensively where you'd rather the enemy attack you over you attack them, but the unit still has offensive value.

The guy who made Warcraft 3 doesn't have to be a strategy master. Certainly the best Starcraft player in the world is not the designer of Starcraft. Warcraft 3 like most RTS has hard counters and the only wiggle room is that hard counters can be countered by APMs. Zealots counter Dragoons unless you click really fast on the Dragoons, then it does not.

At any rate winning and strategy isn't even about the same thing. If I was playing you a turn-based game and I only cared about winning, you'd not be able to even move your units before you lose because I would just attack you with something that has no counter that will also remove your ability to control any unit. Your Warcraft 3 strategy is presumably what you feel pretty comfortable about this force winning so we'll say that has a 99% success rate. Well in a turn based strategy game if I attack you I have either a 100% success rate or I would concede before the fight started.

In Starcraft people concede when they go with one strat and found out the enemy saw through it. Well in a turn based game it'd be over after the first attack, sometimes earlier. One time in a game of HOMM3 I scouted the other guy by turn 3 and saw that he didn't get the resources to put up a level 6 building while I had them. Well technically it took 3 more turns before the champions arrived to win the game but the game was already over at turn 3 because at that point the other player has no strategy that can counter my champions. If we're equally skilled I could just say 'I got champions coming in 2 turns' and the other guy should say 'GG' and forfeit instead of trying to fight me and lose anyway.

Now if we're interested in a game to explore how deep strategy is then I should let the other guy have the 2 more turn he needs to get something that counters my champions, but if we're playing competitively, strategy never enters the picture at all. Incidentally this is why master gremlins are banned on competitive HOMM3 games prior to the patch that weakened them because otherwise you might as well forfeit at the start of the game if the other guy picked Tower and you are not Tower.

 #134971  by Don
 Fri Apr 10, 2009 8:39 pm
You mentioned stuff like Creeps and items. I consider that about as useful as Chess openings. If you're on map X there's always a level 6 creep that does Y damage, and it will always drop either item A or item B. There is clearly some optimal way to approach the said creep and clearly some item A is the most useful that can adversely change the game. Yet this is also 100% predictable. You can trace the initial steps like a giant flowchart just like you can trace out the opening of Chess. There is certainly value in studying this just like there is value in studying build orders but you're basically just executing a series of predictable manuevers that works every time. If you can't do it, it's because you didn't click or move in a certain way. I know I can't do it because I don't really care about Warcraft 3. But obviously anybody playing the game competitively can certainly do it with enough practice.

If you go look at Chess, why do you open with soandso's whatever and they response with someone else's whatever? Because that part of Chess is closed to solved so if you don't do this you're at a big disadvantage. But there is no strategy involved in using known, nearly solved patterns. Deep Blue might be stronger than the strongest human player but nobody will say Deep Blue has good strategy. But it sure can solve Chess.

I suppose you can say then there's creep jacking and stuff but then this is just a matter your APM versus his APM. If you scouted him while he's creeping while he did not scouted you approaching then you have a sizeable advantage, because you probably have some kind of APM advantage on him to see him before he sees you.

And the items and stuff you get from this gives you a predictable advantage. You alluded to Boots of Speed being a big deal. Well in ROTK 11 if I have a Halberd of the Sky and you don't, that basically means whenever I get a critical charge there's a 25% the unit getting hit is now dead regardless of how little damage was inflicted. Is this a strategic advantage? I guess you can say the I-WIN button is a strategic advantage but I really don't agree. I don't consider the guy lucky enough to get a Halberd of the Sky first and then automatically winning by the virtue of having it done anything strategic.

To me strategy is defined as anything that a computer will not slap you silly if it was actually serious. Can you ever have an advantage in creeping against a computer assuming it's not artificially handicapped? Of course not. It will be able to kill every unit in the creep perfectly since it can simulate the outcome of the battle before it even happened. The computer can certainly simulate scouting perfectly even if it doesn't just outright cheat. It's pretty trivial to say pick one peon and have it trace out a danger zone around the army you're using to creep, and obviously the computer wll always see you before you see him since its attention span is way more than any human.

In fact since the computer actually knows how fast every unit is, and how long it needs to kill the creep, it can even see you and determine that it saw your army that is 8 seconds away from the computer, but he would've killed the creep in 7 seconds and got the item already, so there is no risk of being caught by both the creep and your army. Or it can determine that your army will get here while the creep still has 400 HP left which puts it at a disadvantage, so it will just withdraw for now, or do something else to avoid being in a disadvantageous position.

On the other hand it's not clear if the ROTK 11 AI is ever holding back. The computer even on Super does not receive extra resources beyond more starting troop/gold. In combat the AI is clearly forbidden from using any lockdown strategy, namely any strategy that will kill your entire army without you even having a chance to move. But other than that it doesn't hold back. The Hard or higher AI basically fights like any top player will do. Perhaps it doesn't try to absolutely optimize every kind of thing it can do, but it is certainly not giving you any easy breaks.

In fact since the AI plays like a top player, it purposely tries to build slower to ensure you would at least have a manpower edge when you go against it, and then it will not hold back knowing it has given you a head start. It's pretty much exactly the opposite of the AI you see in RTS where the AI is written to suck on purpose (it will never do perfect focus fire which it is clearly capable of) and tries to make up by having twice of your troops. If the AI has the same resources as you do, will they always win or always lose to you? You don't really know. I've seen the fight go either way, just like it could against a human player of equal skill.

In the case of Warcraft 3, it's pretty clear if the computer computed enough things, it will never lose to you in combat (or it will retreat if it cannot win). In ROTK 11, it is not clear to me this is true. It is possible the AI in ROTK 11 can compute enough to always have an advantage at the local level, but it doesn't have the strategic insight to the big picture. For example it can always win but it won't really know like say you only attacked this place to draw out his only character with the Majesty skill and now you can totally abuse him on another front with Divine Spears. Now the AI in Warcraft 3 doesn't necessarily have a good grasp of the big picture either, but it is hard for me to imagine an entity that never loses a fight at the local level could lose to a human player in anyway. Remember that the AI can easily simulate perfect scouting.

 #134979  by Don
 Fri Apr 10, 2009 9:24 pm
Actually, at one point I tried to write a targetting algorithm for Warcraft 3 using the in game tools. My goal was come up with something that will beat someone doing obvious focus firing. The triggers turned out to be too limited to anything sophisticated but by tweaking the AI targetting priority and movement I can come up with an AI that performs increasingly well. If you give the AI 6 units with 3 different armor class/attack type using the Warcraft 3 model, those 18 units will reliably beat 24 units using autoattack. Those units were given infinite range so there is no movement penalty for retargetting. When pitted against a human player controlling the same player, they can always beat me since I suck at this game, but they won't always beat someone who is actually good. However the margin obviously goes in the AI's favor as parameters are tweaked better. It's obvious that there exists some combinatino of parameters where the AI will eventually beat the best human player in the realm of focus firing.

Let's take the unit Dryad. I forgot if Abolish Magic resets your attack cycle, but we'll assume for the moment anything worth being dispelled is worth losing one attack round. Then an algorithm for Dryad can be written as such:

If totalNumberOfDispelsInTheArmyLeft() >= (EMERGENCY_DISPEL_THRESHOLD), then cast Abolish Magic

If enemyInRange() and canBeKilledInOneConcentratedAttack(), attack that unit.

Otherwise, if enemyInRange() and notPoisoned(), attack target.

Otherwise, if enemyNotInRange() and notPoisoned() and targetDoesNotOutrangeMe() and DryadIsNotUnderAttack() and TimeToGetToTarget() <= SOME_CONSTANT, attack that target.

Otherwise, attackUnitWithLowestHPInRange().

If DryadIsAboutToDie(), then retreatToTown(), unless DryadIsGoingToDieAnyway().

Now all these functions need to be tested and tweaked, but at some point this will be a comprehensive list of everything a Dryad is supposed to do. At that point an AI running this algorithm will use a Dryad better than any human player can. Likewise you can do this for just about every unit. Some unit might be harder to do but the problem is definitely solveable, and the AI need not to be perfect at every unit to beat a human player. And I don't think anything that can win by having a perfect algorithm is strategic.

 #134982  by Mental
 Fri Apr 10, 2009 9:31 pm
Play me, or put a sock in it. I'm not going to read all of that. You keep talking like you're God's gift to strategy analysis, and it just comes off as full of it until you reassure me that you can actually play. I can guarantee you that if I beat you it won't be because I "clicked faster", and you will easily be able to tell.

 #134984  by Don
 Fri Apr 10, 2009 9:37 pm
I'm curious why you think I'm analyzing turn-based game and I'm supposed to prove something by beating you in a RTS that I only played a little compared to you playing the game continously for multiple years.

When a War 3 patch changes the damage of a huntress from 15 to 14 do you tell the devs better play me or you can't change that damage to 14? I'm willing to bet there's a good chance you're better at this game than the guy who decided on the change.

 #134985  by Mental
 Fri Apr 10, 2009 9:56 pm
You started out analyzing STARCRAFT. That's not turn-based. Then you switched over to ROTK and the others. And if you can't put up a respectable showing, why bother with the analysis? No one is going to look to you for strategy tips, they're going to go watch top-level players and see what they do. Even if all you're doing is coming at it from a top-level game design standpoint instead of a bottom-level player strat standpoint, you have failed to convince me that you understand the fundamentals of what wins games in WC3, and if you don't, you're going to make mistakes in terms of trying to decide how to balance the game. I cannot, cannot, cannot actually think that anyone who believes that WC3 is a game about "memorizing and clicking faster" actually knows a thing about the game at more than a very, very surface level.

I have very great faith that the devs know what they're doing in terms of balancing the game. I might be able to beat the PROGRAMMERS, but I will bet you that there are at least a few beta testers at Blizzard who would understand what I'm talking about and be able to absolutely regulate all over anyone on this forum, myself included.

I have played some players who just handed me my ass without effort and had 70-80% records on battle.net and I can absolutely tell they are leagues above me in terms of having a sense of exactly how much damage their units will be able to do against any other given unit, a sense of how damaged EVERY one their own units are without having to look at the life bars (and yes, I play with always-visible health bars, the equivalent of holding down ALT), a sense of how to micro to keep retreating units EXACTLY out of ranged fire range during a retreat, and et cetera. There are some games you play where you go up against an army that in 95% of the other games you've played would have been a fair fight and don't even manage to kill a single one while simultaneously having your own units cut down like Steve Buscemi's character in the wood chipper at the end of Fargo. That is due to STRATEGY, and it is due to MICRO. It is due to understanding a wide range of VERY subtle differences, such as the height of the map, the exact amount of time it takes a given army to cover ground, the exact type of damage that each unit has (piercing, melee, magic etc.) and how that relates to the exact type of the armor that every unit you fight has, the exact amount that buffs or jinxes will affect all those things, and a hundred other factors more than just "memorizing and clicking fast". It is due to scouting, and unit sight radius. It is due to knowing which creeps on a map a given player might be more likely to creep early so that you know where to creepjack.

When I started talking about Boots of Speed, the ONLY sense in which I meant that in is with regard to a powered-up Demon Hunter, because it makes him just about uncatchable. It won't do shit for a Keeper of the Grove or Priestess of the Moon who stay at the back of the battles anyway and can usually really easily retreat and/or stay out of danger. And EVERY ITEM in the game is like that and has subtle interactions with EACH HERO that can easily win or lose you a close game on their own. Having Boots of Speed on my DH one game literally allowed him to escape a tight battle with something like fifteen hit points left, whereas if he'd died the enemy heroes would have probably leveled up themselves, and hopefully I don't have to tell you that high-level heroes can practically win battles on their own.

To put it simply, Don, you're talking too much for your skill level. I don't think you'd be able to do a good job of balancing WC3 units. You don't really understand the game, and for some reason it just bugs to see you talking as if you think you could outdo the dev team.

 #134987  by Don
 Fri Apr 10, 2009 10:28 pm
I started out with Starcraft because it was a good quote for 'counters should work like this, except not when this happens'. I don't think it was a good example but it's an element I like to think about.

Micro and strategy are inherently not the same thing. I don't know how you can even confuse the two. If you pulled amazingMicroManuver() and win, and I did not, it's not because I choose to not use that manuever. On a strategic level your choice is always to execute amazingMicroManuver(). You can't always do this because humans don't have infinite attention span/clicking speed. It's just like in Street Fighter 4 the optimal strategy is to execute the most insane combo you could do that will inflict the maximum amount of damage. Most people don't do it because they cannot pull off those combosm, but it is never a decision to not use them. If someone spanked you because he was smart at pulling wounded unit out and you did not, it's not because you at a strategic level decided to leave your unit to their doom because there is some value to let your unit die. The guy outclicked you and was able to move his wounded unit out of the way while you cannot.

At least in Starcraft, which is still way too dependent on fast clicking, has the concept of an APM advantage. For example in the same article it was referencing a game between two top players (Protoss vs Terran), and the Protoss player had 9 Gateways and 3 more expansions than the Terran, and the Protoss player correctly figured that if he just continously send waves of Dragoons and other stuff, the Terran player has to crack eventually. The Terran player was using a ton of micro to remain alive against these odds but eventually he lost. The strategy employed by the Protoss trumped any amount of micro skills the Terran player had.

I also really don't understand why you have this appeal to authority attitude. If a top Warcraft 3 player beat you and then said you know in retrospect this game is totally dumb and I wasted 3 years of my life playing this, does that mean you now believe Warcraft 3 is totally dumb? It's almost certain there must exist someone who is better than you that also believes the game is dumb. Consider how much time it takes to get good in a RTS, it'd be very weird if designers have time to play at a top level. Beta testers, sure, but they're not the one making the decisions.

The lead designer of WoW is a guild leader of a second rate guild in EverQuest that quit due to not being competitive, but clearly the guy knew a lot about MMORPG or at least enough to land him the job. Now I don't necessarily think WoW is great by the virtue of its bigness, but certainly being bad a MMORPG didn't stop a guy from designing content for the most successful MMORPG ever, so he must have known something even if he wasn't the close to the best MMORPG player.

 #134989  by Don
 Fri Apr 10, 2009 11:44 pm
I found the article I was thinking of for Starcraft on how APM directly affects your ability to play:

http://www.netharuka.com/games/starcraf ... ne-decade/

In particular: "Careers as a gaming professional are short, as the brain’s ability to process APMs degrade noticeably with age."

Now obviously professional Starcraft players need to have a good grasp on the strategic element, but if APM didn't dominate what it means to be a pro, then you would not see short careers as a professional Starcraft player. You can know all there is to know about Starcraft but if you can't manage 300 APM, you're probably not going to be competitive against the guys who can do it.

And Starcraft has a far bigger macro picture compared to Warcraft 3, if only due to size difference (far more total units), and that individual units in Starcraft have decidedly less variation compared to Warcraft 3.

Edit: I actually watched the video for Starcraft game. It basically looks like with an APM of a world class player you can turn a RTS into a turn based game. Basically most of the game you see the two people trying to get some kind of minor advantage by using a ton of microing. Every exposed unit is killed very quickly. This is not deep strategy stuff, but the fact that you can do this in real time says a lot about the reflexes of the players involved.

It also looks the inability to entrap escaping units made the fight go on for much longer than it should. If this is a turn-based game with any reasonable escape denial, or better yet something like Warcraft 3 where escaping units are trivially trapped given how good those two guys are, then the game would've ended a long time ago. I also noticed that both players abuse the fact that melee unit needs to come to a complete stop to attack to avoid a lot of damage that should not have been avoidable the whole time. Again I don't consider that strategy. I certainly can't do it, but the ability to abuse pathing isn't a strategy.

 #134997  by Mental
 Sat Apr 11, 2009 12:46 pm
I am now done with this thread other than to offer one final invitation to a Warcraft 3 game. All you're doing here is verbally masturbating. Let me know if you feel like playing instead.