EPTFlinching in the Myth Franchise
![[image loading]](http://i.imgur.com/jSWHn.jpg)
In Myth the vast majority of the units in the game will flinch when they take damage. So if a javelin hits a warrior then:
(1) he will briefly stop moving
(2) any attack he was performing will be delayed by the amount of time it takes for him to recover from the flinch
(3) he will go through a short animation to demonstrate that he has been disrupted by the hit
To make things a bit more precise, let’s consider a unit with the following attributes:
- 1 attack every second
- Flinch recovery time of .3 seconds
If this unit were hit by an arrow midway through its attack animation, then rather than resetting the cooldown on that unit’s attack entirely, it would not count the last attack as complete, and would instead set the cooldown back .3 seconds from “ready to attack.” If the unit were at 0/1 then it would have to wait 1 full second before it can swing again, but if that unit is at .7/1 (as a result of losing .3 from a fully-cooled-down 1/1) then it only need wait .3 seconds before attacking again. So the flinch delays the attack by the flinch recovery time; it does not fully reset the cooldown.
Had this unit just recently completed an attack, however, and was at .5/1 on its attack cooldown when it was hit by the arrow, then it would, as in the last case, simply set back the cooldown by .3 seconds, setting the unit to .2/1. And to reiterate, if this unit were moving when it was hit then it would be stopped for the duration of the flinch recovery time as well.
There are only several units in the Myth games that do not really flinch, and they are the very large and fearsome ones such as trow and forest giants. (They can flinch very briefly if hit by a mortar or something else very potent, but very few things in the game make them flinch and if they do it is largely insignificant.)
A Revised Flinching Mechanic
The flinching system in Myth, in my opinion, is not ideal. I believe it can be improved upon in a few ways to make it more interesting and coherent for a modern, competitive RTS title. The flinching mechanic I would like to see experimented with would retain the following features:
(1) every unit type has a base flinch recovery time
(2) flinching interrupts attacks and abilities
(3) flinch recovery time determines how far back the unit’s cooldowns will be set on their attacks or abilities
(4) every unit type has a stature, which would consist of 3-4 tiers (e.g. small, medium, large, giant)
(5) every unit type has an attack strength, which would consist of 3-4 tiers (the number would have to be identical to the number of stature tiers)
(6) in order for a unit to cause an enemy unit to flinch, its attack strength must be greater than or equal to the stature of the enemy unit (tier 1 attack strength would not make a tier 2 stature unit flinch, but would make a tier 1 stature unit flinch, etc.)
(7) for each tier the attack strength is greater than the target’s stature, the flinch recovery time is increased by .1 seconds (such that a tier 4 attack strength hitting a tier 2 stature with a base flinch recovery of .3 seconds would result in a total flinch time of .5 seconds)
(8) flinching units would not immediately come to a halt if they had been moving, but would instead be slowed for the duration of the flinch recovery time
The reason flinching is based on the relation between attack strength and stature, and is not simply a stature to stature comparison, is that there would be units of a smaller stature with very powerful attacks (such as spellcasters or certain kinds of artillery units). Also, this type of system allows for some strategically interesting unit designs that are not possible otherwise, such as a tier 3 stature, tanky unit with a tier 1 or tier 2 attack strength. This unit would be good at absorbing a lot of damage and dealing with lower tier units, but would not be as strong in a straight up fight against larger units due to its lessened ability to bring about flinching on hit. (It would be similar in spirit to units from Magic the Gathering that have higher defense than attack, such as a 1/3. This could obviously be balanced out by its cost and tech requirements, as it is balanced in Magic by mana cost, to make it an interesting choice at certain stages in a game.)
Overall this system would add an additional layer of complexity in unit compositions, as picking units according to their relative statures and attack strengths would become incredibly important. If all of your melee units are being swatted around but cannot phase the troops of the enemy then the result will be far more one sided than the other attributes alone (health, damage, attack speed) would suggest.
Some of the Implications
So how would this affect skillful micromanagement, the length of battles, the overall look of the game, how the units in the game “handle,” or the extent to which things such as the raw number of units or attack speed scale?
Micromanagement
![[image loading]](http://i.imgur.com/riOhz.jpg)
(1) Flanking becomes more effective. Assume that the units in this crude schematic are all the same type of basic melee unit, and that the red lines indicate the attack path of the red units as they come in from the side. Now this would clearly be a bad engagement for the blue player regardless of whether or not flinching was in the game, but when flinching is involved it becomes even worse. This is because the two unlucky blue warriors that will be hit first by this attack will not only die very quickly, but their damage output will be greatly reduced. If this schematic represented zealots in Starcraft 2, where there is no flinching, the two blue units on the edge would die quickly but they would still be hacking and slashing away at the enemy zealots for however long they were alive. In contrast, if we imagine these as units with a flinching mechanic in play, we can see that the two blue units on the edge would be getting hit by 3-4 units at a time, and so would have very little chance to actually get off an attack of their own. If every hit sets back the cooldown on their attacks by .3 seconds, and they attack once every second, having 3-4 units gang up on this unit will make it so that it attacks very rarely or not at all. The end result of this will be that the red warriors win significantly more easily in a system with flinching than one without, making attacking into an enemy from a side they are not defensively arrayed to deal with that much more effective.
![[image loading]](http://i.imgur.com/emUAY.jpg)
(2) Formations and surface area become more important. In addition to the usual benefit of simply having more of your units attacking than your opponents, flinching makes a scenario such as the above even more favorable for the player with the concave. As you can see in the diagram there are 3 units for blue that are being hit by two of red’s units. So, on the assumption that red’s units have sufficient attack strength to make the blue units flinch, the damage output of those three blue units will be drastically less than it would be if they were fighting one on one. So the concave not only provides the advantage of having more units dealing damage; now it also reduces the damage output of several enemy units that are in the fray as well.
(3) Focus firing can be especially powerful. As noted before, if the units that gang up on one unit have sufficient attack strength to make that unit flinch then they will not only kill that unit quickly, as you would expect in a game like Starcraft 2, but they will also greatly reduce its damage output while it still lives. If enough units of sufficient strength are hitting the unit in question it will not be able to attack at all. This could be especially powerful against units with a lower tier stature and higher tier attack strength and damage output, as standard infantry or ranged units could quickly shut down their offensive capability given proper positioning and targeting.
(4) Retreat is made more costly. Since units are slowed for the duration of the flinch recovery time when they flinch, retreating becomes a trickier business. If several units are hit as they about-face and try to run away then, depending on how fast the pursuing units are, they may easily be hit several times in a row and never be able to escape. The effect would be similar (though briefer) to the concussive shells upgrade on a marauder.
![[image loading]](http://i.imgur.com/53Rnq.jpg)
(5) Incentivizes the abuse of stature/attack strength relations. In order to minimize flinching on your end of the battle and maximize it on the opponent’s, a player must intelligently position, target, and rearrange their units based on knowledge of stature and attack strength relations. In the above example we can see that the blue player’s large units are attacking the small units of red, while the large units of red are attacking the large units of blue. Further, two of blue’s smaller units are attacking the large red units. This will result in the large units of blue flinching every time the large units of red attack, while the large units of red will not be flinching at all. Since these are the bigger damage dealers in the battle this is highly relevant. Blue’s main damage dealers will be interrupted periodically in dealing damage, whereas red’s will not. Further, the two smaller units that blue has attacking the large units of red are only doing damage—they are not causing flinching. The net result of these differences in targeting will be that blue’s army is flinching significantly more often, and thus is dealing less damage, than red’s army.
(6) Incentivizes the abuse of attack speed differences. If an attacking unit can cause its target to flinch then that unit’s attack speed will be of great importance. Units with a greater attack speed will cause the enemy unit to flinch more often, thus dampening its damage output. For this reason attack speed scales far more effectively when flinching is involved than it would in a game like SC2; not only does increasing attack speed increase the damage output of the unit, it also increases the extent to which that unit disrupts the attacks of its foes. This can lead to interesting scenarios where, say, a tier 2 stature and tier 2 attack strength unit with high attack speed will devastate lower tier units but, on account of not being able to make larger units flinch, be highly cost inefficient against a unit of tier 3 or 4 stature. The unit would rely for its potency on its attack speed having this effect, so facing a foe which did not suffer the flinching would greatly reduce its effectiveness in battle.
Battle Length
Another consequence of the flinching mechanic is that battles generally last longer than they would if the same units did not cause one another to flinch. The reason for this, rather straightforwardly, is that many of the units will be spending brief periods of time flinching, and so will not be dealing damage during that time. The most likely consequence of this is just a general lessening of DPS in battles, which gives players slightly more time to make decisions and control their units, and spectators slightly more time to analyze the situation and enjoy the spectacle.
Unit Handling
The effects on the mechanical “feel” of the units would be minimal. There would be more instances where units were briefly slowed as a result of taking damage and times when a unit was incapable of attacking or was effectively perma-slowed as a result of being ganged up on, but there would be nothing in this flinching system to greatly impede constant and effective micromanagement (a la fungal growth).
The Look of the Game
The contribution of flinching animations to the overall look of the game is quite noticeable. Gone would be the days of units standing toe-to-toe, ripping each other to pieces with nary a hint of the bullets/arrows/fireballs/axes/swords that are killing them causing any discomfort. Units would visibly respond to being damaged, and that response could (and should!) increase in severity as the differential between the unit’s stature and the attacker’s strength increases. That is to say, a unit might have a brief, jolted animation when hit by a unit of comparable size, but might buckle at the knees and wobble briefly when hit by a unit two tiers higher.
Further, when most of the units in the game flinch most of the time, it makes those few units that rarely or never flinch that much more intimidating and exciting to watch. An ultralisk not flinching would strike me as a lot more meaningful if the marine that is 1/50th its size wasn’t calmly tanking the ultralisk’s pincers while continuing to fire off rounds. The flinching system gives higher tier units the look and in-game feel of being especially badass. Having attacks not phase the unit suddenly means something, as you are well aware that a lesser unit would be absolutely useless with 6 warriors hacking away on it, while your giant unit can eat the blows and keep on kicking ass.
Finally, there is just a general sense of realism that this system (properly employed with quality animations) lends to the battles. Seeing a mortar round hit a pack of units is made more tense and emotional by those units reacting to the barrage, rather than only noticing their health bars move as they continue unaffected along their original path. Whether or not realism in competitive RTS games is a good thing I leave to your judgment, but the fact that flinching makes things seem more like real warfare doesn’t seem debatable to me.
Concluding Remarks
I find that these are probably the most relevant and straightforward implications of using a flinching mechanic in an RTS game. I’m not completely convinced that this would be a superior system for competitive RTS titles, but, given the considerations laid out here, I think it has a lot of potential. I would really like to see an attempt to implement something along these lines, as the potential for highly intelligent micro, attack angles, and positional play separating the good players from the mediocre more reliably than more common systems is very exciting. (NOTE: I do NOT want this implemented in SC2. This is a theoretical discussion of whether or not it would be worth using in future RTS titles.)
Let me know below what you think of the notion, or if you can think of some important implications that I glossed over. And if you enjoyed this, you might enjoy my previous posts concerning combat systems and unit death.
Thanks for reading!
TL;DR
The flinching mechanic has some serious potential for competitive RTS games.
Good times.
