My 5000th post. Or my treatise on unit micro vs A-move with a little bit on the Oracle and a lot more on the Collosus.
**NEW** This has probably passed out of common circulation already, but here is a Condensed Version/ TLDR Version of the blog if you were put off by the length of the original TLDR Version Spoiled immediately below + Show Spoiler +
Unit Micro: Beyond Spells
If marine splitting vs banelings and unit kiting is your thing and you'd like to see a wider array of these sorts of micro opportunities, then this thread is for you.
I really appreciate that Blizzard is trying very hard to add more cool, interesting micro options in HotS. However, I wonder how much thought is being given to a more fundamental shift in making regular attack units more microable?
What I mean is it seems in order to make any unit interesting, we need to keep adding more and more spells to the game, which all well and good. But I think there is another solution.
Specifically looking at attack-retreat micro. Otherwise known as hold position micro or in the faster variant moving shot micro. The advantage of this micro is it is extremely fast and opens up new tactical options for players. It's similar to early game stalker vs marine micro only faster.
Not only that, but it is extremely spectator friendly. I would argue that one of the things that made BW such a popular competitive viewing experience was this very odd marriage of solid macro-mechanics of RTS with the twitch control of units that would normally be the domain of Fighting Games.
Microbility
What makes a unit more microable without spells? 1) Burst damage (front loaded damage with time in between shots) 2) Very little space between firing and moving or moving and firing again. 3) Speed 4) lower hitpoints (although with SC2, things die pretty fast, so take or leave this
Some Definitions/ Terms by LaLuSh
Moving Shot A series of techniques employed to avoid deceleration when firing. Applied in Starcraft using the following techniques:
Attack command: Right click or a-click on a unit followed by a quick move command to avoid deceleration. If you don’t a-click on a unit or building your units will act like SC2 air units.
Hold position: Move units towards enemy and press H followed by a move command to avoid deceleration. Allows spreading shots and dealing damage more efficiently as opposed to target firing one single unit and wasting damage.
Patrol command: Allows you to fire from a 90° angle without losing speed. Is frequently employed against scourge.
Moving Shot: When the firing animation is shorter than the built in delay for deceleration.
Gliding Shot: When the firing animation is longer than the built in delay for deceleration.
Gliding shot is what we currently have which means we have built in sluggishness for unit control. But this sort of attack-retreat micro (whether moving, patrol, or hold position micro) should be almost every ranged unit.
Now the sluggishness may also partially to do with Battlenet latency. If we really want to the competitive scene to be all it can be and if we really want unit micro to be incomparably excellent, then we also need LAN latency. Many of the old micro tricks could not be performed properly even on Battlenet 1.0 because of the slight delay, but needed private servers that had LAN latency (Fish, iCCup, etc)
Video Explanations and Examples
I have two video explanation on how this micro works. 1) Difference between Hellion and Vulture movement
This is moving shot micro. Not what the Phoenix received which basically kept its gliding movement and turned on it's axis to fire backwards.
2) Burst damage, Collosus, Reaver, and Oracle
Oracle is out of date, but the principles of what makes an interesting harass unit still stands. I was being generous on the Stalker micro (sometimes need to stretch to make SC2 comparisons). There is also an attack delay on the stalker similar to the hellion. I strongly suspect this is an undercurrent issue for a lot of units.
Application: Nony video.
(7:40) Note how many ways the Reaver can be used: harass, defence, offence. And how it's able to operate independent from the army. Aka, away from Deathball like the Oracle was supposed to do for Protoss. And if you have time. Nony (aka Liquid Tyler)
A Final Video if you have time. This isn't strictly about burst damage, but on a couple redundant unit designs that effects the deathball situation in SC2. These units will continue to cause problems for any positional unit/ any unit with big damage and restricted movement just because of their design. That is no matter what new units created to break up deathball and push into a positional game, these units will cause problems to said new units. www.youtube.com/my_videos?feature=mhee
Final Thoughts
I most often hear that the micro of BW was unintentional game 'bugs.' Yet it demonstrably created an awesome viewing experience and allowed skill players cool micro moves. Street Fighter II's combo system was a game design bug, but was incorporated into future games because it was so awesome. Similarly, Quakes strafe-jumping also began as game design oversights, but became incorporated into future games.
I feel like we also had our game changing revolution in BW with twitch control (due to burst damage and fast transition between attacking and moving and moving and attacking.) However, I am afraid we are leaving behind some of our best discoveries. We have some very cool micro opportunities in SC2/HotS beyond spells, but many of them are very sluggish by comparison (Banshee vs Wraith micro, BW muta vs SC2 muta micro. Nony's Carrier video is another good of what we could have example http://www.teamliquid.net/forum/viewmessage.php?topic_id=369313) Now if there is better handling, then naturally the damage would have to be adjusted accordingly.
Furthermore, if we really want to see the top results of this sort of micro, we need LAN latency on Battlenet. That's how precise and fast these movements are. And that's what makes it such great skills to play and watch. It's a must for competitive gaming.
I love the variety of spells we get in HotS, but I would like to see how regular attack units can have better attack-retreat handling. With the basic design of the Microbility 4 points, I'm quite certain any so-called a-move unit can become interesting in the hands of skilled players. It's much faster, requires skill and it is VERY obvious micro for spectators (compared to slow back-up/ adjustment for collosus.)
I hope this is a sufficient summary of my major points.
Table of Contents
Many Causes: TL's Thoughts
Background/Definitions
Spells/Abilities
SC and Fighting Games
The Crux: Microbility
Redundancy and Tankiness
Counter-Argument: "Fighting the Interface"
Categorizing the Old Micro Tricks
Conclusion
Many Causes: TL's Thoughts
There has been a lot of thought in regards to this and I don't think there is one cause.
The spoilered list has a bunch of threads that you've probably read before, but they're worth glancing over if you have time to see what the community has come up over time. + Show Spoiler +
Breadth of Gameplay- Barrin http://www.teamliquid.net/forum/viewmessage.php?topic_id=321242 I really think there's a strong case to be made that the rate of income combined with the macro mechanics has led to more deathball style play. Maxing out 200/200 by 12-15minutes is insane unless you're playing BGH or probably even Fastes Possible Map.
Unit Interactions-Plexa http://www.teamliquid.net/blogs/viewblog.php?id=370905 It's a mistake to think all units/ races need the same thing. As Plexa points out, unit interactions can make a potentially boring unit, very fun. (The collosus does NOT need a set up time like the tank like I've seen suggested in some of the HotS threads.)
Carrier Micro- Liquid-Nony http://www.teamliquid.net/forum/viewmessage.php?topic_id=369313 This is the micro we've all been arguing for. MBS should allow us to do more of this sort of micro. This is what that saved apm is supposed to be spent on. Cool unit control.
On October 23 2012 02:33 sluggaslamoo wrote: I've made some progress on this during my time making a mod. What do you guys think?
Only trigger is disable units pushing other units (one command, its a player flag you can switch off). Also lots of trial and error with changing unit radiuses, accel/decel, formation radius, turn rate, etc. This makes em bump into each other a bit forcing them to need to spread out a bit more.
Philosophy of Design 2 - EternaLegacy http://www.teamliquid.net/forum/viewmessage.php?topic_id=302136 Certain unit spells actually prevent micro. This is most disconcerting when they are low level and prevalent. Late game and rare is a better place for lockdown type spells.
I'm sure there were more, but these are the significant discussions that stood out in my mind. There was one really old one from 2010 about the Triple threat marauder, immortal, and roach, but I can't find it.
First one thing this blog is NOT. **This is not balance/ racial whine. I now have 5K posts on TL. Find one where I've cared about balance beyond it being a passing nuisance that will inevitably be patched. (I doubt you can even find that.)**
Background/Definitions:
This came out of a couple debates/arguments I've had or else read with people insisting that Collosi are in fact not A-move units and that the Deathball is so 2010. I specifically had a quite long debate with a fellow viewer of Nony's channel after Nony had left and that got me thinking. Because the basic argument is that there is positioning involved with Collosus and the Protoss army and the Protoss does need to back up their Collosi to avoid damage. Therefore it can't be an A-move unit.
And they're right that the Collosus requires micro. In a RTS, if the unit moves, then unless you are rallying it across the map to your opponents base, then you are doing some sort of micro. So what then do we mean by a-move units?
I would suggest, that perhaps both sides are over-simplifying what A-move means and this is perhaps because it A-move is so often used in racial whine threads. However, I would say A-move has never meant no micro, but that it can mean the absence of the most intensive micro.
As soon as we move from turn-based strategy we are going have a basic level of micro: moving units to certain parts of the map, positioning your troops ideally before the battle and focus firing, retreating wounded units or re-establishing control during battle. I sincerely doubt that people complaining about a-move units think this sort of micro doesn't exist with these units. This is expected micro in any RTS. Sometimes people argue that the so-called BW elitists simply don't understand that SC2 just has a different sort of micro. SC2 micro is all about positioning and spreading out into concaves, etc. But that's exactly the sort of micro that BW, Warcraft 1-3, AoE, etc also had. But with BW there was something more.
So we have a base level amount of micro. Attack-Move is simply that a unit can be sent straight into battle and attack. (Way to go Captain Obvious.) However, we are not arguing there is no micro. The difference is not whether a unit can be a-moved, but whether the unit design most promotes a-move.
For instance, the siege tank in siege mode is not a-move. (It can't even be moved.) But at some points in the game (when they have an overwhelming advantage), a Terran won't even bother to siege their tanks and just a-move in. So a tank can be a-moved, but it is primarily used for non-a-move purposes.
Sometimes, You just go kill him.
So two things with that. 1) There is an overlap between A-move units and units that require more micro (set-up and position for tanks, moving shot for mutalisks) 2) At certain points in the game, the ideal, but most micro intensive may no longer be worth it. You could set up and attack, or you could just push in, take acceptable losses and destroy them quicker. This is important because people are arguing that we need to wait because pro's still haven't perfected the ideal unit positioning pre-battle. And until that happens, we needn't worry about more micro-able options like moving shot. And it's true that pro's are continually getting better, but we also need to look at the cost-benefit of these fine tuned actions. If it's too fine a line, the very elite may do it, but for everyone else, they will just make more units and ignore the fine tuning. It may never happen and the units continue to be a-move.
For example, if we took away splash from tanks. Terrans might still put them in siege mode because they get the ideal damage from any armoured attackers. But chances are the tanks would be kept in tank mode to a-move around (if they were made at all). There is a slight benefit for putting the tank in siege mode, but is the benefit great enough to be worth the player's time? We can always point to little things that pro-gamers are missing that makes their game 'not perfect,' but barring playing like a machine, do the units provide reward enough to something other than be a-moved forward?
Spells/Abilities
Spells/ Abilities are also not a-move in the sense that you can't just move them in to attack and they'll attack automatically. So it's interesting to note that SC2 and HotS actually has more spells and abilities than BW did.
Spoiled are more specific comparisons of spells/ abilities. + Show Spoiler +
SPELL/Abilities- in the case, I am counting any unit ability that is manually toggle-able (so siege mode counts as spell for Terran in both BW and SC2.) Auto abilities don't count (so Carriers are in both games are considered no-spell.) But abilities that can be automatic or manual do count (managing charge is an important control issue). Workers were counted as no spell for both games. Overlord and Overseer is counted separately as non-spell and spell respectively. Something like Hellion/Hellbat now counts as an ability, but only counts as one same as the Viking or the Tank.
BW has 23 units with no Spell and 16 with Spells Protoss: 10 no Spell 4 Spell Terran: 4 no Spell 9 Spell Zerg: 9 no Spells 3 Spell
WoL has 20 units with no spell and 20 with Spells //HotS has 20 units with no spell and 26 with Spells Protoss: 8 no Spell 7 Spell// 9 Spell Terran: 3 no Spell 7 Spell// 11 Spell Zerg: 8 no Spell 4 Spell// 6 Spell
You could nitpick individually on which units I've included where, but even still SC2 has overwhelmingly more spells and abilities.
HotS Protoss has twice as many units with spells/ abilities on it than BW Protoss. HotS altogether has 10 more spell casters than BW altogether. Spells are easier to use and even core ground units have spells/ abilities- Zealot and Dragoon-analog (Stalker) both have toggable abilities. Spells are everywhere, so how can SC2 armies still be considered more a-move?
Spells and abilities. More than ever.
Furthermore, just looking at BW, Zerg has far fewer units in general and the least number of spells in general (and the Queen was rarely used.) This means that Zerg is massing the most of the same sorts of units. Why then was it the BW Protoss that was considered 1a2a3a?
From my perspective, Blizzard has been continually adding new, 'cool' abilities to all the units in an effort to make them more exciting. Every single early game unit that the Protoss has, has some sort of tricky ability. Zealots- charge, Stalkers- blink, Sentries- FF and Hallucinate.
However, there is a much simpler of way making units more micro-able and less a-movey without having to come up with all these crazy spells... or mundane spells, the Corrupter has it's Corruption spell which is visually pretty meh, but the actual unit is pretty sluggish handling. Similarly, the Thor has it's Strike Cannon which seems to exist solely to get feed-backed and the unit itself is very sluggish to move around.
So what made the BW's so exciting if they didn't have that many spell casters?
SC and Fighting Games: Or How we Overplayed Macro over Micro
"BW is a macro game and WC3 is a micro game." Is often what you would hear and it's true that to win in BW you needed to have really solid macro. Macro is the fundamentals. The dribbling in basketball, the stick handling in hockey. But I think we overplayed macro and short-shifted micro. Micro was just as important to demonstrate skill and extremely important for making it an spectator sport.
Sure people were in awe of NaDa's macro, but they didn't scream that during the game. People screamed for Jaedong's muta micro, for reaver shots, for Jangbi's storms, for Fantasy's vultures. Unit control micro was the slam dunks, the wrist shots, the touchdowns of SC.
Vulture control. Something worth cheering for.
One thing to note, most of these units that had crazy micro actually didn't have special spells/ abilities. Or if they did, it wasn't so much the 'spell' that was awe inspiring, but how the unit was handled. (Here is where we're delving into Day9's baseball vs frisbee analogy.)
The micro-ability was something more than just backing up at the right time, or engaging from the right angle.
The micro tricks of Starcraft BW helped avoid Deathballs and A-move as they relied on hit and fade. We've exchanged rapid, twitch micro for a massive light show. We still have the base-line micro and we have some awesome micro tricks (Marine splitting comes to mind.)
Many of these units were almost a game unto themselves, with special moves that needed certain button combos (Mutalisk micro). In retrospect it comes as no surprise that Boxer actually came from arcade fighting games into Starcraft. Similarly, the other competitive game Artosis likes to watch is fighting games. Now that's anecdotal, but I think there's an an interesting overlap between BW unit's handling and fighting game's twitch control.
And I think this is unique to BW (and to some extent SC2.) I came from turn-based strategy and so this is not something I would have naturally thought necessary in an RTS. But BW is the one RTS that has both the fundamentals of economy management and this highly spectator friendly twitch control normally found in something like fighting games.
The micro tricks, I would almost call Micro Plus. It's in addition to the normal micro found in most RTS's. It's more than forming a concave and backing up when hit points are low. It's even more than filling the entire screen with spells (which can get really clustered.)
The Crux: Micro-ability
At it's core, pretty much any non-spell micro is simply attack, then retreat. Most of the attack-retreat micro is pretty much iterations of the same sort of micro. Stutter-step, moving shot, reaver micro, carrier micro, etc. It's just fire off a shot and retreat out of range from the returning shot.
Blink micro kinda uses this idea, but because blink is on a cool down, stalkers suffer a lot of damage in between. The difference between blink stalker and attack-retreat micro is simple attack-retreat micro requires movement in between ever shot to be effective (early game Stalker vs Marine micro). It's very intensive and requires far more speed to pull off properly.
There's a few things that I would say are necessary for the Micro Plus 1) Burst damage (Front loaded damage, time in between shots) 2) Speed 3) Speed between attacking and moving and moving and attacking 4) Relatively low hit points. (Although with how fast things die in SC2, this point might be moot.)
Based on these four things, I'm pretty sure we could create any number of new units that would have good micro-able potential.
Comparison: Vultures and Hellions Video Explanation by Yours Truly http://www.youtube.com/watch?v=TuQKRYbs04g Now I should say, it's not necessary that every unit has as rapid a response as vulture micro, but the fact that we have no unit with this sort of micro is, I think, a bad thing.
A-Move by Design: The Collosus, Oracle, and Reaver http://www.youtube.com/watch?v=i9_ZIBOlydA I was being generous on the Stalker micro (sometimes need to stretch to make SC2 comparisons). There is also an attack delay on the stalker similar to the hellion. I strongly suspect this is an undercurrent problem in a lot of units.
Follow up vids: Nony: Reavers in action http://www.youtube.com/watch?v=YXFeHfvVVYI#t=7m40s (7:40) Note how many ways the Reaver can be used: harass, defence, offence. And how it's able to operate independent from the army. Aka, away from Deathball like the Oracle was supposed to do for Protoss.
Morrow: Collosus Micro
Morrow shows there is some sort of synergy with the warp-prism, but I'm doubtful it will ever become standard due to the Collosus design/ cost. But to be entirely fair, I am adding this counter example.
So because the Collosi's attack rate is so fast, the attack animation is so long I would say the Collosi design promotes a-moving. To that list we could throw in Corrupters, Vikings, Thors, Broodlords, Carriers, Tempest. It's not that they require no micro (Need to spread those broodlords.) But they either don't have discernible burst damage, or their attacks require them to come to a complete halt to shoot or they are very sluggish to move around.
For instance, the Thor is very sluggish in turning around and moving. That means attack-retreat is not as useful and not as impressive.
Now having said that, not every unit needs to have the crazy micro options. Guardians were in BW after all. But we certainly need more than we currently have.
Redundancy and Tankiness
One further thing that leads to A-move armies is a glut in redundant, tanky units.
Redundant Units
One thing I forgot to mention is that people (including myself) were in a big outcry about the Warhound being too similar to existing units. This is the same problem as the Warhound, but this goes back to the first Beta.
Baller's hydraroach may be a joke, but there is some truth to it. They basically took the hydra and split it into two units. At this point I would just as soon throw out both the Hydralisk and the Roach and start over entirely with a brand new 1 supply ranged, low health, but fast micro-able unit.
Or to show it pictorially: (This before I decided to make a video on it) The other armies (with more of the units. Spoiled for space) + Show Spoiler +
This is of course a simplistic comparison and a bit tongue-in-cheek. But the main point is not so much whether they do the same duty from BW to SC2, but how radically different the units are one to the other... except the Triple Tanky units.
If you lose to roaches, you didn't macro correctly. There is nothing tricky about roaches.
Day9 (Can't remember which Daily.)
The Inevitable Protest/ Counter-Argument
So now that we've gotten to BW micro tricks, I can already hear the hue and cry. "We want to fight people, we don't want to fight the game! You have to break the AI for the micro tricks to work They were just game bugs."
As to the last, just because it was a bug, doesn't mean it is bad. Here we go back to the Fighting Games. In Street Fighter 2, a bug allowed players to combine several attack with no time to recover if time right. Capcom recognized the importance of the combo and programmed it into their next games. Or Strafe-jumping in Quake. That also developed as a bug, but the skill mastery of strafe-jumping led developers to specifically program it in to their next games. Starcraft under Blizzard had our own bug revolution that led to awesome skill masteries, but I feel like we're leaving behind the revolutionary game play.
Combos started as a bug, but revolutionized fighting games
Bugs is a bit of a dirty word and I'm not sure it necessarily is as all encompassing of the micro tricks that was available.
Patrick Wyatt the lead developer for Warcraft 1 and 2 and Starcraft has a Blog where he has been describing the development of the old Blizzard games. He's been answering a questions in the comments and I got mine answered where I specifically asked him about the micro tricks/ bugs.
Awesome read although not being a programmer, certain parts go over my head I wonder, are you familiar with "move-shot micro" such as the attack-retreat micro using the patrol command with the vulture? And if so was this intentional design or incidental unit behaviour due to how they were programmed? (Hold command works also for a bunch units including dragoons, un-sieged tanks, mutalisks, etc.) Because there are so many crazy things that keeps SC awesome even 14 years later.
Patrick Wyatt
There was a lot of emergent behavior (like vulture micro) that players discovered over the course of years; most of it was because -- in the rush to finish the game -- many edge cases weren't discovered by the design, programming and test teams. My favorite discovery by players was the "Korean eraser", where Terrans would irradiate a flying building and use it to wipe Zerg units off the map; it wasn't something we had anticipated. Another boo-boo was that, with careful timing, it was possible for a Terran player to land a Command Center much closer to resources than should have been possible, enabling faster harvesting, which made the Terrans unstoppable in the early game until patched. None of the many "features" that were eventually discovered were intentional.
I think emergent behaviour is a perhaps better than bugs. Unintended consequences perhaps, but there were some that were patching out (CC landing really close to minerals) and others that developed gameplay.
Which leads me to the 'fighting against the AI' argument. I want to break down the bugs/ micro trick in the game into several categories. Because it seems to me, people conflate all the micro tricks into one category. (They're called bugs, therefore they are bad.)
The micro tricks we're interested in is not extra clicking for the sake of clicking, but Micro Plus, the extra moves, the movement combos. What we don't want is the category of 'bugs that hinder or get in the way of what I'm trying to do."
Categorizing the Old Micro Tricks
For the next section I define Fighting the Game as: try to move unit--> unit won't move OR somehow the normal movement is hindered. (aka sporadic dragoon freak-out.)
This list is not exhaustive, but I hope it is comprehensive enough.
Tier 0: Movement Hindering Dragoon bugginess. Freezing/ frantic movement Goliaths (same as Dragoons, but to less extent) Units getting stuck between buildings Building exits blocked
Thoughts: No good. No-one wants these in SC2. The cause of the dragoon freak outs according to Patrick Wyatt
Because Dragoons were larger than other units they needed to find wider paths that weren't obstructed by terrain or other game units.
But this seems to be the level of bugs that people always get hung up on. None of the micro tricks actually fall into this camp.
Tier 1: Attacks with Timers Reaver scarab (random duds) Mine chase (sometimes chase units over long distances) Example
Thoughts: I can see why people wouldn't want it. On the surface, this looks like random duds and this could be considered frustrating and some people are adamantly against anything random. In reality it's not random. The scarab tries to fit through spaces between structures and then must go around. The entire scarab is on a timer, so if it doesn't reach it's target in time it will dud. I (as a Protoss player) actually think it's good. It allows the Reaver to be really powerful, but it won't hit 100% of the time (balance). But there's something really exciting about the unknown. Spectators wait with baited breath to see if it will shoot for maximum effect or whether through clever building dodging or pulling units away fast enough it will dud. The fact that you hear cheers, screams, and groans when the unit is in play should tell you it's doing something right.
Here were some of the ways of either ensuring the scarab hit or else running the timer out so it dudded. 1) If buildings/ mineral lines were in the way, it would be more likely to dud. 2) If the workers were pulled perpendicular to the reaver, it would be more likely to hit 3) If the workers were pulled away from the reaver, it was less likely to hit 4) An entire thread devoted to it: [G] TvP Deterministically Dodge Scarabs.
Both players know this information so these limitations actually creates targets and goals. It became a more more directional attack, adding a layer of decision making on the part of both players. (Where am I going to land the reaver so as to get maximum effect vs where is the best way to run beyond a non-directional away.) And there was enough damage to make the reaver worth it even if it dudded on occasion.
But if it duds too often it would be frustrating, but just often enough to make the spectators lean forward? Priceless. I'd vote yes, but I suspect most would vote no.
Tier 2: Hopping Over Walls (some random) Bugging workers over mineral lines Bugging military units over mineral lines Bugging units over stacked buildings to get up ramps
Examples My video demonstrating some of the above micro tricks
Thoughts: This is actually exciting to see and adds new possible maps. It doesn't effect the unit handling any way outside of giving it an extra ability to hop over the mineral lines. (Can't possibly be argued to be movement hindering as SC2 units can't currently do this at all.) Again, the anti-random people would hate the random part. The randomness could be cut out, but I'd suggest part of what it makes it works is that it's not a guaranteed thing and you can't get an entire armies over.
I was going to say that it could problems with balance, but as no map actually has mineral walls where this would be an issue, it actually wouldn't change much. But it would give map makers something to play with. I think it's exciting, but I could see why people wouldn't want it.
Tier 3: Stacking Worker Drill (see above video) Mutalisk Stacking
Thoughts: No reason to not have these things. There's already a precedent for using minerals to perform micro tricks. It also happens to be called the Worker Drill:
Mutalisk- the stacking in SC2 isn't quite there yet. The grouped units don't quite move like a cohesive whole that get out line occasionally. Each mutalisk still seems quite independent of the other and therefore there is less precise control.
I should also add that Muta's weren't the only ones that could be stacked so that they could be microed as (nearly) one unit. Corsairs and wraiths could also do the same if you trapped a worker or something else in the same hotkey group as the air unit.
Tier 4: Unit Control Hold Position Micro (almost any ranged BW unit) Carrier Micro (see Nony's blog) Reaver-Shuttle Micro (Colossus, Oracle, Reaver video) Scourge vs Air unit micro Move Shot (Mutalisk, Vulture, Wraith)
Thoughts: When people are asking for more micro tricks from BW... it's this level and maybe Tier 3. None of these tricks actually interfere with your ability to A-move across the map. It doesn't hinder movement, stop movement, cause the unit to freak out or any of things. It's the equivalent of the combos in the Fighting Game. The skilled can perfect it, but it's rewarding enough that lower level players will also try it. (A lot of these tricks should be filtering down to Platinum I should think- they just can't do both macro and the crazy micro at the same time.) We probably need better latency to get a lot of these things working however.
We made a big stink about this back in Beta and got the Phoenix. The Phoenix is not moving shot it just rotates on an axis and fires backwards. Furthermore, these sorts of micro tricks should be applied to almost every regular ranged attack unit in SC2, not just experiment on the Phoenix (and get it wrong.)
Moving Shot A series of techniques employed to avoid deceleration when firing. Applied in Starcraft using the following techniques:
Attack command: Right click or a-click on a unit followed by a quick move command to avoid deceleration. If you don’t a-click on a unit or building your units will act like SC2 air units.
Hold position: Move units towards enemy and press H followed by a move command to avoid deceleration. Allows spreading shots and dealing damage more efficiently as opposed to target firing one single unit and wasting damage.
Patrol command: Allows you to fire from a 90° angle without losing speed. Is frequently employed against scourge.
Moving Shot: When the firing animation is shorter than the built in delay for deceleration.
Gliding Shot: When the firing animation is longer than the built in delay for deceleration.
But I can NOT emphasize this enough. These sorts of micro tricks do NOT effect normal pathing.
Dragoon hold position also shows up here, but note that hold position micro is independent of the dragoon thinking it was smaller than it actually was. That Tier 0 bugginess hindered dragoon functionality. Hold position micro increased its functionality. (I'm sure you've already seen Bisu's dragoon micro linked countless times.)
God Tier: Chinese Triangle Chinese Triangle: Mutalisk vs Scourge Example:
Thoughts: I still can't do this with any consistency. I guess it should be under Tier 4, but it's so freaking hard.
Conclusion:
Again, I don't consider this THE solution to solving a-moving units. I really think it's a multiple cause problem. But I suspect it is part of the solution.
A-move units does not mean units entirely devoid of micro. There is always a base level of micro that any RTS has. The sort of positioning and concave army battles in SC2 also existed in BW and countless other RTS's. But there was something more that created the tremendous excitement for spectators and were cools skill for players to master.
Beyond adding spells to every single unit, the simplest method making armies generally and the collosus specifically less a-move-y is to focus on 1) burst damage 2) speed between difference between attacking and moving again or moving and then attacking. 3) speed in general 4) low hit points (depending on rate of fire)
The micro tricks of BW may have been unintended. There were genuine, gameplay hindering bugs. Nobody want those. But there were also emergent behaviours that were awesome to behold. Street Fighter II had their combo systems develop from bugs and Capcom developed it as part of regular gameplay. Quake also had their strafe-jumping orginate from game bugs. BW experienced a similar gameplay revolution through bugs, but I'm afraid we're largely ignoring a significant part of what made SC such a successful competitive game.
Competitive BW is an odd marriage between huge economy management and unit micro control normally reserved for something like a Fighting Game. The macro provided BW a firm foundation as an RTS and the micro a spectacular viewing experience. Spectators were cheering for more than a big light show with explosion filling the entire screen. They cheered for the players' mastered skill of twitch control units.
There. 5000th post done. And so after a massive slow down I'm back to moderating on TL, so watch out.
edit tldr. Watch the videos posted. 4 of them I made specifically for this article. The other ones are decent examples of what is being talked about.
Mutalisk... there's a lot video examples out there. I tried to make a video explaining the difference between BW and SC2 muta micro. But while I can see and feel the difference, I couldn't figure out how to visually break it out down for a video
But here is some partial thoughts:
Two additional points 1) Regardless of what you think of either my BW or SC2 muta micro, it is simply the case that you never see equivalent muta micro where they move in. The micro you see it a much slower, lazy sort of attack back and forth. 2) Balance Concern- when this was argued awhile back, people were afraid of muta's one shotting CC's with 30 muta. To me that's immediately another argument to for limited unit selection. But barring that (because it's unlikely to get traction) the best work around solution I can think of is to make it so that 2-x units (let's say 2-11) will stack properly to be microed as nearly one unit. If you have more than x muta's selected, they unstack thus becoming unwieldy to micro.
Damn, this is some quality analysis. All the omnipotent balance shitposters should take a page (or ten) from this post. Definitely how analysis should be done on this site, and damn interesting at that.
brevity is helpful. I could probably halve the length of this without causing even minimal damage and cut it much more without removing anything essential.
Unfortunately lengthy posts are my bane I suspect a lot of my normal posting is about as long as some people's blogs.
It certainly makes it a much greater task to read. But I don't like leaving loose ends that will quickly turn into useless side arguments when I can cover all my bases at the beginning.
But I definitely agree that brevity would be helpful.
This type of micro would definitely help with the deathball problem. Having to actively perform difficult actions to get the most out of your units means that bigger armies require exponentially more attention, so just getting a bigger army isn't automatically better. It makes small groups of units moving around much more effective, since the deathballing player can't just a-move at the weaker army and win handily, but the split up player can a-move his unit groups that aren't actively fighting other units.
The problem with SC2 is that your army get exponentially better the bigger it is, and no more difficult to use, so you get less utility from two equal halves than from the whole.
This is absolutely brilliant, and really helps to put into words the "feeling" that us BW die-hards miss in SC2. Props for spelling them out, with concrete suggestions for improvement - I don't know if Blizzard will see this and bother trying to change the game at this point, but if SC2 could take some of these wonderful aspects of gameplay from BW, it would increase my interest in the game a lot.
I was wondering whether it would help to have more units with longer cooldown times? It seems to be that in BW alot of interesting micro was done by/against units with relatively longer cooldowns, e.g. dragoons, vultures, reavers, mutas, lurkers, etc.
I think the important cool down is in-between shots. The problems a lot of SC2 units have is the delay between moving and attacking (as the vulture hellion tried to demonstrate) which makes it feel sluggish. If you watch Banshee micro it suffers from the same delay in attack. But of course the front-loaded damage needs to be adjusted to compensate. So less shots overall, but possibly more damage per shot. Then again, people complain about how fast units die, so maybe not more damage. But it's an avenue worth exploring.
The wrong way to go about it is simply having units move while they attack no matter what direction they face (a few SupCom2 units I can think of as well as the Phoenix change.)
Incidently, not only is it my 5000th post, but including the spoilers and formatting, it's also 5000 words (plus an extra couple hundred.)
First, the reaver's scarab. That scarab randomness is IMO, the luckiest feature of the game. To think that a game as professional as StarCraft can have something so random (I mean most sports generally shy away from the luck factor) and be so celebrated. It will be difficult perhaps close to impossible to put something like that in the game and make the players/spectators like it. You do remember all of those "if the scarab didn't dud then X could've won don't you?
Lastly, the bug statement reminded me a lot of Team Fortress' Spy class. Check the trivia part.
But why is this hidden in blogs where Blizzard will never see it
I haven't seen much success analyzing SC2 in SC2 General whenever BW examples are used. The views are less (except the one that was spotlighted), but I find the pace and the tone of the conversation much more reasonable.
Besides I get the feeling Blizzard doesn't actually regularly read TL. They only seem to respond to posts on Battlenet (Although I guess it's home turf, so it's easier that way.)
@mmp That's actually even better as it means there are predictave behaviours that can be countered or else adjusted to avoid the counters.
Hit the nail on the head. The delay before a hellion fires is so painful i hate using the unit. Everytime theres a pack of lings chasing my hellions i wish they were vultures. =[ Same goes for the colossus, its just not worth it to "excessively" micro.
Huh i always thought true "duds" were very rare, most of the time its scarab expiration(like i like to call it). Its true that scarab has imperfect path finding but its simple trial and error learning experience for players. I very rarely saw reaver mistakes from best reaver users, and i think positives WAY outweigh the negatives here. So far ive never seen any mechanic like that besides BW. Instead of sure hit mechanic or random percentage hit/loss bullets most units use in every RTS scarab is actually moving object with physical collisions and can be countered with simple positioning/movement even after its fired which is unique still.
You could say that scarab is kinda reminding of seeker missile if we had to give an sc2 equivalent but really its not. Seeker missile pathing has no point because it can choose straight line because it lacks collisions, scarab cannot phase through building/units.
2nd issue i noticed in sc2 that makes "tactical" micro less efficient and generally makes combat bland is ... perfection of technology. Units of BW had limited fps of animations, its a specially evident in marine micro from both games, you could say that is the only micro mechanic that sc2 excels over bw, because of that fluent movement patterns you can abuse in sc2.
But there starts the problem, while marines vs banelings is incredible (cue in Happy marine micro) it really makes higher tech unit tricks too hard and rarely doable. In BW every race had their +1.5/2 tier unit micro tricks and uses simply because it was action-reaction based behavior, because of low fps on animation you could clearly see it as "fast turn based micro when compared to sc2 fluent movie-like behavior" where action and reaction happens instatenous without turns.
Huh what? Lurker vs marine, is good example... If you look hard you will see that all of this is scattered in BW world, from scv mining, marine micro, zergling, muta till tier 3. The timing was pleasent on brain because you knew that this marine will take his own small intervals while walking to the bunker so you can interrupt it(not mentioning other mechanics, moving shot, collisions, unit frames being bigger etc). In SC2 units behave like a water. Which would be good if it was possible to hold to this potential. But i believe we will not.
This action-reaction based micro and was more highlighted by certain abilities, firing rate, and units synergies of course. For example reaver's shot delay was intended.
I probably over-emphasized the Reaver scarab experation time. I suspect I've heard too much from the anti-Reaver scarab side that I've adopted some of their language on the "randomness." Quite clearly there is more predictability than even I was giving credit for it. I might tighten up some of the Reaver description tomorrow when I'm thinking a little clearer.
I love reading your blogs. I hope you make more in the future.
I think the most important quote that i've ever heard form a game developer is "players will always take the path of least resistance"
Probably the most insightful quote ive heard from Dustin Browder is "[deathball] is the easiest way to play"
This applies to all games, even all sports potentially. Why would you micro when you're better off a-moving and focus more on macro? This is because the gain from microing sc2 units are extremely small compared to microing BW or WC3 units. I don't really think that units like the Colossus aren't micro-ed intensively because they lack micro friendly mechanics such as these.
Beyond adding spells to every single unit, the simplest method making armies generally and the Colossus specifically less a-move-y is to focus on 1) burst damage 2) speed between difference between attacking and moving again or moving and then attacking. 3) speed in general 4) low hit points (depending on rate of fire)
Units like the Colossus just don't have any incentive to be microed like others. A BW example would be the Dragoon. It really didn't have any micro friendly mechanics yet you still had to perfectly micro them against spider mines and tanks otherwise they would just get 1shot.
Compare this to Terran bio. There is huge gain from micro (to the point that it's needed). Great bio micro is vastly superior to decent bio micro. We see all the top Terran players are always improving their bio micro.
There are also units that has good benefit from micro, but the incentive isn't great enough. A prime example is the stalker. When you have relatively few of them in the early-mid game, you'll be microing your ass off trying to save every stalker and pick off every exposed unit. Once you're in the late-mid game or when you just have a shit load of stalkers, micro just slips because perfect control doesn't give that needed incentive. it's evident in many pro games over the past year.
until Blizzard completely redesigns core units of Protoss and Zerg, we'll be stuck in this situation. (thank goodness their making Terran more micro intensive by removing the Warhound and emphasizing the Mine.)
All this feedback has most likely been seen by blizzard at some stage. I personally think they're too deep in denial to admit the game needs serious change.
Great blog, Falling. I especially liked the redundant units clip.
That said, I rather like the Immortal despite disliking the gimmick of the hardened shield (it smacks too much of the hard counters that seems to define so much of SC2). But, I may be biased because I am Protoss, and I like the sounds of the Disruptor Cannons and the Immortal phrase, "My Cannons shall sing". However, the fact that the Immortal cannot be massed (unlike the Marauder and Roach) may be its saving grace, and something Blizzard did right in moving it from Gateway to Robo.
MasterCynical, above, makes a great point regarding the incentives to micro and the requirement to do so for good Terran play.
Great blog! Blogs like that let me wish i would had played starcraft BW! SC2 is allready a complicate game with unbeliveble depht! But BW - boy oh boy!
On October 23 2012 12:37 Falling wrote: Unfortunately lengthy posts are my bane I suspect a lot of my normal posting is about as long as some people's blogs.
It certainly makes it a much greater task to read. But I don't like leaving loose ends that will quickly turn into useless side arguments when I can cover all my bases at the beginning.
But I definitely agree that brevity would be helpful.
I like to say that if you wanna say big things you need big explanations that leave no holes, that's exactly why and how science works, a good way of doing things if you ask me.
Very good post, perhaps could use slightly more organisation but solid analysis overall. If I may nitpick a little bit, I believe glitching through minerals/stacked temples etc is not random as such, seeing as progamers (and pro mapmakers) used it as a feature very consistently. I mean Outsider and Monty Hall were just made on that. It's just we are not quite able to pull it off every time.
On October 23 2012 14:19 mmp wrote: Correction: Scarab behavior is not random.
I'm glad somebody actually knows about that thread, I was hugely impressed when I first read it.
It's really great how the OP references all kinds of posts from all corners of TL. Extremely insightful, deep and intricate analysis. Sometimes I wish Blizzard would just remake BW with better graphics. But then again we would get tired of always playing the same thing for over 14 years... or would we?
Find one where I've cared about balance beyond it being a passing nuisance that will inevitably be patched. (I doubt you can even find that.)
then I started to sit back and enjoy it.
Just a thought: a table of contents might be good and make it easier to digest. Background; strengths of micro (subcategories?); unit design in BW/WoL (I suck at names ); The Inevitable Protest/Counter-Argument [AKA the game is emergent! Yeah, this part was very cool]; categorizing micro; conclusions.
On October 23 2012 23:40 sorrowptoss wrote: It's really great how the OP references all kinds of posts from all corners of TL. Extremely insightful, deep and intricate analysis. Sometimes I wish Blizzard would just remake BW with better graphics. But then again we would get tired of always playing the same thing for over 14 years... or would we?
You do a really good job at analyzing this shit. I come up with the same conclusions, but I only use personal experience so I never have diagrams or images representing what I mean.
Something to really note here is that Capcom took what were bugs in their game, and turned them into features for their sequels. You've already given numerous other examples of this, but the biggest thing to me is that Blizzard said "Oh, technology is different so we can't do it like that now."
That's bullshit. Technology may be different, but that doesn't mean you can't keep those things in there. It's not that it can't be done, it's that Blizzard was too lazy to do it and decided they knew better. It's obvious too when they created SC2 multiplayer with "esports in mind." You didn't create BW with Esports in mind, so why try to d the same with SC2?
there's plenty of interesting micro to be done in SC2
I still think most of the 'problems' are rooted in the resource system/map design that incentivizes 3 base turtling and fighting with gigantic high tech armies. You could add all the small scale tricks you want but they wouldn't be as relevant as in BW just because the scale of the game switches to multi base, huge army so much quicker in SC2.
On October 23 2012 17:42 MasterCynical wrote: I love reading your blogs. I hope you make more in the future.
I think the most important quote that i've ever heard form a game developer is "players will always take the path of least resistance"
Probably the most insightful quote ive heard from Dustin Browder is "[deathball] is the easiest way to play"
This applies to all games, even all sports potentially. Why would you micro when you're better off a-moving and focus more on macro? This is because the gain from microing sc2 units are extremely small compared to microing BW or WC3 units.
I think that was more or less what I was trying to get at with my just a-move tanks example. If the incentive isn't great enough, there might be a more technically complicated way of attacking, but players will take the path of least resistance.
On October 23 2012 17:42 MasterCynical wrote: I don't really think that units like the Colossus aren't micro-ed intensively because they lack micro friendly mechanics such as these.
Beyond adding spells to every single unit, the simplest method making armies generally and the Colossus specifically less a-move-y is to focus on 1) burst damage 2) speed between difference between attacking and moving again or moving and then attacking. 3) speed in general 4) low hit points (depending on rate of fire)
Units like the Colossus just don't have any incentive to be microed like others. A BW example would be the Dragoon. It really didn't have any micro friendly mechanics yet you still had to perfectly micro them against spider mines and tanks otherwise they would just get 1shot.
Dragoons actually do have the burst damage. They have their front loaded damage and (relatively) lengthy time between shots and it has a fast enough attack in between moving so that it is microable. The shot isn't as fast as vulture patrol micro, but it's fast enough for hold position micro. But again, we don't need every unit to have vulture like micro.
And the rest of your post I more or less agree with. But the first 2 points (burst- front loaded damge with decent cooldown/ and speed difference between attack and move) is exactly why the collosus doesn't have the incentive to be microed. (It certainly isn't top speed because the collosus is relatively fast.) The path of least resistance is to back it up slowly or spread out a bit to mitigate damage.
@Snorkels Table of contents would be cool, but I don't know how to do internal links within a post.
@Wax And that is why I think it is a multi-causal problem and linked Barrin's thread at the beginning. But even if you fixed how fast resources come in and how fast armies can be accumulated, you are still left with a large number of units that are generally best used to reposition, spread and back up. But don't have that dominating attack-retreat micro as found in BW. All the small delays whether by unit design or Battlenet latency slows down how fast a unit can be attack retreated.
Not to say there isn't cool micro in SC2. I agree there is. I usually go out of my way to note this. But there can and should be more. Much more.
Lately there seems to be a lot of these analytical posts regarding SC2 on TL. It's good to see that a lot of intelligent users genuinely care about the game.
Some more anecdotal evidence for the crossover of interest between fighting game players and real time strategy game players:
I started out playing RTS games, and then got in to fighting games. My girlfriend started out playing fighting games, and then got in to RTS games. (not through me, but before I met her)
I think every non-Brood War player should read this and look at the videos as it illustrates something that brood war players will try to convey every now and then, but often without success. It can be so hard to explain, just like you illustrated with your muta video. The mutas in SC2 seem a little drunk in comparison to their Brood War counterparts if you attempt the same moves.
this si quite helpful in understanding the goals. incorporating certain elements found in fighting games does sound like a good idea; and it relates to high speed micro. making sure each individual unit has things you can manually do to improve it, and making sure that each one is interesting. Thanks OP, this is quite illuminating.
There's plenty of micro tricks in SC2 that you kinda overlooked.
Yeah, delay before firing is a bitch on hellions, but you could do a reasonable impression of moving shot with banshees, mutas and even brood lords.
There's a of micro involved with HTs in warp prisms. They not only useful for storm drops on mineral lines, they also protect the HTs from snipes/EMPs in big engagements.
Even in deathballs, there's still plenty of micro tricks available. Staggering of siege tanks (death mech ball) is a commonly used one. Spreading units is actually more tedious in SC2 because of the tendency of units to clump. There's marine splitting and blink stalker positioning to snipe vikings. Even in the dreaded colossus vs colossus mass laser battles, there's still subtle micro tricks like staggering your colossi so that only 1 gets hit instead of 3.
On October 23 2012 13:06 Kovaz wrote: This type of micro would definitely help with the deathball problem. Having to actively perform difficult actions to get the most out of your units means that bigger armies require exponentially more attention, so just getting a bigger army isn't automatically better. It makes small groups of units moving around much more effective, since the deathballing player can't just a-move at the weaker army and win handily, but the split up player can a-move his unit groups that aren't actively fighting other units.
The problem with SC2 is that your army get exponentially better the bigger it is, and no more difficult to use, so you get less utility from two equal halves than from the whole.
QFT. Harass and multitask was so much more important in the average game of BW than it is in SC2 for precisely this reason.
This was an excellent read, and hits on many key subjects that are too often overlooked in favor of balance "discussion." Balancing a game, while difficult, is something that can be often be done with numbers. Making a game interesting and dynamic, however, is an intensely complex affair.
I would disagree on one point: the Immortal. Immortals are often treated as heavy duty stalkers, but this is more the fault of the units they commonly interact with. Immortals are actually very interesting and fun to use because of their huge discrepancy in damage to light and armored units. If you're playing a PvT, and you're up against some early marine/marauder pressure, you benefit from quickly selecting your immortals and targeting down marauders. Letting them fire freely can lead to them doing less than half their potential damage. In PvP, Immortal drops are incredibly fun; picking up and dropping immortals to snipe stalkers so you can get away with your warp prism, and the fact that 2 immortals can 1-shot a probe makes it one of the most micro-intensive but potentially devastating harasses in the game. And in PvZ, immortals are heroes when you fight larger roach numbers; I recall a game where HerO killed an absurd number of roaches with a pair of well-microed immortals and a warp prism, with one reaching nearly 50 kills by the end. However, as the game progresses to larger numbers, they simply become durable damage dealers, as there are so many marauders/roaches that it just doesn't make a difference if you tell them to target or not.
By contrast, the Marauder and Roach are boring. On paper, the marauder looks really cool; it can slow down units, so you can have 1 or 2 dart in and peel a few units away from a retreating force, and kite melee units. But when you fight a gateway army, for example, you want to shoot zealots to slow them down, and take advantage of their extra damage against the stalkers. At first, this makes for interesting fights when there are 3 or 4 units on each side, and you're maneuvering around zealots trying to shoot stalkers... but then, as the engagements grow, you kind of want to be shooting everything with the marauders, so you make alot of them and just attack. Roaches are similar; the quick regen when burrowed means that you simply burrow roaches to heal in between fights; by the time you get it, fights are too large for the burrow micro to really be effective. Against force fields, burrow movement is cool, but binary; if you have it, you sijmply move under the forcefields and keep attacking. If not, you get shut down by the forcefields. Not very interesting or fun.
But, remove the marauder/roach, and immortals gain a new identity - a tank-buster type of unit. Actually, this is their original identity, but it was largely forgotten in favor of "Hey, extra DPS against mass roach/marauder." In large fights, they can be a key unit in taking down high profile threats such as tanks/thors/ultralisks(?)/Colossus when focus-fired, but bulky, low-damage paperweights when firing at units like zealots/zerglings (which are usually automatically targeted). They also retain their synergy with warp prisms, picking them up to escape or reposition against melee units, or even moving them closer to targets like tanks or colossus.
On October 24 2012 08:38 snively wrote: omg that chinese triangle thing is the coolest thing ive ever seen i kind of wish i had been old enough to get into competitive BW T_T
On October 24 2012 04:47 Heh_ wrote: There's plenty of micro tricks in SC2 that you kinda overlooked.
Yeah, delay before firing is a bitch on hellions, but you could do a reasonable impression of moving shot with banshees, mutas and even brood lords.
There's a of micro involved with HTs in warp prisms. They not only useful for storm drops on mineral lines, they also protect the HTs from snipes/EMPs in big engagements.
Even in deathballs, there's still plenty of micro tricks available. Staggering of siege tanks (death mech ball) is a commonly used one. Spreading units is actually more tedious in SC2 because of the tendency of units to clump. There's marine splitting and blink stalker positioning to snipe vikings. Even in the dreaded colossus vs colossus mass laser battles, there's still subtle micro tricks like staggering your colossi so that only 1 gets hit instead of 3.
Even banshess, mutas and broodlords have the same problem. Broodlords... I dunno, the same could be said about Guardians probably, but I've never really played much with either. Muta and banshees definitely also have the same sluggishness if you compare it to wraith and muta micro from BW.
There definitely is micro with HT, but Blizzard is throwing it's attention at getting the Oracle into an interesting harass unit. Whereas the HT used to do that very same role. Sure you'll see it on occasion, but not nearly so often. Did people suddenly forget how to storm drop when they switched to SC2? It's at least worth looking at.
And absolutely, there is marine splitting and blink stalker. I do not argue SC2 is currently devoid of micro. But that certain unit designs promotes more a-move. (And a-move itself is not devoid of micro.) But note with the colossi wars you are focusing on subtle tricks such as staggering. The same subtle tricks exist in BW. But on top of that, there were some very obvious, in your face, set people screaming, twitch control of your units that required skill to master and were awesome to behold.
On October 24 2012 04:47 Heh_ wrote: There's plenty of micro tricks in SC2 that you kinda overlooked.
Yeah, delay before firing is a bitch on hellions, but you could do a reasonable impression of moving shot with banshees, mutas and even brood lords.
There's a of micro involved with HTs in warp prisms. They not only useful for storm drops on mineral lines, they also protect the HTs from snipes/EMPs in big engagements.
Even in deathballs, there's still plenty of micro tricks available. Staggering of siege tanks (death mech ball) is a commonly used one. Spreading units is actually more tedious in SC2 because of the tendency of units to clump. There's marine splitting and blink stalker positioning to snipe vikings. Even in the dreaded colossus vs colossus mass laser battles, there's still subtle micro tricks like staggering your colossi so that only 1 gets hit instead of 3.
Even banshess, mutas and broodlords have the same problem. Broodlords... I dunno, the same could be said about Guardians probably, but I've never really played much with either. Muta and banshees definitely also have the same sluggishness if you compare it to wraith and muta micro from BW.
There definitely is micro with HT, but Blizzard is throwing it's attention at getting the Oracle into an interesting harass unit. Whereas the HT used to do that very same role. Sure you'll see it on occasion, but not nearly so often. Did people suddenly forget how to storm drop when they switched to SC2? It's at least worth looking at.
And absolutely, there is marine splitting and blink stalker. I do not argue SC2 is currently devoid of micro. But that certain unit designs promotes more a-move. (And a-move itself is not devoid of micro.) But note with the colossi wars you are focusing on subtle tricks such as staggering. The same subtle tricks exist in BW. But on top of that, there were some very obvious, in your face, set people screaming, twitch control of your units that required skill to master and were awesome to behold.
Well in SC2, the moves that set people screaming are "good" fungals, nice marine splitting, immortal warp prism micro... yeah the standards got lowered. I was just saying that I felt your article was implying that micro is nonexistent in SC2.
An oversight on my part. I am usually much more careful about acknowledging the micro that exists in SC2. On the other hand, believe it or not, but I was actually editing it for length.
nice post, also hadnt seen the day9 vid yet which describes my sentiments about bw units vs sc2 units much better than i could have ever put it myself. I dont quite understand the intention though. If I wanna play a game with all these great aspects I just go and play bw. SC2 is another game where units function differently. Is that bad? Yes I think so, the op thinks so and day9 thinks so too. But there are enough ppl out there who still like and play the game. In order to make things better the whole engine would have to be reworked and even all WoL units would have to be fiddled with. Its not gonna happen.
Well I guess Capcom and idSoftware could've said the same thing about the gameplay that developed from Street Fighter II and whichever Quake it was. But instead they decided to build the combo system and strafe jumping into their sequels because it was awesome. It was awesome for the players and it was awesome for the spectators. Win-win.
There is a point of dragoon micro mastery where players who are extremely good(especially the koreans) could defuse spider mines without the help of observers you really need good reactionary timing and reflexes to pull it off and when they do work . Holy cow you will feel like a boss. I have seen it in action and I felt that little aspects of the game that can't be pulled a the lower level of gameplay is what keeps bringing me back to bw . I haven't seen anything in sc2 that has the same skill level as mention above .
Very good analytical and analysis by falling here he answered most of the response about the AI being the hindrance to the game and when I logically think about it, good players will know how to negate the AI effects from being too much and make the units that are supposedly dumb to the core work at it's best again especially dragoons.
5/5 from me bw had everything micro attraction,macro superiority and spectator friendly units .
On October 24 2012 10:24 Falling wrote: An oversight on my part. I am usually much more careful about acknowledging the micro that exists in SC2. On the other hand, believe it or not, but I was actually editing it for length.
Well, 5000 words is pretty long. I wonder if there's a character limit for posts.. must be like 100000 or something.
good post. Are you by chance in game design at all? You seem to be able to break down game mechanics much better than most people (including some professional game designers I've worked with).
Nope. Not in game designing at all- unless you count home-brew strategy board games. I just think about it a lot and find ways to break it down with examples. The teacher in me?
On October 23 2012 12:16 Falling wrote: We probably need better latency to get a lot of these things working however.
Because units are so unresponsive there are a lot of micro situations that aren't even worth attempting. Phoenix vs mutas, weak stalkers vs marines, hellions vs roaches come to mind.
IIRC there is a minimum fixed latency on every command in multiplayer mode. I want to say its 250ms, I think it was at one point during the WoL beta it changed between 500ms, 250ms, and 125ms. In fighting game terms that's 30 frames, 15 frames, and ~8 frames of input lag.
Try microing stalkers vs ai roaches in single player then try again in multiplayer. Even though it is just you vs AI in both situations the difference is night and day.
That actually doesn't surprise me. I seem to recall from my TL lurking days that people didn't advise muta micro on Battlenet 1.0 because the latency made it much more likely to mess up. I believe it was strictly on private servers with LAN latency or else LAN latency that the most intensive micro could be performed with regular degrees of success. (I could be remembering wrong as it must have been a 2007 or 2008 thread.)
But for SC2, simply out of the gate there is a limiting factor on the micro already available.
Does untapped micro exist in sc2 and in particular god micro? The first thing I can come up with is roach burrow/unburrow micro which is apparent if you play vs insane AI bots in sc2. The limiting factor is off course apm and you could actually calculate the necessary apm threshold for certain Unit v Unit fights. If you know the in-game factors such as movements, attack speed and turning rate you can theoretically calculate the aftermath in UvU fights with a set apm. To be able to replicate in-game battles with an equal or greater than X apm is a different thing which require practice and talent. Physical factors (Human machine interface) such as frame rate and latency adds a barrier for certain battle micro. It would be neat to see a list about what apm is required to execute certain battle micro. It would then be possible to use battle mirco stats for balance and skill cap considerations.
On October 25 2012 09:59 archonOOid wrote: Does untapped micro exist in sc2 and in particular god micro? The first thing I can come up with is roach burrow/unburrow micro which is apparent if you play vs insane AI bots in sc2. The limiting factor is off course apm and you could actually calculate the necessary apm threshold for certain Unit v Unit fights. If you know the in-game factors such as movements, attack speed and turning rate you can theoretically calculate the aftermath in UvU fights with a set apm. To be able to replicate in-game battles with an equal or greater than X apm is a different thing which require practice and talent. Physical factors (Human machine interface) such as frame rate and latency adds a barrier for certain battle micro. It would be neat to see a list about what apm is required to execute certain battle micro. It would then be possible to use battle mirco stats for balance and skill cap considerations.
Who cares if you can burrow some roaches... that's not interesting. What's interesting is when I do something, you react to it, I react to your reaction, and so on forever. When you have that, you have a dance between the two players. Like a simple PvP battle in which both sides have 16 dragoons, 2 reavers, and a shuttle. That battle has an infinite skill ceiling. INFINITE.
Man you've been posting really high quality blogs lately, i would've loved to see them being posted maybe 1 year ago, and i'd love to see this being crossposted on blizz forums and possibly reddit
I can't on Beta forums. But posting on Battlenet forums is kinda frustrating because I have to strip out all my pictures and videos. I did it once for my In Defence of Mech, but it's easy to get buried on Battlenet.
Edit. Actually I can now post on Beta forums But stripping if of pictures and videos would be irritating.
On October 26 2012 10:47 Falling wrote: I can't on Beta forums. But posting on Battlenet forums is kinda frustrating because I have to strip out all my pictures and videos. I did it once for my In Defence of Mech, but it's easy to get buried on Battlenet.
Edit. Actually I can now post on Beta forums But stripping if of pictures and videos would be irritating.
You can't post youtube links on the Bnet forums ? Obviously it wouldn't be embedded but a simple link people could click maybe ? Having said that tho it's not like at blizz cares ... I mean they don't listen and even if they did ... well we've seen how they gave us "moving shot" with the phoenix ... and they come out and say they're surprised to see terran splitting marines against banes when BW pros have been splitting mnm against lurkers for years ...
Well for better or for worse, I made a condensed version and posted on Beta forums. I've also included this version in the OP (spoilered of course) in case people stumble across this and find the original appallingly long. Also have added a table of contents.
I've been trying to pin point what's different about unit control for a long time. I've probably made some not 100% correct claims in trying to figure it out previously. For example, I myself don't think it completely necessary to have different definitions of "gliding shot" and "moving shot" anymore. Units glide in both games, and behave essentially the same given the same behaviors, and given they face and travel directly towards their target upon firing. The only definition necessary in my opinion, is the one in the two paragraphs below.
After 2 years what I can say is: The only fundamental difference between SC2 and BW unit control seems to be that SC2 air units rotate around their axis to "lock on" to their targets while still gliding in their original direction, whereas BW air units must face and travel towards their target before being allowed to fire. You can see it in the chinese god tier muta micro; the mutas have to fully turn around. They very briefly travel toward the scourge before firing and turning back.
The rest all comes down to attack animations, turn speeds, acceleration etc. All of which can be emulated through the SC2 editor. It's just been Blizzard's choice to design immobile and clumsy air units with long animations and relatively slower turn speeds.
The cases where you are most likely to fuck up moving shot micro in SC2, is when you are not perfectly aligned with the target you're firing at. Why? Because the air units are rotating around their axis while gliding in a slightly different direction than they're facing. Before they can be "snapped out" of their movement in a fluid manner, they must first rotate back to the direction they were originally gliding towards. It's noticeable in your SC2 muta micro video. The instances where the mutas are the most non responsive and pause the most before acting out your next move command, is when they were not perfectly aligned with their target before firing.
People use this feature the most when microing, say, banshees. Move towards a direction you want to be travelling towards, fire the shot, issue a move command in the same direction as previously (the same direction you'll be gliding towards). And the banshee almost has time to rotate back before slowing down.
You will notice this when vikings are microed as well. If they travel sidewards relative to their target. They can sometimes rotate back in time to the direction they were travelling towards and make it look like a decent moving shot. Conversely, if they are travelling in the opposite direction of their target before being told to fire, they will never have enough time to rotate back before having slowed down to a stop.
MavercKs SCBW mod has been somewhat succesful in removing attack animations, increasing turn speed and acceleration. I've been meaning to make a video showcasing it. But busy with an exam period right now. Cheers, nice blogs. Your next one deserves to be posted in the forums, don't be modest man.
Just came here from the link in the Balance Map thread.
I wish I'd seen this ages ago, as it is a well-thought out and presented thread. My disagreement comes as a randomness hater, however. There needs to be as little randomness as possible, with the only exception being critical hit rating in a game that supports it.
I do, however, agree that glitches can be the very source of high-skilled play. The best example is the Super Smash Bros. series. While unknown to casual players in the series, Super Smash Bros Melee is one of the most micro-intensive and precise game ever played competitively. One of the things that enabled this level of play is something that was a glitch in an earlier game, but was a somewhat difficult maneuver to pull off. This glitch was adapted to intentionally be part of the next installment, and balanced so that it affected all characters equally (L-canceling, if you're wondering).
There is no reason a glitch that allows for additional skill ceiling should not be adapted into a game. In your example, that glitch could be mineral jumping with army units. As long as it is smoothed out (not random) and affects all units (or at least all types- Ling, 'Rine, and 'Lot can perform it equally, if controlled correctly) I see no reason not to increase a game's micro ceiling.
EPT
![[image loading]](http://i.imgur.com/9faDJ.png)
![[image loading]](http://i.imgur.com/amQK3.jpg)
![[image loading]](http://i.imgur.com/5GKCU.jpg)
![[image loading]](http://i.imgur.com/aEguW.png)
![[image loading]](http://i.imgur.com/gqC3y.png)
![[image loading]](http://i.imgur.com/CMFgf.png)
![[image loading]](http://i.imgur.com/2AZNT.jpg)
![[image loading]](http://i.imgur.com/OgEuy.png)
); The Inevitable Protest/Counter-Argument [AKA the game is emergent! Yeah, this part was very cool]; categorizing micro; conclusions. 
