There is a second point I would also like to touch upon, and that is to say thanks to the DevTeam for unveiling their core ideas on what they consider the Core GTA/AA unit for Mech players should be, it is hard for me to express and be thankful for this because, on my eyes, doing such a move leaves the ideas DevTeam has exposed to public opinion and attack, and I'm sure DevTeam is pretty damn tired of being belittled all the time, and even when they have some tough skins, creating more attack vectors for themselves in order to improve the game is something which I consider I and the rest of the Community should be maybe not thankful for, but recognize and appreciate.
Once again, the weekly disclaimer that all of the things and writings on this series of Blogs are coming from my 6 years of experience as a Community Melee Mapmaker for StarCraft alongside working on other RTS GameDev projects over the years.
This week’s balance update
Today, we'll be adding most of the changes we’ve been discussing to balance testing. Please check out the changes and let’s get more discussions going where needed.
We would also like to thank everyone for the great discussions regarding what to do with the Cyclone. This was one of those few times where there wasn’t a clear consensus to whether we should keep the Cyclone or revert it, but we agreed with the slight majority of players in that the current version is probably the more helpful option towards pushing Mech play.
Factory AA
One other thing we wanted to discuss regarding the Cyclone AA feedback was Mech AA. As some of you have also pointed out, the Thor is much more of a core option against air in testing. Not only did the Thor receive a buff vs. light air units, but also the armored air units that used to outrange the Thor now have equal or lower range as the Thor.
The main reason why we would like to push the Thor more so than the Cyclone as the core AA option is very simple: We believe the core Mech end game composition should consist of very powerful but less mobile units such as the Thor, Siege Tanks, BCs, etc. This is because the high mobility gameplay option is already provided well through Bio. For example, when being harassed by Mutalisks, I can stim pack and quickly chase them off. Using fast-moving cyclones to deal with them would feel very similar. In contrast, Thors battle Mutalisks in a different way: they need to be in position to really get strong damage against Mutalisks trying to fly in, which is more fitting for the Mech fantasy.
Today, we'll be adding most of the changes we’ve been discussing to balance testing. Please check out the changes and let’s get more discussions going where needed.
We would also like to thank everyone for the great discussions regarding what to do with the Cyclone. This was one of those few times where there wasn’t a clear consensus to whether we should keep the Cyclone or revert it, but we agreed with the slight majority of players in that the current version is probably the more helpful option towards pushing Mech play.
Factory AA
One other thing we wanted to discuss regarding the Cyclone AA feedback was Mech AA. As some of you have also pointed out, the Thor is much more of a core option against air in testing. Not only did the Thor receive a buff vs. light air units, but also the armored air units that used to outrange the Thor now have equal or lower range as the Thor.
The main reason why we would like to push the Thor more so than the Cyclone as the core AA option is very simple: We believe the core Mech end game composition should consist of very powerful but less mobile units such as the Thor, Siege Tanks, BCs, etc. This is because the high mobility gameplay option is already provided well through Bio. For example, when being harassed by Mutalisks, I can stim pack and quickly chase them off. Using fast-moving cyclones to deal with them would feel very similar. In contrast, Thors battle Mutalisks in a different way: they need to be in position to really get strong damage against Mutalisks trying to fly in, which is more fitting for the Mech fantasy.
This is a very, very hard topic for me to tackle on text, there is lots of nuance which I'm afraid might be lost on this communication way, and without having great knowledge on past decisions made by DevTeam regarding the position and function of the Thor inside the greater Mech composition I hope not to be missing too many of what should be obvious points. What I'm trying to get at, is that this might be a rough ride, so better buckle up!
Thor...
Yes, on my eyes, DevTeam is 100% correct regarding that the unit does indeed fill the Mech fantasy of having strong & rather immobile units. Yet, there is other thing which fills that niche better than the Thor does, and that is the Missile Turret. Tanks and Turrets, now that's something I feel better fits the Mech fantasy, and it does so on a better fashion than Thors can and do.
DevTeam, mentions that Thors battle Air units on a different way than Marines or Cyclones do, and that's right, yet, Thors from a design perspective heavily overlap with what in my eyes is what truly fulfills the Mech fantasy, Turrets. Not only that, but because the Thor isn't a reliable DPS GTA dealer, Thors work as support units, support units to bolster and (fulfilling DevTeam's Thor fantasy) protect enemy fliers from picking near by friendly units, yet when the battle occurs, if Terran players find themselves with 3 to 4 Thors and nothing else do deal with enemy air units, the said Terran most often than not will not be on a commanding position during the battle.
Once again, yes, I do fully agree with DevTeam on the fact that Thors do work in order to fulfill the Mech fantasy DevTeam has wrote about here, yet the Thors do not fully fulfill the role Mech as a whole lacks, it does fits into its AA Mech role, but it does so poorly.
Thors, by design as DevTeam stated are meant to be units which are very strong and non-mobile (like a bigger Turret), yet, this means that a player with air units will not want to engage a Thor, nor get near it, nor pick off enemy units if the Thor is around, therefore reducing the amount of player interactions/engagements that occur on the game meaning Thors also reduce the amount of action and player's spot light moments overall.
Now going into the Thor itself, the unit suffers from the very same issues Colossus suffer, the unit is mighty expensive, be it counting just resources, supply, or factory time and it is not granular enough, you won't have 20 Thors lying around. The unit itself in order to follow its concept of being a “colossal and bulky” war machine, is unreliable and unwieldy to the user. As previously mentioned, the Thor, is not the kind of unit a player wants to have Mid-Battle, it is simply too unreliable and not a stable DPS dealer, yet it is the kind of unit that players want a couple of in order to shoo away pesky Mutalisk balls from picking units off. These in my eyes are the features of a support unit.
As a support unit, Thors are absolutely wonderful, yes they are expensive and slow, but as just I mentioned they have an incredible range and avoid your units from being picked off, getting a Thor on an army will give you peace of mind that your army won't be constantly be pecked at/picked off. Yet, and I apologize to repeat myself so many times, the unit, by concept, design and application, is too unreliable for it to work as a Core GTA DPS dealer for Mech armies, yes, it fits the general idea of Mech well, yet it falls short on too many areas for it to be a reliable GTA work horse.
One of these areas, which I outlined previously and is of great importance for Mech is that Thors are simply not granular enough, a player can't easily leave 1 or 2 Thors per base in order to fend off enemy air raids if the air units decide to commit, the Thor does not have DPS for that. Because of that Mech players naturally will build Missile Turrets, the Turrets will work as the main GTA work force, and a lone Thor on a Turret defended base will work as support for the DPS Turrets dish out in case the enemy Air decides to engage, all the while protecting the Turrets from being picked off by concentrated/stacked air forces. But this is simply not applicable to more than 1 or 2 bases, and on LotV where players tend to be much more spread out, tying 1 Thor to most of one's bases is not an efficient way to play StarCraft.
Following the granularity and Thor as support GTA unit, we enter into the unit's speed, Thors are not fast, and that's OK, as previously mentioned it does indeed fit well into the immobile Mech idea, yet, this Mech fantasy is something that for the most part applies to how the mech composition works as a whole, and not for each individual unit.
Mech. Ground mech, in my eyes and experience as a Mapmaker is in need of a reliable & rather fast GTA work horse, that can help with keeping and supporting the main Missile Turret defenses, not simply be a bigger turret for the army, and it needs to do that all the while being a reliable GTA DPS dealer when big battles ensue. Thor's concept, sadly does not fit these needs. The unit itself, by design is not a fast reliable support, but big, bulky area denial. In other words, the Thor is a bigger, costlier, and less reliable slow moving Missile Turret. Mech already has Missile Turrets.
I hope, I made the points clear, yet not smeared them on the DevTeam's face, that's not my intention, far, far from it, I want to aid, not “down talk”, yet on text is hard to convey emotion.
Now, with the Thor talk over, lets delve into what I consider the Cyclone has the potential to become.
I really dislike forcing my things & ideas on anybody, I consider arguing and seeing things from different angles to finally arrive a consensus is a much better strategy, yet, I think must post once again the video I made for last weeks Mapmaker's Perspective, and further explain my thought process behind it, or my thought process behind the concepts and capabilities of the Cyclone in question.
Just like I mentioned talking about the Thor, and agreed with DevTeam that the Thor does indeed very much fits on the Core ideas the Mech composition has. While the unit does not work well as a main GTA work horse, it works very, very well as a support unit to a greater GTA force, like Marines or a healthy amount of Widow Mines, or Missile Turrets, or even a Liberator-Viking force, yet by themselves, 3 to 4 Thors on an army simply are not a reliable GTA force.
One very, very important point that should also be discussed, is the self-sustainability of Mech, stronger GTA Cyclones, do and will overlap with Marines, they will give a similar player experience, and they will overshadow the Thor's AA attack. Yet, we know that Mech atm does not have a good way of dealing with air, and we know that the reason this is so. So I ask, is it truly worth it to have Mech, not be a self-sustainable army composition, in order to avoid having two units (Marines/Cyclone) partially overlap? This is far from an easy decision, yet, I do consider that if DevTeam wants to have Mech be on a better place, on a Metagame and structural sense than it has been on the past (overshadowed by Bio). Mech should also have the capabilities to be self reliant, be self sustainable, and independent of needing considerable investment into Starport or Barrack units, in order to at least partially be able to deny enemy air and map control via a strong more or less mobile ground based AA unit.
Now, all of this is much, much, much easier said than done. I do not have most of the timeframes and pressures DevTeam has, and as such, I can give some of these thoughts more merits than they deserve, I mean, it is hard to even think of redesigning the Thor and Cyclone when one has Producers breathing on one's neck. Nonetheless, I sadly have my own pressures, and can't invest all my weekend onto these Blogs, yet this time, I can still give it a go.
Going back into the Cyclone video above, and Cyclones in general.
Cyclones on the test version of the game work at their core as mobile anti-armored, with light anti-air capabilities, they are not core GTA, and they barely count as GTA support at all (Too unreliable/User dependent). Like discussed on the past Cyclone focused Blog, they are very mobile units with an insane vs armored DPS, I'm sure the DPS is for testing/exploitation purposes, and partially because the unit was designed to be non-microable as a personality and balance trait, so higher DPS is kind of warranted, yet a DPS of 42 on such a mobile unit, is not Ok. 6 Cyclones can take down a enemy Town Hall in less than 6 seconds. And I also know that I don't really need to go very deep regarding my concerns this can cause in the game, because I have already outlined them on the previous Blog (For those new coming into this series of blogs, be sure to read that one).
Cyclones working as a very mobile anti-armored skirmisher, leaves the unit very out of place regarding its function on the Mech Terran army, and sadly no, simply because this unit comes out the Factory and has tracks it does not make it a “Mech” unit on the eyes of this “Mildly Open Minded Hardliner”. Siege Tanks are the Core Mech unit, which all others orbit around, Test Cyclone on their highly un-microable and high damage capacity by their very existence and function risk over-stepping their “unit fantasy” boundaries into Siege Tank territory (Main ground anti-armored unit).
This is partially why on the ideal Cyclone video I made the unit have the Reaper Mine as a utility support spell, so the unit wouldn't only work as a GTA Core force or main ground anti-armored Skirmisher, but also have other uses for Mech, all the while giving me an excuse to heavily nerf its GTG damage, so the unit simply would be unable to work on its own (or with light support) as a GTG strong skirmisher. And instead by its rather fast speed would work as a light skirmisher and support unit, planting mines over important areas of the map in order to disrupt enemy forces and gain vision. Or at least that's the “ideal” purpose, tho the mines don't do enough dmg for that in the iteration showcased, and I was afraid of overlapping too much with Widow Mines.
Sadly on this crude vision of Mech and the Cyclone, Thors get put aside, the unit would simply be outclassed by this new Cyclone, tho arrangements could be made, such as choosing making the Cyclones GTA be single target and leaving the splash capabilities on the Thor, or the inverse which I find more appealing, leaving a single target auto-attack GTA Cyclone with +light while having the bulky and heavy Thor with the +armored or +massive attack tag, so as to still have Thors work as the go to ground Factory unit vs a large air army. The interaction between both units could also be changed in favor of the Cyclone on a much more stronger fashion by removing the GTA capabilities of the Thor altogether, and having the unit become much stronger on the ground on extreme late-game scenarios, I'm sure DevTeam has seen some really interesting non-Tank overlapping spells and capabilities on the Thor over the years. I consider that if DevTeam decides to test this, some of these spells and capabilities might be very interesting for this new Thor.
DevTeam might even want to simply phase out the Thors, re-structure the Meka into a Mothership-esq unit, allowing it to fulfill a much more niche and fantasy fulfilling spot. But I'm getting off track here, Cyclones on my mind can't work as a strong ground skirmisher and harassment unit, by creating such a unit DevTeam would be weakening the fantasy of StarCraft Mech, where this unit simply (be it Live or Test) does not reliable fit into. Instead I humbly would suggest Cyclones be reworked into a support unit for filling the other spaces SC Mech as a concept still has left to fill.
Factory AA
But that doesn’t mean we can’t increase the current effectiveness of the new Cyclone’s AA capabilities as long as it isn’t the main AA option throughout the whole game. We feel that there could be room here to increase its effectiveness, and we’ll start testing out some numbers as per your suggestion.
But that doesn’t mean we can’t increase the current effectiveness of the new Cyclone’s AA capabilities as long as it isn’t the main AA option throughout the whole game. We feel that there could be room here to increase its effectiveness, and we’ll start testing out some numbers as per your suggestion.
I will be very happy to test a more reliable AA Cyclone, at the moment, Cyclones are being used on TvZ to hit some serious Timing attacks alongside Hellbats, and by the considerable DPS Cyclones have on the test map, Zerg players have adapted into using Mutalisks to fend those timings off, and because the current iteration of the Cyclones does not have a reliable AA attack, those Mutalisks are incredibly successful when it comes to fully shutting down said Timings. From a design perspective I don’t consider DevTeam had this development in mind while giving Cyclones the current GTA attack they have.
Next balance changes
Aside from the potential Cyclone AA damage changes, we would like to take a pass at Ultralisk armor and really focus on tuning of the new Tempest ability. With Ultralisk armor, we can try the popular suggestion of increasing the base armor by 1 point and reducing the armor the upgrade provides by 2 points for a slight buff to the base unit + a nerf to the overall unit after upgrades.
Aside from the potential Cyclone AA damage changes, we would like to take a pass at Ultralisk armor and really focus on tuning of the new Tempest ability. With Ultralisk armor, we can try the popular suggestion of increasing the base armor by 1 point and reducing the armor the upgrade provides by 2 points for a slight buff to the base unit + a nerf to the overall unit after upgrades.
This will be a very interesting change to test, damage done to Ultralisks by Marines will be virtually duplicated, yet from the small testing I have done today it is not very noticeable. I just don't have the energy left to write 4 other paragraphs going about Ultralisk timings and how Zerg aligns its production to fit upgrades, best case example here being Infestors and the Pathogen Glands upgrade, so the +1 armor buff might in my eyes not be the best way to make un-upgraded Ultras actually be used by Zergs who are not already against the cords and are forced to use their un-upgraded Ultras on less than ideal conditions, but instead nerfing the time of the Carapace upgrade, while slightly reducing the build time of the Ultralisk Cavern might be an interesting path to think and delve on. I'm sure DevTeam does not need me telling them this, yet I do consider that it might still be important to mention.
Next balance changes
For the Tempest, we would like to ask you guys to help us out on testing this ability against heavy Siege Tank based compositions and Hydra/Lurker based armies.
For the Tempest, we would like to ask you guys to help us out on testing this ability against heavy Siege Tank based compositions and Hydra/Lurker based armies.
It will be fun to test, sadly not many Protosses are playing the test map. I consider that maybe considering testing the Stalker buff, and then rolling it back a month or two later in order to lure more Protoss players into the test map might be a reasonable tactic to further increase the interest on that part of the playerbase.
Regarding the Hydra/Lurker composition, on the very, very little testing I have done, Tempests don't work well vs that composition during Midgame/early-Lategame timings. It takes time to get a decent amount of Tempests, and they are a considerable time and resource investment even if the Protoss player opened Phoenix, the unit still is mighty expensive. Now, this is not a bad thing on my eyes, only that I need to try and focus on seeing how well does the unit behave on more Lategame oriented scenarios.
KR feedback regarding SC2’s main goal
We’ve been seeing a lot of discussion from the KR community on two points: making the game easier, and toning down harassment so that the general pace of the game is slower.
We agree heavily with many players in our global community that SC2 is one of very few games where you are solely responsible for whether you win, or lose, and that these results are tied to the time, dedication, and skill which players put into mastering the game. The feeling of practicing and mastering a part of the game, and directly seeing my personal increase in skill, is truly unlike any other game in the world.
Our main goal for StarCraft 2 is to create the best game of its type that it can ever be, and not necessarily selling more copies of the game or increasing the playerbase. Those are also great secondary goals, but we don’t feel that we should be in a place where we start to hurt the main, most important goal of StarCraft 2.
We’ve been seeing a lot of discussion from the KR community on two points: making the game easier, and toning down harassment so that the general pace of the game is slower.
We agree heavily with many players in our global community that SC2 is one of very few games where you are solely responsible for whether you win, or lose, and that these results are tied to the time, dedication, and skill which players put into mastering the game. The feeling of practicing and mastering a part of the game, and directly seeing my personal increase in skill, is truly unlike any other game in the world.
Our main goal for StarCraft 2 is to create the best game of its type that it can ever be, and not necessarily selling more copies of the game or increasing the playerbase. Those are also great secondary goals, but we don’t feel that we should be in a place where we start to hurt the main, most important goal of StarCraft 2.
It is very hard bordering on impossible for me to assets anything the Korean Coaches might have said regarding the game, because I simply wasn’t there to hear it, yet, there are many, many things on SC2 which can be easily misread and classified into the “game's too hard” bin. Gamepacing, Macro Mechanics, and more importantly than any other Army DPS density, which is heavily dependent of the game's path-finding and unit clumping, I fully realize that DevTeam has stated over and over again that they are not on a position to look into increasing the unit separation on the pathing algorithm, yet, if Korean Coaches and Progamers consider that this piece of the game is increasing the game's volatility so much that Korean ProGamers are having trouble enjoying the game because of it, I seriously consider that DevTeam should look into what options there are on the table for possible changes to the pathing algorithm. Now days with a much stronger Siege Tank, Baneling & Lurker it might be the best possible time to do this change.
Yes, spell's area might need to be tweaked, yes, SC2 Pathfinding must be one of the 7 wonders of Video Game software, yes, it is indeed a good feature for new players to have, yet, just like I mentioned on the previous paragraph, if pathfinding is creating such a high volatility that Korean pro's consider that the game is simply too hard and volatile, it is something that on my eyes should be looked at and reviewed. Now it doesn't need to be a huge sweeping change, the engine and SC2 Path-finding have a fair amount of tweaking points that can be touched in order to reduce the Army DPS Density problem, or at least have armies that just received a command not clump as much.
Other thing that I think has mild value is when this test patch is over how well will it sound for the DevTeam to announce: “Thanks to the close work with our community and feedback received we have done large sweeping changes to StarCraft II, including even changing the very Pathfinding algorithms of the game in order to create a better and more stable gameplay for both New and Experienced players alike!”... Now I know that such a thing might be a pipe dream, yet this Open Minded Hardliner, like with many other things thinks that it is something worth saying, and hopefully worth reading too,
Sadly, I'll need to cut this week's Mapmaker's Perspective a bit short by time reasons, that said, I do find quite fun to air my thoughts this way, I just hope to not be too much of a contrarian hitting too hard on what I consider unit designs which could be improved by straight forward changes, which DevTeam might agree with me only to be constrained by Timeframe issues. That aside, I hope to be able to keep writing those as long as time and will allow it.
Previous Blogs:
♦ Oct 7/14
♦ Sep 30