The swarm host was under a lot of fire in the LoTV forum. People wanted it to receive large changes, small changes, and some wanted to see the unit completely gone. However, after experimenting with the swarm-host in some ZvZs before the beta shut down I realized that its one of the most interesting units ever to exist in starcraft. It is not a deathball unit: it is more of a harass/position-based unit. The swarm-host's use comes in it's ability to massively buff the strength of a small hit-squad, poke down defenses, snipe bases, or strengthen the zerg's army when breaching a location (or buying time on defense). From my experience, the swarm-host can only be added on well into the mid-late game when both players are on around 4 bases. My tactic is simple: I make 6-12 swarmhosts (although 8 is my usual number), an adequate amount of drop-upgraded overlords and have my swarmhosts hover around the map. When I want to mount an offensive on an enemy location or base, I drop my swarmhosts nearby, throw out the locusts, and preemptively fly away to the next location I want to attack.
Some would argue that they chew up too much supply, cost too much and the down-time is just too high. This is not the case especially in ZvZ, which is a very trade-heavy, push-and-pull matchup: you are never maxed in ZvZ. If you are, you aren't attacking enough. Also, ZvZ is very anti-deathball once lurkers are out. In a deathball meta, these kinds of units would be terrible, since it subtracts from your own deathball and the utility is just not worth it. But in a game where battles happen all across the map, and it is constant unit-trading, it is possible to slowly add on swarmhosts. You will be a bit behind at first, but they will quickly be worth it after 2-3 locust volleys if used right. The damage and damage buffer that the locust provide against lurker lines is invaluable: the enemy is forced to take a very unfavorable fight, or abandon the position (very important in the new positionally-based metagame). The swarm-host acts as an earlier, mineral based siege-breaker as opposed to a later, gas intensive viper or a less reliable ravager. Proper swarm-host usage is also a skill players will learn: in order to make the swarm-host useful, you MUST use those locust every 47~ seconds or else its a wasted investment. And making too many swarm-hosts (probably) makes you have too weak of a mobile ground army, and the enemy would probably just be able to steamroll you. And too many swarm-hosts lategame seems too clunky to be used efficiently. Do you see the cool, unique and niche unit that is the LoTV swarmhost? What other units are as unique as it? The metagame will dictate it's exact uses, but so far it seems to be an incredibly designed utility unit. ZvZ is shaping up to be a very challenging, deep and fun matchup to play.
EDIT: I haven't tried it out against protoss ground or mech yet, but I can't imagine it would be as good. Also, I think colossi will get a resurgence in the metagame against bio and maybe Z. I theorize that protoss players will eventually split up their armies more to deal with the multi-pronged attacks of bio and zerg. Because disruptors are much weaker against lower unit counts (big hits much less guaranteed), colossi will provide an important, reliable source of AOE damage. Probably 1-2 per army, since you still want some disruptors.
|
United Kingdom20326 Posts
Is the Swarmhost the best designed unit in LoTV?
Played 250-300 games in high master (lotv archon GM) and never had one built against me, not even once
time will tell how good they are in pro level play but there's a long list of annoying and strong plays that SH didn't manage to make it to during beta
-----------
Also, I think colossi will get a resurgence in the metagame against bio and maybe Z
Colossi used to deal 1.25x more damage than they do now and meanwhile, the options for terran and zerg have become outright stronger. Ravagers. Lurkers. Liberators - it would be harder for protoss even with the old colossi that hit 1.25x harder.
Building 2 colossi with thermal lance is a few hundred minerals and some seconds more expensive than building 4 disruptors, so i doubt they'll take off in low numbers. They were never the staple support unit - they're the unit you try to get to, not the one that you use to try to get to something else.
While disruptors can't crush lower unit counts, if the enemy is using small packs of units then that's advantageous to you in some ways; protoss infantry has always scaled very poorly against large packs of units, needing AOE support to stand their ground. Against 8 hydralisks or 16 marines, you can blob them with gateway units such as upgraded adepts or newly buffed chargelots. At 150/150, you can also afford to have disruptors spread around a bit and maybe have a couple in a warp prism
|
EPT
