|
It is hard to make a unit with such a good survivability good in low numbers, because it is hard to make something like this bad in big numbers. Also it would actively work against the swarmy Zerg picture. Either they reduce the survivability in exchange for making them better in low numbers (would become a problem if someone is good at keeping them alive) or they make it impossible to get alot of this unit (Roach Supply Syndrome).
For me the Swarmhost is just an harassment and slow down unit and it does its job there. And they work like tanks, the more shots you get off the better, just that they shoot really slow, but have insane range. Sadly they made the Locust a Miniroach, which makes it hard for other Zerg units to move together with the Locust and also because of their high damage the tanking ability had to be reduced.
I like them anyway, burrow ... shoot ... run away ... burrow ... shoot ... run away. Sounds like using a Lurker against tanks. Just that its more important to use your army along this. Never used them in high numbers though, so they never went do or die with them. But I would really like to see the Locust become more of a tank. Would dislike if the Swarmhost would be turned into a unit that can be used alone, like the Roach after all the changes it got, to become either to use for players.
|
United Arab Emirates439 Posts
I. The Swarm Host is largely ineffective in small numbers (4-8), but becomes exponentially more effective as you reach a critical mass of the unit (12-15+). This is a case of increasing returns.
II. Investing into a critical mass of Swarm Hosts represents a huge investment. It is extremely difficult to transition out of this composition, as the investment essentially puts you "all in." Battles can become very lopsided very quickly, much like in lategame ZvP with Archon toilet vs Broodlord Infestor. One side often wins very convincingly here, and comebacks are not possible.
I really don't agree with (most of) this and I think it's really an issue of how most people use them right now that has people believing this. I do agree going for huge amounts of Swarm Hosts is more or less an "all in" in most situations. Going for this mass Swarm Host strategy is a good way of "economically leaning on" your opponent. Forcing them to either counter the composition quickly enough or die slowly to waves of Locusts. But if they do get up enough Splash (or an Air army perhaps) then they can crush the Swarm Host army and often win the game (I suppose base trading is an option too). I 100% agree with you on this part.
Where I disagree is that Swarm Hosts are worse in small numbers. I think they just operate very differently than the Mass Swarm Host strategy, you aren't "leaning on" your opponent in the same way. Instead of basically forcing your opponent to counter your army in time or die, you are forcing your opponent to get a good position soon or die/take damage. Bouncing around with Swarm Hosts pressuring different locations over and over is very strong against a player who has no map control or vision. Especially against a Mech Terran who will primarily rely on Siege Tanks to deal with Locusts. If you gain map control and then bounce around between 2/3 locations with only ~8 Swarm Hosts constantly it is very hard for a Terran to deal with if they don't have vision of the Swarm Hosts to see where they are moving. The only way to deal with it without vision of your Swarm Hosts is to get a high enough Siege Tank count to split between 2 (or 3 on some maps) places. This is a way to force your opponent into a certain army composition with less investment on your part than on his, not to mention before he has the resources to do this you often will do at least a bit of damage. During this time it can also open up opportunities for counter-attacks where his Splash units aren't.
|
On December 16 2012 06:59 ZjiublingZ wrote:Show nested quote +I. The Swarm Host is largely ineffective in small numbers (4-8), but becomes exponentially more effective as you reach a critical mass of the unit (12-15+). This is a case of increasing returns.
II. Investing into a critical mass of Swarm Hosts represents a huge investment. It is extremely difficult to transition out of this composition, as the investment essentially puts you "all in." Battles can become very lopsided very quickly, much like in lategame ZvP with Archon toilet vs Broodlord Infestor. One side often wins very convincingly here, and comebacks are not possible. I really don't agree with (most of) this and I think it's really an issue of how most people use them right now that has people believing this. I do agree going for huge amounts of Swarm Hosts is more or less an "all in" in most situations. Going for this mass Swarm Host strategy is a good way of "economically leaning on" your opponent. Forcing them to either counter the composition quickly enough or die slowly to waves of Locusts. But if they do get up enough Splash (or an Air army perhaps) then they can crush the Swarm Host army and often win the game (I suppose base trading is an option too). I 100% agree with you on this part. Where I disagree is that Swarm Hosts are worse in small numbers. I think they just operate very differently than the Mass Swarm Host strategy, you aren't "leaning on" your opponent in the same way. Instead of basically forcing your opponent to counter your army in time or die, you are forcing your opponent to get a good position soon or die/take damage. Bouncing around with Swarm Hosts pressuring different locations over and over is very strong against a player who has no map control or vision. Especially against a Mech Terran who will primarily rely on Siege Tanks to deal with Locusts. If you gain map control and then bounce around between 2/3 locations with only ~8 Swarm Hosts constantly it is very hard for a Terran to deal with if they don't have vision of the Swarm Hosts to see where they are moving. The only way to deal with it without vision of your Swarm Hosts is to get a high enough Siege Tank count to split between 2 (or 3 on some maps) places. This is a way to force your opponent into a certain army composition with less investment on your part than on his, not to mention before he has the resources to do this you often will do at least a bit of damage. During this time it can also open up opportunities for counter-attacks where his Splash units aren't.
Against a mech terran it's better to just blinding cloud and be aggresive with a hydra roach army.
A few units can deny a swarm host from ever dealing damage. Literally any army composition pretty much shits on swarm hosts unless you're seriously massing them. It's not like a siege tank where the damage is guaranteed and WILL whittle you down.
In zvp in particular a lot of times you're sitting there sending locusts up to an enemy army and they can just sit there forever with colossus forever outranging them. And if they opened stargate? Welp. You may as well have just went hydras in the first place.
|
The SH is just another poorly made unit that further indicts the developers. They aren't their predecessors, that's for sure. They might as well put in the lurker and tweak how much aoe it does, but they decided to make a semi lurker knockoff that wasn't nearly as well made. Look what happened.
The biggest issue for me is how much you have to commit to SH, it's so irritating when you think about adding a couple to your army and then realize you can't do it. A lot of times they ARE YOUR ARMY. It would be like going mass lurker in order to make lurkers work. They also suffer from the long respawn time of locusts which compromises the SH in small numbers, they just don't threaten enough and are too often unable to attack at all because of the respawn time. Then you realize you can't have it be a 3 second CD or anything because then the locusts would obviously be OP and pile up. Then you'd have to nerf the duration of locusts but they'd become worthless because they wouldn't have enough time do actually do damage since they're slow.
|
On December 16 2012 07:05 Glurkenspurk wrote: It's not like a siege tank where the damage is guaranteed and WILL whittle you down.
This is the issue I have with them, same issue I have with Broodlords. Lurkers were a better unit because their attacks were stealth and they hit a defined target (or at least a defined area). You could use them for defense for this reason, same as tanks. Locusts take time to spawn and time to get into position. The delay makes it so that you have to mass SH's to get any real effect out of them. Locusts and broodlings getting in the way of other units is also annoying. I'd much rather see broodlings kamikaze their targets and not run around on the ground, but that's another topic entirely.
But what I really want SH's to be is allowing static defense and map contain. I don't need them to be an offensive unit per se, just one I can use to gain map control across the board. But the rate at which locusts spawn, and their speed makes it so that this isn't a real option. In small numbers you run the risk of there being no attack output and it's too easy to break. At larger numbers they become an offensive weapon which imho they shouldn't really be.
If it was me I would speed up the locusts spawn, and increase burrow time of SH's to compensate. You could also make locusts burrow up to their rally point as well if you so choose. But if you want them to work in low numbers you need to have a more consistent attack, like a tank would. And if you don't want to turn them into an offensive powerhouse you need to make it so that they are not as mobile, IE increased burrow time so that massing them for offense is not as feasible.
|
So on a similar topic of this thread, for all the people saying swarm hosts are good. Can anyone show me ANY recent games where swarm hosts were used to deny space in any meaningful way? I have yet to see space control being useful AT ALL in a game where most players pretty much only move out slightly in front of their new bases up until they are actually going to hit a timing attack. Small skirmishes do not happen enough for units like this to fit into the game in a way that makes sense, and as a siege unit they are clearly too weak to supplement a real army.
|
What if the locust had, say for easy math, 100 energy. Spawning a pair of locusts costs 20 energy with the energy generation of 15 or so per locust spawn. This way, every time you spawn you go -5 energy. If z stays for an extended assault with locusts, eventually he will run out, and be forced to move back(or be killed). This would limit the amount of mass SH, burrow attack and win, or attack and lose. Z will have to retreat after a short assault. There would have to be other changes made to the SH, to change their effectiveness in low numbers, but that should limit the "massing" SH power. A proper defence can hold until SH energy is gone, then the counter attack can easily pick off no energy swarm hosts.
|
Russian Federation4295 Posts
On December 16 2012 07:54 HumpingHydra wrote: What if the locust had, say for easy math, 100 energy. Spawning a pair of locusts costs 20 energy with the energy generation of 15 or so per locust spawn. This way, every time you spawn you go -5 energy. If z stays for an extended assault with locusts, eventually he will run out, and be forced to move back(or be killed). This would limit the amount of mass SH, burrow attack and win, or attack and lose. Z will have to retreat after a short assault. There would have to be other changes made to the SH, to change their effectiveness in low numbers, but that should limit the "massing" SH power. A proper defence can hold until SH energy is gone, then the counter attack can easily pick off no energy swarm hosts. Ghosts with emp will be useful now against swarm host. Love that idea as zerg
|
United Arab Emirates439 Posts
On December 16 2012 07:05 Glurkenspurk wrote:Show nested quote +On December 16 2012 06:59 ZjiublingZ wrote:I. The Swarm Host is largely ineffective in small numbers (4-8), but becomes exponentially more effective as you reach a critical mass of the unit (12-15+). This is a case of increasing returns.
II. Investing into a critical mass of Swarm Hosts represents a huge investment. It is extremely difficult to transition out of this composition, as the investment essentially puts you "all in." Battles can become very lopsided very quickly, much like in lategame ZvP with Archon toilet vs Broodlord Infestor. One side often wins very convincingly here, and comebacks are not possible. I really don't agree with (most of) this and I think it's really an issue of how most people use them right now that has people believing this. I do agree going for huge amounts of Swarm Hosts is more or less an "all in" in most situations. Going for this mass Swarm Host strategy is a good way of "economically leaning on" your opponent. Forcing them to either counter the composition quickly enough or die slowly to waves of Locusts. But if they do get up enough Splash (or an Air army perhaps) then they can crush the Swarm Host army and often win the game (I suppose base trading is an option too). I 100% agree with you on this part. Where I disagree is that Swarm Hosts are worse in small numbers. I think they just operate very differently than the Mass Swarm Host strategy, you aren't "leaning on" your opponent in the same way. Instead of basically forcing your opponent to counter your army in time or die, you are forcing your opponent to get a good position soon or die/take damage. Bouncing around with Swarm Hosts pressuring different locations over and over is very strong against a player who has no map control or vision. Especially against a Mech Terran who will primarily rely on Siege Tanks to deal with Locusts. If you gain map control and then bounce around between 2/3 locations with only ~8 Swarm Hosts constantly it is very hard for a Terran to deal with if they don't have vision of the Swarm Hosts to see where they are moving. The only way to deal with it without vision of your Swarm Hosts is to get a high enough Siege Tank count to split between 2 (or 3 on some maps) places. This is a way to force your opponent into a certain army composition with less investment on your part than on his, not to mention before he has the resources to do this you often will do at least a bit of damage. During this time it can also open up opportunities for counter-attacks where his Splash units aren't. Against a mech terran it's better to just blinding cloud and be aggresive with a hydra roach army. A few units can deny a swarm host from ever dealing damage. Literally any army composition pretty much shits on swarm hosts unless you're seriously massing them. It's not like a siege tank where the damage is guaranteed and WILL whittle you down. In zvp in particular a lot of times you're sitting there sending locusts up to an enemy army and they can just sit there forever with colossus forever outranging them. And if they opened stargate? Welp. You may as well have just went hydras in the first place.
DId you even read my post...? First off, I wasn't talking about whether or not Swarm Hosts as a whole are the best composition vs anything, I was disagreeing with the idea that Swarm Hosts in small numbers accomplish very little and you have to get big numbers of them to have any effect. Swarm Hosts can work in some situations. I'm not talking about if Roach/Hydra Viper is better as a general strategy (which everyone already knows at this point it is...)
Second,
In zvp in particular a lot of times you're sitting there sending locusts up to an enemy army
Once again, I have to ask, did you even read my post? If you're just sitting there, that's your fault. I specifically talked about how that's why people have this bad impression of them, because they just sit there with them. Use them to spread the Protoss out.
|
I honestly think they should remove it and put the lurker back in.
This unit has become the new Hydra, very situational and rarely used unless all-ining.
|
On December 16 2012 09:01 ZjiublingZ wrote:Show nested quote +On December 16 2012 07:05 Glurkenspurk wrote:On December 16 2012 06:59 ZjiublingZ wrote:I. The Swarm Host is largely ineffective in small numbers (4-8), but becomes exponentially more effective as you reach a critical mass of the unit (12-15+). This is a case of increasing returns.
II. Investing into a critical mass of Swarm Hosts represents a huge investment. It is extremely difficult to transition out of this composition, as the investment essentially puts you "all in." Battles can become very lopsided very quickly, much like in lategame ZvP with Archon toilet vs Broodlord Infestor. One side often wins very convincingly here, and comebacks are not possible. I really don't agree with (most of) this and I think it's really an issue of how most people use them right now that has people believing this. I do agree going for huge amounts of Swarm Hosts is more or less an "all in" in most situations. Going for this mass Swarm Host strategy is a good way of "economically leaning on" your opponent. Forcing them to either counter the composition quickly enough or die slowly to waves of Locusts. But if they do get up enough Splash (or an Air army perhaps) then they can crush the Swarm Host army and often win the game (I suppose base trading is an option too). I 100% agree with you on this part. Where I disagree is that Swarm Hosts are worse in small numbers. I think they just operate very differently than the Mass Swarm Host strategy, you aren't "leaning on" your opponent in the same way. Instead of basically forcing your opponent to counter your army in time or die, you are forcing your opponent to get a good position soon or die/take damage. Bouncing around with Swarm Hosts pressuring different locations over and over is very strong against a player who has no map control or vision. Especially against a Mech Terran who will primarily rely on Siege Tanks to deal with Locusts. If you gain map control and then bounce around between 2/3 locations with only ~8 Swarm Hosts constantly it is very hard for a Terran to deal with if they don't have vision of the Swarm Hosts to see where they are moving. The only way to deal with it without vision of your Swarm Hosts is to get a high enough Siege Tank count to split between 2 (or 3 on some maps) places. This is a way to force your opponent into a certain army composition with less investment on your part than on his, not to mention before he has the resources to do this you often will do at least a bit of damage. During this time it can also open up opportunities for counter-attacks where his Splash units aren't. Against a mech terran it's better to just blinding cloud and be aggresive with a hydra roach army. A few units can deny a swarm host from ever dealing damage. Literally any army composition pretty much shits on swarm hosts unless you're seriously massing them. It's not like a siege tank where the damage is guaranteed and WILL whittle you down. In zvp in particular a lot of times you're sitting there sending locusts up to an enemy army and they can just sit there forever with colossus forever outranging them. And if they opened stargate? Welp. You may as well have just went hydras in the first place. DId you even read my post...? First off, I wasn't talking about whether or not Swarm Hosts as a whole are the best composition vs anything, I was disagreeing with the idea that Swarm Hosts in small numbers accomplish very little and you have to get big numbers of them to have any effect. Swarm Hosts can work in some situations. I'm not talking about if Roach/Hydra Viper is better as a general strategy (which everyone already knows at this point it is...) Second, Show nested quote +In zvp in particular a lot of times you're sitting there sending locusts up to an enemy army Once again, I have to ask, did you even read my post? If you're just sitting there, that's your fault. I specifically talked about how that's why people have this bad impression of them, because they just sit there with them. Use them to spread the Protoss out.
Once you unborrow they just kill you...
|
On December 16 2012 06:43 mercurial_mind wrote:Show nested quote +On December 16 2012 06:05 rembrant wrote: From the very first SH video we saw I knew this unit was gonna be a problem, its a unit that's either ridiculously op or very useless and I still have no clue how it can be balanced without drastically changing how it works.
The only suggestion which I think could maybe help with minimal changes would be if the SH built up a larger total locust count but it took longer to replenish in whole and you could fire off locusts at any time, basically working like the carrier does with interceptors. This way if you build up locusts by waiting a bit you can get a larger overall burst of units at a time, something along those lines. I want to echo this as a possible improvement, because I was thinking the same thing from the beginning of the thread. Let me try to flesh this out a little. Right now part of the "critical mass" problem is the fact that, when you assault an enemy position, the defenses are able to take out a certain number of locusts before they do any damage. Let's imagine that you are assaulting an intrenched terran who has the dps to take out a maximum of 20 locusts before any damage is taken. This means that you need to have at least 11 swarm hosts to accomplish anything at all, and once you have 11, each additional swarm host adds hugely to the amt of damage that can be done. Now let's imagine that swarm hosts function as follows: they still produce locusts at the same rate as presently. For the sake of the example, let's just say they presently produce 2 every 20 seconds. The new swarm host will produce 1 every 10 seconds. There will be a maximum number of locusts per swarm host. Let's just say: 8. These locusts get stored up and can be released all at once. In addition, you can place a burrowed swarm host on "auto-release" so that every time a new locust is produced, it will be released. This allows you to continue chipping steadily away at a position once you have broken its defenses entirely. This change would make a huge difference, but would have to be balanced carefully. In the initial scenario, you now only need 3 swarm hosts to build 24 locusts at once, breach the 20-locust threshold, and begin to do damage. But it will take 80 seconds to achieve this. Additionally, the locusts still time out in such a way that the maximum potential damage for a mass release is capped by the damage output of the locusts over their life expectancy. Now the obvious danger of this change is: what if you went ahead and built 10+ swarmhosts anyway. That could potentially be devastating. But this would be a risk because you would have to wait to build up all the locusts and you would only get one shot to do this massive damage every 80 seconds. In the meantime you would be vulnerable to attacks that the present SH is not vulnerable to since its continuous small waves of locusts provide a kind of screen against attacking units. Here are some other changes that might need to accompany this for balancing purposes. It might be found that, since total damage output is (locust #)x(dps)x(endurance time), the endurance time of locusts makes the potential damage of a big wave just too high. On the other hand, if you decrease endurance time, then you cannot siege from as great a distance. Perhaps this would call for much faster locusts that can arrive at a distant siege location quickly but that last for much less time. The speed increase would only affect the sorts of attack positions the locusts could reach while the shorter endurance would nerf total damage output. Another consideration is that it might be too powerful to allow unburrowed SH to kite away after releasing a big wave and still be producing new locusts while getting to safety. This could be balanced by requiring SH to be burrowed in order to build up the inventory of locusts. Finally, I think larger "volleys" of locusts would make re-position micro a lot more interesting with SH, and they would become a much more interesting and sneaky sort of siege unit, the likes of which we have never seen. Imagine 3 swarm hosts releasing 24 locusts to level a couple of missile turrets and immediately withdrawing while a flock of mutas swoops in for worker harass.
As a whole, I really like this idea It would also be great if the Locusts had a kind of "leap" that the Raptor has in the HotS campaign that launches it into range of the enemy army. Plus, the notion of tiny units charging at you in waves would be "Zergy". It would be guaranteed damage and would increase the viability of a few Swarmhost.
|
I am unconvinced by the OP's argument. Some of the reasoning seems flawed like in order for them to be effective it takes a lot of investment. Well this is true for a lot of things in this game like carriers or broodlords. It sounds like he wants lurkers while swarm hosts fulfill a different function. It is not wrong for a unit to be designed to be primarily used in the main army. This unit as it stands allows Zerg to build a different type of army. An army that you can not turtle against that is not hive tech.
Oh we could tweak things here and there to make it more robust for different uses but the unit seems fine in concept. We could make the locusts move faster or slower, give it more or less range, give it more or less hp, or more or less timed life. Make swarm hosts move faster or slower, burrow faster or slower, respawn units faster or slower, make them spawn more or less units, or give them more or less hp.
You can mix and match any of these things and get the unit to become worse at some things or better at others.
|
On December 16 2012 10:28 porygon361 wrote:Show nested quote +On December 16 2012 06:43 mercurial_mind wrote:On December 16 2012 06:05 rembrant wrote: From the very first SH video we saw I knew this unit was gonna be a problem, its a unit that's either ridiculously op or very useless and I still have no clue how it can be balanced without drastically changing how it works.
The only suggestion which I think could maybe help with minimal changes would be if the SH built up a larger total locust count but it took longer to replenish in whole and you could fire off locusts at any time, basically working like the carrier does with interceptors. This way if you build up locusts by waiting a bit you can get a larger overall burst of units at a time, something along those lines. I want to echo this as a possible improvement, because I was thinking the same thing from the beginning of the thread. Let me try to flesh this out a little. Right now part of the "critical mass" problem is the fact that, when you assault an enemy position, the defenses are able to take out a certain number of locusts before they do any damage. Let's imagine that you are assaulting an intrenched terran who has the dps to take out a maximum of 20 locusts before any damage is taken. This means that you need to have at least 11 swarm hosts to accomplish anything at all, and once you have 11, each additional swarm host adds hugely to the amt of damage that can be done. Now let's imagine that swarm hosts function as follows: they still produce locusts at the same rate as presently. For the sake of the example, let's just say they presently produce 2 every 20 seconds. The new swarm host will produce 1 every 10 seconds. There will be a maximum number of locusts per swarm host. Let's just say: 8. These locusts get stored up and can be released all at once. In addition, you can place a burrowed swarm host on "auto-release" so that every time a new locust is produced, it will be released. This allows you to continue chipping steadily away at a position once you have broken its defenses entirely. This change would make a huge difference, but would have to be balanced carefully. In the initial scenario, you now only need 3 swarm hosts to build 24 locusts at once, breach the 20-locust threshold, and begin to do damage. But it will take 80 seconds to achieve this. Additionally, the locusts still time out in such a way that the maximum potential damage for a mass release is capped by the damage output of the locusts over their life expectancy. Now the obvious danger of this change is: what if you went ahead and built 10+ swarmhosts anyway. That could potentially be devastating. But this would be a risk because you would have to wait to build up all the locusts and you would only get one shot to do this massive damage every 80 seconds. In the meantime you would be vulnerable to attacks that the present SH is not vulnerable to since its continuous small waves of locusts provide a kind of screen against attacking units. Here are some other changes that might need to accompany this for balancing purposes. It might be found that, since total damage output is (locust #)x(dps)x(endurance time), the endurance time of locusts makes the potential damage of a big wave just too high. On the other hand, if you decrease endurance time, then you cannot siege from as great a distance. Perhaps this would call for much faster locusts that can arrive at a distant siege location quickly but that last for much less time. The speed increase would only affect the sorts of attack positions the locusts could reach while the shorter endurance would nerf total damage output. Another consideration is that it might be too powerful to allow unburrowed SH to kite away after releasing a big wave and still be producing new locusts while getting to safety. This could be balanced by requiring SH to be burrowed in order to build up the inventory of locusts. Finally, I think larger "volleys" of locusts would make re-position micro a lot more interesting with SH, and they would become a much more interesting and sneaky sort of siege unit, the likes of which we have never seen. Imagine 3 swarm hosts releasing 24 locusts to level a couple of missile turrets and immediately withdrawing while a flock of mutas swoops in for worker harass. As a whole, I really like this idea It would also be great if the Locusts had a kind of "leap" that the Raptor has in the HotS campaign that launches it into range of the enemy army. Plus, the notion of tiny units charging at you in waves would be "Zergy". It would be guaranteed damage and would increase the viability of a few Swarmhost.
I think the next issue for swarm hosts is that the locusts take up space. Less surface area for melee units which equals less efficiency for them, so in order for swarm hosts to be useful as a support to all units the locust needs to be altered.
Burrow movement locusts who are revealed with detection or upon attacking would help zerglings and ultralisks gain the surface area without actually making the swarm host benefit only ranged units. With the burrow movement, you can scout the opponent and possibly do some worker harassment if they lack detection. Obviously lowering hp, length of time or speed of the locusts would be needed to help balance it. Or maybe even an upgrade to burrow the locusts.
Just my thought onto actually making it a support unit for all zerg units rather then just ranged ones.
|
Russian Federation4295 Posts
I think the next issue for swarm hosts is that the locusts take up space. Less surface area for melee units which equals less efficiency for them, so in order for swarm hosts to be useful as a support to all units the locust needs to be altered.
Burrowed locusts can be op. But air aren't much.
|
United Arab Emirates439 Posts
+ Show Spoiler +On December 16 2012 06:43 mercurial_mind wrote:Show nested quote +On December 16 2012 06:05 rembrant wrote: From the very first SH video we saw I knew this unit was gonna be a problem, its a unit that's either ridiculously op or very useless and I still have no clue how it can be balanced without drastically changing how it works.
The only suggestion which I think could maybe help with minimal changes would be if the SH built up a larger total locust count but it took longer to replenish in whole and you could fire off locusts at any time, basically working like the carrier does with interceptors. This way if you build up locusts by waiting a bit you can get a larger overall burst of units at a time, something along those lines. I want to echo this as a possible improvement, because I was thinking the same thing from the beginning of the thread. Let me try to flesh this out a little. Right now part of the "critical mass" problem is the fact that, when you assault an enemy position, the defenses are able to take out a certain number of locusts before they do any damage. Let's imagine that you are assaulting an intrenched terran who has the dps to take out a maximum of 20 locusts before any damage is taken. This means that you need to have at least 11 swarm hosts to accomplish anything at all, and once you have 11, each additional swarm host adds hugely to the amt of damage that can be done. Now let's imagine that swarm hosts function as follows: they still produce locusts at the same rate as presently. For the sake of the example, let's just say they presently produce 2 every 20 seconds. The new swarm host will produce 1 every 10 seconds. There will be a maximum number of locusts per swarm host. Let's just say: 8. These locusts get stored up and can be released all at once. In addition, you can place a burrowed swarm host on "auto-release" so that every time a new locust is produced, it will be released. This allows you to continue chipping steadily away at a position once you have broken its defenses entirely. This change would make a huge difference, but would have to be balanced carefully. In the initial scenario, you now only need 3 swarm hosts to build 24 locusts at once, breach the 20-locust threshold, and begin to do damage. But it will take 80 seconds to achieve this. Additionally, the locusts still time out in such a way that the maximum potential damage for a mass release is capped by the damage output of the locusts over their life expectancy. Now the obvious danger of this change is: what if you went ahead and built 10+ swarmhosts anyway. That could potentially be devastating. But this would be a risk because you would have to wait to build up all the locusts and you would only get one shot to do this massive damage every 80 seconds. In the meantime you would be vulnerable to attacks that the present SH is not vulnerable to since its continuous small waves of locusts provide a kind of screen against attacking units. Here are some other changes that might need to accompany this for balancing purposes. It might be found that, since total damage output is (locust #)x(dps)x(endurance time), the endurance time of locusts makes the potential damage of a big wave just too high. On the other hand, if you decrease endurance time, then you cannot siege from as great a distance. Perhaps this would call for much faster locusts that can arrive at a distant siege location quickly but that last for much less time. The speed increase would only affect the sorts of attack positions the locusts could reach while the shorter endurance would nerf total damage output. Another consideration is that it might be too powerful to allow unburrowed SH to kite away after releasing a big wave and still be producing new locusts while getting to safety. This could be balanced by requiring SH to be burrowed in order to build up the inventory of locusts. Finally, I think larger "volleys" of locusts would make re-position micro a lot more interesting with SH, and they would become a much more interesting and sneaky sort of siege unit, the likes of which we have never seen. Imagine 3 swarm hosts releasing 24 locusts to level a couple of missile turrets and immediately withdrawing while a flock of mutas swoops in for worker harass.
@ Mercurial_Mind: What you are saying is based on a false premise, or at least an incomplete picture, right here:
Right now part of the "critical mass" problem is the fact that, when you assault an enemy position, the defenses are able to take out a certain number of locusts before they do any damage. Let's imagine that you are assaulting an intrenched terran who has the dps to take out a maximum of 20 locusts before any damage is taken. This means that you need to have at least 11 swarm hosts to accomplish anything at all, and once you have 11, each additional swarm host adds hugely to the amt of damage that can be done.
This is only true if you are assaulting one position, constantly. Let's take a look at your example: If you are fighting a Terran that has the 4+ ranged DPS to take out 20 locusts before they get in range to attack, how many Swarm Hosts do you actually need to do any damage? If you are attacking one area, 11. If you are attack two areas, 6. If you are attacking three areas, 4. etc. So you can almost triple the effectiveness of your Swarm Hosts by assaulting different areas.
Now of course it's more complicated then this. If the Terran is using splash to kill these Swarm Hosts, then the smaller engagements favor the Swarm Host'ing player, because they can spread their Swarm Hosts/Locusts out more easily, negating some splash. So where they may have had the DPS to take out 20 locusts before, now divided in two they might only be able to take out 8 per group. Of course the narrowness of chokes plays a role here in preventing the Zerg from spreading their Locusts out. This isn't even considering the fact that Terrans/Protoss depend on being more supply efficient with their splash units, so using Swarm Hosts to spread them out, while the rest of your army can attack in one location, generally favors the Zerg.
(I do understand it's even still more complicated - fighting for air control to spot Locust movements, using Scans potentially [though that is some damage itself], and Sensor Towers especially are useful for this if the terrain allows one to be forward enough - SHs do have amazing range).
@KT(Rolster)HaunteR
If you are in a situation where your opponent can move out on the map, tank your Locust shots, and run in and kill your Swarm Hosts just because you unburrowed them, then either 1) It wasn't a good situation to make Swarm Hosts in the first place or 2) You made too many of them and went for the "all in" Swarm Host strategy that OP is talking about in his OP, and he countered your 1 dimensional composition.
|
Well I was complaining about the swarm host in HotS for a while and saying pretty much the same thing in this thread and after thinking about it for a while I came to the conclusion that if swarm hosts had further range slightly less damage and locusts fly it would make the unit more interesting and work better. The problem at the moment is that it blocks pathing which makes the unit a lot harder to stop in ZvZ in particular, swarm hosts are pretty much the ultimate army composition until broodlords come out and most of the time you die if you don't mass swarm hosts yourself.
So what im suggesting is making the swarm host kinda like a lower damage carrier light harass from range, extra DPS without being out front with the rest of the units, no path blocking so you would have to protect them correctly and have the locusts very microable.
|
I agree, there is an inherent problem with units that spawn free units. It lends to massing these units up, because why only make 2 free units when you can make 40? I feel like the idea of free unit-spawners needs to be removed, or at least redesigned, because it creates situations exactly like those the OP is referencing. Scrap the SwarmHost, replace with a cost efficient zerg unit.
|
I think it's not possible to know how the unit will be used in even a year if it remains in its current state. To take an example, look at ZvZ. For at least a year after wol release, no one would make mutas in zvz except if he was going for some kind of cheese and so mutas were considered to be something like a 2 port banshee. You couldnt transition out of it. If you used them, you were all-in unless you did terrible damage. Etc, etc. Making only a handfull of mutas? Why would you do that? They wouldnt even kill a queen+ a spore, and the investment in spores is less than what you are investing in mutas!
Then, month after month, and with new maps (bigger maps), a new strategy emerged, mutas en masse followed by mass lings or even speedbanelings to killl the hydras that were considered the counter. It was before infestor buff happened. You would deny the third then threaten basetrade as soon as you opponent would leave his base.
I would say a few months to a year ago we saw the mutas becoming a standard of zvz in an entirely different fashion, as said above. You would rush lair, make 8 mutas and be done with it. You wouldnt hope to kill your opponent with them. It's simply that the almighty infestor rised in popularity at the cost of 2 base roach timings that were so common, with every Z rushing to infestor on 2 base. And guess what? A few infestors dont do very well against a little pack of mutas.
So we saw three entirely different ways to use the same unit without a single patch affecting the said unit. No one could imagine the evolution of the metagame before it happened before our eyes. I honestly think that if some of the posters of this thread read this thread in a year they will simply smile.
TM;DR There are a ton of creative ways and builds using SH that will make sure their obvious weakness do not matter anymore. It's simply too soon to know where is it heading.
Edit: to take an example, I'd love to see how SH would be doing against the immortal all-in (Parting style). It could be their role, be a niche unit you would use to defend at certain key moments in the game and if you scouted certains builds. I would be totally fine with it even if it was the only use they would ever have in zvp.
|
Natalya,
I absolutely understand where you are coming from when you say we really can't know right now what kind of impact that this unit will have on the game. Your Mutalisk example was a very good one, and there are many other units from Wings that would fit this description.
That being said, I believe that the unit has the potential to be much more interesting than it is right now, and promote better gameplay. One of the main points of the Beta is to provide Blizzard with as much feedback as possible so that they may make changes as necessary. In some cases, units were even removed (Warhound) because they didn't quite make the cut. I don't think it is necessary to remove the Swarm Host, and I think the unit has a lot of potential.
By posting this thread here, and more importantly over at Battle.net, I am hoping to start a discussion on the unit and how we can make it function better. I think many people would agree that the Warhound wasn't functioning the way it was supposed to, that the unit design was flawed, and that changes were necessary. Although it may not be as obvious of an example, I feel the Swarm Host meets this description as well, and that there is still time to have our voices heard by Blizzard and to have positive changes implemented on the unit before the HOTS release date.
If you prefer a wait-and-see approach, by all means feel free to do so
|
|
|
|