|
4 attacks will reduce the damage output against heavy-armored targets. That's one of the reasons why high-payload is a single shot. I also thought you don't like randomness?
Anyway, I removed homing, didn't change the splash range though. I was able to kill an unmicroed Thor with a single mutalisk. It was a lot of clicking though. I think the no-homing would work better if the Thor fried at a predicted location of the unit, taking its velocity (at missile launch) into an account. This way, if a fast moving target doesn't change its vector, it takes full damage.
|
On July 31 2015 03:32 BlackLilium wrote:4 attacks will reduce the damage output against heavy-armored targets. That's one of the reasons why high-payload is a single shot. I also thought you don't like randomness? In my mod the Viking wasnt supposed to be used against heavy armored targets. It was to be used against many lighter units. The banshee had gotten the ability to attack both ground and air and shot only 1 missle instead of 2.
There is also 2 kinds of randomness: The bad kind: There is a random chance of nothing happening. You wasted your time. But on the other hand something big happens. This makes interactions very binary and frustrating on either side. The good kind: Different things can happen but they are all equally good. Which event happens is random. This is not bad because each outcome will give you a gain and hurt your opponent but each one might do it differently. It can give a little bit more depth to a situation.
On July 31 2015 03:32 BlackLilium wrote:Anyway, I removed homing, didn't change the splash range though. I was able to kill an unmicroed Thor with a single mutalisk. It was a lot of clicking though. I think the no-homing would work better if the Thor fried at a predicted location of the unit, taking its velocity (at missile launch) into an account. This way, if a fast moving target doesn't change its vector, it takes full damage. If you manage to implement that then go for it. I would have done it this way too but I didnt see a way to do it. I didnt really try very hard though so there might very well be a way.
|
Experimental Branch changes
- Thor
- Removed high payload mode.
- Javelin Missiles
- No homing (RoomOfMush idea)
- Range: 10 -> 11
- Number of attacks: 4 -> 2
- Refire rate: 3s -> 2s
- Damage (and DPS): 14 + 4 vs Light
- 3 splash ranges: 100% 0.5, 50% 1.0, 25% 1.5
- Reintroduced 250mm Strike Cannons from WoL (will work on that ability later)
- Mutalisk
- Removed faster regeneration rate
|
No homing makes Thor super useless against fast targets!! What is the point of range increase, if it can deal damage only to targets which are close?
|
On August 01 2015 16:15 BrokenSegment wrote: No homing makes Thor super useless against fast targets!! What is the point of range increase, if it can deal damage only to targets which are close? I am not that happy with no-homing either I am currently trying to implement a predicted shot, so that if the target moves at a constant velocity the missile will connect perfectly. This should encourage more microing around the Thor to minimize the damage, while not making them completely useless. PurpleStreak is helping with it a lot, but he doesn't have a perfect solution just yet. However, if you are experienced map/mod maker and have an idea (or maybe already implemented it in the past?) - please let us know!
In the meantime:
Experimental Branch Changes
- Hydralisk: Supply cost: 4 -> 3 (equivalent to 1.5)
- Immortal attack: targets air
- Immortal attack: 20+30 vs Armored -> 25 + 15 vs Armored
- Viking range: 9 -> 8
The Immortal is now a bit less of a hard-counter to armor and a bit more as a beefy core unit. It is still slow expensive, and produced from Robotics, making it hard to mass and even harder to form an Immortal+Colossus ball (competing slot in production). I hope it is still different enough from a Stalker due to lack of mobility and lack of warp-in.
I hope the supply change to hydralisks will allow us to see a bit more of them used for defense purposes. Also getting a higher number of them should be a little bit less expensive. I don't think a squishy unit should take that much supply as hydralisk previously did.
Viking range reduction is coupled with a reduction of damage point, allowing it to be microed better. I hope it will lead to more interesting battles with this unit, plus it aligns with a general air-to-air nerf we want to make. We will have a look into the Colossus soon as well, hopefully making it a bit less of an issue for Terran - allowing it to be a bit less dependent on Vikings. Maybe, with the change to Immortal, Siege Tanks could be a decent response to Colossus as well?
I checked if Phoenix vs Light damage could be reduced, but it causes many negatives against lifted ground units: it takes more shots to kill workers, marines and hydralisks. I am not sure if such a nerf to a tactic which even now is not that popular (right?) is warranted.
|
I disagree with the removal of Mutalisk tissue regeneration. I think you should've given a chance to the idea of changing Mutalisk regeneration to be like Reaper regeneration + Zerg passive regeneration. I understand that their health is being used as a resource, but I think that the slight level of recklessness that tissue regeneration afforded Zerg players was actually a benefit in creating a slightly more unique style of play with the unit.
|
On August 03 2015 20:18 Pontius Pirate wrote: I disagree with the removal of Mutalisk tissue regeneration. I think you should've given a chance to the idea of changing Mutalisk regeneration to be like Reaper regeneration + Zerg passive regeneration. I understand that their health is being used as a resource, but I think that the slight level of recklessness that tissue regeneration afforded Zerg players was actually a benefit in creating a slightly more unique style of play with the unit. In what way does super speedy hp regeneration benefit the game? Its just a way to be sloppy without getting punished.
|
Frankly, I don't see much practical difference between high regeneration (Zerg) and high regeneration with delay (Reaper). Mutalisks are not annoying because they can overheal the damage they take. They are problematic because they can retreat, heal they woulds quickly and return to battle. That's why ground AA neither kills nor properly deters Mutalisk harassment. Currently, in order to get rid of Mutalisks you need to actually kill them. And to kill a Mutalisk, you need a persistent damage dealer - that is - air.
As a compensation, we could consider changing the Queen healing ability so that it would bounce onto next 2 or 3 units with some reduced healing effect. This way a Zerg would have a chance to heal a moderately damaged flock of Mutalisks. I am not convinced however that it is absolutely necessary....
|
Experimental Branch changes
- Thor
- Javelin Missiles now shoot at a predicted unit location based on its movement vector.
To clarify: when Thor is shooting, it takes 0.1s to measure the velocity of the targetted unit. It then fires his missile at a location where he thinks the unit and his missiles will connect. However, the missiles are not homing: if the unit alters its heading it may be able to dodge the missiles - at least partially. The higher the travel distance, the easier it is to dodge. Due to high speed of Javelin missiles, it is not an easy task, but we might see some interesting plays from top tier players.
On implementation side - I want to really thank PurpleStreak for spending his time trying to solve the prediction task in the data editor. Unfortunately, it required a lot of effects and a series of enumerated actions. I ended up writing the logic using triggers. However, if we find a better solution in the future, I may swap the implementation.
Any other comments, ideas, suggestions in the domain of Ground-to-Air and Air-to-Air? I was thinking about reducing the range of a Viking even further, but I don't want to fall into a trap of nerfing units into obscurity...
|
Good to hear, that you managed to do it. I personally think that this is a great mechanic that will greatly improve the possibilities for micro battles while not making it one-sided or a hard counter.
I would not try to touch the viking much more without proper testing. I think you have already made plenty of changes, few of which seem to me to be much more then what you originally wanted to do. Something that I would certainly recommend is looking at unit supply counts, for example the supply costs of ultralisks, thors, carriers, cattlebruisers and siege tanks. (perhaps others too) I was always baffled why SC2, the newer game, had LESS units on the field compared to SC:BW. We should definitely try to make 200/200 armies an exception instead of the norm. If 200/200 is too easy to hit there will never be a reason not to wait until 200/200 to move out. (I mean from a psychological point of view)
You should also look at the attack scan ranges, that is something that even blizzard has fixed in LotV because of community demand.
|
It will be much harder to attack Protoss from air with all those buffs. Are you sure you don't want to nerf Phoenix somehow?
|
Experimental Branch Balance Changes
- Viking range reduced 8 -> 7.5
- Phoenix fire cooldown 1.1s -> 1.6s
Phoenix damage cannot be reduced because crucial numbers would be altered: e.g. a lifted worker would not longer be 2-shot. However, changing fire cooldown allows to reduce persistent DPS (by around 31%) without affecting those crucial numbers above. I also ended up reducing Viking range further, but just by a small bit.
If there are no other comments on this subject, I will be closing this topic soon.
On August 05 2015 23:12 RoomOfMush wrote: Something that I would certainly recommend is looking at unit supply counts...
Adding it to the TODO list
On August 05 2015 23:12 RoomOfMush wrote: You should also look at the attack scan ranges, that is something that even blizzard has fixed in LotV because of community demand.
Experimental Branch Micro Changes Scan range set to be +1 compared to actual weapon range. The following units have been affected + Show Spoiler + Stalker Banshee Battlecruiser Broodlord Carrier Colossus Corruptor Sentry Ghost Hellion Hydralisk Immortal Infested Terran Tempest Marauder Marine Missile Turret Mothership Core (despite we remove it) Photon Cannon Point Defense Drone Oracle Mothership Queen Siege Tank Spine Crawler Lurker (the unit is actually there in HotS, just not enabled) Spore Crawler Thor Viking Void Ray
|
Thats quite some nerfs to Vikings and Phoenix. Especially with the Phoenix you should be careful not to nerf them to oblivion. They are already not used too much. Perhaps you can buff them in some other way to increase their usefulness, maybe make them beefier.
About the critical damage: Remember, there are upgrades in the game. If you reduce their damage by 1 and they suddenly need more attacks to kill basic targets, thats all the more reason to research +1 weapons when you want to use them.
|
On August 07 2015 09:09 RoomOfMush wrote: Thats quite some nerfs to Vikings and Phoenix. Especially with the Phoenix you should be careful not to nerf them to oblivion. They are already not used too much. Perhaps you can buff them in some other way to increase their usefulness, maybe make them beefier. These changes alone could be concerning. However, please remember that it is a part of a bigger plan of making GtA stronger and AtA weaker. Still, air-to-air has unique capability to chaise their targets everywhere on the map. I doubt they will disappear completely unless we really nerf them (which we don't plan to). For Vikings, the reduced range will make the game more micro intensive, but with the reduced damage point - high skill should be more rewarding. Some concern may be regarding Terran response to Collosi, but I hope to address that on the Collosi side - a topic which I would like to discuss next. In case of Phoenix, its lift-kill-and-run tactic should not be affected much, hopefully. However, a longer, more persistent interaction against air (e.g. against mutalisks) is weakened.
On August 07 2015 09:09 RoomOfMush wrote: About the critical damage: Remember, there are upgrades in the game. If you reduce their damage by 1 and they suddenly need more attacks to kill basic targets, thats all the more reason to research +1 weapons when you want to use them. You are right about that - I didn't pay much attention to it. I don't think however that +1 attack should be a prerequisite for efficient phoenix worker killing.
|
Protoss Deathball After dealing with the more general game design issues, we would like to now focus on particular units. The first to "attack" are the key Protoss units that form what is known as "Protoss Deathball": the Colossus and Void Ray. These untis have high DPS. Colosi have a nice splash, Void Rays scale up very well. Both can stack well over each other and other ground units.
Colossus This unit is complained about very often. Regarded as "boring", 1-A-victory with limited micro potential from either side (operator and receiver). We have seen many suggestions on how to change Colossus. Most revolve around adding new mechanic or activated abilities, e.g. the Hadronic Protection Field - activated 50% reduction shield over 5 seconds. However, we believe that the Colossus can be made more interesting without really adding any new behavior into the unit. Instead, let us try to change what the unit is already doing.
The most interesting domain of Colossus is how it is doing damage. Currently attack works as follows (with bit simplicifaction)
- Draw an invisible line L forward from the Colossus.
- Pick any point X on that line in the attack range.
- Draw a perpendicular line M at that point.
- Damage is done alone line M, within small distance from X.
This attack pattern has several properties:
- The actual attack is fast preventing the opponent reacting to it.
- Distance doesn't matter. Attack is as effective at range 3 as it is at range 9. Consequently, keeping the maximum distance against enemy is always the most optimal play.
- Getting a surround/concave around Colossi is not as effective as against other units. Some say that the attack pattern "nullifies" the concave.
Consequently Colossi are always held back in the army and the main ways to defeat them is to either simply overwhelm the ground defences and kill everying, or get a long-range anit-air and snipe the colossi.
We would like to propose a change: Collosi fires along the view direction line L, similarly to a Lurker. The attack always starts at a minimum distance (2 or 3) and moves at moderate speed till the maximum range. The idea is not new, similar concepts have been suggested before:
While similar, to the original Colossus, this actually changes the attack properties significantly:
- The attack is slow and the opponent has a change to react by moving units away from the beam path.
- The time for enemy raction scales with shot distance. An attack at range 3 cannot be avoided, but at range 9 shot can be avoided easily.
- Positioning Colossus close to enemy group gives the beam a chance to fry more targets in one shot. Sticking to maximum range renders Colossi quite ineffective.
- Consequently, long range attacks remain as safe as before, but their effectiveness is limited. A more aggressive Colossi placement can give higher rewards, but is also more risky - hopefully introducing an interesting interaction between both sides.
- Splitting an army and getting a surround/concave around Colossi, especially with a single row of units can diminish Colossi effectivenes. In contrary, when Colossi hits a ball of units, or better - flank into an already established battle, it can do a lot of damage.
If this works, we can then tweak the numbers for a better balance. For the moment, consider a case where effective DPS of a single beam is the same as before. However, only a unit in the center, between the beams, are hit by both of them.
And lastly: we believe it would look way cooler with the beams scorching earth parallel to each other, rather than crossing each other.
Void Ray Back in the early days of WoL Void Rays seemed to be a unit with an interesting charge-up mechanic designed for killing high HP targets. It soon ended up begin a gimmick: VRs would shoot at rocks or neutral/friendly units to get a charge, and afterwards unleash their full power at unsuspecting enemy - regardless if targetting heavy units or just a group of light units. In HotS they changed the unit, giving it a boring ability: "do more damage". No drawback, no situational behavior, just a straight stat increase.
Regardless of the above, VRs remain problematic due to their scalability. In few numbers they are rather weak and insignificant. But as their number grows, they become a killing machine that is hard to stop. One reasonable counter is a hard-counter in the form of anti-air AoE effects, but that is not always accessible. This problem has not been addressed at all, ever.
We would like to propose to step back to the interesting and unique WoL mechanic, with one important change: changing target resets the chargeup. This not only resolves the gimmicky use in WoL, but also addresses the scalability problem. Why? If you focus all VRs on a single target, it will die quickly resetting the charge. Targetting different enemies, however, is possible only when the VR count is low. Consequently, these units remain useful in low numbers when doing a mission or their own or with a small support. However they would not be so effective when massing up or when used in a deathball. Chargeup becomes an anti-massing mechanic, rather than only an incentive to focus on big targets (which didn't really work).
The basic VR shot would be generally weaker than current, while charged-up version noticeably stronger.
|
I did that Colossus change in my own mod. I personally think its a good change, perhaps not the best solution but definitely an improvement.
About the Void Ray: (I wrote that before in some thread) The Void Ray seems to be an air unit that is designed as a ground unit. In SC all air units fall into a certain category. They are either fast moving but fragile harassment (Wraith, Scout (upgraded), Mutalisk, Banshee, Phoenix, Oracle) or anti-air support (Valkyrie, Corsair, Scourge, Devourer, Viking, Corrupter), a capital ship (BC, Carrier, Tempest, BL) or a Spellcaster (Science Vessel, Arbiter, Queen, Raven, Oracle, Viper). The Void Ray however is slow, has the same single target attack against both ground and air and is cheap in terms of supply cost. It is basically a flying dragoon.
In my opinion the categories of air unis mentioned above make sense. Basic air units can be used for harassment, but since they die so quickly they are no good for actual army engagements. Capital ships can be used as a main army but they are very slow to build up and the enemy gets plenty of time to prepare (if we assume he scouts properly). This is good because air units are so much more powerful then ground units because they can stack and move freely.
The fact that Void Rays dont fit makes them either overpowered or nerfed to oblivion.
|
On August 09 2015 22:42 RoomOfMush wrote: About the Void Ray: (I wrote that before in some thread) The Void Ray seems to be an air unit that is designed as a ground unit. In SC all air units fall into a certain category. They are either fast moving but fragile harassment (Wraith, Scout (upgraded), Mutalisk, Banshee, Phoenix, Oracle) or anti-air support (Valkyrie, Corsair, Scourge, Devourer, Viking, Corrupter), a capital ship (BC, Carrier, Tempest, BL) or a Spellcaster (Science Vessel, Arbiter, Queen, Raven, Oracle, Viper). The Void Ray however is slow, has the same single target attack against both ground and air and is cheap in terms of supply cost. It is basically a flying dragoon.
In my opinion the categories of air unis mentioned above make sense. Basic air units can be used for harassment, but since they die so quickly they are no good for actual army engagements. Capital ships can be used as a main army but they are very slow to build up and the enemy gets plenty of time to prepare (if we assume he scouts properly). This is good because air units are so much more powerful then ground units because they can stack and move freely.
The fact that Void Rays dont fit makes them either overpowered or nerfed to oblivion.
I remember that. You are probably referring to this: http://www.teamliquid.net/forum/closed-threads/489939-a-protoss-redesign-proposal
Although the thread is now locked, there are useful points of yours.
However, I think that the core reason why Void Rays are so hard to balance is their scalability. Either they are too weak in low numbers or too strong in big numbers. That's why I am attacking this particular aspect and not redesigning the unit completely to fit a different role.
On August 09 2015 22:42 RoomOfMush wrote: I did that Colossus change in my own mod. I personally think its a good change, perhaps not the best solution but definitely an improvement. Yes, I remember that now. I should mention you too in the opening discussion post!
|
The scalability problem of Void Rays is a direct result of the problem I mentioned. They are designed (stat wise) like a core army unit. But because they are also flying they scale way better then any ground unit can. What is their role? What is their purpose? In what situation do you build Void Rays? Right now, it is for a core army unit. Powerful, beefy, scaling really well with numbers.
|
I agree with your reasoning that Void Ray is atypical among air. However, just because of that it is not necessarily a bad unit just yet. Instead, I am looking at particular problems and how it can be fixed. After identifying that scalability is problematic, I am looking a way to resolve just that, without changing the unit completely. If that is proven impossible, we can of course look at more drastic changes.
|
Void Ray suggestion:
I was thinking about the role of VR and it seems that it supposed to be good against big, armored targets - Broodlords, Ultralisks, Carriers, BCs, Thors etc also including buildings. In reality instead of being surgical units snipeing key targets VR in HotS are all around main fighters as You both mentioned above. We should rework VR to make really important what they target and force Protoss players to be more precise with them. Proposal:
- Make VR ability "Prismatic Alignment" be able to cast on unit instead of straight stim-like buff.
- PA increases VR damage against targeted unit with every second VR is attacking. In other words the longer VR is attacking the more dps it inflicts.
- PA is now channel-like ability. If protoss players will attack with VR with activated PA, other unit - it stops to work and cooldown starts. Also if targeted unit move out or range of VR - PA stops. If targeted unit dies it stops etc.
- VR has 6 range by default. To help to snipe key units more range might be needed. That's why I suggest for PA to increase also range of VR by 2-3 for duration of PA.
- Cast Range = VR default range + 2-3 as mentioned above.
- Also if needed, "leash" range (not sure about the name?) might be also increased.
- **We might try to allow PA to also target structures but it'd probably be hard to balance
This changes would promote selecting individual Void Rays to cast Prismatic Alignment on key big units. There would be more decisons: do I want to select all my VR to kill 1 key unit very fast OR I want to use whole potential and target enemys units with VR in 1:1 ratio? Additionally You might aslo use PA against smaller units like Roach/Marauder/Hydra and still benefit to small extent.
Void Ray might become more specialized unit and hopefully replace Tempest as a unit designed to deal with broodlords/colossus/other massive units.
|
|
|
|