Tips for playing in new tower meta
AManApart123Kill the structure, not the other team. If your team focuses it down they can remove the threat quickly. If you don't have the sustained damage to do that, well, maybe you should have drafted some.
If you do get targeted, LEAVE! There's a delay before the tower actually shoots you. Once it resets, you can come back in.
If you do have structure shutdown, consider holding off using it until a teammate gets targeted, to force a reset. Use minions/mercs/objectives for the main push.
I can confirm that Gazlowe is incredible, but then, he always was.
FuenteThe biggest hurdles for me when playing with these changes internally were these:
- Learn and internalize which abilities your hero has that will hit enemy heroes and adjust your play to use them effectively while avoiding aggro.
- Disengage! If you get tower aggro, it is very easy to just leave and drop it now since the slow is gone and one or two tower shots is unlikely to kill you (but watch for enemy CC and Garrosh).
- If your friends have aggro'd the tower, you can safely attack enemy heroes, but do so carefully -- the tower could switch to you if its target goes out of range (or dies)
Definitely takes a bit of getting used to, but I had a lot of fun playing around and learning this anomaly as we were developing and testing it, so I hope you do as well!
Blizz_JeffB15-04-2020 23:11:21DelminI like the changes in theory, but I think it could use some tweaking.
Regarding point 1: I feel that incidental damage shouldn't aggro the towers. Targeted attacks and skillshots and stuff, sure; if you flame strike the tower and hit a hero and the tower starts hitting you that's your own fault. But things that you can't control like burning rage-type talents are an issue. Someone can just walk next to you to take the damage and then force the tower to focus you, which is unfortunate.
A compounding issue is that heroes with burning rage are also largely melee (except like Rexxar?) too, so it could be a bit harder to disengage from the tower than if you accidentally hit an enemy hero with Valla's multishot.
Another thing is summons I feel; I think if you're eg Anub'arak, and your beetles just happen to hit an enemy hero, the tower will automatically start firing at you. If it's a "directed" summon, like Banshee or Hydralisk that's fair game, but if it's one that has its own aggro table it shouldn't trigger tower aggro.
FuenteI think this is good feedback and there are definitely iterations we can pursue if some of these things become a problem.
For persistent AoEs, we have some options here like adding toggles to these abilities (I think Anub'arak can already do this with his beetles, for instance). Summons may end up being something we address on a per hero basis as well since (as you point out) the differing levels of autonomy that summoned units have may prevent this from being a clear, unilateral decision.
Thanks for taking the time to write up this constructive feedback!
Blizz_JeffB16-04-2020 02:35:49PscythicAnub'arak can only toggle the beetles created by [[Legion of Beetles]]; the rest spawn every time he casts an ability no matter what.
Instead of adjusting abilities on a per-hero basis and ending up with a bunch of spaghetti code, I suggest adding a damage threshold to the tower aggro: Track how much hero damage a player deals while under a tower, and if that damage exceeds a small amount (I'm thinking between 100 and 150, +4% per level) within 3-4 seconds, then the tower aggros onto them.
This would drastically reduce "accidental" tower aggro from small summons, burning rage effect, and similar cantrips without the need for ugly case checking and spaghetti code. Additionally, with such a small damage threshold of 100-150, almost any earnest attack on a hero will catch aggro just the same. Towers serve their purpose just as well, but certain abilities and summons are no longer liabilities to their caster.
Also you guys should nerf Sylvanas. She is the single best assassin in the game now. A simple damage nerf would suffice.
FuenteThis is an interesting suggestion that definitely could address some of the issues outlined above, if we think it is necessary to do so.
In my experience with this type of problem (a gameplay state change that has to be conveyed to the player, i.e., the tower wasn't attacking me and now it is), it is better to be consistent and binary, rather than analog, so the point at which the state changes is explicit and can be played around and accurately predicted.
You can imagine that as an attacker you would want to be able to know exactly when a tower will turn on you -- knowing if you can tower-dive for a kill and survive is a skill you can develop.
As a defender, you want to be able to know exactly what a tower will protect you from -- knowing how to bait a tower-dive for a kill is also a skill you can develop.
Thanks for the suggestions and feedback!
Blizz_JeffB16-04-2020 20:09:42alch334what's the testing process like? do you guys just scrimmage all day every day on test builds? How many anomalies get tested? Do you just figure one out on paper before trying to enact it?
FuenteWe have teamwide playtests fairly often, but we have too much to do to "just scrimmage all day every day." In addition, the designers play very often as they are iterating on all the changes we make to the game (including this anomaly).
Every anomaly was tested by the team for gameplay feedback and every anomaly was tested by QA for functional issues (and more gameplay feedback). This is true for every feature we implement!
The engine is pretty flexible, so it supports a designer just going in and
trying out a feature like this very quickly. How much initial paper design is required is really up to the individual designer and the complexity of the feature.