Life And Manasteal Rework

If you are using life- or manasteal in your build you are probably using super slow attackspeed weapon, or items that give you negative attack tiers. With super fast attackspeed the ls/ms triggering chance is 5.8% which makes it pretty much useless unless you have ALOT of it. With superfast(atleast for me) the manasteal triggers about every 8-10 seconds when using melee on enemies. The thing on ms and ls (/4s) means that it should trigger about every 4 seconds, but it works like that only on super slow and very slow weapons.
Also with superslow you have to hit mobs once every 2-4 seconds to get ls/ms, but with superfast you have to be hitting mobs 4 times a second and it still wont be as good. Thats why I think ls/ms need to be changed. Here are my suggestions on how to make superfast(almost) as good as superslow.

Suggestion 1

• Give lifesteal and manasteal 100% triggering chance on all attack speeds, BUT give it a 4 second cooldown. This wouldn't really change superslow at all, it would make superfast much better. You would no longer need to be constantly hitting mobs to get mana. You would only need to hit mobs every 4 seconds.

Suggestion 2

• Make ls trigger on every hit, but make it divided depending on what attack speed you are using. What I mean by this is, for example superslow gets 0.51 melee hits per second. Which means that it hits every ~2 seconds(2 hits every ~4 seconds). If you have 1000/4s lifesteal and 10/4s manasteal you get +500 hp and +5 mana every time you hit a mob. And if you have the same amount of ls and ms with slow speed(1.5 hits per second) (6 melee hits in 4 seconds) you will get ~166hp every time time you hit a mob and 1.6 mana(you will get 1 mana and 2 mana on every 2 hits) . With superfast(4,3 hits every second) you will get 53 hp when hitting a mob and 0.5mana(meaning you will get 1 mana every 2 hits). This wouldn't buff superfast as much, but I think it would be better than what we have now.

I think that suggestion 2 is cooler and more fun, but suggestion 1 would better in balancing the gap between superfast and superslow.

(also sorry if if my english is bad)

 

Atlas Inc. does not approve
(btw this does not meant I am part of Atlas Inc.! I am not mad enough)

________________________________

For me personally I don't really mind any of these, although 2nd option is a bit better

 

As someone who prefers glassy dps builds, life steal is kind of mandatory but I also hate slow weapons so it's honestly kind of frustrating to be forced to use them. I personally hope this is implemented, either one works for me - though #2 will be harder to code I imagine - so I don't really mind as long as it's added xD

 

Actually much easier. Just divide the amount by the atk. speed modifier

 

Huh. Well I have no knowledge about coding so I'll take your word for it. (I just assumed because the second one seemed more complicated xD) But as I said, I don't really care which one, just as long as one of them gets added :P

 

To clarify, here is some pseudocode:
1st:
IF no cooldown:
. proc ls/ms
. ls/ms = amount of ls/ms
. set cooldown to 4s (idk how hard is to make a timer)
ELSE
. skip


2nd:
proc ls/ms
ls/ms = amount of ls/ms divided by atk speed modifier (which already exist)

 

Option 1 is how its intended to work.
Salted changed it just before gavels lunch because it was cuasing lag issues with the way they were coding it.

I think the issue was mainly jumjum code, not the way it intended to work.

----

That being said, I'm not sure why you disaprove of it. (Or they) mind you, I don't think Atlas is as right as many people on the forums think they are, they are often, highly, biased.

 

First part would be you. Assuming you have perfect cps to be attacking as fast as the cooldown for attacks is with super fast, you would be getting ms/ls every 4 seconds or around there anyways, it is actually balanced like that but the thing is that requires a very very high cps and also is hard to keep up, plus if you're meleeing constantly you have no time for spellcasting.

You'd be surprised to hear then that most of us would actually prefer to have alternatives to atlas and/or a nerf to it, except guess what there isn't any alternatives so the use rate remains high.

 

I meant the group...

 

Suggestion 1 wouldn't work as that would mean super fast is just as good as super slow, giving no benefit to super slow attack speed. While this is probably how the gme was intended to work, it would probably tick off a lot of class builders atlas inc

 

Not true.
First off this would ignore the fact that it would equalize them roughly, not leave one behind in some way.
Second off, slower attack speeds still have the benefit of more easily getting through element defense.
That's how the element system was designed, faster attack speeds can take higher advantage of elemental weakness, were as slower attack speeds could take advantage of elemental strengths.

Even then I don't see the issue.
All this does is open up super fast attack speed to a different playstyle, while also better balancing the game, not sure why or who would complain about that.

 

yeah but ele defs aren't really taken into consideration when class building

 

Ok, so how is this a negative? How exactly would this upset class builders?
Again, this generally just opens up super fast to an id that, frankly, had relatively no use on it.
It's better balanced by almost all metrics.

 

people don't like change, lots of people think it's a good mechanic

 

Well, if it would be done Atlas would be useless, right? Yes, ls/ms good, but the attack speeds give no advantage. Amd since it is literally a cult of Atlas... you get the joke?

 

3ms bracelet with no reqs and extra 200ls is already really good. Also I dont think that the negative attack tiers were ever meant to give an advantage.

 

Get atk speed down -> better chance
(for super-slow builds it doesn't matter of course)

 

yeah just like how people think greed is a good mechanic /s

 

It is good enough

 

There's a better way.

Make a HashMap<UUID,Integer> lastProc that maps players to the last time they attacked
When they attack:
check if their last attack was 80 ticks ago
If that's true, proc mana steal and life steal and set the players current lastProc to the current time in ticks

You would also have to have a player join event that adds the new player to the hashmap and a leave event that removes them. I am specifying a hashmap because they are pretty efficient for this size.

On players, maybe not, but bosses can easily have hundreds to thousands of ele def. You can easily find yourself dealing no damage when you are supposed to deal thousands.