Click here to go to the first staff post in this thread. Thread: v2.6.14 Soulstealer Ultimate Change

Page 1 of 4 1234 LastLast
Results 1 to 20 of 61
  1.   Click here to go to the next staff post in this thread.   #1
    Offline
    Account Icon
    Chat Symbol
    Join Date
    Jul 2009
    Location
    Southern California
    Posts
    11,095

    v2.6.14 Soulstealer Ultimate Change

    soulstealer/ability_04/ability.entity

    Code:
    <ability
    	name="Ability_Soulstealer4"
    
    	icon="icon.tga"
    	statuseffecttooltip="State_Soulstealer_Ability4"
    	
    	maxlevel="3"
    	requiredlevel="6,11,16"
    	
    	anim="ability_4"
    	casttime="1400"
    	castactiontime="1400"
    
    	actiontype="target_self"
    	targetradius="1250"
    	casteffecttype="Magic"
    	targetscheme="enemy_units"
    	
    	manacost="150,175,200"
    	cooldowntime="120000,110000,100000"
    >
    	<onimpact>
    		<spawnaffector name="Affector_SoulStealer_Ability4" count="0.5" countb="30" countop="mult" distribute="even" target="source_position" distance="75,100,125" />
    		<spawnaffector name="Affector_SoulStealer_Ability4_State" target="source_position" param="owner_counter"/>
    		<playeffect effect="effects/cast.effect" />
    	</onimpact>
    </ability>

    soulstealer/ability_04/affector_state.entity

    Code:
    <affector
    	name="Affector_SoulStealer_Ability4_State"
    
    	radiusstart="50"
    	radiusend="1250"
    	
    	
    	lifetime="1000"
    	maximpactspertarget="1"
    	targetselection="all"
    	targetscheme="enemy_units"
    	effecttype="Magic"
    	persist="true"
    >
    	<onimpact>
    		<applystate name="State_Soulstealer_Ability4" duration="5000" />
    		
    		<distance target="target_entity" source="source_entity" />
    		
    		<setvar0 a="result" b="483,483,483" op="add" />
    		<setvar0 a="577778,866667,1155560" b="var0" op="div" />
    		<setvar0 a="-333,-500,-667" b="var0" op="add" />
    		
    		<setvar1 a="20,30,40" b="param" op="mult" />
    		<setvar1 a="var1" b="100,200,300" op="add" />
    		
    		<setvar2 a="20,30,40" b="30" op="mult" />
    		<setvar2 a="100,200,300" b="var2" op="add" />
    		
    		<setvar0 a="var0" b="var1" op="mult" />
    		<setvar0 a="var0" b="var2" op="div" />
    		
    		<compare a="var0" b="var1" op="lt">
    			<damage effecttype="Magic" amount="1" b="var0" op="mult" />
    		</compare>
    					
    		<compare a="var0" b="var1" op="ge">
    			<damage effecttype="Magic" amount="1" b="var1" op="mult" />
    		</compare>
    	</onimpact>
    </affector>

    soulstealer/ability_04/state.entity

    Code:
    <state
    	name="State_Soulstealer_Ability4"
    
    	
    	
    	icon="icon.tga"
    	passiveeffect=""
    	
    	effecttype="StatusDebuff"
    	allowtransfer="true"
    
    	movespeedslow="0.15,0.2,0.25"
    	basedamagemultiplier="-0.15,-0.20,-0.25"	
    >
    </state>

    Damage Formula

    Damage = [C/(Distance+D)+E]*[(A+B*Souls)/(A+B*30)]

    Where
    A = 100, 200, 300
    B = 20, 30, 40
    C = 577778, 866667, 1155560
    D = 483, 483, 483
    E = -333, -500, -667
    Distance = distance between soulstealer and target, center to center
    Souls = number of souls

    If the formula results in damage greater than A+B*30 (occurs within 50 radius at max level), then damage is reduced to A+B*30.



    The existing Soulstealer ultimate mechanic had a few sub-optimal attributes that we wanted to improve upon. The focus on this change is two-fold: to increase reliability and to further reward incrementally better positioning.

    First, the damage distribution as a function of angle with respect to Soulstealer was highly unpredictable. Here is a chart showing how the torrents were distributed around SS with max souls:

    Due to the torrent-per-souls mechanic, this distribution gets very problematic with less than maximum souls. For example, if you had only 8 souls, you would shoot four torrents; one at 0 degrees, one at 90 degrees, one at 180 degrees, etc. This means that there were large gaps (the fewer the souls, the larger the gaps) where the enemy may not take any damage whatsoever even if relatively close to Soul Stealer. There were also premium locations where enemies would take double damage. Depending on the relative angle to Soulstealer, an enemy further away may very well take more damage than an enemy close up. In contrast, the new method is formulaic, meaning that damage is very predictable and only varies with distance and number of souls. The luck factor of the relative angle is removed completely.

    Second, the damage distribution as a function of distance was nearly binary. In other words, enemies standing on top of Soulstealer would be instantly killed within a very, very (read: very) small "critical damage zone." Enemies beyond this circle would then take approximately the same damage, on average, regardless of positioning. Here is a graph of weight averaged expected damage as a function of distance from Soulstealer:

    As you can see, an enemy within about 25 unit distance would take catastrophic damage (this is fine; rewarding positioning is the goal of this ultimate!). However, an enemy only 300 distance away (still VERY close by) would take only 500 damage; this substantial damage drop off makes Soulstealer's ultimate quite unreliable, which we feel is preventing him from truly shining at a competitive level. Furthermore, notice how the damage remains essentially flat for much of the distance. An enemy at a whopping 900 distance will still take about 350 damage, on average. This means that beyond the critical damage zone, incrementally better positioning was largely inconsequential.

    The new mechanic balances this contradiction. Soulstealer's ultimate should reward positioning at all distances; the closer you are to the target the more they should feel the pain. Here is a graph of the new mechanic's expected damage output with maximum souls. For your convenience the old graph is retained and shaded for direct comparison:

    Yes, we are aware the maximum damage potential for enemies closer than 25 distance is reduced. Instead of 2400 damage Soulstealer can now only do (a still huge) 1500 damage. However, notice the significant buff between the ranges of 50 and 600. Enemies a mere 300 distance away will now take 800 damage instead of just under 500, on average. The idea here is that positioning is not a hit-or-miss ideal; incrementally better positioning now yields incrementally better results.

    Finally, the previous Soulstealer mechanic relied 100% on souls in the bank. If Soulstealer was recently killed, his ultimate was nearly useless as his damage cap would not only be cut, but the distribution of the torrents would be spread out and unpredictable and unreliable (see first point). To ameliorate this, Soulstealer's ultimate potential is now a mixture of souls AND a fixed damage amount. At max level, 20% of the damage comes from the fixed amount while 80% of the damage comes from his soul count. Basically, even if you have zero souls your ultimate will still be 20% as effective.

    This means that this change is an even larger buff when Soulstealer is at less than maximum souls. Here is the expected damage output for a stash of 10 souls:

    While the previous mechanic decreased by 66%, the new mechanic only decreased by 53.6%.

    Summary:
    -The "critical damage zone" damage has been capped at 1500 (nerf)
    -The new mechanic is very reliable, removing uneven damage distribution as a function of angle (buff)
    -The new mechanic rewards positioning better than old mechanic for distances beyond the "critical damage zone" (buff)
    -The new mechanic deals more damage between 50 and 600 range (buff)
    -The new mechanic deals significantly more damage between 100 and 400 range (buff)
    -The new mechanic has a fixed damage amount, reducing the complete dependence on soul count (buff)
    -The new mechanic applies the slow/damage debuff at the full 1250 radius instead of 700 radius (buff)

    Note:
    The old critical damage zone had a size of 24 unit radius. For comparison, this is less one-sixth the attack range of melee heroes, center to center. This scenario required heroes to be literally overlapped, only possible with out-of-shroud activation, PK + ghost marchers, or with assistance of allied push/pull abilities. (If you can find a replay of a competitive match where the full 2400 damage was applied without the aid of a Tempest I'll be absolutely shocked).

    Beyond this phenomenally short radius this change is an effective buff.

    Last edited by MacroHard; 08-11-2012 at 03:25 PM.
    When I was younger I had an imaginary friend.
    Now, with internet gaming, I have hundreds of imaginary friends.

    -American proverb

  2. #2
    good post, good read, nice changes
    hated the fact that SS ult became useless if you couldn't port directly on someone's ass

  3. #3
    Offline
    Account Icon
    Chat Symbol
    Join Date
    Aug 2011
    Location
    Islamabad, PK
    Posts
    305
    I'm gonna miss one-hitting enemy heroes with my shroud.. Never made PK on SS.

    Oh well.

  4. #4
    Nice post.

  5. #5
    1. Nice casual buff IMO - it will be easier to deal good damage but impossible to deal insane damage. Less rewarding if you actually would stand on top of people. I'm gonna miss oneshotting still-standing farmers with shroud..

    2. I am, as often nowadays, amazed by the information the community gains from S2. Every change gets explained and discussed. None to do it better than our graphmaster. Salutations!
    /Thorns, tThorns, Twistling & TwistIt

    Follow my stream and watch my streamathons each week!

  6. #6
    Prefer the old one

  7. #7
    Think the strongest part of this change is the significant increase in usefulness below max souls. As someone who rarely uses shroud on Soulstealer, I guess I shouldn't be seeing a huge change in my damage output as most of my ults probably fall in contact range.

    Edit: oh also he one shots every support in a team fight.
    Last edited by `Haunter; 08-10-2012 at 11:21 AM.

  8. #8
    Quote Originally Posted by TheEnigma View Post
    I'm gonna miss one-hitting enemy heroes with my shroud.. Never made PK on SS.

    Oh well.
    Yeah, *shrug* at least he can be "effective" using his ult while invisible (FINALLY), it was ridiculous when one was trying to ult someone and well, just cancel cause they already ran away...

  9. #9
    if its so hard to do why not let the rape circle still do 2400 damage? now its less rewarding to have a tempest on your team along with ss, which was one of the ultimate dota combos... i like the changes dont get me wrong but i dont know man....

  10.   Click here to go to the next staff post in this thread.   #10
    Offline
    Account Icon
    Chat Symbol
    Join Date
    Jul 2009
    Location
    Southern California
    Posts
    11,095
    Quote Originally Posted by powerofbr View Post
    if its so hard to do why not let the rape circle still do 2400 damage? now its less rewarding to have a tempest on your team along with ss, which was one of the ultimate dota combos... i like the changes dont get me wrong but i dont know man....
    If you look at the last image spoiler, you'll see that the mythical 2400 damage rape circle was near impossible to achieve: less than one-sixth of melee attack range.

    Visualize melee attack range.
    Divide that by 6.

    Most of the times people felt like they insta-gibbed somebody they were actually within the ~1375 damage range (which is still insanely huge damage). After reviewing dozens of competitive matches with Soulstealer we found that not once he achieved the full 2400 damage potential of his ultimate. Not once.
    When I was younger I had an imaginary friend.
    Now, with internet gaming, I have hundreds of imaginary friends.

    -American proverb

  11. #11
    Quote Originally Posted by powerofbr View Post
    if its so hard to do why not let the rape circle still do 2400 damage? now its less rewarding to have a tempest on your team along with ss, which was one of the ultimate dota combos... i like the changes dont get me wrong but i dont know man....
    Most top tier players say tempest is pretty bad with Soulstealer mainly because if you want to run a soulstealer, you don't want a jungler. Obviously, you can run suicide tempest or something or have tempest as pseudotrilane while jungling part of their jungle but those line ups are situational based on what the enemy runs. Soulstealer just ramps up so much more easily with stacked jungle camps for him.

  12. #12
    Anyone have any ideas why the original Nevermore (Soulstealer) was coded in such an unintuitive fashion? When it comes to things such as this usually the reason is the limited W3 engine, but damage as a function of distance is a trivial thing to code so I don't think it's that.

    Reason I ask is maybe there was a balance rationale involved that is apparently not involved with this decision or discussion surrounding it.
    Last edited by AngryAtheist; 08-10-2012 at 12:19 PM.

  13. #13
    Offline
    Account Icon
    Chat Symbol
    Join Date
    Aug 2011
    Location
    Islamabad, PK
    Posts
    305
    Quote Originally Posted by MacroHard View Post
    If you look at the last image spoiler, you'll see that the mythical 2400 damage rape circle was near impossible to achieve: less than one-sixth of melee attack range.

    Visualize melee attack range.
    Divide that by 6.

    Most of the times people felt like they insta-gibbed somebody they were actually within the ~1375 damage range (which is still insanely huge damage). After reviewing dozens of competitive matches with Soulstealer we found that not once he achieved the full 2400 damage potential of his ultimate. Not once.
    Man... competitive games? :s

    Watch some public games. I've used that ulti perfectly so many times...

    Imagine using your judgement... Someone running in a direction... You wait and then press R as they're running towards you. BAM. Insta-kill.

  14. #14

  15. #15
    Could you please make a damage distribution graph for the new ultimate as you had for the first image in regards to the old ultimate? I'm aware it's linear (close to linear?) but it's a nice comparison and spoiler tagging the graph means it's a very small amount of space in the post.

    It's nice to compare the two and I don't have your graphing program whilst I would like it in the first post as well.
    #reinstateapostate

  16. #16
    I'm amused that this otherwise very professional presentation you've given us uses the term "rape circle" so many times, I hope nobody gets TOO offended :P

    I take it as a prime example that to gamers, rape doesn't mean rape

  17.   Click here to go to the next staff post in this thread.   #17
    Offline
    Account Icon
    Chat Symbol
    Join Date
    Jul 2009
    Location
    Southern California
    Posts
    11,095
    I added the scripts as well as a summarized damage formula to the first post.

    Quote Originally Posted by herro`kiti View Post
    I'm amused that this otherwise very professional presentation you've given us uses the term "rape circle" so many times, I hope nobody gets TOO offended :P

    I take it as a prime example that to gamers, rape doesn't mean rape
    This is, unfortunately, the established lingo for this phenomenon. =(
    When I was younger I had an imaginary friend.
    Now, with internet gaming, I have hundreds of imaginary friends.

    -American proverb

  18. #18
    Offline
    Account Icon
    Join Date
    Jul 2011
    Posts
    612
    This explains why when I ulted the other day two heroes who were within 2 hero body lengths of me in the trees received almost zero damage when I had full souls and was a couple levels above them.

    Without testing, I do feel like this is a slight nerf in one sense because while the 2400 damage may be near impossible to pull off, I feel like it is likely that many good team fight ults are in the 1800+ish range of damage when executed properly. Frankly, aside from one instance I have never had any problem with SS ult doing not enough damage when executed properly. When you land a properly positioned ult it always seems to destroy the enemy team even if they are not exactly on top of you from a Flux or Tempest ult. I am not sure why this change was really a necessary change when it seems like all it does is make the hero easier to play. Unfortunately, the reason the hero is considered so difficult to play in the first place has little to do with landing properly positioned ults.

  19. #19
    Offline
    Account Icon
    Join Date
    Jul 2011
    Posts
    612
    Quote Originally Posted by Gilly View Post
    Have you even read this thread?

    Macrohard already stated that getting the old "rape circle" damage was near impossible, most the time you would end up 1 shot people with the ~1375 damage ulti. This change makes him much more viable in team fights, and a stronger hero.

    An example of this would be a quick by back to get into a team fight, with the change to the souls you lose to dying and the fact your ult now does a decent amount of damage without souls, you could still get of a team fight winning ulti off. Where as it would be a completely different story pre buffs.

    I'm not saying the hero was useless prior to these recent buffs, but he is certainly alot better now.
    I disagree that he is a lot better now. It sounds like this change overall isn't going to do much to SS and is one of those changes designed to remove randomness from the game.

    The problem with the hero in general has more to do with newer heroes all having an escape mechanism/gap closer/stun/etc. while having strong burst damage. In other words, there are better heroes than SS that farm just as well or almost as well nowdays. SS is too vulnerable to ganks, has extremely low mobility, lacks any crowd control, and is very squishy. In addition, heroes like MK can flat out dominate him mid 1v1. This change sounds like it might be a good change for him, but I dont think it is going to change much overall in terms of how often he is played competitively or in public games.

    The real nail in the coffin for SS was the change that disallowed refilling the bottle with your courier. For a hero that relied so heavily on a filled bottle, this change essentially limited him significantly in public games. In competitive and very high level games you have supports and roamers securing the runes with you, but in random games it is simply too dangerous for a SS to grab a rune unless you are absolutely sure nobody else is going to be contesting it (yes, I realize in random low level games runes are not heavily contested). SS is one of the weakest rune control heroes in the game when it comes to actually contesting the runes. No escape, no crowd control, very low health, etc means if he runs into almost anyone in the river it is instant death. Especially if he runs into the opposing mid. Any decent mid hero run by a competent player can give up one creep wave to the tower at most when SS pushes the lane to secure the rune. After they win the first rune, SS loses the ability to push the lane at the xx:45 min mark and rune control is totally lost. Suddenly all of his strengths mid are gone. Ferrying bottle alleviated this.

    SS is one of my favorite heroes, but he is certainly underpowered at this stage of the game. I can see him being picked up as an ult bot in competitive games again, but I think what everyone will find out is that in order for a SS to be successful in competetive games he needs to be babysat by pretty much the entire team and it's just not worth the cost. To be successful SS needs the runes secured with his team, he needs perfect vision and wards or he can't leave his safe zone to even farm without extremely high risk, he needs the creep camps stacked early, and he needs one or more heroes to set his ult up properly. Why do all of that when there are other options that don't need to be babysat the entire game and actually are capable 1v1 heroes that can take some risks throughout the game? I realize that all of that is done anyways in competitive games, but SS absolutely REQUIRES it more so than any other hero.
    Last edited by greenman101; 08-10-2012 at 01:26 PM.

  20. #20
    Quote Originally Posted by greenman101 View Post
    I disagree that he is a lot better now. It sounds like this change overall isn't going to do much to SS and is one of those changes designed to remove randomness from the game.

    The problem with the hero in general has more to do with newer heroes all having an escape mechanism/gap closer/stun/etc. while having strong burst damage. In other words, there are better heroes than SS that farm just as well or almost as well nowdays. SS is too vulnerable to ganks, has extremely low mobility, lacks any crowd control, and is very squishy. In addition, heroes like MK can flat out dominate him mid 1v1. This change sounds like it might be a good change for him, but I dont think it is going to change much overall in terms of how often he is played competitively or in public games.

    The real nail in the coffin for SS was the change that disallowed refilling the bottle with your courier. For a hero that relied so heavily on a filled bottle, this change essentially limited him significantly in public games. In competitive and very high level games you have supports and roamers securing the runes with you, but in random games it is simply too dangerous for a SS to grab a rune unless you are absolutely sure nobody else is going to be contesting it (yes, I realize in random low level games runes are not heavily contested). SS is one of the weakest rune control heroes in the game when it comes to actually contesting the runes. No escape, no crowd control, very low health, etc means if he runs into almost anyone in the river it is instant death. Especially if he runs into the opposing mid. Any decent mid hero run by a competent player can give up one creep wave to the tower at most when SS pushes the lane to secure the rune. After they win the first rune, SS loses the ability to push the lane at the xx:45 min mark and rune control is totally lost. Suddenly all of his strengths mid are gone. Ferrying bottle alleviated this.

    SS is one of my favorite heroes, but he is certainly underpowered at this stage of the game. I can see him being picked up as an ult bot in competitive games again, but I think what everyone will find out is that in order for a SS to be successful in competetive games he needs to be babysat by pretty much the entire team and it's just not worth the cost. To be successful SS needs the runes secured with his team, he needs perfect vision and wards or he can't leave his safe zone to even farm without extremely high risk, he needs the creep camps stacked early, and he needs one or more heroes to set his ult up properly. Why do all of that when there are other options that don't need to be babysat the entire game and actually are capable 1v1 heroes that can take some risks throughout the game?

    Don't get me wrong, I agree with alot that your saying. I do miss the days of SS dominance. However this is definitely a step in the right direction. I believe to make him truly viable again, they should revert the change of his ultimate cast time, as ever since then he has dropped off slowly and steadily. But this is a good start, as well as the soul loss when dying buff.

    I do believe that he is underpowerd, and is beaten by a ton of heroes 1v1, and most definitely 2v1 mid. However getting a ring of the teacher and having him play the safe lane can also be ridiculously strong. The mana regen from the RoT can do wonders for SS where he can delve into the forest clear a stack when the lane is pushed and then get back into the lane. As long is the pull camp is stacked and pulled alot, it can really work well. Sure its nothing compared to ferrying bottles back and forward from mid, however it can be still effective. He obviously can only be built around certain line ups, and cant outcarry the likes of a sil, tdl, mb however it would be still a decent pick. As you say he needs camps stacked etc, I do believe that in top tier play that is no issue and the camps should mostly be stacked ever 53 anyway.

    But all in all +1 to your post

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •